Legal Aspects of Electronic Commerce - Rules of Evidence, Contract Formation and Online Performance

LEGAL ASPEctS OF ELEctRONIc COMMERcE: RULES OF EVIDENcE, CONtRAct FORMAtION AND ONLINE PERFORMANcE
by
JOSE ANGELO EStRELLA FARIA
CONtENtS
Biographical Note List of Principal Publications Introduction International Trade and Electronic Commerce Liability and Standards of Conduct for Information Service Providers Unfair Competition, Deceptive Trade Practices and Consumer Protection Privacy and Data Protection in Electronic Commerce Protection of Intellectual Property Rights Cybercrime Jurisdiction and Applicable Law Scope of this Study
135 136 139 142 145 146 147 148 149 150 155
Part One Legal Value of Electronic Communications

Chapter I. Rules of Evidence, Form Requirements and Electronic Communications A. Traditional Recording and Authentication Methods and Their Electronic Equivalents 1. Function and Nature of Rules of Evidence and Form Requirements (a) Authentication and Attribution (b) Practical Consequences of Form Requirements 2. Problems Posed by Electronic Communications (a) Intangibility and Alterability (b) Limitations in Retention and Retrievability B. Conditions for Legal Equivalence between Electronic Communications and Paper Records 1. Work of UNCITRAL in the Area of Electronic Commerce (a) Recommended Legislative Principles for Electronic Commerce (i) Functional Equivalence (ii) Media and Technology Neutrality
The Xiamen Academy of International Law, Collected Courses 2009, Volume 2, 131339 2009 Koninklijke Brill NV. ISBN 978 90 04 18093 2. Printed in the Netherlands.
157
157 157 157 158 163 166 166 166 169 170 172 173 173
Collected Courses 2009, Volume 2
(iii) Party Autonomy (b) Legal Value of Electronic Communications (i) Conditions for Meeting Form and Evidentiary Requirements (ii) Conditions for the Use of Electronic Communications 2. Implementation and Application of Electronic Commerce Laws (a) Authentication and Attribution of Electronic Records (b) Ability to Meet Legal Signature Requirements Chapter II. Electronic Signature and Authentication Methods A. Notion of Electronic Authentication and Signature 1. Functions of Electronic Authentication and Electronic Signature 2. Main Methods of Electronic Signature and Authentication (a) Digital Signatures Relying on Public Key Cryptography (i) Digital Signature Creation and Verification (ii) Public Key Infrastructure and Certification Services Providers (b) Practical Problems in Public Key Infrastructure Implementation 3. Biometrics 4. Passwords and Hybrid Methods 5. Scanned Signatures and Typed Names B. Legal Treatment of Electronic Authentication and Signatures 1. Minimalist Approach 2. Technology-Specific Approach 3. Two-Tiered or Two-Pronged Approach
174 175 175 177 177 179 184 188 189 190 193 193 194 195 197 198 200 200 201 203 205 206
Part Two Electronic Communications in Contract Formation and Performance 208

Chapter I. Contract Formation through Electronic Communications A. Applicable Law in International Electronic Contracting 1. Elements to Determine the Applicable Law (a) Location of Information Systems (b) Domain Names and Electronic Addresses (c) The Continued Usefulness of the Traditional Notion of Place of Business (d) A Duty to Disclose the Place of Business? 2. Substantive Applicable Law: Sales Law and Virtual Goods (a) Computer Software under the UN Sales Convention (b) A Special Legal Category for Virtual Goods? 208 211 212 213 217 218 219 220 220 225
132
Legal Aspects of Electronic Commerce: Rules of Evidence, Contract Formation and Online Performance
B. Contract Formation through Electronic Communications: Substantive Issues 228 1. Qualification of Parties Intent: Offers and Invitations to Make Offers 229 (a) Offers and Advertisements in Electronic Commerce 229 (b) Special Cases: Click-Wrap Agreements and Internet Auctions 232 2. Time of Receipt and Dispatch of Electronic Communications and Contract Formation 235 (a) Rules on Contract Formation 235 (b) Timing of Dispatch and Receipt of Electronic Communications 238 (i) The Rule in Article 15 of the MLEC 238 (ii) Electronic Communications in Domestic Enactments of the MLEC 240 (iii) Electronic Communications in Other Domestic Laws 244 (iv) The Debate during the Negotiation of the EEC 248 3. Automated Information Systems 255 (a) Responsibility for Automated Information Systems 256 (b) Errors in Messages and Communications 257 (i) Human Errors 257 (ii) Errors Generated by Information Systems 259 4. Incorporation and Availability of Contract Terms 261 (a) Incorporation of Terms and Conflicting Contract Terms 261 (b) Availability of Contract Terms 263 Chapter II. Particular Issues in Government Contracting 265 A. General Issues in Connection with the Use of Electronic Communications in the Procurement Process 269 1. Publication of Procurement-Related Information 269 2. Use of Electronic Communications in the Procurement Process 271 (a) Electronic Qualification and Suppliers Lists 272 (b) Electronic Supply of Solicitations and Other Documents 275 (c) Form of Other Communications during the Procurement Process 275 (d) Electronic Submission of Tenders, Proposals and Quotations 276 (e) Procurement Contracts and Electronic Signatures 280 B. Legal Issues arising out of Specific Electronic Procurement Techniques 281 1. Electronic Reverse Auctions 281 (a) The Extent of Use of ERAs 282 (b) Procedural Aspects of ERAs 285 (i) Auction Stage 285 (ii) Bidding Requirements 286 (iii) Disclosure of Information during the Auction 287 (iv) Contract Award 288 (c) Benefits and Concerns 290
133
2. Dynamic Purchasing Systems 292 (a) Scope and Duration of Dynamic Purchasing Systems 293 (b) The Operation of Dynamic Purchasing Systems 294 (c) Benefits and Concerns 295 Chapter III. Contract Performance through Electronic Means 298 A. Traditional Methods for Transfer and Perfection of Rights 298 1. Transfer of Rights in Tangible Goods and Other Rights 299 (a) Transfer by Consent 299 (b) Transfer by Registration 300 (c) Transfer by Delivery 301 (d) Transfer by Symbolic Delivery 301 2. Security Interests in Tangible Goods and in Intangible Property 302 (a) Perfection by Possession 303 (b) Perfection by Registration 303 (c) Other Methods 304 B. Transfer or Creation of Rights by Electronic Means of Communication 304 1. General Legal Obstacles 304 2. Electronic Equivalents of Negotiable Instruments and Documents of Title 305 (a) Basic Conditions for Electronic Negotiability 305 (b) Practical Experience: Electronic Registry Systems 310 (i) Dematerialized Securities 310 (ii) Electronic Registries as a Replacement of Paper Bills of Lading 313 (iii) The International Registry of Security Interests on Mobile Equipment 316 (c) Limitations and Problems of Registry Systems 317 (i) Liability for System Failure 318 (ii) Effectiveness of Electronic Communications for Transfer of Rights 320 Conclusion 327 Bibliography 328
134
BIOGRAPHIcAL NOtE
Jos Angelo Estrella Faria is the Secretary-General of the International Institute for the Unification of Private Law (Unidroit). He worked at the secretariat of the United Nations Commission on International Trade Law (UNCITRAL), in Vienna, between February 1996 and October 2008. He was the Secretary of UNCITRAL Working Group I (Privately Financed Infrastructure projects) during the preparation of the UNCITRAL Legislative Guide and the UNCITRAL Model Legislative Provisions on Privately Financed Infrastructure Projects, adopted in 2001 and 2003, respectively; he was also the Secretary of and UNCITRAL Working Group IV (Electronic Commerce) during the negotiation of the UN Convention on the Use of Electronic Communications in International Contracts, adopted in 2005. As a Senior Legal Officer, between October 2005 and October 2008, he supervised the secretariat support to UNCITRAL Working Group III (Transport Law) during the final years of negotiation of the UN Convention on Contracts for the International carriage of Goods Wholly or Partly by Sea (Rotterdam Rules). Prior to his work with UNCITRAL, Mr. Faria had served at the General Legal Division of the UN Office of Legal Affairs, in New York, from 1992 to 1996. Before joining the UN, Mr. Faria had worked as an attorney in Brazil, specialising in commercial and trade law matters. Mr. Faria graduated from the Federal University of Rio Grande do Sul (Porto Alegre, Brazil) and holds a Master on European Law from the Europa Institut der Universitt des Saarlandes (Saarbrcken, Germany). He has published various articles and books on legal harmonisation, commercial law and international law.
135
LISt OF PRINcIPAL PUBLIcAtIONS

Estrella Faria, Jos Angelo. Selbstbeschrnkungsabkommen im GATT und im EWG Wettbewerbsrecht, Universitt des Saarlandes, Saarbrcken, 1988 (206 pages); Estrella Faria, Jos Angelo. Integrao econmica na Amrica Latina: sairemos do discurso?, Revista de Direito Mercantil, ano XXIX (1990), No. 79 (July/ September) p. 64-83; Estrella Faria, Jos Angelo. Aplicao extraterritorial do direito da concorrrncia, Revista de Informao Legislativa, Vol. 27, No. 105 (January/March 1990), pp. 19-46; Estrella Faria, Jos Angelo. Princpios, Finalidade e Alcance do Tratado de Assuno, Ministrio das Relaes Exteriores, Braslia, 1993 (193 pages); Estrella Faria, Jos Angelo. The work of the United Nations Commission on International Trade Law in 1996, Uniform Law Review, vol. 1 (New Series), 1996-3, pp. 476-493; Estrella Faria, Jos Angelo. UNCITRAL Legislative Guide on Privately Financed Infrastructure Projects, EBRD, Law in Transition, Spring 2001, pp. 29-31; Estrella Faria, Jos Angelo. Establishing a Legal Framework for Electronic Commerce: The work of the United Nations Commission on International Trade Law (UNCITRAL) (jointly with Renaud Sorieul and Jennifer R. Clift), The International Lawyer, Vol. 35, No. 1 (Spring 2001), pp. 107-122; Estrella Faria, Jos Angelo. The UNCITRAL Legislative Guide on Privately Financed Infrastructure Projects, The Journal of World Investment, Vol. 3 No. 2 (April 2002), pp. 212-229; Estrella Faria, Jos Angelo. International Harmonization of e-Commerce Law: The Way Ahead, Treasury Management International, October 2002, p. 1; Estrella Faria, Jos Angelo. Perspectivas da Arbitragem Comercial no Mundo Contemporneo, I Seminrio Internacional sobre Direito Arbitral, 2002 [proceedings], Belo Horizonte, Cmara de Arbitragem de Minas Gerais, 2003, pp. 180-199; Estrella Faria, Jos Angelo. La CNUDCI et les autres organisations internationales, Petites affiches- Le quotidian juridique, No. 252, 18 December 2003, pp. 21-27; Estrella Faria, Jos Angelo. Electronic Commerce and International Legal Harmonization: Time to Go beyond Functional Equivalence?, South African Mercantile Law Journal, v. 16(4), 2004, pp. 529-555; Estrella Faria, Jos Angelo. O Controle de concentraes de empresas estrangeiras e a Lei no. 8.884: a extraterritorialidade revisitada, Revista de Direito da Concorrncia, Brasilia, 2004 No. 2, pp. 11-65;
136
Estrella Faria, Jos Angelo. Legal Certainty for Electronic Transactions: The Role of the Uncitral Model Law on Electronic Commerce (1996), in Andrea Schulz (ed.), Legal Aspects of an E-Commerce Transaction. International Conference in The Hague, 26 and 27 October 2004, Sellier European Law Publishers GmbH, 2006, pp. 159-169; Estrella Faria, Jos Angelo. Harmonizing the Law of International Electronic Contracting: Adjust the Rules but Dont Rewrite Them Andrea Schulz (ed.), Legal Aspects of an E-Commerce Transaction. International Conference in The Hague, 26 and 27 October 2004, Sellier European Law Publishers GmbH, 2006, pp. 74-98; Estrella Faria, Jos Angelo. Relationship between Formulating Agencies in International Legal Harmonization: Competition, Cooperation or Peaceful Coexistence? A Few Remarks on the Experience of UNCITRAL, Loyola Law Review, vol. 51 (Summer 2005), pp. 253-285; Estrella Faria, Jos Angelo. The United Nations Convention on the Use of Electronic Communications in International Contracts. An Introductory Note, International and Comparative Law Quarterly, vol. 55 (July 2006), Part 3, p. 689-694; Estrella Faria, Jos Angelo. Convention des Nations Unies sur lutilisation de communications lectroniques dans les contrats internationaux, Journal du Droit International, vol 133, issue 1 (2006), pp. 393 -403; Estrella Faria, Jos Angelo. Posibilidades y lmites de la armonizacin del derecho en materia de contratacin electrnica, DeCITA; Derecho del comercio internacional, temas y actualidades (Buenos Aires) 5-6 (2006), pp. 225-260; Estrella Faria, Jos Angelo. Online Contracting: Legal Certainty for Global Business - The New U.N. Convention on the Use of Electronic Communications in International Contracts, Uniform Commercial Code Law Journal, vol. 39 (2006), No. 1 pp. 25-73; Estrella Faria, Jos Angelo. Problemas Jurdicos dos Mtodos Eletrnicos de Criao e Transferncia de Direitos sobre Bens Corpreos e Imateriais, in Ricardo A. L. Camargo (ed.) Ttulos de Crdito Concorrncia e Mercosul Estudos em Memria do Professor Werter R. Faria, Porto Alegre, Srgio Antnio Fabris, 2008, pp. 23-48; Estrella Faria, Jos Angelo. O Contrato de Compra e Venda Internacional no Mercosul: Da Disparidade de Leis a um Regime Uniforme? in Elisabeth Accioly (org.), Panorama do Direito no Sculo XXI: Estudos em Homenagem a Werter R. Faria. Curitiba, Juru, 2007 (to be issued); Estrella Faria, Jos Angelo. Drafting and Negotiating History of the Electronic Communications, in Amelia Boss and Wolfgang Kilian (eds.), The United Nations Convention on the Use of Electronic Communications in International Contracts: An In-Depth Guide and Sourcebook, Austin Boston Chicago New York The Netherlands, Wolters Kluwer: 2008, pp. 17-42. Estrella Faria, Jos Angelo. La responsabilidad del porteador en el proyecto de convenio de la CNUDMI sobre el transporte de mercancas: breve comparacin
137
con el Convenio de Bruselas y las Reglas de Hamburgo, paper presented at 12th Congress and XX th Anniversary of the Iberoamerican Maritime Law Institute, Seville (Spain), 14 November 2007 (available as of 29 January 2008 at <http:// www.institutoiberoamericanodederechomaritimo.com/Conferencias/02.%20 JOSE%20ANGELO%20ESTRELLA%20FARIA.pdf,>); Estrella Faria, Jos Angelo. Uniform Law for International Transport at UNCITRAL: New Times, New Players, and New Rules, paper presented at the symposium Transport Law For The 21st Century: The New UNCITRAL Convention (Austin, 28 mars 2008) (to appear in Texas Journal of International Law; vol. 44 (2009); Estrella Faria, Jos Angelo. Legal Aspects of Electronic Commerce, Collected Courses of The Xiamen Academy of International Law, The Brill and Martinus Nijhoff, The Hague/London/New York, 2009 (to be issued).
138
LEGAL ASPEctS OF ELEctRONIc COMMERcE: RULES OF EVIDENcE, CONtRAct FORMAtION AND ONLINE PERFORMANcE*
Jose Angelo Estrella Faria**
Introduction
A few decades ago, it was said that the world had entered a so-called post-modern age. This was the dawn of a post-industrial era where intensive application and
* This article draws on years of research done at the secretariat of the United Nations Commission on International Trade Law (UNCITRAL) in the area of electronic commerce part of which has been separately published and reproduces materials from various notes by the U.N. Secretariat, in particular from the following: Possible future work on electronic commerce Comprehensive reference document on elements required to establish a favorable legal framework for electronic commerce: sample chapter on international use of electronic authentication and signature methods (U.N. document A/CN.9/630, of 25 April 2007 and addenda 1-5); Possible revisions to the UNCITRAL Model Law on Procurement of Goods, Construction and Services the use of dynamic purchasing systems in public procurement (U.N. document A/CN.9/WG.I/WP.44, of 21 February 2006, and addendum 1); Possible revisions to the UNCITRAL Model Law on Procurement of Goods, Construction and Services issues arising from the use of suppliers lists (U.N. document A/CN.9/WG.I/WP.45, of 21 February 2006 and addendum 1); Possible revisions to the UNCITRAL Model Law on Procurement of Goods, Construction and Services issues arising from the use of electronic communications in public procurement (U.N. document A/CN.9/WG.I/WP.34, of 13 December 2004, and addenda 1-2); Possible revisions to the UNCITRAL Model Law on Procurement of Goods, Construction and Services issues arising from the use of electronic communications in public procurement Comparative study of practical experience with the use of electronic (reverse) auctions in public procurement (U.N. document A/CN.9/WG.I/WP.35, of 16 February 2005, and addendum 1); Legal aspects of electronic commerce Electronic contracting: background information (U.N. document A/CN.9/WG.IV/WP.104, of 8 September 2003, and addenda 1-5); Possible future work on electronic commerce Transfer of rights in tangible goods and other rights (U.N. document A/CN.9/WG.IV/WP.90, of 20 December 2000). Nonetheless, the views expressed in this article are those of the author and do not necessarily reflect the views of the United Nations. The texts of domestic laws and regulations citied in this article and for which no source is specifically indicated are on file with the UNCITRAL secretariat. This article was submitted by the author on 13 December 2007.
139
wide use of information and communication technology (ICT) made of knowhow the principal force of production. In the information society1 in which we now living, the majority of employees work in information jobs, which means that they have to deal more with information, signals, symbols, and images than with energy and matter. Knowledge itself becomes a commodity. The dual function of information as means of production and commodity has created a special information economy, that is, a wide range of economic activities involving trade in ICT goods and services and ICT application for economic purposes, including the supporting infrastructure and hardware elements.2 Electronic commerce is both a narrower and broader concept, depending on how it is defined. As used in this article, electronic commerce means transactions for the sale of goods and provision of services where an order is placed by the buyer or price and terms of sale are negotiated over electronic means,3 whether or not pay is made online.4 Where their objects relate to ICT, electronic transactions are measurable as part of the information economy. In other cases, however, electronic commerce is nothing more than old economy using electronic means, as a manifestation of the broader information society. Measuring electronic commerce is not always easy, and different results may be arrived at depending on the way electronic is defined. As understood in this article, electronic commerce is estimated by the U.S. Department of Commerce to account for a total value of US$ 2.400 billion in shipments and sales in 2005 or 12.25 % of total shipments, sales, and revenues.5 On a global scale, electronic commerce has been estimated at some US$ 12.8 trillion of total
** LL.B. UFRGS (Brazil); LLM. Saarbrcken (Germany); Senior Legal Officer, International Trade Law Division, U.N. Office of Legal Affairs; Secretary, UNCITRAL Working Group IV (Electronic Commerce) and supervisor of the secretary, UNCITRAL Working Group III (Transport Law).
1 2
Jean-Franois Lyotard, La Condition Post-Moderne (Paris, ditions de Minuit, 1979). See United Nations Conference on Trade and Development (UNCTAD), Information Economy Report, 2006, New York Geneva, United Nations, 2006; OECD Working Party on Indicators for the Information Society, Sector Definitions Based on the International Standard Industry Classification (Isic 4) (OECD document DSTI/ICCP/IIS(2006)2/FINAL, 5 March 2007) <www.oecd.org> (29 November 2007). For example, the Internet, an extranet, Electronic Data Interchange (EDI) network, or other online system. Definition adopted by OECD Working Party on Indicators for the Information Society (Guide to Measuring the Information Society OECD document DSTI/ICCP/IIS(99)4/FINAL, 7 October1999) and the US Department of Commerce E-Stats <www.census.gov/eos/ www/2005/2005reportfinal.pdf> 19 November 2007)
US Department of Commerce, E-Stats, 25 May 2007 <www.census.gov/eos/ www/2005/2005reportfinal.pdf> (19 November 2007).
140
shipments in 2006.6 Admittedly, this is only a small part of the worlds output,7 but the bottom line is that electronic commerce keeps growing and is far from reaching its potential.8 The advantages of enabling the use of ICT applications for business transactions should be obvious. Besides the evident efficiency and productivity gains offered by the speed and worldwide reach of electronic communications, the potential savings of replacing paper-based trade documentation by electronic means can be illustrated by data provided by the International Air Transport Association (IATA), according to which the average cost of processing paper airway bills is US$ 30 per bill, which, by an average of 35million airway bills being issued every year, amounts to a total cost of some US$1billion every year in this industry alone.9 As would be expected, a closer look at existing statistics shows that most e-commerce happens in specific industries in a few economies of the world, mainly in North America, Europe, and Japan.10 The concentration of electronic commerce users and revenues is one of the aspects of the so-called digital divide between industrialized countries and less developed ones. This digital divide results from a combination of various factors that influence the development of information and communication technologies (ICT). They include infrastructure factors, such as connectivity, access to information technology, policy aspects, and telecommunications traffic levels.11 The digital divide also means that countries have different priorities and policy concerns,
These are the figures arrived at by the well-known global research firm Forrester Research <www.forrester.com> (12 November 2007). Estimated by the World Bank to have been some US$ 48, 244 billion in 2006 (see World Development Indicators Database, World Bank, 1 July 2007, <www.worldbank.org> (under Data and research, World development indicators) (20 November 2007)). There are, however, several conceptual and practical difficulties in measuring electronic commerce, which may lead to widely diverging estimates depending on the methodology used (see Measuring Electronic Commerce in UNCTAD, E-Commerce and Development Report 2001, United Nations Publication Sales No. E.01.II.D.30, (New York and Geneva, 2001) <www.unctad.org> (29 November 2007), p. 3). Cited by United Nations Economic Commission for Europe, Roadmap Towards Paperless Trade (U.N. document ECE/TRADE/371) <www.unece.org/cefact/publica/publi_index. htm> (29 November 2007). Although no reliable statistics exist for electronic commerce worldwide, studies done by UNCTAD suggest that the participation of developing countries in electronic commerce revenues is marginal (UNCTAD, E-Commerce and Development Report 2004 (United Nations, New York and Geneva, 2004), chapter I. <www.unctad.org> (29 November 2007). UNCTAD has developed a basket of indices to assess the level of each countrys development in the field of information and telecommunications technology (see UNCTAD, Information and Communications Technology Development Indices (2003) <www.unctad.org> (29 November 2007).
141
10
11
depending on their level of ICT development. The less developed countries are typically concerned with establishing the basic infrastructure for ICT use, whereas developed countries,12 such as Australia, Canada, most of the members of the European Union, Japan, New Zealand, and the United States, have long passed that level of concern and focus on issues related to more sophisticated business applications of ICT, such as electronic payments, privacy protection, intellectual property rights.13 Laws dealing with electronic transactions, electronic signatures, cybercrime and data protection are a constant component of most domestic strategies to develop electronic commerce. This means that most countries that have passed the initial stages of developing the basic infrastructure for electronic commerce will sooner or later pay some attention to the interplay between ICT use and the existing legal framework. The use of electronic communications for the negotia tion and performance of contracts has however highlighted the inadequacy of existing laws to accommodate new technology or to address some of the issues it raised. This applies both to public and private law, domestic and international. The legal implications of electronic commerce extend to various areas of law and may demand an extensive law reform effort.14 The main areas concerned are briefly summarized below. International Trade and Electronic Commerce The international trade liberalization regime that developed in the course of the 20th century, initially under the auspices of the General Agreement on Tariffs and Trade (GATT) and later under the umbrella of the World Trade Organization (WTO), for exempla, was based on a strict distinction between trade in goods and trade in services. Technological evolution made it possible, however, not only to reach out for external markets with nearly unlimited accessibility, but also to deliver some goods entirely online, without any supporting tangible medium, which had thus far constituted an essential element of the legal notion
Of course, the labels of developed and developing countries are an oversimplification of the worlds picture, as several developing countries rank higher than some developed countries in several indicators of ICT development, and the level of internet penetration and readiness for electronic commerce varies widely among developing countries (see UNCTAD, Information and Communications Technology Development Indices (2003)). Most of these topics have been the subject of extensive work at various international forums, including global and regional, intergovernmental, and non-governmental organizations. For a general overview, see Current work by other international organizations in the area of electronic commerce Note by the Secretariat (U.N. document A/CN.9/57 <www.uncitral. org/uncitral/en/commission/ sessions/38th.html>). For an overview of the wide range of issues affected by electronic commerce, see Ugo Draetta, Internet et commerce lectronique en droit international des affaires, 314 Recueil des Cours de lAcadmie de Droit International de La Haye (2005) 99.
12
13
14
142
of goods. Recognizing that global electronic commerce is growing and creating new opportunities for trade, trade ministers at the Second Ministerial Conference of World Trade Organization (WTO), which took place in Geneva from 17 to 20 May 1998, adopted a declaration,15 to commence a work programme on the subject in the General Council of the WTO, for making recommendations to the following Ministerial Conference. One of the first issues raised by WTO member States was the classification of virtual goods for the purposes of the existing multilateral trade regime, namely how electronic transmissions should be characterized, that is, as goods, services or something else. This is a question of significant practical and economic importance, as the types of commitments and extent of liberalization vary greatly between trade in goods under the General Agreement on Tariffs and Trade (GATT) and the regime established by General Agreement on Trade in Services (GATS). Although it appears that most countries within the WTO recognize that a large majority of e-commerce activities correspond to services (such as financial or professional services), there has been extensive argument about how to classify certain specific electronic transmissions or deliveries, in particular the electronic equivalents of goods such as CDs, photos, postcards, video tapes, software, video games, and books.16 Most developing countries seem to be concerned that ensuring the openness of electronic commerce will exacerbate the technological gap between them and developed countries. There are also fears that they might become mere consumers of e-commerce goods and services, rather than providers. Some countries also oppose the idea of open e-commerce in order to preserve the prerogative to regulate the flow of foreign goods, services, and culture within their boundaries. Thus, developing countries are generally inclined to regard transactions involving virtual goods as trade in services. Some of them would even prefer to see virtual goods treated as trade in intellectual property rights,17 a category that would offer the least in terms of market access and trade liberalization. While most developed countries want to ensure that there will be no future barriers to electronic commerce, because they are often the providers of services that are delivered electronically, their positions are not exactly uniform. The European Union maintains the position that the fact that information and instructions can
15
WTO Document WT/MIN(98)/DEC/02 <www.wto.org/english/tratop_e/ecom_e/ mindec1_e.htm> (29 November 2007). For an overview of the debate within the WTO, see Stewart A. Baker, Peter Lichtenbaum, Maury D. Shenk and Matthew S. Yeo, E-Products and the WTO, 35 Intl Law. (2001) 5. See Communication of Indonesia and Singapore (WTO document WT/GC/W/247, 9 July 1999) <www.wto.org/english/tratop_e/ecom_e/ecom_e.htm> (29 November 2007)
143
16
17
be put and transmitted on a physical support does not transform them into goods for the purpose of the WTO regimes. While the physical support itself (e.g. the disk) is seen as a good, and is, therefore, subject to GATT rules, the provision of the corresponding service should remain subject to the GATS.18 Other developed countries such as Japan19 and the United States,20 seem to focus on the result of maximum possible liberalization rather than on the classification of various types of transactions. But at least at earlier stages of the WTO discussions, the United States openly advocated the classification of e-commerce as trade in goods.21 The divergences among countries are significant, as are the economic interests involved in the WTO discussions,22 which is why the participating countries do not seem to be anywhere near reaching a workable consensus on the matter.
18
Thus, there is no reason to artificially turn the electronic delivery of a service into a good so as to bring it under the scope of the GATT 1994 (see Submission from the European Communities (WTO document WT/GC/W/497, 9 May 2003) <www.wto.org/english/ tratop_e/ecom_e/ecom_e.htm> (29 November 2007)). Japan agrees that the GATS disciplines should be applied to the acts of supplying digital contents by electronic means. However, as it is not entirely clear what disciplines should be applied to the digital contents themselves, for example, software, further consideration is needed. Such consideration should be directed so that the GATT principles of the most-favoured-nation treatment, national treatment and the general elimination of quantitative restrictions will apply to such digital contents (Communication from Japan, WTO document WT/GC/W/253, 14 July 1999 <www.wto.org/english/tratop_e/ecom_e/ ecom_e.htm> (29 November 2007)). the means of delivery of such products may change but the downloadable products functional characteristics do not change merely by a difference in delivery. Trade rules or commitments should not prejudice which business model is optimal for developing or delivering these products. Thus, the focus should not be on how to classify these products, but rather how to treat them for trade purposes with the goal being the most liberal treatment irrespective of how such products are classified. (Submission from the United States, WTO document WT/GC/W/493/Rev.1, 8 July 2003 <www.wto.org/english/tratop_e/ ecom_e/ecom_e.htm> (29 November 2007). See Contribution by the United Sates, WTO Document WT/GC/16, 12 February 1999 <www.wto.org/english/tratop_e/ecom_e/ecom_e.htm> (29 November 2007). However divergent the views of States may be, at least the position of the worlds largest software producer is clear: States should keep treating software embedded on a tangible medium as goods; they should neither create a category of virtual goods for on-line software transmissions, nor reclassify them as trade in services, but rather build upon the TRIPs regime (see WTO and Electronic Commerce: Issues for World Trade, A Microsoft White Paper, 8 September 1999 <www.microsoft.com/issues/essays/1999/11-15wto-b. mspx>(29 November 2007)). The United States Council for International Business (USCIB) is somewhat less emphatic, but points out that its members do not support, at least at this juncture, a determination that differentiates a physical delivery of a product as a good and an electronic delivery of a product as a service (USCIB Response to the European Commission Questionnaire on the Services Aspects of the WTO Work Programme on Electronic
19
20
21
22
144
Liability and Standards of Conduct for Information Service Providers Information service providers play an essential role in the functioning of the Internet. Typically, they act as intermediaries who transmit or host third party content but do not take part in the decision to disseminate particular material. Liability may arise from theories of direct and indirect or contributory infringement in national tort law, criminal law, and intellectual property law. Most cases arise from the fact that service providers take part in the technical process of transmitting or storing information for third party content of any kind. Responsibility for unlawful content or unlawful acts of their users is related to the opportunity and extent of control that information service providers are able to exert. The possibilities of storage and transmission of data files in data networks have multiplied the opportunities of unlawful behaviour and reduced chances of detection and control. Hence, imposing general liability for service providers would amount to establishing duties to monitor and filter all transmitted or stored content a burdensome task for information service providers for technical and economic reasons as well as unacceptable for other reasons. As a result, many countries have perceived a need for limiting liability of information services. However, the interest in limiting liability of service providers has to be weighed against the interests of injured parties in enforcing their rights and holding all contributing parties responsible. It does not seem to be necessary that the approaches be identical: they may differ depending on the particular circumstances and legal traditions in any given country. But they should be interoperable if global networks and electronic commerce are to develop smoothly. An additional set of legal issues relates to the possible liability of information service providers for failures that occur during transmission of messages (delivery delay or loss of information), or for malfunctioning of data storage systems (loss of stored data or unauthorized access by third parties). Typically, these matters would be dealt with at a contractual level, through general conditions of contract of information service providers. However, the extent to which information service providers may disclaim liability for loss or damage caused by service failure, or may limit their liability in those cases, is likely to vary from country to country. Lack of appropriate rules, guidelines or voluntary codes of conduct, or even the perception of insufficient legal protection, undermine confidence in electronic commerce and constitute an obstacle to its development. Conflicting standards across borders may also affect the offer of goods and services, as business entities operating under a less developed or excessively tolerant framework may enjoy
Commerce, 30 June 1999 <www.uscib.org/index.asp?documentID=1318> (29 November 2007).

145
an unfair competitive advantage, as compared to companies required to comply with more stringent requirements. In some cases, operations under a more lenient legal framework may be favoured by business entities interested in shielding themselves from liability that may arise under more stringent regimes. Unfair Competition, Deceptive Trade Practices and Consumer Protection Electronic communication permits new forms of advertising and marketing that may affect the interests of consumers as well as the functioning of the competition process. Unfair competition law will protect these interests but legal evaluation of practices in conventional commerce cannot always be transferred to the electronic environment. Online advertising includes forms of banner advertising with remuneration calculated on the basis of page impressions or ad clicks. Other forms of advertisements include information that loads between two content pages, either as small format pop-ups or full-page advertisements. Depending on the manner they are used, such techniques may raise issues of the separation of advertising from editorial parts of media, or may mislead customers and users to purchase services not originally intended. Unfair practices may also involve search engines, which have become the main service for users to cope with the enormous amount of information present on the Internet, or use of hyperlinks for misappropriation or deceptive comparative advertising. Again, lack of appropriate rules, guidelines or voluntary codes of conduct, or even the perception of insufficient legal protection, undermine confidence in electronic commerce and constitute an obstacle to its development. Conflicting standards may also distort international competition, as business entities operating under a less developed or excessively tolerant framework may enjoy an unfair competitive advantage, as compared to companies required to comply with more stringent requirements. Closely related to the issue of unfair competition and trade practices are rules intended for consumer protection. Domestic rules on consumer protection are typically based on concerns about information asymmetries as well as a lack of negotiating power on the side of the consumer. While media such as the Internet offer convenient alternatives to traditional purchase methods, one of the main barriers to electronic commerce taking off has been the lack of consumer confidence due to uncertainties in the use of electronic media for contracting. Information asymmetries are exacerbated in electronic commerce, as consumers lack vital information concerning the product, which the consumer cannot inspect physically. Consumers also have virtually no information concerning vendors and have little means to verify their identities and the standing of their business. Moreover, the features of the technical means used for the transaction may not be familiar to the consumer, resulting in unintended communications. Also, there are fears that a vendor providing the technical system may be able to
146
construe key features in its favour leaving the consumer in a weaker position in the transaction process. Legal uncertainties in cross-border transactions arise with respect to the applicable law and efficient ways to assert consumer claims. Various international and regional organizations have recognized the importance of improving the legal framework for consumer protection in electronic commerce, including OECD,23 the European Union24 and APEC.25 Privacy and Data Protection in Electronic Commerce Data protection and privacy are concepts that have been acknowledged in most parts of the world, sometimes even on a constitutional law level. With the advent of the computer there was a first wave of data protection initiatives in the seventies. With the spreading use of the Internet and the increased technical potential for collecting and transmitting data in electronic commerce, the protection of personal data has gained renewed attention. Practices like data mining or data warehousing as well as the placement of cookies are widely used in electronic commerce. Data protection and privacy rules may serve the interests of user as well as of business but also have to be weighed against conflicting interests. The lack of consumer trust and confidence in the privacy and security of online transactions and information networks is seen as an element possibly preventing economies from gaining all of the benefits of electronic commerce. On the other hand, regulatory systems restricting the flow of information can have adverse implications for global business and economies. The key elements in the international discussion on principles of data protection are concerned with consent to data collection, adequate relation to the purpose, time limitation of storage, adequate level of protection in third countries to which transmission takes place, information and correction claims for users, and enhanced protection for sensitive data. New issues and restrictions on data protection arise from international security concerns, which have led to legislative actions directed at data retention. With a growing stock of international rules these do not only become more heterogeneous but also make it more difficult for companies to comply. As these standards consider conflicting interests, the delineation of the field of application of these instruments as well
23
See the OECD Guidelines on Consumer Protection in the Context of Electronic Commerce of 9 December 1999 <www.oecd.org/document/51/0,3343,en_2649_37441_1824435 _1_1_1_37441,00.html> (27 November 2007). Directive 97/7/EC of the European Parliament and of the Council of 20 May 1997 on the protection of consumers in respect of distance contracts (Official Journal No. 144, of 4 June 1997). Voluntary Consumer Protection Guidelines for the On-line Environment (<www.apec.org>, under Documents and reports (27 November 2007)).
147
24
25
as which of the interests protected will prevail in a specific case are of gaining growing importance. Important work in setting standards for privacy protection has been done by international and regional organizations, including OECD,26 the European Union,27 the Council of Europe,28 APEC29 and the Commonwealth.30 Protection of Intellectual Property Rights Modern means of communication have had a significant impact in the way some intellectual property rights are defined and have challenged traditional enforcement mechanisms. Copyright has been closely intertwined with the features of the production, reproduction, and distribution of works from the outset. Hence, the advent of a uniform digital format as well as digital networks poses a challenge for the specific characteristics of copyright as to subject matter, scope of rights, and enforcement as new technological possibilities and related innovative business models develop. All kinds of protected materials are now distributed and traded over digital networks. The first challenge for the legal framework is to adapt to new technological and economic developments. This concerns the scope of
26
See OECD Guidelines on the Protection of Privacy and Transborder Flows of Personal Data, adopted on 23 September 1980 <www.oecd.org/document/18/0,3343,en_2649_37 441_1815186_1_1_1_37441,00.html> (26 November 2007). See further the OECD Privacy Policy Statement Generator <www.oecd.org/ document/39/0,3343,en_2649_3 7441_28863271_1_1_1_37441,00.html> (26 November 2007). Beginning with Directive 95/46/EC of 24 October 1995 on the protection of individuals with regard to the processing of personal data and on the free movement of such data (Official Journal 1995 L 281/31, 23 November 1995). This work was supplemented by Commission Decision 2001/497/EC of 15 June 2001 on standard contractual clauses for the transfer of personal data to third countries under Directive 95/46/EC (Official Journal J L 181/19 of 4 July 2001) and the Directive 2002/58/EC of the European Parliament and of the Council of 12 July 2002 concerning the processing of personal data and the protection of privacy in the electronic communications sector (Official Journal L 2001 201/37 of 31 July 2002). See Council of Europe, Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data, done at Strasbourg on 28 January 1981, E.T.S. 108, <conventions.coe.int/> (29 November 2007). The APEC Privacy Framework promotes a consistent approach to information privacy protection across APEC member economies and also avoids the creation of unnecessary barriers to information flows (see <www.apec.org>). The Commonwealth Model Privacy Law deals with the collection, use, disclosure and retention of personal information as well as establishing a Privacy Commissioner and a system of investigation of complaints of breaches of privacy <www.thecommonwealth.org/ shared_asp_files/uploadedfiles/ {82BDA409-2C88-4AB5-9E32-797FE623DFB8}_protection%20of%20privacy.pdf> (27 November 2007).
27
28
29
30
148
rights with respect to digital distribution as well as the extent of limitations to copyright. Also, certain kinds of information goods may obtain increased importance in an electronic environment calling for increased protection. The protection of databases can be seen as an example. Digital networks pose a threat to traditional distribution channels and economic models as well as to existing systems of collective management. Finally, moral rights that were not in the focus of the earlier phases of computerization, which emphasized software protection, are now increasingly gaining importance with respect to the creation and distribution of works over the Internet. Trademarks have an important function in commerce that is equally present in electronic commerce. While there is consensus that trademark law should apply to electronic commerce the same way as to traditional means of communication, problems arise from the fact that the provisions of trademark law and protection of related signs are not tailored to the features of the new medium. Further issues deriving more from conventional use of trademarks and related to the issue of cross-border communication as opposed to the territorial nature of trademark systems include the acquisition as well as infringement of trademark rights through use of signs on the Internet. Another illustration of the impact of electronic commerce on the traditional system for protecting intellectual property rights concerns domain names. Domain names are a necessity of todays user-friendly information retrieval in the Internet. The economic value of a concise and characteristic domain cannot be underestimated. Due to this, many conflicts over certain Internet-domains have arisen. Patent law is another area affected by modern means of communication, with software patents playing an increasing role in electronic commerce. Countries interested in developing an appropriate legal framework for electronic commerce would be well advised to consider carefully the intellectual property implications of the use of modern information and communication technologies. WIPO is the driving force in the international field for developing a framework for the protection of intellectual property. Due to technical developments much of the activity is now related to the electronic environment. WIPO has a comprehensive working agenda on all aspects of intellectual property in electronic commerce. The organizations expertise and universal membership ensures the broad acceptability of the international standards set by WIPO. Cybercrime Use of modern information and communication technologies has provided new means for criminal, fraudulent or indecent activities, such as embezzlement of funds, slander, industrial espionage, violation of trade secrets or dissemination of child pornography. At the same time, new types of criminal conduct have emerged, such as identity theft, dissemination of computer viruses, or intentional breakdown of computer and information services. Besides their criminal character,
149
all these activities may significantly affect international trade by causing physical loss or moral damage to individuals and business entities and by undermining business and consumer confidence in electronic commerce. The establishment of an effective legal framework for preventing and prosecuting computer crime and cybercrime, for example, as provided for in the Convention on Cybercrime31 adopted by the Council of Europe for the purpose of developing a common criminal policy aimed at the protection of society against cybercrime, inter alia, by adopting appropriate criminal legislation and fostering international cooperation is therefore an essential component of domestic and international strategies to promote electronic commerce. The Commonwealth has also formulated a model law on computer and computer-related crimes.32 Jurisdiction and Applicable Law Cyberspace it has been said, is not some mystical incantation capable of warding off the jurisdiction of courts built from bricks and mortar. Thus, just as traditional notions of jurisdiction have proven adaptable to other [economic changes], so too are they adaptable to the transformations wrought by the Internet.33 While this conclusion is certainly true, as a general proposition, the Internet has given rise to very intricate arguments over proper jurisdiction, which are far from being entirely solved. A few examples should suffice to make this point. In a famous case, two French associations fighting anti-Semitism and racism sued an American citizen who had created an Internet auction platform to offer Nazi memorabilia and Nazi products. The relevant web site could be accessed from anywhere in the world, including France, where, however, the sale of Nazi memorabilia is forbidden as war crimes apology. The French court affirmed its competence over the publishers of offensive material pursuant to French criminal
31
Council of Europe, Convention on Cybercrime (Budapest, 23 November 2001), E.T.S. 185, <conventions.coe.int/>(29 November 2007). The Commonwealth Model Law on Computer and Computer Related Crime establishes offences in relation to certain computer crimes including illegal access, interfering with data or with a computer system, the illegal interception of data, illegal devices and child pornography <www.thecommonwealth.org/shared_asp_files/uploadedfiles/ {DA109CD2-5204-4FAB-AA77-86970A639B05}_Computer%20Crime.pdf> (29 November 2007). Gorman v. Ameritrade Holding Corp., 293 F.3d 506, 510 (C.A.D.C. 2002). See also Metcalf v. Lawson, 148 N. H. 35, 38 (2002): It can be difficult to apply long-standing jurisdictional principles in cases involving Internet contacts. Nevertheless, while the Internet undoubtedly challenges the territorial-based concepts that courts have traditionally applied to problems of personal jurisdiction, it is equally true that traditional constitutional requirements of foreseeability, minimum contacts, purposeful availment, and fundamental fairness must continue to be satisfied before any activity including Internet activity can support an exercise of personal jurisdiction.
32
33
150
law, regardless of their nationality or domicile, to the extent that the offensive materials are accessible from French territory, and despite the defendants allegations that the publication or offer of such content was not illegal in the country of origin, the United States.34 The court ordered the internet services provider that hosted the web site in question (Yahoo!) to render impossible access by persons in France to certain content on servers based in the United States.35 Anticipating a request by the French plaintiffs to enforce the French decision in the United States, Yahoo! sought from a U.S. court a declaration that the order of the French court was unenforceable in the United States because it contravened the First Amendment to the constitution and other laws of the United States. The French defendants subsequently moved to dismiss for lack of personal jurisdiction. The district court denied the defendants motions to dismiss. The district court proceeded to hear the case and entered summary judgment for Yahoo!, declaring the order to be unenforceable in the United States as contrary to the guarantee of free speech under the First Amendment to the U.S. constitution. The district court noted that the French orders content and viewpoint-based regulation of the web pages and auction site on Yahoo.com was entitled to great deference as an articulation of French law, but held that it would be inconsistent with the First Amendment if mandated by a court in the United States.36 The court added that, [a]lthough France has the sovereign right to regulate what speech is permissible in France, this Court may not enforce a foreign order that violates
34
Tribunal de Grande Instance de Paris, Ordonnence de rfer, 22 May 2000, Union des Etudiants Juifs de France (UEJF) et La Ligue Contre le Racisme et lAntisemitisme (LICRA) c/. Yahoo! Inc. et Yahoo France <www.juriscom.net/txt/jurisfr/cti/tgiparis20000522.htm> (3 December 2007). The website was published at Yahoo.com, an American corporation, who argued that the French tribunal had no jurisdiction to hear the case. The tribunal decided however, that the crime was committed both in France and abroad because the website could be accessed in any jurisdiction. French court issued an order requiring Yahoo! to (1) disallow French citizens access to the offending offerings on Yahoo! Auctions; (2) eliminate French citizens access to web pages on yahoo.com that contain offending material; (3) post a warning on yahoo.fr stating that searches through yahoo.com may lead to sites and material prohibited by R645-1 of the French Criminal Code and that viewing such material may cause prosecution under the statute; and (4) eliminate from search directories accessible in the French Republic links to negationist materials which violate the criminal statute Believing that the order imposed technologically impossible requirements, Yahoo! requested, without success, the French court to modify the order. The French court gave Yahoo! three months to come into compliance as originally set forth. Yahoo! took steps to prevent Nazi propaganda from being accessible on www.yahoo.fr, but made no such effort on the U.S. site. Yahoo! Inc. v. La Ligue Contre le Racisme et lAntisemitisme, 169 F. Supp. 2d 1181, 1192 (N.D. Calif. 2001).
151
35
36
the protections of the United States Constitution by chilling protected speech that occurs simultaneouslywithin [the United States].37 The two French associations appealed to the U.S. Court of Appeals for the Ninth Circuit, reasserting their claim that the U.S. court lacked personal jurisdiction over them. The initial decision was eventually reversed, main on procedural grounds. The Court of Appeals conceded that, had the French associations attempted to enforce the order of the French court in the United States, jurisdiction would be proper. As no such enforcement had yet been sought, and the French associations contacts with California were not continuous and systematic, U.S. courts lacked personal jurisdiction over them.38 This decision was later confirmed, although not entirely on the same grounds, by a judgment en banc of the Court of Appeals.39 This case generated much controversy and extensive debate both in favour and against the decisions taken by the various courts involved from the point of view of ensuring international harmonization and legal predictability for what is essentially a borderless activity.40 Perhaps the most significant lesson to be learned from it may sound like a truism: not all that is illegal at home is forbidden abroad; neither is everything allowed abroad not forbidden at home. In the absence of universal standards, the Internet emphasizes more than ever then the need for mutual understanding of legal traditions and values. The interest to protect nationals against criminal activity conducted by electronic means seems to lead courts in various countries to assert jurisdiction on the basis of the accessibility of the offensive content. Australian courts have held that a defamation action could be brought wherever the information is
37
Id. Yahoo! Inc. v. La Ligue Contre le Racisme et lAntisemitisme, 379 F.3d 1120 (9th Cir. 2004). Yahoo! Inc. v. La Ligue Contre le Racisme et lAntisemitisme, 399 F.3d 1010 (9th Cir. 2005). See Marc H. Greenberg, A Return to Lilliput: The LICRA v. Yahoo! Case and the Regulation of Online Content in the World Market, 18 Berkeley Tech. L.J. (2003) 1191. For a view of the later developments, see Robert T. Razzano, Error 404 Jurisdiction Not Found: The Ninth Circuit Frustrates The Efforts Of Yahoo! Inc. To Declare A Speech-Restrictive Foreign Judgment Unenforceable, 73 Cincinatti L. Rev. (2005) 1743. See also the views of Michael Geist (We are beginning to see courts moving toward an effects based analysis for Internet jurisdiction); Joel R. Reidenberg (Companies will have to comply with the laws where they target business); Pierre Trudel (Les mesures de censure sont une mauvaise faon de rguler les conflits engendrs par la circulation de linformation) and Yves Poulet (Cest laune de ses propres principes constitutionnels et du respect d la libert dexpression dans un pays dmocratique que ltat ou le juge interviendra de manire exceptionnelle) at < www. juriscom.net/uni/doc/20010131.htm> (3 December 2007)
38
39
40
152
viewed.41 A similar conclusion was reached by the Italian Court of Cassation in a slender case, which the court saw as falling within the jurisdiction of the plaintiff s place of residence.42 Courts that opt for this broad understanding of domestic jurisdiction seem to take the view that the person who sets up the website can be regarded as potentially committing a delict in any country where the website can be seen, in other words in any country in the world.43 The opposite view, however, advocates restraint the exercise of jurisdiction. Thus, for example, noting that internet publications are accessible from virtually anywhere, a French court expressed the view that it would be unreasonable to expose the author to all laws of the world, which would lead to total legal uncertainty in the exercise of the freedom of thought, which is also protected by law. The court therefore advocated the use of objective criteria to exercise jurisdiction, in particular to confer jurisdiction on the place where the web site is located.44 Similarly, in the United States, courts have distinguished between various types of Internet sites, according to their level of interactivity in order to assess the reasonableness of exercising personal jurisdiction.45 When a defendant
41
Gutnick v. Dow Jones & Co Inc. [2001], Supreme Court of Victoria, VSC 305, 28 August 2001 <www.austlii.edu.au/au/cases/vic/VSC/2001/305.html> (30 November 2007). Corte di Cassazione, Civil Third Section, Ordinance No. 659, 18 May 2002, V. Giustiniani, President, A. Segreto, and Relator <www.infoius.it/sentenze/cass_2002/cassazione_ o6591_02.asp> (30 November 2007). This case involved allegations of copyright infringement commited through a web site maintained outside Scotland. Speaking for the Scottish Court of Session, Lord Drummond Young took the view that the critical question for present purposes is the location of a wrong that is said to have been committed by way of the Internet. I do not think that it is helpful to adopt the defenders analysis that a website is nothing more than material stored on a particular computer, with the result that any delict could only be committed where the computer containing the material was located. [The Internet] is, in essence, a system of communication that allows a verbal or graphic message to be set up on a computer, which may be situated anywhere in the world, and conveyed to another computer, which may likewise be situated anywhere in the world. [W]hen a website is set up, it is self-evident that the intention is that other persons should be able to obtain access to it via the Internet. The person who creates the website is linking himself into a process of communication, with the obvious intention that communication should take place by way of that process (Bonnier Media Ltd v. Greg Lloyd Smith and Kestrel Trading Corporation, Scotland, Court of Session, Outer House, 2002 SCLR 977, 1 July 2002). Presdok et Siline Gmbh D.J. v. F.C.O. fiduciaire SA, Court dappel de Paris, 11th Chamber, Section A, 10 November 1999, <www.juriscom.net/documents/caparis19991110.pdf> (3 December 2007). In the particular case, however, the court asserted jurisdiction on the basis of a provision in the French Penal Code (article 113-6) that grants French court universal jurisdictions for offences committed by French citizens. Machulsky v. Hall, 210 F. Supp. 2d 531(D.N.J. 2002).
153
42
43
44
45
merely posts information or advertisements on a website, personal jurisdiction over such defendant is not found to be proper. Jurisdiction could only be exercised if the acts were aimed at producing effects in the forum.46 Even regarding purely commercial transactions, which do not normally rise to the same level of sensitivity as criminal cases, there is argument as to what conditions justify the exercise of jurisdiction in Internet-related cases. Courts in the United States have transposed the analysis of the level of interactivity of web sites to the commercial environment. However, the dividing line between one and the other category seems at times rather fluid. Courts that have considered the issue of whether web presence creates personal jurisdiction in a particular forum have categorized Internet use into three areas for the purpose of determining whether the exercise of personal jurisdiction is permitted. At one end of the spectrum are cases where individuals can directly interact with a company over their Internet site, download, transmit or exchange information, and enter into contracts with the company via computer. In such cases, the exercise of jurisdiction is appropriate, particularly when combined with evidence of sales from the forum state.47 Exercise of personal jurisdiction is justified, for example, where the Internet presence of a company is clearly more than passive, extending, for example to maintaining a website through which its customers can engage in electronic transactions, open accounts online, transmit funds to their accounts electronically and use those accounts to buy and sell products.48 At the other end of the spectrum are cases where the defendant has only advertised on the Internet, and where another medium such as the telephone or mail is necessary to contact the seller; in the case of such passive sites, personal jurisdiction usually does not lie.49 However, if a passive web site by itself is not sufficient to confer personal jurisdiction, a passive website in conjunction with
46
Mere allegations that a plaintiff feels the effects of a defendants tortious conduct in the forum, simply because plaintiff resides there, do not satisfy the effects test for purposes of determining when a nonresident defendants publication of allegedly libelous material can subject such defendant to personal jurisdiction (Machulsky v. Hall, supra note 45, 540-541). See also Shamsuddin v. Vitamin Research Prods., 346 F. Supp. 2d 804 (D. Md. 2004). Compuserve, Inc. v. Patterson, 89 F.3d 1257, 1261 (6th Cir. 1996). Gorman v. Ameritrade Holding Corp., 293 F.3d 506, 5112-513,. (C.A.D.C. 2002). Bensusan Restaurant Corp. v. King, 937 F. Supp. 295, (S.D.N.Y 1996) (holding that mere uploading a web site onto a server located outside the forum that was accessible to forum residents was insufficient, without more, to subject web masters to personal jurisdiction where site not intended to sell product to forum residents.). See also Cybersell, Inc. v. Cybersell Inc., 130 F. 3d 414, (9th Cir. 1997) (declining jurisdiction where a web site is passive and the non-resident does not do something more to target the residents of the forum).
47 48 49
154
something more, such as conduct directly targeting the forum, may well do so.50 It is not clear what additional off-line activity would render the operation of a passive web site capable of justifying the exercise of personal jurisdiction. For instance, some courts have held that a companys maintenance of a passive web site accessible to forum residents and a toll-free telephone number justified jurisdiction,51 while other countries have expressly denied it.52 The middle ground between the two extremes involves sites where parties can interact with the defendant company, but may not be able to contract with the company or make purchases over the Internet site; in such situations, most courts determine whether jurisdiction is proper by examining the level of interactivity and commercial nature of the exchange of information that occurs on the Web site.53 In these cases, it is normally held that there must be more than a fortuitous contact with the forum to justify personal jurisdiction. For example, the mere offer of goods through a platform such as e-bay, is not sufficient to subject a non-resident to the forums jurisdiction.54 Even for such middle ground web sites, exercise of jurisdiction may be based on findings that the contacts with the forum are continuous and intense.55 Interactivity alone is not sufficient.56 Scope of this Study Information technologies have introduced profound and lasting changes in the way people, enterprises and governments communicate with each other and conduct their business. The task of legislators and policy makers around the world is therefore to ensure that existing laws, conceived against the background of other means of communication and premised upon strictly territorial application, can properly operate in the new environment created by the information society. In doing so, legislators must ensure that the legal framework for domestic and
50 51 52
Rio Properties, Inc. v. Rio International Interlink, 284 F.3d 1007 (9th Cir. 2002). Inset Systems Inc. v. Instruction Set, Inc., 937 F. Supp. 161 (D. Conn. 1996). See Fix My PC, LLC d/b/a Fixx My PC v. N.F.N. Associates, Inc., (48 F. Supp. 2d 640 (N.D. Tex. 1999); American Homecare Federation, Inc. v. Paragon Scientific Corp. et al., 27 F. Supp. 2d 109, (D. Conn. 1998); Edberg v. Neogen Corporation, 17 F. Supp. 2d 104 (D. Conn. 1998). Zippo Manufacturing Company v. Zippo Dot Com, Inc., 952 F. Supp. 1119, (W.D. Pa. 1997). See also Search Force v. Dataforce International, 112 F. Supp. 2d 771, (S.D. Ind. 2000). Winfield Collection Ltd. v. Mccauley, 105 F. Supp. 2d 746, 751(E.D. Mich. 2000); see also Metcalf v. Lawson, 148 N. H. 35, 39-40 (2002). Mieczkowski v. Masco Corp., 997 F. Supp. 782, (E.D. Tex. 1998). Millennium Enterprises, Inc. v. Millennium Music, LP, 33 F. Supp. 2d 907, (D. Ore. 1999). See also On-Line Techs. v. Perkin Elmer Corp., 141 F. Supp. 2d 246, (D. Conn. 2001).
155
53
54
55 56
international trade does not hinder or diminish the gains promised by the use of new technologies. This brief introduction has shown the variety of legal issues raised by electronic commerce. Attempting to deal with all of them would by far exceed the limits of this paper, which will, therefore, be limited to issues affecting electronic contracting. For that purpose, Part One discusses the interplay between form requirements and electronic communications. Part Two considers matters related to contract formation and performance in an electronic environment.
156
PARt ONE LEGAL VALUE OF ELEctRONIc COMMUNIcAtIONS
Chapter I. Rules of Evidence, Form Requirements and Electronic Communications

A. Traditional Recording and Authentication Methods and Their Electronic Equivalents
The Internet presents many of the same issues as other transnational technologies, but by facilitating the ability to communicate anywhere, any time, and reducing the importance of geographical and economic boundaries and locations, it facilitates changes to the way companies do business. One of the initial difficulties, however, results from the fact that paper documents have been the basis for rules on form and evidence of legal acts in most countries. Indeed, most legal systems use concepts that to a lesser or greater extent presuppose or even require that information of legal significance be recorded in a tangible medium capable of being displayed and stored for subsequent use. Notions such as writing, document, instrument, and other related concepts such as original or signature have been traditionally associated with hand-written or typed signs impressed by ink on a piece of paper. The medium of electronic communications is by nature intangible, and this characteristic gave rise to concerns that existing law may not support electronic communications or may even be an obstacle to its use. 1. Function and Nature of Rules of Evidence and Form Requirements At least in theory, the law is generally medium neutral. Contracts being the result of human agreement, the law in principle does not require any formalities for their validity. This is true both in civil and in common law countries. In practice, however, there are numerous exceptions to the general principle of freedom of form. The first type of exception is made of requirements of a particular form (usually the written form) as a condition for the validity of a contract. The second type of exception comprises rules of the admissibility and hierarchy of evidence in judicial proceedings. Even if this rule may not per
157
se deprive a contract of its validity, it may seriously restrict its enforceability whenever the required type of evidence (typically documentary evidence), for whatever reason, is not available. (a) Authentication and Attribution Legal writing and signature requirements as a condition for the validity of certain acts in common law jurisdictions are typically found in the British Statute of Frauds57 and its versions in other countries.58 In its origins, the Statute of Frauds had an essentially evidentiary function,59 and it is generally understood as merely rendering a contract not complying therewith unenforceable, but not invalid. The notions of writing or document have broader application than within the narrow confines of form requirements, such as those set forth in the Statute of Frauds. Indeed, those concepts are an essential element of rules on evidence, both in civil and criminal proceedings. A writing or document generally encompasses anything in which information of any description is recorded.60 This would include, for example, such things as photographs of tombstones and houses,61 account books62 and drawings and plans.63 The relevancy of a document as a piece of evidence is established by connecting it with a person,
57
The Statute of Frauds was originally passed in Great Britain in 1677. When it was originally enacted, it covered five types of contract: (i) a contract by an executor or administrator to accept personal liability; (ii) a contract in consideration of marriage; (iii) a contract for the disposition of land; (iv) a contract where one promises to be a surety; and (v) a contract to be performed more than one year after its formation. Most of its provisions were repealed in the United Kingdom during the twentieth century. J.H. Baker, An Introduction to English Legal History (London: Butterworths-LexisNexis, 2002) 350. For example, section 2-201, subsection 1, of the Uniform Commercial Code of the United States, which has expressed the Statute of Frauds as follows: Except as otherwise provided in this section, a contract for the sale of goods for a price of $500 or more is not enforceable by way of action or defense unless there is some writing sufficient to indicate that a contract for sale has been made between the parties and signed by a party against whom enforcement is sought or by his authorized agent or broker. The purpose of the Statute of Frauds is said to have been [f]or the prevention of many fraudulent practices which are commonly endeavoured to be upheld by perjury and subordination of perjury. Henry Reed, A Treatise on the Law of the Statute of Frauds and of Other Like Enactments in Force in the United States of America and in the British Empire (Philadelphia: Kay & Brother, 1884) 1-3. United Kingdom, Civil Evidence Act 1995, chapter 38, section 13. Lyell v. Kennedy (No. 3), UK, Court of Appeal [1881-1885] All ER Rep 814, 8 April 1884. Hill v. Regem, UK, Kings Bench Division [1945] KB 329, 1 March 1945. J. H. Tucker & Co., Ltd. v. Board of Trade, UK, Chancery Division [1955] 2 All ER 522, 19 May 1955.
58
59
60
61
62 63
158
place or thing, a process which in some common law jurisdictions is referred to as authentication.64 Under the common law on civil evidence, a record or document is regarded as authentic if there is evidence that the document or record is what its proponent claims.65 Signing a document is a common albeit not exclusive means of authentication, and, depending on the context, the terms to sign and to authenticate may be used as synonyms.66 Thus, the purpose of statutes that require a particular document to be signed by a particular person is to confirm the genuineness of the document.67 The legal notion of signature, in turn, is understood as any name or symbol used by a party with the intention of constituting it his signature.68 The paradigm case of signature is the signatorys name, written in the signatorys own hand, on a paper document (ahandwritten or manuscript signature).69 However, the handwritten signature is not the only conceivable type of signature in the common law tradition. Since courts regard signatures as only a mark, where a statute merely requires that a document shall be signed, the statute is satisfied by proof of the making of a mark upon the document by or by the authority of the signatory.70 Thus, for example, the printed name of the party who is required to sign the document may be enough, or the signature may be impressed upon the document by a stamp engraved with a facsimile of the ordinary signature of the person signing, and the courts do not require evidence of the giving of the authority pursuant to which the stamp was so impressed, any more than it would do so if the signature were apparently handwritten.71
Farm Credit Bank of St. Paul v. William G. Huether, 454 N.W.2d 710, 713 (N. Dak. 1990). United States of America, Federal Rules of Evidence, rule 901, subdivision (a): The requirement of authentication or identification as a condition precedent to admissibility is satisfied by evidence sufficient to support a finding that the matter in question is what its proponent claims. In the context of the revised article 9 of the United States Uniform Commercial Code, for example, authenticate is defined as (A) to sign; or (B) to execute or otherwise adopt a symbol, or encrypt or similarly process a record in whole or in part, with the present intent of the authenticating person to identify the person and adopt or accept a record. Lobb v. Stanley, UK, Queens Bench, (1844) 5 Q.B. 574, 114 E.R. 1366. Alfred E. Weber v. Dante De Cecco, 1 N.J. Super. 353, 358 (1948). Lord Denning in Goodman v. Eban, Queens Bench Division, [1954] Q.B.D. 550 at 56: In modern English usage when a document is required to be signed by someone that means that he must write his name with his own hand upon it. R. v. Moore: ex parte Myers (1884) 10 V.L.R. 322, 324 (cited in The Queen v. Joseph Marijancevic, Supreme Court Of Victoria, Court Of Criminal Appeal, 1991 VIC LEXIS 452, 21 June 1991). The Queen v. Joseph Marijancevic, 1991 VIC LEXIS 452.
159
64
65
66
67 68 69
70
71
With time, courts have tended to interpret form requirements, in particular those set forth in the Statute of Frauds liberally, out of recognition that its strict rules were conceived against a particular background72 and that their unconditional enforcement unnecessarily deprives contracts of legal effect.73 Thus, since the 19th century, common law jurisdictions have seen an evolution of the concept of signature from an original emphasis on form to a focus on function.74 The admissibility of identification methods that could be assimilated to the hand written-signatures, such as crosses75 initials,76 pseudonyms77 and identifying phrases,78 printed names,79 signatures by third parties,80 and rubber stamps,81 were admitted by English courts by drawing an analogy with a manuscript signature. Thus, it could be said that against a background of some
72
The Statute of Frauds was passed at a period when the legislature was somewhat inclined to provide that cases should be decided according to fixed rules rather than to leave it to the jury to consider the effect of the evidence in each case. This, no doubt, arose to a certain extent from the fact that in those days the plaintiff and the defendant were not competent witnesses. (J. Roxborough in Leeman v. Stocks, UK, Chancery Division, [1951] 1 Ch 941, 17 April 1951,pp. 947-948) citing approval for the views of J. Cave in Evans v. Hoare, UK Divisional Court [1892] 1 QB 593, 15 March 1892, p. 597). As explained by Lord Bingham of Cornhill It quickly became evident that if the seventeenth century solution addressed one mischief it was capable of giving rise to another: that a party, making and acting on what was thought to be a binding oral agreement, would find his commercial expectations defeated when the time for enforcement came and the other party successfully relied on the lack of a written memorandum or note of the agreement. (Actionstrength Limited v. International Glass Engineering, UK, House of Lords, 3 April 2003, [2003] UKHL 17). Chris Reed, What is a signature?, 3 J. Infor. L. & Tech. (2000) 223, and reference to case law therein <www2.warwick.ac.uk/fac/soc/law/elj/jilt/2000_3/reed/#fn6> (3 December 2007). Baker v. Dening (1838) 8 A. & E. 94 (cited by Lord McDermott L.C.J. in Fulton v. Kee, Court of Appeal (Civil Division), [1961] NI 1, 30 June 1960) United Kingdom, Adolphus and Ellis Queens Bench Reports). Hill v. Hill, Court of Appeal, [1947] Ch 231, 5 December 1946. Redding, in re (1850) 14 Jur. 1052, 2 Rob. Ecc. 339 (cited in Chris Reed, supra note 74). Cook, In the Estate of (Deceased) Murison v. Cook and Another [1960] 1 All ER 689 (id.,). Brydges v. Dicks (1891) 7 T.L.R. 215 (cited in Brennan v. Kinjella Pty Ltd., Supreme Court of New South Wales, 24 June 1993, 1993 NSW LEXIS 7543, 10). Typewriting has also been considered in Newborne v. Sensolid (Great Britain), Ltd. [1954] 1 QB 45 (United Kingdom, Law Reports, Queens Bench). France v. Dutton, 24 April 1891 [1891] 2 QB 208 (cited by Parker L.J. in London County Council v. Vitamins Ltd., Court of Appeal [1955] 2 QB 218, [1955] 2 All ER 229, 31March 1955). Goodman v. J. Eban Ltd., [1954] 1 QB 550 (cited by Denning L.J. in Lazarus Estates, Ltd. v. Beasley, Court of Appeal [1956] 1 QB 702, 24 January 1956).
73
74
75
76 77 78 79
80
81
160
rigid general form requirements, courts in common law jurisdictions have tended to develop a broad understanding of what the notions of authentication and signature mean, focusing on the intention of the parties, rather than on the form of their acts. The approach to form requirements and the understanding of terms such as authentication and signature in civil law jurisdictions is not in all respects identical to the common law approach. Most civil law jurisdictions follow the rule of freedom of form for contractual engagements in private law matters, either expressly82 or impliedly,83 subject, however, to a more or less extensive catalogue of exceptions depending on the jurisdiction concerned. This means that, as a general rule, contracts need not be in writing or signed in order to be valid and enforceable. Nonetheless and here lies a significant difference as compared to the common law approach statutory form requirements in civil law countries are usually understood as a necessary condition for the validity of a contract, and not only for its enforceability. As regards the evidentiary aspects, there are civil law jurisdictions that generally require a writing to prove the contents of contracts, except in commercial matters.84 In contrast to common law jurisdictions, civil law countries tend to interpret evidentiary rules rather strictly. Typically, rules on civil evidence establish a hierarchy of evidence for proving the content of civil and commercial contracts. Highest in such ranking are documents issued by public authorities, followed by authentic private documents. Often, such hierarchy is conceived in such a way that the notions of document and signature, although formally distinct,
82
This is recognized, for instance, in article 11, paragraph 1, of the Code of Obligations of Switzerland. Similarly, section 215 of the Civil Code of Germany provides that agreements are only invalid where they failed to observe a form prescribed by law or agreed upon by the parties. Except for such specific instances, it is generally understood that private law contracts are not subject to specific form requirements. Where the law expressly prescribes a particular form, that requirement is to be interpreted strictly.
83
In France, for instance, freedom of form is an implication within the basic rules on contract formation under the Civil Code. According to article 1108 of the Civil Code of France, the validity of a contract requires the consent of the promisor, his or her legal capacity, a certain object and a licit cause; once these have been met, the contract is law between the parties according to article 1134. This is also the rule in Spain under articles 1258 and 1278 of the Civil Code. Italy also follows the same rule, although less explicitly (see Civil Code of Italy, articles 1326 and 1350). Article 1341 of the Civil Code of France requires a writing for the proof of contracts exceeding a certain value, but article 109 of the Commercial Code admits various types of evidence, without a particular hierarchy. This led the Court of Cassation of France in 1892 to recognize the general principle of freedom of evidence in commercial matters (Cass. civ. 17 mai 1892, DP1892.1.604; cited in Luc Grynbaum, Preuve, Rpertoire de Droit Commercial Dalloz (June 2002), sections 6 and 11).
161
84
may become nearly inseparable.85 Some civil law jurisdictions positively link the notion of document to the existence of a signature.86 This does not mean that a document that has not been signed is necessarily deprived of any value as evidence, but such a document would not enjoy any particular presumption and is generally regarded as a beginning of evidence.87 Authentication in most civil law jurisdictions is a concept that is rather narrowly understood to mean that the authenticity of a document has been verified and certified by a competent public authority or a notary public. In civil procedure it is common to refer instead to the notion of originality of documents. As is the case under the common law, the paradigm of a signature in civil law countries is the handwritten one. As regards the signature itself, some jurisdictions tend to admit various equivalents, including mechanical reproductions of signatures, despite an otherwise formalist approach to evidence.88 Other jurisdictions, however, admit mechanical signatures for commercial transactions,89 but until the advent of computer technologies, continued to require a handwritten signature for the proof of other types of contract.90 By way of conclusion, it could be said that against a general background of freedom of form for the conclusion of business contracts, civil law countries tend to apply strict standards to assess the evidentiary value of private documents
85
Thus, for instance, under German law a signature is not an essential element of the notion of document (Urkunde) (Gerhard Lke and Alfred Walchshfer, Mnchener Kommentar zur Zivilprozessordnung (Munich, Beck, 1992), section 415, No. 6). Nevertheless, the hierarchy of documentary evidence established by sections 415, 416 and 419 of the Code of Civil Procedure of Germany clearly links the signature to the document. Indeed, section 416, on the evidentiary value of private documents (Privaturkunden), provides that private documents constitute full proof for the information they contain as long as they are signed by the author or by a notarized signature). As nothing is provided for documents without a signature, it seems that they share the sort of defective documents (i.e. garbled, damaged), whose evidentiary value is freely established by the courts (Code of Civil Procedure of Germany, section 419). Thus, in France, a signature is an essential element of private documents (actes sous sein priv ) (see Recueil Dalloz, Preuve, no. 638). This is the situation in France, for example (see Recueil Dalloz, Preuve, nos. 657-658). Commentators of the Code of Civil Procedure of Germany point out that requiring a handwritten signature would mean excluding all forms of mechanically made signs, a result that would run counter to ordinary practice and technological progress. See Lke and Walchshfer, Mnchener Kommentar zur Zivilprozeordnung 1. Auflage (1992 Verlag: C.H. Beck), section 416, No. 5. For example, France (see Recueil Dalloz, Preuve, no. 662). In France, for instance, the signature could not be replaced with a cross or other signs, by a seal or by fingerprints (see Recueil Dalloz, Preuve, no. 665).
86
87 88
89 90
162
and may be dismissive of documents whose authenticity is not immediately recognizable on the basis of a signature. (b) Practical Consequences of Form Requirements The above discussion shows not only that the notions of signature and authentication are not uniformly understood, but also that the functions they fulfil vary across legal systems and even under the same legal system, depending on the context of the particular form requirement. Despite these divergences, a few general common elements can be found. The notions of authentication and authenticity are generally understood in law to refer to the genuineness of a document or record, that is, that the document is the original support of the information it contains, in the form it was recorded and without any alteration. Signatures, in turn, perform three main functions in the paper-based environment: signatures serve to identify the signatory (identification function); signatures provide certainty as to the personal involvement of that person in the act of signing (evidentiary function); and signatures associate the signatory with the content of a document (attribution function). Signatures can be said to perform various other functions as well, depending on the nature of the document that was signed. For example, a signature might attest to the intent of a party to be bound by the content of a signed contract; the intent of a person to endorse authorship of a text (thus displaying awareness of the fact that legal consequences might possibly flow from the act of signing); the intent of a person to associate him or herself with the content of a document written by someone else; and the fact that, and the time when, a person has been at a given place. It should be noted, however, that even though the authenticity is often presumed by the existence of a signature, a signature alone does not authenticate a document. The two elements may even be separable, depending on the circumstances. A signature may retain its authenticity even though the document to which it is affixed is subsequently altered. Likewise, a document may still be authentic even though a signature it contains was forged.91 Furthermore, the authority
91
Some areas of the law recognize both the inherent insecurity of handwritten signatures and the impracticability of insisting on strict form requirements for the validity of legal acts, and admit that in some instances even the forgery of a signature would not deprive a document of its legal effect. Thus, for example, article 7 of the Uniform Law on Bills of Exchange and Promissory Notes annexed to the Convention Providing a Uniform Law for Bills of Exchange and Promissory Notes (Geneva, 7 June 1930) provides that if a bill of exchange bears the signatures of persons incapable of binding themselves by a bill of exchange, or forged signatures, or signatures of fictitious persons, or signatures which for any other reason cannot bind the persons who signed the bill of exchange or on whose behalf it was signed, the obligations of the other persons who signed it are none the less valid (League of Nations Treaty Series, vol. CXLIII, No. 3313).
163
to intervene in a transaction and the actual identity of the person in question, while important elements to ensure the authenticity of a document or signature, are neither fully demonstrated by the signature alone, nor are they sufficient assurance of the authenticity of the documents or of the signature. This observation leads to another aspect of the issue presently discussed. Regardless of the particular legal tradition, a signature, with very few exceptions, is not self-standing. Its legal effect will depend on the link between the signature and the person to whom the signature is attributable. In practice, various steps may be taken to verify the identity of the signatory. When the parties are all present at the same place at the same time, they may simply recognize one another by their faces; if they negotiate over the telephone, they may recognize each others voices and so on. Much of this happens as a matter of course and is not subject to specific legal rules. However, where the parties negotiate by correspondence, or where signed documents are forwarded along a contracting chain, there may be few means of establishing that the signs that appear on a given document were indeed made by the person to whose name they appear to be linked and whether indeed only the duly authorized person was the one who produced the signature supposed to bind a particular person. As has been indicated above, in many legal systems, commercial contracts need not always to be contained in a document or evidenced by a writing to be valid. Even where a writing exists, a signature is not necessarily mandatory in order for the contract to be binding on the parties. Of course, where the law requires contracts to be in writing or to be signed, failure to meet those requirements would render the contract invalid or unenforceable. Perhaps more significant than form requirements for purposes of validity of contracts, are form requirements for evidentiary purposes. The difficulty of proving oral agreements is one of the main reasons why commercial contracts are reflected in written documents or documented by correspondence, even if an oral agreement would be otherwise valid. Parties whose obligations are recorded in signed documents are unlikely to succeed in attempts to negate the content of their obligations. Strict rules on documentary evidence typically aim at affording a high degree of reliability to the documents that meet them, which is generally believed to raise legal certainty. At the same time, however, the more elaborate the evidentiary requirements, the greater the opportunity a party has to invoke formal defects with a view to invalidating or denying enforceability to obligations they no longer intend to perform, for instance because the contract has become commercially disadvantageous. It should also be noted that although a manual signature is a familiar form of authentication and serves well for transaction documents passing between known parties, in many commercial and administrative situations a signature is relatively insecure. The person relying on the document often has neither the names of persons authorized to sign nor specimen signatures available for
164
comparison. This is particularly true of many documents relied upon in foreign countries in international trade transactions. Even where a specimen of the authorized signature is available for comparison, only an expert may be able to detect a careful forgery. Where large numbers of documents are processed, signatures are sometimes not even compared except for the most important transactions. Trust is one of the basic foundations of international business relations. Therefore, most legal systems have special procedures or requirements that are intended to enhance the reliability of handwritten signatures. Some procedures may be mandatory in order for certain documents to produce legal effects. They may also be optional and available to parties that wish to preclude possible arguments concerning the authenticity of certain documents. Typical examples include notarization,92 attestation;93 or seals.94 Apart from these special situations, however, handwritten signatures have been used in commercial transactions, both domestic and international, for centuries without any particularly designed legislative or operational framework. The addressees or holders of the signed documents have assessed the reliability of signatures on a case-by-case basis depending on the level of trust enjoyed by the signatory. In fact, the vast majority of international written contracts if there is writing at all are not necessarily accompanied by any special formality or authentication procedure. This commercial reality is often overlooked by legislators and scholars who advocate the imposition of the highest technical standards for the recognition of
92
In certain circumstances, the act of signing has a particular formal significance due to the reinforced trust associated with a special ceremony. This is the case, for instance, with notarization, i.e. the certification by a notary public to establish the authenticity of a signature on a legal document Attestation is the act of watching someone sign a legal document and then signing ones name as a witness. The purpose of attestation is to preserve evidence of the signing. By attesting, the witness states and confirms that the person whom he or she watched sign the document in fact did so. Attesting does not extend to vouching for the accuracy or truthfulness of the document. The witness can be called on to testify as to the circumstances surrounding the signing (Adrian McCullagh, Peter Little and William Caelli, Electronic Signatures: Understand the Past to Develop the Future, 21 New S. Wales L.J. (1998) 452 <www.austlii.edu.au/au/journals/UNSWLJ/1998/59.html> (28 November 2007), see especially chapter III, section D, on the concept of witnessing). Seals are used in several countries in eastern Asia, such as China and Japan. Signing or sealing may, for example, provide evidence of the identity of the signatory; that the signatory agreed to be bound by the agreement and did so voluntarily; that the document is final and complete; or that the information has not been altered after signing. See Mark Sneddon, Legislating to Facilitate Electronic Signatures and Records: Exceptions, Standards and the Impact of the Statute Book, 21 New S. Wales L.J. (1998) 334 (see especially part 2, chapter II, on policy objectives of writing and signature requirements).
165
93
94
electronic equivalents of handwritten signatures. The result has been legislation that forces the use of particular applications that commercial parties would not have freely chosen. 2. Problems Posed by Electronic Communications As we have seen, notions such as writing, document, instrument, and other related concepts such as original or signature have been traditionally associated with hand-written or typed signs impressed by ink on a piece of paper. These notions and the resulting requirements evolved over time as answers to particular legal needs (to proof the content of contracts, to establish when a right had been granted, etc.). New means of communication, however, do not fit squarely within those notions. (a) Intangibility and Alterability Paper documents have been the basis for rules on form and evidence of legal acts in most countries. Accordingly, concepts that support the operation of form requirements and evidentiary rules presuppose to a lesser or greater extent or even require that information of legal significance be recorded in a tangible medium capable of being displayed and stored for subsequent use. The medium of electronic communications however, is by nature intangible, that is, its original form consists of a number of binary digits, or bits that make up the computer file on which the information is recorded. The so-called hard copy, or printout, as the name already suggests, is but a copy of the intangible original computer file. Besides being intangible, electronic communications are also easily alterable and in many cases the changes made to the original information may be hardly, if at all, detectable. These characteristics gave rise to concerns that existing law may not support electronic communications or may even be an obstacle to its use. Those concerns were based on recognition that requirements such as written form or the very notion of document supposed the recording of information of a medium capable of being displayed so as to allow for the information to be read, and which offered some assurance as to the content of the information at the time it was produced. While various technologies can be used to offer some assurance of integrity to electronic records, not all techniques are necessarily comparable so that it is not possible to accept all electronic records as legally equivalent to paper documents. The legislative challenge consists in formulating reasonable criteria for establishing such equivalence. (b) Limitations in Retention and Retrievability Closely related to the above discussion are concerns about the instability of computer files and their relatively short lifecycle. Experience has shown that, in the
166
end, all we can preserve in an electronic context are bits, but it is very difficult to keep a set of bits indefinitely.95 This circumstance, too, makes it difficult to equate all electronic records with paper documents. Although the authentication tools that were used in the past (such as handwritten signatures, seals, stamps, fingerprints etc.) are also subject to reformatting (e.g. microfilming) because of the obsolescence of the paper medium, they never become completely useless after reformatting, since [t]here is always at least a copy available that can be compared with other original authentication tools.96 The main difficulty results from the risk that the original electronic records may become unreadable or unreliable over time, typically because of the obsolescence of the software, the equipment or both. The rapid changes in text processing software over the last twenty years whether or not they always qualify as an evolution for example, give a clear indication of the problem. Without periodic reformatting and conversion, most records produced even less than a decade ago would be unreadable by todays standard software. This makes the long-term retention of electronic records generally problematic and makes electronic records prima facie less reliable than paper documents. Even if the software or hardware as such has not become obsolete, the security devices and techniques used to ensure the integrity of records may become insecure, for example, as a consequence of scientific advances in cryptanalysis.97 Sometimes, the very technique that is used to ensure integrity of electronic records becomes the cause of the records unreliability over time. Digital signatures,98 for example, were for some time believed to be essential for archival purposes. However, they are not immune to long-term risks either. Since every alteration to the record after the time when the signature was created will cause the verification of the signature to fail, reformatting operations intended to keep a record legible for the future (such as migration or conversion) may affect the durability of the signature. Indeed, doubts about the usefulness of digital signatures as a means to ensure data integrity over time appeared as soon as it became clear that, when using digital signatures, control of the integrity is only possible if the electronic data remain completely unchanged at the bit
95
Jos Dumortier and Sofie Van den Eynde, Electronic Signatures and Trusted Archival Services, <www.law.kuleuven.ac.be/icri/publications/172DLM2002.pdf?where> (25 November 2007), p. 5. Id. Jean-Franois Blanchette, Defining Electronic Authenticity: An Interdisciplinary Journey, <polaris.gseis.ucla.edu/blanchette/papers/dsn.pdf> (26 November 2007) (paper published in a supplemental volume of the 2004 International Conference on Dependable Systems and Networks (DSN 2004), Florence, Italy, 28 June-1 July 2004), pp. 228-232. See infra, chapter II.
167
96 97
98
level.99 This raises a problem when archivists want to migrate electronic data to new formats or software platforms in order to keep them accessible and legible. Some experts therefore have concluded that digital signatures are therefore not useful and hence not relevant for archival purposes. If a digital signature secures a document before it enters the archive, the signature should be stripped and translated into metadata. From a legal point of view, however, in order for signed documents to keep their value over time, it is important that the original electronic signature is still present.100 Moreover, digital signatures work well as a means to verify signatures that are created during the period of validity of a certificate. However, once the certificate expires or is revoked, the corresponding public key loses its validity, even if the key pair was not compromised. Accordingly, a digital signatures scheme would need a signature management system to ensure the availability of the signature over time. In fact, data security devices and methods, such as digital signatures, were conceived more for providing security for the communication of information than for the preservation of information over time.101 Initiatives to overcome this problem have not yet resulted in a durable solution.102
99
Dumortier and Van den Eynde, supra note 95 at 3. Id. In 1999, archivists from various countries launched the International Research on Permanent Authentic Records in Electronic Systems (InterPARES) project with the aim of developing the theoretical and methodological knowledge essential to the long-term preservation of authentic records created and/or maintained in digital form (see <www. interpares.org>). The draft report issued on 28 October 2001 by the Authenticity Task Force, which was part of the first phase of the project (InterPARES 1, concluded in 2001), indicated that digital signatures and public key infrastructures (PKI) are examples of technologies that have been developed and implemented as a means of authentication for electronic records that are transmitted across space. Although record-keepers and information technology personnel place their trust in authentication technologies to ensure the authenticity of records, these technologies were never intended to be, and are not currently viable as, a means of ensuring the authenticity of electronic records over time (emphasis added) <www.interpares.org/documents/atf_draft_final_report.pdf> (27 November 2007). (The final report of InterPARES 1 was available on 27 November 2007 at <www.interpares.org/book/index.htm>). The European Electronic Signature Standardization Initiative (EESSI), for example, was created in 1999 by the Information and Communications Technology Standards Board, a collaborative group of organizations concerned with standardization and related activities in information and communications technologies established to coordinate the standardization activity in support to the implementation of European Union Directive on electronic signatures (Official Journal of the European Communities, L 13/12). The EESSI consortium (a standardization effort which seeks to translate the requirements of the European directive on electronic signatures into European standards) has sought to address the need for ensuring the long-term preservation of cryptographically signed documents through its standard on Electronic Signature Formats (Electronic Signature
100
101
102
168
B. Conditions for Legal Equivalence between Electronic Communications and Paper Records
The considerations above should not lead to pessimism as regards the ability of electronic communications to fulfil essentially the same function as paper records. After all, electronic communications are merely a medium for doing business. By and large, the electronic medium has no impact on the nature of the transactions being conducted. Thus a contract for the sale of goods by printed purchase orders, phone conversation, or exchange of e-mails has essentially the same nature and should not be treated differently. New technologies and the emergence of electronic commerce have not rendered old law obsolete. All that is needed is to ensure that concepts currently used in the law, by implying the use of a certain medium (paper records), would not lead to the exclusion of another (electronic records). As legislators and policymakers around the world started to focus on measures to adapt existing laws to accommodate modern means of communication, it soon became clear that law reform intended to remove possible legal barriers to electronic commerce under domestic law required adequate international harmonization to avoid the creation of another set of barriers to international electronic commerce through conflicting domestic standards. As a global organization, the United Nations Commission on International Trade Law (UNCITRAL)103 was chosen to propose uniform private law standards for electronic commerce.
Formats ES 201 733, ETSI, 2000). The designers of these electronic signature formats were concerned with the security threat to the validity of the signature that results from decay in cryptographic strength. To guard against this threat of decay, EESSI signatures are regularly time stamped afresh, with signing algorithms and key sizes appropriate to state-of-the-art cryptanalytic methods. The problem of software longevity has been addressed in a 2000 report by EESSI, which introduced trusted archival services, a new type of commercial service that would be offered by yet to be specified competent bodies and professions, in order to guarantee the long-term preservation of cryptographically signed documents. (see Blanchette, supra note 97). A follow-up study on the EESSI recommendation on trusted archival services by the Interdisciplinary Centre for Law and Information Technology of the Katholieke Universiteit Leuven (Catholic University of Leuven), Belgium, entitled European Electronic Signature Standardization Initiative: Trusted Archival Services (Phase 3, final report, 28 August 2000) is available at <www.law.kuleuven. ac.be/icri/publications/91TAS-Report.pdf?where=> (12 April 2007). EESSI was closed in October 2004. Systems to implement these recommendations do not seem to be currently in operation (see Dumortier and Van den Eynde, Electronic Signatures, supra note 95 pp. 7 et seq.).
103
UNCITRAL was established by the U.N. General Assembly in 1966 with the general mandate to promote harmonization and unification of international trade law. UNCITRAL has 66 member states elected by the General Assembly for terms of six years, the terms of half of the members expiring every three years. Membership is structured so as to be
169
1. Work of UNCITRAL in the Area of Electronic Commerce The transnational nature of electronic commerce and its disregard for traditional jurisdictional borders, with markets migrating from geographic space to cyberspace,104 clearly required international solutions rather than individual state initiatives. International harmonization was the logical approach for dealing with the legal implications of technological developments and to fill what was perceived as a legal vacuum. As a first step to that end, UNCITRAL adopted, in 1996, the UNCITRAL Model Law on Electronic Commerce (hereafter MLEC).105 The MLEC was prepared by the UNCITRAL Working Group on Electronic Commerce106 between 1992 and 1996. It was followed, five years later, by a more specific text, the UNCITRAL Model Law on Electronic Signatures (hereafter MLES).107 Both instruments, in particular the first MLEC, have been very successful and represent a broadly accepted basis for international legal harmonization. The negotiation of the second MLES proved to be more difficult, as member States could not easily reach a common understanding of the legal issues relating to various electronic signature techniques. While some of the signature legislation
representative of the worlds various geographic regions and its principal economic and legal systems. Besides member states, all other states and invited international organizations may participate as observers in UNCITRALs work. UNCITRAL has implemented its mandate by developing texts on a number of topics including sale of goods, arbitration and conciliation, carriage of goods by sea, banking and finance law, procurement, crossborder insolvency, and electronic commerce (for further information, see www.uncitral. org).
104
Stephen J. Kobrin, Economic Governance in an Electronically Networked Global Economy, in R. Hall and T. Biersteker (eds.), The Emergence Of Private Authority: Forms Of Private Authority And Their Implications For Global Governance, <www-management.wharton. upenn.edu/kobrin/Research/revision% 201.pdf> (30 November 2007), p. 11. For the drafting history of the MLEC, see UNCITRAL Model Law on Electronic Commerce with Guide to Enactment, New York, 1999, United Nations publication sales No. E.99.V4 <www.uncitral.org/uncitral/en/uncitral texts/electronic commerce.html> (30 November 2007), paras. 123-150 (hereafter Guide to Enactment of the MLEC). UNCITRAL Working Groups are composed of representatives from member and observer states who may be government officials, academics, practicing lawyers or other experts, depending upon the subject matter. UNCITRAL also invites international, governmental, and non-governmental organizations to participate in its meetings as observers (in electronic commerce, this included organizations such as the United Nations Conference on Trade and Development (UNCTAD), the World Intellectual Property Organization (WIPO), the Organization for Economic Cooperation and Development (OECD), the European Union, the Commonwealth Secretariat, the African Development Bank, the International Chamber of Commerce (ICC), the International Bar Association (IBA), the Internet Law and Policy Forum). Observer States and organizations traditionally take an active role in the preparation of UNCITRAL instruments. See Guide to Enactment of the MLEC, supra note 105, paras. 12-25.
105
106
107
170
around the world had initially focused upon digital signature techniques used in the context of public key infrastructures (PKI),108 it became increasingly clear that PKIs would be only one of several possible methods for electronic authentication. However, PKI models had strong supporters in governments less concerned with preserving party autonomy than with raising security levels. There were also differing views between the United States and members of the European Union (EU) and also between other countries as regards the adequate level of regulation of electronic signatures. The resulting consensus was a flexible set of rules to ensure the continuing usefulness and applicability of the MLES and not to hinder the development of new techniques. The MLES affirms the principle of party autonomy and allows private agreements to be taken into account in assessing whether the nature of the authentication methods used is reasonable or appropriate for the purpose of the particular transactions to which they relate. The MLES also offers basic provisions on cross-border recognition that aim at ensuring legal interoperability. Even if the first two instruments took the form of model laws rather than a treaty, it was clear that the work to be done by UNCITRAL had to aim at promoting the enactment of new legislation. Early exploratory work had shown the need for a set of principles that would provide a basic legal framework for electronic commerce. Some legal issues associated with electronic communications could be addressed by contractual arrangements between the parties to the electronic commerce relationship. However, it was clear that contractual frameworks, such as trading partner agreements,109 which were then being proposed for users of electronic commerce, relied to a large extent upon the structures of local law, which made them inadequate for international use.110 Moreover, a purely contractual framework would not be sufficient to address mandatory
108
Such as the Utah Digital Signature Act 1995, 46 Utah Code Annotated 3-101 to 46-3504; the German Digital Signatures Act 1997 (Gesetz zur Digitalen Signatur or Signaturgesetz, enacted as article 3 of Gesetz zur Regelung der Rahmenbedingungen fur Informations-und Kommunikationsdienste or Informations-und Kommunikationsdienste-Gesetz, 13 June 1997); or the Malaysian Digital Signature Act 1997. A trading partner agreement is an agreement used by parties that agree to exchange information electronically which is used to structure the electronic communications relationship by dealing with various issues that arise in the course of such communications, such as business issues that need to be made in structuring the communications relationship and a number of legal issues that are ordinarily addressed by the communications agreement (see Amelia Boss, Electronic Data Interchange Agreements: Private Contracting Toward a Global Environment, 13 Nw. U. L.J. Intl. L. & Bus. 31, 37-38 and the discussion of issues typically covered in trading partner agreements beginning at 45. See the summary of early international e-commerce initiatives and contractual standards in Eric A. Caprioli and Renaud Sorieul, Le Commerce International Electronique: Vers Lemergence De Regies Juridiques Transnationales, 2 Journal Du Droit International (1997) 323.
171
109
110
requirements in national legislation relating to handwritten signatures or written records or form of legal acts, or to effectively provide rules enforceable against third parties. The need for binding rules was even clearer in connection with international contracts, in view of the divergences in domestic laws and the existence of form requirements established in international conventions. These two circumstances eventually led UNCITRAL to prepare an international treaty, the United Nations Convention on the Use of Electronic Communications in International Contracts (ECC),111 which was adopted by the General Assembly of the United Nations in 2005112 to offer uniform solutions for issues raised by electronic communications in international contracting practice. Despite this evolution, the work of UNCITRAL has maintained a clear and consistent line of policy advice. Its fundamental principles and basic rules are summarized below. (a) Recommended Legislative Principles for Electronic Commerce There is extensive literature regarding both the MLEC113 and its domestic enactments.114 The following paragraphs therefore will not discuss in detail the provisions of the MLEC but only highlight some of its main aspects as a background for discussing UNCITRALs later work. In line with the general mandate of UNCITRAL to promote legal harmonization of international trade law, the MLEC focuses on electronic messages used in the context of commercial activities. This does not mean, however, that an enacting state could not use the MLEC in other areas as well. The MLEC is not conceived for an exclusively business-to-business environment. Consumer transactions are not specifically excluded from the scope of the MLEC, since it was felt that consumers might benefit from the enhanced legal certainty provided by
111
For the drafting history of the ECC, see United Nations Convention on the Use of Electronic Communications in International Contracts, Explanatory Note by the UNCITRAL Secretariat (U.N. sales publication No. E.07.V.)<www.uncitral.org/uncitral/en/uncitral_texts/electronic _commerce/ 2005Convention.html> (30 November 2007), paras. 21-43. See General Assembly Resolution A/RES/60/21, Annex. See, for instance, Renaud Sorieul, Jennifer R. Clift and Jos Angelo Estrella-Faria, Establishing a Legal framework for Electronic Commerce: The Work of the United Nations Commission on International Trade Law, 35 Intl. Law. (2001); see also Amelia H. Boss, Electronic Commerce and the Symbiotic Relationship Between International and Domestic Law Reform, 72 Tulane L. Rev. (1998) 1931; A. Brooke Overby, Model Law on Electronic Commerce:Will Cyberlaw Be Uniform? An Introduction to the UNCITRAL Model Law on Electronic Commerce, 7 Tulane J. Intl & Comp. L. (1999) 219.
112 113
114
For the United States, see, for instance, Amelia H. Boss, 37 Idaho L. Rev. (2001) 275; and R. J. Robertson, Jr., Electronic Commerce on the Internet and the Statute of Frauds, 49 S.C. L. Rev. (1998) 787.
172
the MLEC. Nevertheless, the MLEC was drafted without special attention being given to consumer protection issues.115 Therefore, article 1 of the MLEC includes a footnote to the effect that the MLEC does not override any rule of law intended for the protection of consumers. (i) Functional Equivalence The basic assumption of the work of UNCITRAL in the area of electronic commerce is that traditional legal notions such as document, writing, signature, original form and the like, need not be replaced by entirely new ones. From a legislative perspective, all that may be necessary and prudent to do is to identify the circumstances under which the same function envisaged by the law, for example for a written contract, may be fulfilled by the exchange of communications in electronic form. The approach taken by UNCITRAL has been called a functional equivalence approach. It involves an analysis of basic functions fulfilled by a writing, a signature, or an original in the world of paper documents in order to determine how those functions could be transposed, reproduced, or imitated in a dematerialized environment. (ii) Media and Technology Neutrality UNCITRAL advocates the adoption of neutral rules; that is, rules that do not depend on or presuppose the use of particular types of technology and could be applied to communication and storage of all types of information. Technological neutrality is particularly important in view of speed of technological innovation. It helps ensure that legislation remains capable of accommodating future developments and would not become dated too soon. Accordingly, the various instruments formulated by UNCITRAL carefully avoid any reference to particular technical means of transmission or storage of information in electronic form. The concern to promote media neutrality raises other important points. The impossibility of guaranteeing absolute security against fraud and transmission error is not limited to the world of electronic commerce; it applies to the world of paper documents as well. When formulating rules for electronic commerce, legislators are often inclined to aim at the highest level of security offered by existing technology. The practical need for applying stringent security measures to avoid unauthorized access to data, ensure the integrity of communications, and protect computer and information systems cannot be questioned. However, from the perspective of private business law, it may be more appropriate to graduate security requirements in steps similar to the degrees of legal security encountered
115
Mainly in view of the difficulty of achieving a universally accepted definition of consumer and the existence in some countries of special consumer protection laws that may govern certain aspects of the use of information systems.
173
in the paper world. In the paper world, businessmen are, in most cases, free to choose among a wide range of methods to achieve integrity and authenticity of communications (for example, the different levels of handwritten signature seen in documents of simple contracts and notarized acts). In other words, as a general rule, it is the tradesman, and not the legislator that bears the risk that the contents of a contract cannot be proved because the means chosen do not provide a reliable record of the transaction. Apart from legal form requirements, businessmen make their choices in the light of various factors, including the level of confidence in their trade partners, the type and value of the transactions, and the costs entailed by any particular authentication method. There seems to be no compelling reason why, in purely commercial transactions, that freedom should not exist in the electronic world as well. Hence the flexible notion of reliability appropriate for the purpose for which the electronic communication was generated,116 which UNCITRAL proposes as a general parameter for assessing the quality of electronic communication. (iii) Party Autonomy UNCITRAL has consistently recognized the importance of contract and party autonomy. The importance of allowing trade partners to choose the solutions that best fit their needs should be clear when one considers that most electronic commerce happens through business-to-business transactions.117 Accordingly, the entire third chapter of the MLEC118 contains only default rules and leaves the parties free to organize the use of electronic commerce among themselves. Even some of the provisions in the MLEC that are intended for mandatory application in an enacting country allow, for example, agreements concluded between the parties to be taken into consideration in assessing whether the nature of the methods used to ensure, for example, the security of messages, is reasonable or appropriate for the purpose. The later ECC goes even further and allows the parties to derogate from any of its provisions.119
116
As set out in article 7 of the MLEC; in article 6 of the MLES and article 9, paragraph 3 of the ECC. 92.7% of total e-commerce in 2004 in the United States was B2B with manufacturers leading all industry sectors and accounting for 23.4% of total shipments (US Department of Commerce, E-Stats). See article 11 (Formation and validity of contracts); article 12 (Recognition by parties of data messages); article 13 (Attribution of data messages); article 14 (Acknowledgement of receipt); and article 15 (Time and place of dispatch and receipt of data messages). EEC, article 3.
117
118
119
174
(b) Legal Value of Electronic Communications On the basis of those general principles, UNCITRAL offered various criteria under which electronic communications and records may fulfil existing form requirements for legal transactions. These provisions constitute the core of the MLEC and are intended for mandatory application once implemented domestically. Provision for exceptions is made, however, so that formal requirements for paper can be maintained in certain circumstances. A number of national laws indeed set out transactions where paper is to be maintained. Common examples include wills or other testamentary dispositions; negotiable instruments; trusts and powers of attorney; contracts for disposition or acquisition of real or immovable property; documents of title; affidavits; and court process.120 (i) Conditions for Meeting Form and Evidentiary Requirements UNCITRAL has defined the basic standard to be met by an electronic communication in order to satisfy a requirement that information be retained in writing or that it be contained in a document or other paper-based instrument. While a number of functions are traditionally performed by writings, UNCITRAL focused upon the notion of information being stored in a stable and readable format. These two notions are expressed by UNCITRAL as an objective test that information in electronic form be accessible so as to be usable for subsequent reference.121 The signature issue gave rise to lengthy discussions within UNCITRAL during the preparation of the MLEC and eventually led to another specific MLEC: the 2001 MLES. While signatures perform many functions, all legal systems recognize that a signature serves, at the very least, to (a) identify a person and provide certainty as to the personal involvement of that person in the act of signing; and (b) associate that person with the content of a document. UNCITRAL concentrated upon these two basic functions. In order to satisfy a legal requirement for a signature, UNCITRAL recommends not only that a method must be used that both identifies the originator and confirms the originators approval of the content of the message but also that the method of identification to be used should be as reliable as was appropriate for the purpose for which it was used.122 A number of legal, technical, and commercial factors that might be relevant to determining appropriateness,
120
See a list of examples of such common exceptions in Legal Aspects of Electronic Commerce Electronic Contracting:Provisions for a Draft Convention (U.N. document A/CN.9/WG.IV/ WP.95, of 20 September 2001), Annex II <www.uncitral.org/uncitral/en/commission/ working_groups/4Electronic_Commerce.html> (30 November 2007). MLEC, article 6; EEC, article 9, paragraph 2. MLEC, article 7; MLES, article 6, paragraph 1; EEC, article 9, paragraph 3.
175
121 122
such as the nature of the trade activity being undertaken, the course of trading between the parties, relevant trade customs and practice, and the importance and value of the information contained in the message.123 The rules proposed by UNCITRAL rely on the reasonableness of the parties and the need to strike a balance between the chosen method of identification and the purpose of the message, leaving it to national judges to apply a precise definition. In adopting a flexible test for the way in which these functions should be fulfilled, UNCITRAL deliberately refrained from identifying particular technologies that could be used to achieve functional equivalence. UNCITRAL was of the view that it would be inappropriate to promote any specific technology, be it by mandating its use or by giving it a more favourable treatment as compared to other products or technologies. What constitutes an electronic equivalent of an original document is a matter that according to UNCITRAL should be assessed in the light of the integrity of information and the ability to present it, when this is a requirement, as forming the essence of the concept of originality. As with the notion of writing, here, too, UNCITRAL advocates a flexible test of requiring that the method of assurance as to integrity must be reliable. Reliability is assessed by reference to the purposes for which the information was generated and to relevant circumstances. Factors to be considered in determining reliability would include whether the recording of the information was systematic, whether it was recorded without gaps or errors, and how the information was protected against alteration.124 In order to ensure that, once such conditions are met, electronic records will be truly equivalents of paper documents, UNCITRAL also recommends that in any legal proceedings, nothing in the application of the rules of evidence shall apply so as to deny the admissibility of an electronic communication on the sole ground that it is not in its original form if it is the best evidence that the person adducing it could reasonably be expected to obtain.125 For the purpose of assessing the evidential weight of information in the form of electronic communication, courts are invited to consider the reliability of the manner in which the electronic communication was generated, stored or communicated, to the reliability of the manner in which the integrity of the information was maintained, to the manner in which its originator was identified, and to any other relevant factor.126 UNCITRAL also proposed rules that allow current requirements relating to storage of information (such as accounting or tax records) to be adapted to
123 124
See, Guide to Enactment of the MLEC, supra note 105, paragraph 61.
MLEC, article 8, subparagraphs 3(a) and 3(b); ECC, article 9, subparagraphs 4(a) and 4(b). MLEC, article 9, paragraph 1. MLEC, article 9, paragraph 2.
125 126
176
the needs of electronic commerce.127 Here, functional equivalence should be recognized if the information contained in the electronic record (a) is accessible so as to be usable for subsequent reference, (b) was retained in the format in which it was generated, sent or received (or in a format which can be demonstrated to represent accurately the information generated, sent or received, and such information, if any) (c) information has been retained as enables the identifica tion of the origin and destination of an electronic communication and the date and time when it was sent or received. However, an electronic communication does not need to be retained unaltered as long as the information stored accurately reflects the electronic communication as it was sent.128 (ii) Conditions for the Use of Electronic Communications The general rules on functional equivalence are supplemented by a number of rules concerning the actual exchange of electronic communications. The rules formulated by UNCITRAL are to some extent drawn from, or inspired by, certain rules contained in model contracts and interchange agreements. They deal with formation and validity of contracts; recognition of electronic communications by the parties; attribution of electronic communications; acknowledgement of receipt of electronic communications; and time and place of dispatch and receipt of electronic communications.129 2. Implementation and Application of Electronic Commerce Laws One of the main objectives of the MLEC and the MLES was to preempt disharmony and possible over-regulation by offering general criteria to establish the functional equivalence between electronic and paper-based signature and authentication methods. The MLEC has found widespread acceptance, and an increasing number of countries have used it as a basis for their e-commerce legislation130 The MLEC
127 128 129
MLEC, article 10. Guide to Enactment of the MLEC, supra note 105, para. 73.
The most significant of these rules are discussed in Part Two, chapter I. As of November 2007, legislation implementing provisions of the MLEC had been adopted in at least the following countries: Australia (Electronic Transactions Act 1999); China (Electronic Signatures Law, 2004); Colombia (Ley de comercio electrnico, 1999); Dominican Republic (Ley sobre comercio electrnico, documentos y firmas digitales, 2002); Ecuador (Ley de comercio electrnico, firmas electrnicas y mensajes de datos, 2002); France (Loi 2000-230 portant adaptation du droit de la preuve aux technologies de linformation et relative la signature lectronique, 2000); India (Information Technology Act 2000); Ireland (Electronic Commerce Act 2000); Jordan (Electronic Transactions Law 2001); Mauritius (Electronic Transactions Act 2000); Mexico (Decreto por el que se reforman y adicionan diversas disposiciones del cdigo civil para. el distrito federal en materia federal, del Cdigo federal
177
130
has also served as a basis for the domestic harmonization of e-commerce legislation in countries organized on a federal basis, such as Canada131 and the United States of America.132
de procedimientos civilies, del Cdigo de comercio y de la Ley federal de proteccin al consumidor, 2000); New Zealand (Electronic Transactions Act 2002); Pakistan (Electronic Transactions Ordinance 2002); Panama (Ley de firma digital, 2001); Philippines (Electronic Commerce Act 2000); Republic of Korea (Framework Act on Electronic Commerce 2001); Singapore (Electronic Transactions Act 1998); Slovenia (Electronic Commerce and Electronic Signature Act, 2000); South Africa (Electronic Communications and Transactions Act 2002); Sri Lanka (Electronic Transactions Act 2006); Thailand (Electronic Transactions Act, 2001); United Arab Emirates (Dubai) (Electronic Transactions and Commerce Law, 2002); Venezuela (Ley sobre mensajes de datos y firmas electrnicas, 2001); and Viet Nam (Law on Electronic Transactions, 2006). The MLEC has also been adopted in the British crown dependencies of the Bailiwick of Guernsey (Electronic Transactions (Guernsey) Law 2000), the Bailiwick of Jersey (Electronic Communications (Jersey) Law 2000) and the Isle of Man (Electronic Transactions Act 2000); in the overseas territories of the United Kingdom of Great Britain and Northern Ireland of Bermuda (Electronic Transactions Act 1999), the Cayman Islands (Electronic Transactions Law 2000) and the Turks and Caicos (Electronic Transactions Ordinance 2000); and in Hong Kong Special Administrative Region (SAR) of China (Electronic Transactions Ordinance (2000)). Unless otherwise indicated, references made hereafter to statutory provisions of any of these countries refer to provisions contained in the statutes listed above.
131
The domestic enactment of the MLEC in Canada is the Uniform Electronic Commerce Act, adopted by the Uniform Law Conference of Canada in 1999 <www.chlc.ca/en/poam2/ index.cfm?sec=1999&sub=1999ia> (30 November 2007). The Act has since been enacted in a number of provinces and territories of Canada, including Alberta, British Columbia, Manitoba, New Brunswick, Newfoundland and Labrador, NovaScotia, Ontario, Prince Edward Island, Saskatchewan and Yukon. The Province of Quebec enacted specific legislation (the Act to Establish a Legal Framework for Information Technology (2001)), which, although being broader in scope and drafted very differently, achieves many of the objectives of the Uniform Electronic Commerce Act and is generally consistent with the MLEC. Updated information on the enactment of the UETA Act may be found at <www. chlc.ca/en/us/> (30 November 2007). In the United States of America, the National Conference of Commissioners on Uniform State Law used the MLEC as a basis for preparing the Uniform Electronic Transactions Act, which it adopted in 1999 (the text of the Act and the official commentary is available at <www.law.upenn.edu/bll/ulc/uecicta/eta1299.htm> (30 November 2007)). The Uniform Electronic Transactions Act has since been enacted in the District of Columbia and in the following 46 states: Alabama, Alaska, Arizona, Arkansas, California, Colorado, Connecticut, Delaware, Florida, Hawaii, Idaho, Indiana, Iowa, Kansas, Kentucky, Louisiana, Maine, Maryland, Massachusetts, Michigan, Minnesota, Mississippi, Missouri, Montana, Nebraska, Nevada, New Hampshire, New Jersey, New Mexico, NorthCarolina, North Dakota, Ohio, Oklahoma, Oregon, Pennsylvania, Rhode Island, SouthCarolina, South Dakota, Tennessee, Texas, Utah, Vermont, Virginia, West Virginia, Wisconsin and Wyoming. Other states are likely to adopt implementing legislation in the near future, including the state of Illinois, which had already enacted the UNCITRAL MLEC through
132
178
Yet it cannot be assumed that the principles of the MLEC have achieved universal application. The attitude taken by various jurisdictions in relation to electronic signatures and authentication typically reflects the general approach of the jurisdiction to writing requirements and the evidentiary value of electronic records. (a) Authentication and Attribution of Electronic Records The use of electronic methods of authentication involves two aspects that are relevant for the present discussion. The first aspect relates to the general issue of attribution of a message to its purported originator. The second relates to the appropriateness of the identification method used by the parties for the purpose of meeting specific form requirements, in particular legal signature requirements. The MLEC deals with attribution of electronic communications in its article 13. This provision is intended to apply where there is a question as to whether an electronic communication was really sent by the person who is indicated as being the originator. In the case of a paper-based communication, the problem would arise as the result of an alleged forged signature of the purported originator. In an electronic environment, an unauthorized person may have sent the message, but the authentication by code, encryption or similar means would be accurate. The purpose of article 13 is not to attribute authorship of an electronic communication or to establish the identity of the parties. Rather, it establishes the conditions under which a party may rely on the assumption that an electronic communication was actually from the purported originator. Article 13, paragraph 1, of the MLEC recalls the principle that an originator is bound by an electronic communication if it has effectively sent that message. Paragraph 2 refers to a situation where the message was sent by a person other than the originator who had the authority to act on behalf of the originator. Paragraph 3 deals with two kinds of situations in which the addressee could rely on an electronic communication as being that of the originator: first, situations in which the addressee properly applied an authentication procedure previously agreed to by the originator; and second, situations in which the electronic communication resulted from the actions of a person who, by virtue of his or her relationship with the originator, had access to the originators authentication procedures.
the Electronic Commerce Security Act (1998). Updated information on the enactment of the UETA Act may be found at <www.nccusl.org/nccusl/uniformact_factsheets/ uniformacts-fs-ueta.asp> (30 November 2007).
179
A number of countries have adopted the rule in article 13 of the MLEC, including the presumption of attribution established in paragraph 3 of that article.133 Some countries expressly refer to the use of codes, passwords or other means of identification as factors that create a presumption of authorship.134 There are also more general versions of article 13, in which the presumption created by proper verification through a previously agreed procedure is rephrased as an indication of elements that may be used for attribution purposes.135 However, other countries have adopted only the general rules in article 13, namely that an electronic communication is that of the originator if it was sent by the originator him or herself, or by a person acting on the originators behalf or by a system programmed by or on behalf of the originator to operate automatically.136 In addition, several countries that have implemented the MLEC have not included any specific provision based on article 13.137 The assumption in those countries was that no specific rules were needed and that attribution was better left to ordinary methods of proof, in the same way as attribution of documents on paper.138 Other countries have preferred to take the provisions of the MLEC on attribution separately from provisions on electronic signatures. This approach is based on the understanding that attribution in a documentary context serves the
133
Colombia (art. 17); Ecuador (art. 10); Jordan (art. 15); Mauritius (sect. 12, subsect. 2); Philippines (sect. 18, para. 3); Republic of Korea (art. 7, para. 2); Singapore (sect. 13, subsect. 3); Thailand (sect. 16); and Venezuela (Bolivarian Republic of) (art. 9). The same rules are also contained in the laws of the British crown dependency of Jersey (art. 8) and the British overseas territories of Bermuda (sect. 16, para. 2) and Turks and Caicos (sect. 14). Mexico art. 90, para. I. For example, the Uniform Electronic Transactions Act of the United States provides in section 9, subsection (a), that an electronic record or electronic signature is attributable to a person if it was the act of the person. The act of the person may be shown in any manner, including a showing of the efficacy of any security procedure applied to determine the person to which the electronic record or electronic signature was attributable. Section 9, subsection (b), provides further that the effect of an electronic record or electronic signature attributed to a person under subsection (a) is determined from the context and surrounding circumstances at the time of its creation, execution, or adoption, including the parties agreement, if any, and otherwise as provided by law. Australia (sect. 15, para. 1); essentially in the same manner, India (sect. 11); Pakistan (sect. 13, subsect. 2); Slovenia (art. 5); the British crown dependency of the Isle of Man (sect. 2); and Hong Kong SAR of China (sect. 18). For example, Canada, France, Ireland, New Zealand and South Africa. It has been noted that a person who wishes to rely on any signature takes the risk that the signature is invalid, and this rule does not change for an electronic signature (UECA, Official Commentary <www.ulcc.ca/en/poam2/index.cfm?sec=1999&sub=1999ia#1>, section 10).
134 135
136
137 138
180
primary purpose of providing a basis for reasonable reliance, and may include broader means than those more narrowly used for identifying individuals. Some laws, such as UETA in the United States, emphasize this principle by stating, for example, that an electronic record or electronic signature is attributable to a person if it was the act of the person, which may be shown in any manner, including a showing of the efficacy of any security procedure applied to determine the person to which the electronic record or electronic signature was attributable.139 Such a general rule on attribution does not affect the use of a signature as a device for attributing a record to a person, but is based on the recognition that a signature is not the only method for attribution.140 According to the UETA commentary: Certain information may be present in an electronic environment that does not appear to attribute but which clearly links a person to a particular record. Numerical codes, personal identification numbers, public and private key combinations all serve to establish the party to whom an electronic record should be attributed. Of course security procedures will be another piece of evidence available to establish attribution. The inclusion of a specific reference to security procedures as a means of proving attribution is salutary because of the unique importance of security procedures in the electronic environment. In certain processes, a technical and technological security procedure may be the best way to convince a trier of fact that a particular electronic record or signature
139
United States, Uniform Electronic Transactions Act (1999), section 9. Paragraph 1 of the official comments to section 9 offer the following examples where both the electronic record and electronic signature would be attributable to a person: a person types his/her name as part of an e-mail purchase order; a persons employee, pursuant to authority, types the persons name as part of an e-mail purchase order; or a persons computer, programmed to order goods upon receipt of inventory information within particular parameters, issues a purchase order which includes the persons name, or other identifying information, as part of the order. Id. Paragraph 3 of the official comments to section 9 states: The use of facsimile transmissions provides a number of examples of attribution using information other than a signature. A facsimile may be attributed to a person because of the information printed across the top of the page that indicates the machine from which it was sent. Similarly, the transmission may contain a letterhead that identifies the sender. Some cases have held that the letterhead actually constituted a signature because it was a symbol adopted by the sender with intent to authenticate the facsimile. However, the signature determination resulted from the necessary finding of intention in that case. Other cases have found facsimile letterheads NOT to be signatures because the requisite intention was not present. The critical point is that with or without a signature, information within the electronic record may well suffice to provide the facts resulting in attribution of an electronic record to a particular party.
181
140
was that of a particular person. In certain circumstances, the use of a security procedure to establish that the record and related signature came from the persons business might be necessary to overcome a claim that a hacker intervened. The reference to security procedures is not intended to suggest that other forms of proof of attribution should be accorded less persuasive effect. It is also important to recall that the particular strength of a given procedure does not affect the procedures status as a security procedure, but only affects the weight to be accorded the evidence of the security procedure as tending to establish attribution.141 It is also important to bear in mind that a presumption of attribution would not of itself displace the application of rules of law on signatures, where a signature is needed for the validity or proof of an act. Once it is established that a record or signature is attributable to a particular party, the effect of a record or signature must be determined in light of the context and surrounding circumstances, including the parties agreement, if any and of other legal requirements considered in light of the context.142 Against the background of this flexible understanding of attribution, the courts in the United States seem to have taken a liberal approach to the admissibility of electronic records, including e-mail, as evidence in civil proceedings.143 Courts in the United States have dismissed arguments that e-mail messages were inadmissible as evidence because they were unauthenticated and parol evidence.144 The courts have found instead that e-mails obtained from the plaintiff during the discovery process were self-authenticating, since the production of documents during discovery from the parties own files is sufficient to justify a finding of selfauthentication.145 The courts tend to take into account all available evidence and do not reject electronic records as being prima facie inadmissible. In countries that have not adopted the MLEC, there seem to be no specific legislative provisions dealing with attribution in an analogous fashion. In those countries, attribution is typically a function of the legal recognition of electronic signatures and the presumptions attached to records authenticated with particular types of electronic signature. Concerns about the risk of manipulation in electronic records have, for instance, led courts in some of those countries to dismiss the value of e-mails as evidence in court proceedings, on the grounds
141 142 143
Id., official comments on section 9. Id., paragraph 6 of the official comments on section 9. Commonwealth Aluminum Corporation v. Stanley Metal Associates, 186 F. Supp. 2d 770 (W.D. Ky. 2001); and Central Illinois Light Company (CILCO) v. Consolidation Coal Company (Consol), 235 F. Supp. 2d 916 (C.D. Ill. 2002). Sea-Land Service, Inc. v. Lozen International, LLC, 285 F.3d 808 (9th Cir. 2002). Superhighway Consulting, Inc. v. Techwave, Inc., (N.D. Ill. 1999 LEXIS 17910).
144 145
182
that e-mails do not offer adequate guarantees of integrity.146 Further examples of a more restrictive approach to the evidentiary value of electronic records and attribution can be found in cases involving Internet auctions in Germany, in which courts have applied a high standard for attribution of electronic communications. Those cases have typically involved suits for breach of contract on the grounds of lack of payment for goods allegedly purchased in Internet auctions. Claimants maintained that the defendants were the buyer, as the highest bid for the goods had been authenticated with the defendants password and had been sent from the defendants e-mail address. German courts have found that those elements were not sufficient to firmly conclude that it was in fact the defendant who had participated in the auction and submitted the winning bid for the goods. The courts have used various arguments to justify that position. For example, passwords were found not to be reliable because anyone who knew the defendants password could have used its e-mail address from anywhere and participated in the auction using the defendants name,147 a risk that some courts estimated as very high on the basis of expert evidence regarding security threats to Internet communications networks, in particular through the use of trojan horses capable of stealing a persons password.148 According to those courts, the risk of unauthorized use of a persons identification device (password) should be borne by the party that offered goods or services through a particular medium, as there was no legal presumption that messages sent through an Internet website with recourse to a persons access password to such website were attributable to that person.149 German courts held that such a presumption might conceivably be attached to an advanced electronic
146
Germany, Amtsgericht (District Court) Bonn, Case No. 3 C 193/01, 25 October 2001, JurPC Internet-Zeitschrift fr Rechtsinformatik und Informationsrecht, JurPC Web-Dok. No. 332/2002 <www.jurpc.de/rechtspr/20020332.htm> (11 September 2003). Germany, Amtsgericht (District Court) Erfurt, Case No. 28 C 2354/01, 14 September 2001, JurPC Internet-Zeitschrift fr Rechtsinformatik und Informationsrecht, JurPC WebDok. No. 71/2002 <www.jurpc.de/rechtspr/20020071.htm> (25 August 2003); see also Landesgericht (Land Court) Bonn, Case No. 2 O 472/03, 19 December 2003, JurPC, Internet-Zeitschrift fr Rechtsinformatik und Informationsrecht, JurPC Web-Dok. No.74/2004 <www.jurpc.de/rechtspr/20040074.htm> (2 February 2007). Germany, Landesgericht (Land Court) Konstanz, Case No. 2 O 141/01 A, 19 April 2002, JurPC Internet-Zeitschrift fr Rechtsinformatik und Informationsrecht, JurPC Web-Dok. No. 291/2002 <www.jurpc.de/rechtspr/20020291.htm> (25 August 2003). Germany, Landesgericht (Land Court) Bonn, Case No. 2 O 450/00, 7 August 2001, JurPC Internet-Zeitschrift fr Rechtsinformatik und Informationsrecht, JurPC Web-Dok. No. 136/2002 <www.jurpc.de/rechtspr/20020136.htm> (25 August 2003).
183
147
148
149
signature, as defined by German law, but the holder of a simple password should not bear the risk of it being misused by unauthorized persons.150 (b) Ability to Meet Legal Signature Requirements In some countries, the courts have been inclined to interpret signature requirements liberally. As previously indicated this has been typically the case in some common law jurisdictions in connection with Statute of Frauds requirements that certain transactions must be in writing and bear a signature in order to be valid. Courts in the United States for instance, have also been receptive to legislative recognition of electronic signatures, and have been inclined to assess the adequacy of the authentication in the light of the dealings between the parties, rather than using a strict standard for all situations. Thus, where the parties had regularly used e-mail in their negotiations, United States courts have found that the originators typed name in an e-mail satisfied statutory signature requirements.151 A persons deliberate choice to type his name at the conclusion of all e-mails has been considered to be valid authentication.152 The readiness of the United States courts to accept that e-mails and names typed therein are capable of satisfying writing requirements153 follows a liberal interpretation of the notion of signature, which is understood as encompassing any symbol executed or adopted by a party with present intention to authenticate a writing so that, in some instances, a typed name or letterhead on a document is sufficient to satisfy the signature requirement.154 Where the parties do not deny having written or received communications by email, statutory signature requirements would be met, since courts have long recognized that a binding signature may take the form of any mark or designation thought proper by the party to be bound, provided that the author intends to bind himself .155
150
Germany, Oberlandesgericht (Court of Appeal) Kln, Case No. 19 U 16/02, 6 September 2002, JurPC Internet-Zeitschrift fr Rechtsinformatik und Informationsrecht, JurPC Web-Dok. No.364/2002 <www.jurpc.de/rechtspr/20020364.htm> (25 August 2003). Cloud Corporation v. Hasbro, Inc., 314 F. 3d 296 (7th Cir. 2002). Jonathan P. Shattuck v. David K. Klotzbach, 14 Mass. L. Rep. 360 (2001). Central Illinois Light Company v. Consolidation Coal Company, 235 F. Supp. 2d 916 (C.D. Ill. 2002). Id., at 919: Internal documents, invoices and e-mails can be used to satisfy the Illinois [Uniform Commercial Code] statute of frauds. In the concrete case, however, the court found that the alleged contract failed to satisfy the statute of frauds, not because the e-mails as such could not validly record the terms of a contract, but because there was no indication that the authors of the e-mails and the persons mentioned therein were employees of the defendant. Roger Edwards, LLC v. Fiddes & Son, Ltd., 245 F. Supp. 2d 245 (D. Me. 2003).
151 152 153
154
155
184
Courts in the United Kingdom have taken a similar approach, generally considering the form of a signature to be less relevant than the function it serves. Thus, courts would consider the fitness of the medium both to attribute a record to a particular person, and to indicate the persons intention with respect to the record. E-mails may therefore constitute documents, and names typed on the e-mails may be signatures.156 Some courts have declared that they have no doubt that if a party creates and sends an electronically created document then he will be treated as having signed it to the same extent that he would in law be treated as having signed a hard copy of the same document and that [t] he fact that the document is created electronically as opposed to as a hard copy can make no difference.157 In Singapore, too, the courts are inclined to admit e-mails and names typed therein as sufficient proof of identity for the purpose of meeting legal signature requirements.158 On occasion, courts have rejected arguments that e-mails constituted signed contracts for the purposes of the Statute of Frauds, mainly because the intent to be bound by the signature was lacking. There seems to be no precedent, however, where courts in common law jurisdictions would have denied a priori the ability of e-mails and names typed therein to meet statutory writing and signature requirements. In some cases, it was found that the requirements of the Statute of Frauds were not met because the e-mails in question only reflected ongoing negotiations and not a final agreement, for instance because during the negotiations one of the parties had contemplated that a binding contract would be entered into once a deal memo had been signed, and not before.159 In other cases courts have suggested that they might have been inclined to admit as a signature the originators name or initials at the end of the e-mail or anywhere else in the body of the e-mail, but held that the automatic insertion of a persons e-mail address after the document has been transmitted by either the sending and/or receiving [Internet service provider] was not intended for
156 157
Hall v. Cognos Limited (Hull Industrial Tribunal, Case No. 1803325/97) (unreported). Mehta v. J. Pereira Fernandes S.A., Chancery Division, [2006] EWHC 813 (Ch), [2006] 2 Lloyds Rep 244, 7 April 2006 Given that no real distinction can be drawn between a typewritten form and a signature that has been typed onto an e-mail and forwarded with the e-mail to the intended recipient of that message, the signature requirement in this case was satisfied by the inscription of the defendants representatives name next to his e-mail address at the top of his e-mails (Sm Integrated Transware Pte Ltd v. Schenker Singapore (Pte) Ltd, High Court, [2005] SGHC 58, 30 March 2005). Pretty Pictures Sarl v. Quixote Films Ltd., Queens Bench [2003] EWHC 311 (QB), 30 January 2003.
185
158
159
a signature.160 Although British courts seem to interpret the writing requirements of the statute of frauds more strictly than their United States counterparts, they are generally inclined to admit the use of any type of electronic signature or authentication method, even outside any specific statutory authorization, as long as the method in question serves the same functions as a handwritten signature.161 Courts in civil law jurisdictions tend generally to follow a more restrictive approach, arguably because for many of those countries the notion of document ordinarily implies the use of some form of authentication, thus becoming hardly dissociable from a signature as seen above. Courts in France, for instance, had been reluctant to accept electronic means of identification as equivalent to handwritten signatures until the adoption of legislation expressly recognizing the validity of electronic signatures.162 A slightly more liberal line is taken by decisions that accept the electronic filing of administrative complaints for the purpose of meeting a statutory deadline, at least as long as they are subsequently confirmed by regular correspondence.163 In contrast to their restrictive approach to the attribution of electronic communications in the formation of contracts, German courts seem to have been liberal in the acceptance of identification methods as equivalent to handwritten signatures in court proceedings. The debate in Germany has evolved around the increasing use of scanned images of legal counsels signature to authenticate computer facsimiles containing statements of appeals transmitted directly from a computer station via modem to a courts facsimile machine. In earlier cases,
160
Mehta v. J. Pereira Fernandes S.A. supra note 157 Mehta v. J. Pereira Fernandes S.A., supra note 157, No. 25: It is noteworthy that the Law Commissions view in relation to [the European Union Directive on electronic commerce (2000/31/EC)] is that no significant changes are necessary in relation to statutes that require signatures because whether those requirements have been satisfied can be tested in a functional way by asking whether the conduct of the would-be signatory indicates an authenticating intention to a reasonable person. Thus, as I have already said, if a party or a partys agent sending an e-mail types his or her or his or her principals name to the extent required or permitted by existing case law in the body of an e-mail, then in my view that would be a sufficient signature for the purposes of [the statute of frauds]. The Court of Cassation of France rejected the receivability of a statement of appeal signed electronically, because there were doubts as to the identity of the person who created the signature and the appeal had been signed electronically before entry into force of the law of 13 March 2000, which recognized the legal effect of electronic signatures (Cour de cassation, Deuxime chambre civile, 30 April 2003, St Chalets Boisson c/ M. X. <www. juriscom.net/jpt/visu.php?ID=239> (29 November 2007). France, Conseil dtat, 28 dcembre 2001, No. 235784, lections municipales dEntreDeux-Monts <www.rajf.org/article.php3?id_article=467> (3 December 2007).
161
162
163
186
courts of appeal164 and the Federal Court (Bundesgerichtshof )165 had held that a scanned image of a handwritten signature did not satisfy existing signature requirements and offered no proof of a persons identity. Identification might conceivably be attached to an advanced electronic signature, as defined in German law. Generally, however, it was for the legislator and not the courts to establish the conditions for the equivalence between writings and intangible communications transmitted by data transfers.166 That understanding was eventually reversed in view of the unanimous opinion of the other high federal courts that accepted the delivery of certain procedural pleas by means of electronic communication of an electronic communication accompanied by a scanned image of a signature.167 It is interesting to note that even courts in some civil law jurisdictions that have adopted legislation favouring the use of PKI-based digital signatures, such as Colombia,168 have taken a similarly liberal approach and confirmed, for example, the admissibility of judicial proceedings conducted entirely by electronic communications. The submissions exchanged during such proceedings were found to be valid, even if they were not signed with a digital signature, since
164
For instance, Oberlandesgericht (Court of Appeal) Karlsruhe, Case No. 14 U 202/96, 14November 1997, JurPC Internet-Zeitschrift fr Rechtsinformatik und Informationsrecht, JurPC Web-Dok. No. 09/1998 <www.jurpc.de/rechtspr/19980009.htm> (29 November 2007). Germany, Bundesgerichtshof (Federal Court of Justice), Case No. XI ZR 367/97, 29 September 1998, JurPC Internet-Zeitschrift fr Rechtsinformatik und Informationsrecht, JurPC Web-Dok. No. 05/1999 <www.jurpc.de/rechtspr/19990005.htm (29 November 2007). Id. In a decision on a case referred to it by the Bundesgerichtshof of Germany (Federal Court of Justice), the Gemeinsamer Senat der obersten Gerichtshfe des Bundes (Joint Chamber of the Highest Federal Courts of Germany) noted that form requirements in court proceedings were not an end in themselves. Their purpose was to ensure a sufficiently reliable (hinreichend zuverlssig) determination of the content of the writing and the identity of the person from whom it emanated. The Joint Chamber noted the evolution in the practical application of form requirements to accommodate earlier technological developments such as telex or facsimile. The Joint Chamber held that accepting the delivery of certain procedural pleas by means of electronic communication of a data message with a scanned image of a signature would be in line with the spirit of existing case law (Gemeinsamer Senat der obersten Gerichtshfe des Bundes, GmS-OGB 1/98, 5 April 2000, JurPC Internet-Zeitschrift fr Rechtsinformatik und Informationsrecht, JurPC Web-Dok. No. 160/2000 <www.jurpc. de/rechtspr/20000160.htm (29 November 2007). For example, Colombia, which has adopted the MLEC, including the general provisions of its article 7, but has established a legal presumption of authenticity only in respect of digital signatures (Ley de comercio electrnico, supra note 130, art. 28).
187
165
166 167
168
the electronic communications used methods that allowed otherwise for the identification of the parties.169 Case law on electronic signatures is still rare and the small number of court decisions to date does not provide a sufficient basis to draw firm conclusions. Nevertheless, a brief review of existing precedents reveals several trends. It seems that the legislative approach taken to electronic signatures and authentication, as discussed below, has influenced the attitude of courts on this issue. Arguably, the legislative focus on electronic signatures, without an accompanying general rule on attribution, has led to excessive attention being paid to the identity function of authentication methods. This has, in some countries, engendered a certain degree of mistrust visvis any authentication methods that do not satisfy the statutory definition of an electronic signature. It is therefore doubtful that the same courts that have adopted a liberal approach in the context of judicial or administrative appeals would be equally liberal in respect of signature requirements for the validity of contracts. Indeed, while in a contractual context a party might be faced with the risk of repudiation of the agreement by the other party, in the context of civil proceedings it is typically the party using electronic signatures or records that is interested in confirming its approval of the record and its contents.
Chapter II. Electronic Signature and Authentication Methods

Information and computer technology have developed various means for linking information in electronic form to particular persons or entities, for ensuring the integrity of such information, or for enabling persons to demonstrate their entitlement or authorization to obtain access to a certain service or repository of information. These functions are sometimes referred to generically either as electronic authentication or electronic signature methods. Sometimes, however, distinctions are made between electronic authentication and electronic signature. The use of terminology is not only inconsistent, but to some extent misleading. In a paper-based environment, the words
169
Colombia, Juzgado Segundo Promiscuo Municipal Rovira Tolima, Juan Carlos Samper v. Jaime Tapias, 21 July 2003, Rad. 73-624-40-89-002-2003-053-00. The Court found that the process undertaken via electronic means was valid notwithstanding that the e-mails were not digitally signed because (a) the sender of the data messages could be fully identified; (b) the sender of the data messages consented to and affirmed the content of the data messages sent; (c) the data messages were safely kept in the Tribunal; and (d) the messages could be reviewed at any time <www.camara-e.net/_upload/80403 0-7diaz082003.pdf> (3 December 2007).
188
authentication and signature and the related actions of authenticating and signing do not have exactly the same connotation in different legal systems and have functionalities that may not necessarily correspond to the purpose and function of the so-called electronic authentication and signature methods. Furthermore, the word authentication is sometimes generically used in connection with any assurance of both authorship and integrity of information, but some legal systems may distinguish between those elements. A short overview of differences in terminology and legal understanding is therefore necessary with a view to narrowing the limits of the present discussions.
A. Notion of Electronic Authentication and Signature

The terms electronic authentication or electronic signature are used to refer to various techniques currently available on the market or still under development for the purpose of replicating in an electronic environment some or all of the functions identified as characteristic of handwritten signatures or other traditional authentication methods. A number of different electronic authentication and signature techniques have been developed over the years. Each technique aims at satisfying different needs and providing different levels of security, and entails different technical requirements. Electronic authentication and signature methods may be classified in three categories: those based on the knowledge of the user or the recipient (e.g. passwords, personal identification numbers (PINs)), those based on the physical features of the user (e.g. biometrics) and those based on the possession of an object by the user (e.g. codes or other information stored on a magnetic card).170 A fourth category might include various types of authentication and signature methods that, without falling under any of the above categories, might also be used to indicate the originator of an electronic communication (such as a facsimile of a handwritten signature, or a name typed at the bottom of an electronic message). Technologies currently in use include digital signatures within a PKI, biometric devices, PINs, user-defined or assigned passwords, scanned handwritten signatures, signature by means of a digital pen, and clickable OK or I accept boxes.171 Hybrid solutions based on the combination of different technologies are becoming increasingly popular, such as, for instance, in the case of the combined use of passwords and TLS/SSL (transport layer security/secure sockets layer), which is a technology using a mix of public
170
See Report of the Working Group on Electronic Commerce on the Work of Its Thirty-second Session (Vienna, 19-30 January 1998) (U.N. document A/CN.9/446), paras. 91 and ff. Guide to Enactment of the MLEC, supra note 105, para. 33.
189
171
and symmetric key encryptions. The features of the main techniques currently used are described below. 1. Functions of Electronic Authentication and Electronic Signature As is often the case, technology developed long before the law entered this area. The resulting gap between law and technology leads not only to varying levels of expert knowledge, but also inconsistent use of terminology. Expressions that were traditionally used with a particular connotation under national laws started to be used to describe electronic techniques whose functionality did not necessarily coincide with the functions or characteristics of the corresponding concept in legal usage. The usage in the information technology industry, which evolved essentially around concerns over network security, however, does not necessarily apply the same categories as legal writings. In some cases, the expression electronic authentication is used to refer to electronic techniques that, depending on the context in which they are used, may involve various elements, such as identification of individuals, confirmation of a persons authority (typically to act on behalf of another person or entity) or prerogatives (for example, membership in an institution, or subscription to a service) or assurance as to the integrity of information. In some cases, the focus is on identity only,172 but sometimes it extends to authority,173 or a combination of any or all of those elements.174
172
The Technology Administration of the United States Department of Commerce, for example, defines electronic authentication as the process of establishing confidence in user identities electronically presented to an information system (United States, Department of Commerce, Electronic Authentication Guideline: Recommendations of the National Institute of Standards and Technology, NIST Special Publication 800-63, version 1.0.2 (Gaithersburg, Maryland, April 2006)) <csrc.nist.gov/publications/nistpubs/800-63/ SP800-63V1_0_2.pdf> (4 April 2007). For example, the Government of Australia developed an electronic authentication framework that defines electronic authentication as the process of establishing a level of confidence in whether a statement is genuine or valid when conducting a transaction online or by phone. It helps build trust in an online transaction by giving the parties involved some assurance that their dealings are legitimate. These statements might include: identity details; professional qualifications; or the delegated authority to conduct transactions (Australia, Department of Finance and Administration, Australian Government e-Authentication Framework: An Overview (Commonwealth of Australia, 2005) <www.agimo.gov.au/infrastructure/ authentication/agaf_b/overview/introduction#e-authentication> (4 April 2007). The Principles for Electronic Authentication prepared by the Government of Canada, for instance, define authentication as a process that attests to the attributes of participants in an electronic communication or to the integrity of the communication. Attributes in turn are defined as information concerning the identity privilege or rights of a participant or other authenticated entity (Canada, Industry Canada, Principles for Electronic
173
174
190
Neither the MLEC nor the MLES use the term electronic authentication, in view of the different meaning of authentication in various legal systems and the possible confusion with particular procedures or form requirements. The MLEC uses instead the notion of original form to provide the criteria for the functional equivalence of authentic electronic information.175 The definition of electronic signature in UNCITRAL texts is deliberately broad, so as to encompass all existing or future electronic signature methods. As long as the methods used are as reliable as was appropriate for the purpose for which the electronic communication was generated or communicated, in the light of all the circumstances, including any relevant agreement,176 they should be regarded as meeting legal signature requirements. UNCITRAL texts relating to electronic commerce, as well as a large number of other legislative texts, are based on the principle of technological neutrality and therefore aim at accommodating all forms of electronic signature. Thus, UNCITRALs definition of electronic signature would cover the entire spectrum of electronic signature techniques, from higher-level security, such as cryptographically based signature assurance schemes associated with a PKI scheme (a common form of digital signature to lower levels of security, such as unencrypted codes or passwords. The simple typing of the authors name at the end of an e-mail message, which is the most common form of electronic signature, would, for instance, fulfil the function of correctly identifying the author of the message whenever it was not unreasonable to use such a low level of security. Apart from this, UNCITRAL has not otherwise dealt with issues related to access control or identity verification. This was also in keeping with the fact that, in a paper-based environment, signatures may be signs of identity but are necessarily attributive of identity. The MLEC deals, however, with the conditions under which the addressee of an electronic communication is entitled to assume that the message actually originated from its purported originator. Indeed, article 13 of the MLEC provides that as between the originator and the addressee, an electronic communication is deemed to be that of the originator if it was sent: (a) by a person who had the authority to act on behalf of the originator in respect of that data message; or (b) by an information system programmed by, or on
Authentication: a Canadian Framework (Ottawa, May 2004) <strategis.ic.gc.ca/epic/site/ ecic-ceac.nsf/en/h_gv00240e.html> (4 April 2007).
175
According to article 8 of MLEC, where the law requires information to be presented or retained in its original form, that requirement is met by a data message if: (a) There exists a reliable assurance as to the integrity of the information from the time when it was first generated in its final form, as a data message or otherwise; and (b) Where it is required that information be presented, that information is capable of being displayed to the person to whom it is to be presented. MLEC, article 7, subparagraph 1 (b).
191
176
behalf of, the originator to operate automatically. As between the originator and the addressee, an addressee is entitled to regard an electronic communication as being that of the originator, and to act on that assumption, if: (a) in order to ascertain whether the electronic communication was that of the originator, the addressee properly applied a procedure previously agreed to by the originator for that purpose; or (b) the electronic communication as received by the addressee resulted from the actions of a person whose relationship with the originator or with any agent of the originator enabled that person to gain access to a method used by the originator to identify electronic communications as its own. As a whole, these rules allow a party to infer someone elses identity, whether or not the message was electronically signed and whether or not the method used for attributing the message to the originator could be validly used for signature purposes. This conforms to current practice in the paper-based environment. Checking someone elses voice, physical appearance or identity papers (for example, a national passport) may suffice to conclude that the person is who he or she purports to be for the purpose of communicating with the person concerned, but would not qualify as a signature of such person under most legal systems. Besides the confusion that has been caused by the fact that technical and legal usage of terms in the paper-based and in the electronic environment do not coincide, the various techniques mentioned earlier can be used for different purposes and provide a different functionality, depending on the context. Passwords or codes, for example, may be used to sign an electronic document, but they may also be used to gain access to a network, a database or another electronic service, in much the same way as a key may be used to unlock a safe or open a door. However, while in the first instance the password is a proof of identity, in the second instance, it is a credential or sign of authority, which, while ordinarily linked to a particular person, is also capable of being transferred to another. In the case of digital signatures, the inappropriateness of the current terminology is even more patent. The digital signature is widely regarded as a particular technology for signing electronic documents. However, it is at least questionable whether, from a legal point of view, the application of asymmetric cryptography for authentication purposes should be referred to as a digital signature, as its functions go beyond the typical functions of a handwritten signature. The digital signature offers means both to verify the authenticity of electronic messages and guarantee the integrity of the contents.177 Furthermore, digital signature technology does not merely establish origin or integrity with respect to individuals as is
177
Babette Aalberts and Simone van der Hof, Digital Signature Blindness: Analysis of Legislative Approaches Toward Electronic Authentication, (November 1999), p. 8 <rechten. uvt.nl/simone/ Digsigbl.pdf> (3 December 2007).
192
required for signing purposes, but it can also authenticate, for instance, servers, websites, computer software, or any other data that is distributed or stored digitally, which gives digital signatures much broader use than an electronic alternative for handwritten signatures.178 2. Main Methods of Electronic Signature and Authentication For the purposes of this discussion, four main signature and authentication methods will be discussed: digital signatures, biometric methods, passwords and hybrid methods, and scanned or typed signatures. (a) Digital Signatures Relying on Public Key Cryptography Digital signature is a name for technological applications using asymmetric cryptography,179 also referred to as public key encryption systems, to ensure the authenticity of electronic messages and guarantee the integrity of the contents of these messages.180 Public key cryptography, is based on the use of algorithmic functions to generate two different but mathematically related keys181 which is why this technology is called asymmetric cryptosystem. One such key is used for creating a digital signature or transforming data into a seemingly unintelligible form, and the other key is used for verifying a digital signature or returning the message to its original form. Although the keys of the pair are mathematically related, it is virtually impossible to derive the private key from knowledge of the public key.182
178 179
Id. Cryptography is the branch of applied mathematics that is concerned with transforming information into a seemingly unintelligible form and then back into their original form. While the use of cryptography is one of the main features of digital signatures, the mere fact that a digital signature is used to authenticate a message containing information in digital form should not be confused with a more general use of cryptography for purposes of confidentiality. Confidentiality encryption is a method used for encoding an electronic communication so that only the originator and the addressee of the message will be able to read it. In a number of countries, the use of cryptography for confidentiality purposes is limited by law for reasons of public policy that may involve considerations of national defense. However, the use of cryptography for authentication purposes by producing a digital signature does not necessarily imply the use of cryptography to make any information confidential in the communication process, since the encrypted digital signature may be merely appended to a non-encrypted message. Typically large numbers produced using a series of mathematical formulae applied to prime numbers. Other mathematical techniques are currently used or under development, such as cryptosystems relying on elliptic curves, which may offer a high degree of security through the use of significantly reduced key-lengths. Certain existing standards refer to the notion of computational unfeasibility to describe the expected irreversibility of the process, that is, the hope that it will be impossible to
193
180
181
182
(i) Digital Signature Creation and Verification The private key, which should be kept secret, is used by the signatory to create the digital signature, while the public key, which is ordinarily more widely known, is used by a relying party to verify the digital signature. The private key is likely to be kept on a smart card or to be accessible through a personal identification number (PIN) or a biometric identification device, such as thumbprint recognition. To sign a document or any other item of information, the signatory first determines precisely the limits of the information that is to be signed. Then a hash function183 in the signatorys software establishes a hash result unique to the information to be signed. The signatorys software transforms the hash result of the electronic communication into a digital signature by encrypting it with the signatorys private key. The resulting digital signature is thus unique to both the information being signed and the private key used to create the digital signature. Typically, a digital signature is attached to the message and stored or transmitted with that message. However, it may also be sent or stored as a separate data element, as long as it maintains a reliable association with the corresponding message. Since a digital signature is unique to its message, it is inoperable if permanently disassociated from the message. Verification of a digital signature is accomplished by establishing a new hash result for the original message, by means of the same hash function used to
derive a users secret private key from that users public key. Computationally unfeasible is a relative concept based on the value of the data protected, the computing overhead required to protect it, the length of time it needs to be protected, and the cost and time required to attack the data, with such factors assessed both currently and in the light of future technological advance. (American Bar Association, Digital Signature Guidelines: Legal Infrastructure for Certification Authorities and Secure Electronic Commerce 9 (Chicago, American Bar Association, 1 August 1996), note 23 <www.abanet.org/scitech/ec/isc/ dsgfree.html> (5 April 2007).
183
A hash function is a mathematical process, based on an algorithm that creates a digital representation or compressed form of the message (often referred to as a message digest or fingerprint of the message), in the form of a hash value or hash result of a standard length that is usually much smaller than the message but nevertheless substantially unique to it. Any change to the message invariably produces a different hash result when the same hash function is used. In the case of a secure hash function, sometimes called a one-way hash function, it is virtually impossible to derive the original message from knowledge of its hash value. Another basic feature of hash functions is that it is also virtually impossible to find another binary object (i.e. different from the one from which the digest was originally derived) producing the same digest. Hash functions therefore enable the software for creating digital signatures to operate on smaller and more predictable amounts of data, while still providing robust evidentiary correlation to the original message content, thereby efficiently providing assurance that there has been no modification of the message since it was digitally signed.
194
create the digital signature. Using the public key and the new hash result, the verifier checks whether the digital signature was created using the corresponding private key and whether the newly established hash result matches the original hash result that was transformed into the digital signature during the signing process. The verification software will confirm the digital signature as verified from a cryptographic viewpoint if (a) the signatorys private key was used to sign digitally the message, which is known to be the case if the signatorys public key was used to verify the signature because the signatorys public key will verify only a digital signature created with the signatorys private key; and (b) the message was unaltered, which is known to be the case if the hash result computed by the verifier is identical to the hash result extracted from the digital signature during the verification process. (ii) Public Key Infrastructure and Certification Services Providers To verify a digital signature, the verifier must have access to the signatorys public key and have assurance that it corresponds to the signatorys private key. However, a public-key and private-key pair has no intrinsic association with any person; it is simply a pair of numbers. An additional mechanism is necessary to associate reliably a particular person or entity to the key pair. The required level of confidence may exist between parties who trust each other, who have dealt with each other over a period of time, who communicate on closed systems, who operate within a closed group or who are able to govern their dealings contractually, for example in a trading partner agreement. In a transaction involving only two parties, each party can simply communicate the public key of the key pair each party will use. However, the same level of confidence may not be present when the parties deal infrequently with each other, communicate over open systems such as the Internet, and are not bound by a prior agreement, such as a trading partner agreement. One solution to some of these problems is the use of one or more third parties to associate an identified signatory or the signatorys name with a specific public key. That third party is generally referred to as a certification authority or certification services provider in most technical standards and guidelines. To associate a key pair with a prospective signatory, a certification services provider (or certification authority) issues a certificate, which is an electronic record that lists a public key together with the name of the certificate subscriber as the subject of the certificate, and which may confirm that the prospective signatory identified in the certificate holds the corresponding private key. The principal function of a certificate is to bind a public key with a particular signatory. A recipient of the certificate desiring to rely upon a digital signature created by the signatory named in the certificate can use the public key listed in the certificate to verify that the digital signature was created with the corresponding private key.
195
To assure the authenticity of the certificate with respect to both its contents and its source, the certification services provider digitally signs it. The issuing certification services providers digital signature on the certificate can be verified by using the public key of the certification service provider listed in another certificate by another certification services provider (which may, but need not, be on a higher level in a hierarchy), and that other certificate can in turn be authenticated by the public key listed in yet another certificate, and so on, until the person relying on the digital signature is adequately assured of its genuineness. In a number of countries, certification services providers are organized as a public key infrastructure (PKI). Certification services providers within a PKI can be established in a hierarchical structure, where some certification services providers only certify other certification authorities, which provide services directly to users. In such a structure, some certification authorities are subordinate to other certification authorities. PKIs organized in a hierarchical structure are scalable in the sense that they may incorporate entire new PKI communities simply by having the root authority establish a trust relationship with the new communitys root.184 The root authority of the new community may be incorporated directly under the root of the receiving PKI, thus becoming a subordinate certification services provider within that PKI or be incorporated at a lower level. A drawback of hierarchical PKIs is that, as a consequence of reliance on a single trust point, if the root authority is compromised, the entire PKI is compromised. Furthermore, some countries have found it difficult to select one single entity as a root authority and to impose such a hierarchy on all other certification services providers.185 The so-called mesh PKI is an alternative to a hierarchical PKI. Under this model, certification services providers are connected in a peer-to-peer relationship. Certification services providers will issue certificates to each other; the pair of certificates describes their reciprocal trust relationship. The lack of hierarchy in such a system means that certification services providers cannot impose conditions governing the types of certificate issued by other certification services providers. If a certification services provider wishes to limit the trust extended to other certification services providers, it must specify these limitations in the
184
William T. Polk and Nelson E. Hastings, Bridge Certification Authorities: Connecting B2B Public Key Infrastructures, National Institute of Standards and Technology, September 2000, <csrc.nist.gov/groups/ST/crypto_apps_infra/documents/B2B-article.pdf > (3 December 2007). Id. Note that in the United States of America, it was very difficult to single out one agency of the Federal Government to assume the overall authority over the federal PKI.
185
196
certificates issued to its peers,186 which creates a risk of disharmony in conditions and limitations of mutual recognition. A third alternative structure is built around the so-called bridge certification services provider. Unlike a certification services provider in a mesh PKI, a bridge certification services provider does not issue certificates directly to users. Instead, the bridge certification services provider establishes peer-to-peer trust relationships with the different user communities, thus allowing the users to keep their natural trust points within their respective PKIs. The bridge of trust that joins two or more PKIs through their mutual relationship with a bridge certification services provider enables users from the different user communities to interact with each other through the bridge certification services provider with a specified level of trust.187 (b) Practical Problems in Public Key Infrastructure Implementation Setting up a PKI is a way to provide confidence that (a) a users public key has not been changed and in fact corresponds to that users private key; and (b) the cryptographic techniques being used are sound. To provide the confidence described above, a PKI may offer a number of services, including the following: (a) managing cryptographic keys used for digital signatures; (b) certifying that a public key corresponds to a private key; (c) providing keys to end users; (d) publishing revocation information of public keys or certificates; (e) managing personal tokens (e.g. smart cards) that can identify the user with unique personal identification information or can generate and store an individuals private keys; (f) checking the identification of end users and providing them with services; (g) providing time-stamping services; and (h) managing cryptographic keys used for confidentiality encryption where the use of such a technique is authorized.
186 187
Id. The bridge certification services provider was the structure eventually chosen to set up the PKI system for the United States Federal Government (Polk and Hastings, supra note 184). This was also the model followed to develop the PKI system of the Government of Japan.
197
Despite the considerable knowledge of digital signature technologies and the way they function, the implementation of public key infrastructures and digital signature schemes has, in practice, faced some problems that have kept the level of use of digital signatures below expectations. Another area where digital signatures and PKI schemes may give rise to practical problems concerns data security and privacy protection. Certification services providers must keep safe the keys used to sign certificates issued to their customers and may be exposed to attempts by outsiders to gain unauthorized access to the keys. Furthermore, certification services providers need to obtain a series of personal data and business information from persons applying for certificates. This information needs to be stored by the certification services provider for future reference. Certification services providers must take the necessary measures to ensure that access to such information is in accordance with applicable data protection laws.188 However, unauthorized access remains a real threat. 3. Biometrics A biometric is a measurement used to identify an individual through his or her intrinsic physical or behavioural traits. Traits that may be used for recognition in biometrics include DNA, fingerprint, iris, retina, hand or facial geometry, facial thermogram, ear shape, voice, body odour, blood vessel pattern, handwriting, gait and typing patterns. The use of biometric devices typically involves capturing a biometric sample of a biological feature of an individual in digital form. Biometric data is then extracted from that sample to create a reference template. Eventually, the biometric data stored in the reference template is compared with the one extracted from the end user for the purpose of verification, so that it is possible to indicate whether or not an identification or verification of identity has been achieved.189
188
See OECD Guidelines on the Protection of Privacy and Transborder Flows of Personal Data (Paris, 1980) <www.oecd.org/document/18/0,2340,en_2649_34255_1815186_1_1_ 1_1,00.html> (30 November 2007); Council of Europe Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data (E.T.S. 108) <conventions.coe.int/ Treaty/ en/Treaties/Html/108.htm> (30 November 2007); United Nations, Guidelines for the regulation of computerized personal data files (General Assembly resolution 45/95) <193.194.138.190/html/menu3/b/71.htm> (7 February 2007); and Directive 95/46/EC of the European Parliament and of the Council of 24 October 1995 On the Protection of Individuals With Regard to the Processing of Personal Data and on the Free Movement of Such Data (Official Journal of the European Communities, L 281, 23 November 1995) <eur-lex.europa.eu/ LexUriServ/LexUriServ.do?uri=CELEX:3 1995L0046:EN:HTML> (30 November 2007). International Association for Biometrics (iAfB) and International Computer Security Association (ICSA), 1999 Glossary of Biometric Terms < www.afb.org.uk/docs/glossary.
189
198
A number of risks relate to the storage of biometrical data since biometric patterns are typically not revocable. When biometric systems have been compromised, the legitimate user has no recourse but to revoke the identification data and switch to another set of uncompromised identification data. Therefore, special rules are needed to prevent unauthorized access or misuse of biometrics databases. The accuracy of biometric techniques cannot be absolute since biological features tend to be inherently variable and any measurement may involve deviation. In this respect, biometrics are not considered unique identifiers but rather semi-unique identifiers. To accommodate those variations, the accuracy of biometrics may be manipulated by setting the threshold for matching the reference template with the extracted sample. However, a low threshold may bias the test towards false acceptance while a high threshold may tend towards false rejections. Nevertheless, the accuracy of authentication provided by biometrics may be adequate in the majority of commercial applications. Moreover, data protection and human rights issues arise in relation to the storage and disclosure of biometrical data. Although data protection laws may not refer expressly to biometrics, they aim at protecting personal data relating to natural persons. It is precisely the processing of such data both in their raw form and as templates that is at the core of biometrics technology.190 Moreover, measures may be required to protect individuals against risks generated by the private use of biometric data, as well as in case of identity theft. Other legal domains, including labour and health law, may also come into play.191 Technical solutions might assist in addressing some concerns. For instance, storage of biometrical data on smart cards or tokens may protect from unauthorized access, which could occur if the data is stored on a centralized computer system. Moreover, best practices have been developed to reduce risks in different areas such as scope and capabilities; data protection; user control of personal data; and disclosure, auditing, accountability and oversight.192 Despite these problems, biometric devices are generally considered as offering a high level of security. While they are compatible with a range of uses, their
htm > (30 November 2007).
190
Paul de Hert, Biometrics: Legal Issues and Implications, background paper for the Institute for Prospective Technological Studies of the European Commission (European Communities, Directorate General Joint Research Centre, 2005), p. 13 <http://cybersecurity.jrc.es/ pages/ ProjectlibestudyBiometrics.htm > (30 November 2007). For instance, in Canada, the use of biometrics was discussed with respect to the application of the Personal Information Protection and Electronic Documents Act (2000, c. 5) in the workplace (see Turner v. TELUS Communications Inc., 2005 FC 1601, 29 November 2005 (Federal Court of Canada)). See, for an example of best practices, the International Biometric Group BioPrivacy Initia tive, Best Practices for Privacy-Sympathetic Biometric Deployment <www.bioprivacy.org>.
199
191
192
current main scope is on Government applications, particularly law enforcement applications such as immigration clearance and access controls. Commercial applications have also been developed, where often biometrics are used in the context of a two-factor authentication process requiring provision of an element in possession of the individual (biometrics) and an element in the knowledge of the individual (typically, a password or a PIN). Moreover, applications were developed to store and compare the characteristics of a persons handwritten signature. Digital-based pen tablets record the pen pressure and duration of the signing process. The data are then stored as an algorithm to be used for comparison against future signatures. 4. Passwords and Hybrid Methods Passwords and codes are used both for controlling access to information or services and for signing electronic communications. In practice, the latter use is less frequent than the former, because of the risk of compromising the code if it is transmitted in non-encrypted messages. Passwords and codes are however the most widely used method for authentication for purposes of access control and identity verification in a broad range of transactions, including most Internet banking, cash withdrawals at automatic teller machines and consumer credit card transactions. It should be recognized that multiple technologies can be combined to authenticate an electronic transaction. Several technologies or several uses of a single technology can be utilized for a single transaction. For example, signature dynamics for authentication can be combined with cryptography for message integrity. Alternatively, passwords can be passed over the Internet, using cryptography (e.g. SSL in browsers) to protect them, in conjunction with the use of biometrics to trigger a digital signature (asymmetric cryptography), which, on receipt, generates a Kerberos ticket (symmetric cryptography). 5. Scanned Signatures and Typed Names The main reason for legislative interest in electronic commerce in the private law area has been concern about how new technologies may affect the application of rules of law that were conceived for other media. This attention to technology has often led, deliberately or not, to a focus on sophisticated technologies that offer a higher level of security for electronic authentication and signature methods. It is often neglected, in that context, that a very large number, if not the majority, of business communications exchanged throughout the world do not make use of any particular authentication or signature technology. In day-to-day practice, companies around the world are often satisfied, for instance, with exchanging e-mails without the use of any form of authentication or signature other than the typed name, title and address of parties at the bottom of their communications. Sometimes a more formal appearance is given by the
200
use of facsimile or scanned images of handwritten signatures, which of course constitute only a copy in digitalized form of a handwritten original. Neither typed names on unencrypted e-mails nor scanned signatures offer a high level of security or can definitely prove the identity of the originator of the electronic communication in which they appear. Nevertheless, business entities freely choose to use these forms of authentication in the interest of ease, expediency and cost-effectiveness of communications.
B. Legal Treatment of Electronic Authentication and Signatures

A significant volume of electronic commercial transactions is performed in closed networks, that is, groups with a limited number of participants accessible only to previously authorized persons or companies. Closed networks support the operation of a single entity or an existing closed user group, such as financial institutions participating in the inter-bank payment system, securities and commodities exchanges, or an association of airlines and travel agents. In these cases, participation in the network is typically restricted to institutions and companies previously admitted to the group. Most of these networks have been in place for several decades, use sophisticated technology and have acquired a high level of expertise in the functioning of the system. The rapid growth of electronic commerce in the last decade has led to the development of other network models, such as supply chains or trade platforms. Although these new groups were originally structured around direct computer-to-computer connections, as were most of the closed networks already in existence at that time, there is an increasing trend towards using publicly accessible means, such as the Internet, as a common connection facility. Even under these more recent models, a closed network retains its exclusive character. Typically, closed networks operate under previously agreed contractual standards, agreements, procedures and rules known by various names such as system rules, operation rules, trading partner agreements or voluntary codes of conduct that are designed to provide and guarantee the necessary operational functionality, reliability and security for the members of the group.193 These
See, for example, Economic Commission for Europe, United Nations Centre for Trade Facilitation and Electronic Business, recommendation No. 32, entitled E-commerce selfregulatory instruments (codes of conduct) (ECE/TRADE/277) <www.unece.org/ cefact/recommendations/ rec_index.htm> (28 March 2007. Many initiatives at the national and international levels aim at developing model contracts. See, for example, Economic Commission for Europe, Working Party on the Facilitation of International Trade Procedures, recommendation No. 26, entitled The Commercial Use of Interchange Agreements for Electronic Data Interchange (TRADE/WP.4/R.1133/Rev.1); and United
201
193
rules and agreements often deal with matters such as recognition of the legal value of electronic communications, time and place of dispatch or receipt of electronic communications, security procedures for gaining access to the network and authentication or signature methods to be used by the parties.194 Within the limits of the contractual freedom under applicable law, such rules and agreements are usually self-enforcing. However in the absence of contractual rules, or to the extent that applicable law may limit their enforceability, the legal value of electronic authentication and signature methods used by the parties will be determined by the applicable rules of law, in the form of default or mandatory rules. The various options used in different jurisdictions to develop a legal framework for electronic signatures and authentication are discussed in the present chapter. In developing legal and policy frameworks to deal with these technologies, consideration should be given to the role of multiple technologies. Legal and policy frameworks for electronic authentication will need to be flexible enough to cover hybrid technology approaches, as those that focus on specific technologies could impede the use of multiple technologies.195 Technology neutral provisions would facilitate the acceptance of such hybrid technology approaches. Stringent requirements for electronic authentication and signature, in particular the imposition of a particular method or technology, may inadvertently cast doubt as to the validity and enforceability of a significant number of transactions that are entered into every day without the use of any particular kind of authentication or signature. That, in turn, may stimulate parties acting in bad faith to avoid the consequences of obligations they freely assumed by questioning the authenticity of their own electronic communications. It is unrealistic to expect that imposing a certain high level of authentication and signature requirements would eventually lead all parties to actually use them on a daily basis. Recent experience with sophisticated methods, such as digital signatures, has shown that concerns about cost and complexity often limit the practical use of authentication and signature techniques.
Nations Centre for Trade Facilitation and Electronic Business, recommendation No. 31, entitled Electronic Commerce Agreement (ECE/TRADE/257), both available at <www.unece. org/cefact/recommendations/rec_index.htm> (30 November 2007).
194
For a discussion of issues typically covered in trading partner agreements, see Amelia H. Boss, Electronic Data Interchange Agreements: Private Contracting Toward a Global Environment, 13 Nw. J. Intl L. & Bus. (1992) 45. Foundation for Information Policy Research, Signature Directive Consultation Compilation, 28 October 1998, which provides a compilation of responses made during consultations on the European Union draft directive on electronic signatures, prepared at the request of the European Commission <www.fipr.org/publications/sigdirecon.html> (30 November 2007).
195
202
Electronic authentication legislation and regulation has taken many different forms at the international and domestic levels. Three main approaches for dealing with signature and authentication technologies can be identified: (a) the minimalist approach; (b) the technology specific approach; and (c) the two-tiered or two-pronged approach.196 1. Minimalist Approach Some jurisdictions recognize all technologies for electronic signature, following a policy of technological neutrality.197 This approach is also called minimalist, since it gives a minimum legal status to all forms of electronic signature. Under the minimalist approach, electronic signatures are considered to be the functional equivalent of handwritten signatures, provided that the technology employed is intended to serve certain specified functions and in addition meets certain technology-neutral reliability requirements. The MLEC provides the most widely used set of legislative criteria for establishing a generic functional equivalence between electronic and handwritten signatures: the signature method must be able to identify the signatory and must offer a level of reliability commensurate with the purposes for which the electronic communication was generated.198 This provision contemplates the two main functions of handwritten signatures: to identify the signatory, and to indicate the signatorys intent with respect to the signed information. Any technology that can provide these two functions in electronic form should, according to the MLEC, be regarded as satisfying a legal signature requirement. The MLEC is therefore technologically neutral; that is, it does not depend on or presuppose the use of any particular type of technology and could be applied to the communication and storage of all types of information. This general principle has been incorporated into the laws of many countries. The principle of technological neutrality also permits accommodation of future technological developments. Furthermore, this approach gives prominence to the freedom of the parties to choose technology that is appropriate to their needs. The onus is then placed on the parties ability to determine the level of
196
Susanna F. Fischer, Saving Rosencrantz and Guildenstern in a virtual world? A Comparative Look at Recent Global Electronic Signature Legislation, 7 B.U. J. Sci. & Tech. L. (2001) 229, 234. For example, Australia and New Zealand. Article 7, paragraph 1, of the MLEC provides: (1) Where the law requires a signature of a person, that requirement is met in relation to a data message if: (a) a method is used to identify that person and to indicate that persons approval of the information contained in the data message; and(b) that method is as reliable as was appropriate for the purpose for which the data message was generated or communicated, in the light of all the circumstances, including any relevant agreement.
203
197 198
security that is adequate for their communications. This may avoid excessive technological complexity and its associated costs.199 Except in Europe, where legislation has been primarily influenced by directives issued by the European Union,200 most countries that have legislated in relation to electronic commerce have used the MLEC as their template. With very few exceptions,201 countries enacting the MLEC have preserved its technologically neutral approach and have neither prescribed nor favoured the use of any particular technology. Both the MLES, which was adopted in 2001, and the more recent ECC follow the same approach, although the MLES contains some additional language. When legislation adopts the minimalist approach, the issue of whether electronic signature equivalence has been proven normally falls to a judge, arbitrator or public authority to determine, generally by means of the so-called appropriate reliability test. A wide array of legal, technical and commercial factors may be taken into account in determining whether, under the circumstances, a particular authentication method offers an appropriate level of reliability. They include any of the following: (a) the sophistication of the equipment used by each of the parties; (b) the nature of their trade activity; (c) the frequency with which commercial transactions take place between the parties; (d) the nature and size of the transaction; (e) the function of signature requirements in a given statutory and regulatory environment; (f) the capability of communication systems; (g) compliance with authentication procedures set forth by intermediaries;
199
Stephen Mason, Electronic Signatures in Practice, 6 J. High Tech. L. (2006) 148, 153. In particular, Directive 1999/93/EC of the European Parliament and of the Council on a Community framework for electronic signatures (Official Journal of the European Communities No. L 13, of 19 January 2000, p. 12) (hereafter the Electronic Signatures Directive). This directive was followed by a more general one, Directive 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on certain legal aspects of information society services, in particular electronic commerce, in the Internal Market (Electronic Commerce Directive) (Official Journal of the European Communities, No. L 178, 17 July 2000, p. 1), dealing with various aspects of the provision of information technology services and some matters of electronic contracting. Colombia, Dominican Republic, Ecuador, India, Mauritius, Panama and South Africa.
200
201
204
(h) the range of authentication procedures made available by any intermediary; (i) compliance with trade customs and practice; (j) the existence of insurance coverage mechanisms against unauthorized messages; (k) the importance and the value of the information contained in the electronic communication; (l) the availability of alternative methods of identification and the cost of implementation; and (m) the degree of acceptance or non-acceptance of the method of identification in the relevant industry or field both at the time the method was agreed upon and the time when the electronic communication was communicated.202 2. Technology-Specific Approach Under a technology-specific approach, the law imposes the use of a specific technology to fulfil the legal requirements for the validity of an electronic signature. This is the case, for instance, where the law, aiming at a higher level of security, demands PKI-based applications.203 Since it prescribes the use of a specific technology, it is also called the prescriptive approach. The disadvantages of the technology-specific approach are that, in favouring specific types of electronic signature, it risks excluding other possibly superior technologies from entering and competing in the marketplace.204 Rather than
202 203
Guide to Enactment of the MLEC, supra note 105, para. 61. One of the earliest examples was the Utah Digital Signature Act, which was adopted in 1995, but was repealed effective 1 May 2006 by State Bill 20 <www.le.state.ut.us/~2006/ htmdoc/sbillhtm/sb0020.htm> (30 November 2007). The technology bias of the Utah Act can also be observed in a number of countries where the law only recognizes digital signatures created within a public key infrastructure (PKI) as a valid means of electronic authentication, which is the case, for example, under the laws of Argentina, Ley de firma digital (2001) and Decreto No. 2628/2002 (Reglamentacin de la Ley de firma digital); Estonia, Digital Signatures Act (2000); Germany, Digital Signature Act, enacted as article 3 of the Information and Communication Services Act of 13 June 1997; India, Information Technology Act 2000; Israel, Electronic Signature Law (2001); Japan, Law concerning Electronic Signatures and Certification Services (2001); Lithuania, Law on Electronic Signatures (2000); Malaysia, Digital Signature Act 1997; Poland, Act on Electronic Signature (2001); and Russian Federation, Law on Electronic Digital Signature (2002). Stewart Baker and Matthew Yeo, in collaboration with the secretariat of the International Telecommunication Union, Background and Issues Concerning Authentication and the ITU, briefing paper presented to the Experts Meeting on Electronic Signatures and Certification
205
204
facilitating the growth of electronic commerce and the use of electronic authentication techniques, such an approach may have an opposite effect. Technology specific legislation risks fixing requirements before a particular technology matures.205 The legislation may then either prevent later positive developments in the technology or become quickly outdated as a result of later developments. A further point is that not all applications may require a security level comparable with that provided by certain specified techniques, such as digital signatures. It may also happen that speed and ease of communication or other considerations may be more important for the parties than ensuring the integrity of electronic information through any particular process. Requiring the use of an overly secure means of authentication could result in wasted costs and efforts, which may hinder the diffusion of electronic commerce. Technology-specific legislation typically favours the use of digital signatures within a PKI scheme. The way in which PKIs are structured, in turn, varies from country to country according to the level of Government intervention. 3. Two-Tiered or Two-Pronged Approach In this approach, the legislation sets a low threshold of requirements for electronic authentication methods to receive a certain minimum legal status and assigns greater legal effect to certain electronic authentication methods (referred to variously as secure, advanced or enhanced electronic signatures, or qualified certificates). At the basic level, legislation adopting a two-tiered system generally grants electronic signatures functional-equivalence status with handwritten signatures, based on technologically neutral criteria. Higher-level signatures, to which certain rebuttable presumptions apply, must comply with specific requirements that may relate to a particular technology. Currently, legislation of this type usually defines such secure signatures in terms of PKI technology. This approach is typically chosen in jurisdictions that consider it important to address certain technological requirements in their legislation, but wish, at the same time, to leave room for technological developments. It can provide a balance between flexibility and certainty in relation to electronic signatures, by leaving it to the parties to decide, as a commercial judgement, whether the cost and inconvenience of using a more secure method is suitable to their needs. These texts also provide guidance as to the criteria for the recognition of electronic signatures in the context of a certification authority model. It is generally possible to combine the two-tiered approach with any type of certification model (whether
Authorities: Issues for Telecommunications, Geneva, 9 and 10 December 1999, Document No. 2 <www.itu.int/osg/spu/ni/esca/meetingdec9-101999/briefingpaper.html> (30 November 2007).
205
However, in view of the fact that PKI is today fairly mature and established, some of these concerns may no longer apply with the same force.
206
self-regulated, voluntary accreditation or a Government-led scheme), in much the same way as might be done under the technology-specific approach. Thus, while some rules may be flexible enough to accommodate different electronic signature certification models, some systems would only recognize licensed certification services providers as possible issuers of secure or qualified certificates. The first jurisdictions to have passed legislation adopting the two-tiered approach include Singapore206 and the European Union.207 They were followed by a number of others.208 The MLES allows an enacting State to set up a two-tiered system through regulations, even though it does not actively promote it.
206
Section 8 of the Electronic Transactions Act of Singapore admits any form of electronic signature, but only secure electronic signatures that meet the requirements of section 17 of the Act (i.e. those which are (a) unique to the person using it; (b) capable of identifying such person; (c) created in a manner or using a means under the sole control of the person using it; and (d) linked to the electronic record to which it relates in a manner that if the record was changed the electronic signature would be invalidated) enjoy the presumptions listed in section18 (inter alia, that the signature is of the person to whom it correlates and that the signature was affixed by that person with the intention of signing or approving the electronic record). Digital signatures supported by a trustworthy certificate that complies with the provisions of section 20 of the Act are automatically considered to be secure electronic signatures for the purposes of the Act. Like the Electronic Transactions Act of Singapore, the EU Directive on Electronic Signatures, distinguishes between an electronic signature (defined in art. 2, para. 1, as data in electronic form which are attached to or logically associated with other electronic data and which serve as a method of authentication) and an advanced electronic signature (defined in art. 2, para. 2, as an electronic signature that meets the following requirements: (a) it is uniquely linked to the signatory; (b) it is capable of identifying the signatory; (c) it is created using means that the signatory can maintain under his sole control; and (d) it is linked to the data to which it relates in such a manner that any subsequent change of the data is detectable). The Directive, in article 5, paragraph 2, mandates the States members of the European Union to ensure that an electronic signature is not denied legal effectiveness and admissibility as evidence in legal proceedings solely on the grounds that it is in electronic form, or not based upon a qualified certificate, or not based upon a qualified certificate issued by an accredited certification-service-provider, or not created by a secure signature-creation device. However only advanced electronic signatures which are based on a qualified certificate and which are created by a secure-signature-creation device are declared to (a)satisfy the legal requirements of a signature in relation to data in electronic form in the same manner as a handwritten signature satisfies those requirements in relation to paper-based data; and (b) are admissible as evidence in legal proceedings. (see art. 5, para. 1, of the Directive). For example, Mauritius and Pakistan.
207
207
208
PARt TwO ELEctRONIc COMMUNIcAtIONS IN CONtRAct FORMAtION AND PERFORMANcE
Chapter I. Contract Formation through Electronic Communications

Already at the time of the drafting of the UNCITRAL Model Law on Electronic Signatures (MLES), there were calls for another round of legislation, an international convention on electronic commerce, to achieve further harmonization of national laws. Underlying those proposals is the recognition that, despite the wide acceptance of the MLEC, it could not simply be assumed that its principles have already achieved universal application through domestic legislation. Another justification for further work in the form of an international convention was that only a binding instrument could effectively remove obstacles to electronic commerce that might derive, for example, from form requirements contained in other international conventions.209 Furthermore, the earlier UNCITRAL Model Law on Electronic Commerce (MLEC) does not address aspects of contract formation and performance that may be affected by the ways in which electronic transactions are currently structured and by the ways in which those structures are being changed to facilitate electronic commerce. For example, does the fact that the contract is formed by interaction between a person and an automated information system (electronic agent210) have any influence on the rules of contract formation? How should the law deal with mistakes and errors that may occur in the course of dealings between a person and an automated information system or between two automated information systems?
209
See Legal Aspects of Electronic Commerce, Proposal by France (United Nations document A/CN.9/WG.IV/WP.93, 1 March 2001). An electronic agent has been defined in section 2 (6) of the United States Uniform Electronic Transactions Act (UETA) as a computer program or an electronic or other automated means used independently to initiate an action or respond to electronic records or performances in whole or in part, without review or action by an individual. A similar definition is also used in section 19 of the Uniform Electronic Commerce Act of Canada (UECA).
210
208
Also, should there be special rules governing offers of goods and services over open networks, for example as regards to the availability and accessibility of contract terms? How do international conventions on trade-related matters whose field of application is typically conceived in strictly geographical terms apply to electronic commerce transactions in which national boundaries are increasingly fluid? Also, how are documentary and other formal requirements that might be provided in such conventions to be met? Needless to say, proposals to deal with these issues in a new international instrument were met with some scepticism from circles concerned with the dangers of excessive or rigid state-driven regulation who warned against the possible disadvantages of limiting party autonomy through an international binding instrument.211 The narrow scope of this paper does not allow for a discussion of the various arguments in favour and against state regulation of electronic commerce and Internet activities.212 It is sufficient to point out the complementary nature of legislation and contractual rules in private law matters, a fact that has traditionally been acknowledged by UNCITRAL through the recognition of party autonomy in nearly all its instruments. However, party autonomy has its limits: An offer posted on the Internet is open to everyone everywhere, and it is it is difficult to know another partys location, especially when the contract will be performed entirely electronically. Yet it can be difficult or nearly impossible to tailor ones online practices to conform to the laws of every state and nation in which one could conceivably be haled into
211
See, for example, the comments by an expert group set up by the International Chamber of Commerce to consider the desirability of an international uniform instrument on electronic contracting by UNCITRAL (U.N. document A/CN.9/WG.IV/WP.96, 11 December 2001). See Timothy D. Casey and Jeff Magenau, A Hybrid Model of Self-regulation and Governmental Regulation of Electronic Commerce, 19 Sta. Clara Comp. & High Tech. L. J. (2002) 1, 23. The authors point out that belief in the continued existence of self- regulation may simply be an unrealistic expectation in many areas of electronic commerce. Congress, which once chanted the mantra were not going to regulate the Internet, has changed its tune and proposed and implemented a whole slew of new invasive regulations. Additionally, doubts can be raised about the effectiveness of industry self-regulation. The case of Toysmart.com, which attempted, as part of its bankruptcy proceeding, to sell its customer lists in violation of its own privacy policy, suggests some of the ways that selfregulation can break down. A cause of particular dismay for advocates of self-regulation was the failure of third party certification agency TRUSTe to stop one of its licensees from violating their agreement. (Casey and Magenau, A Hybrid Model of Self-regulation , pp. 28-29.)
209
212
court. These difficulties impose serious legal transaction and litigation costs as well, especially for small online businesses.213 Following a series of studies and other preparatory work undertaken by its secretariat, in 2002 UNCITRAL started to consider two new topics in the area of electronic commerce: a possible international convention to deal with selected issues of electronic contracting,214 and proposals for removing legal obstacles to electronic commerce in existing international trade-related instruments.215 The UNCITRAL Working Group on Electronic Commerce devoted six sessions to the consideration of these topics, which soon were merged into one single project: a new international convention dealing with electronic communications in contract formation and performance.216 The final draft of the new convention was circulated to governments and international organizations for comments217 and, with some final amendments, was approved by UNCITRAL at its 38th session (Vienna, 4-15 July 2005)218 and officially adopted by the U.N. General Assembly with the title United Nations Convention on the Use of Electronic Communications in International Contracts (hereafter the ECC) on 23 November 2005.219 The ECC was opened for signature on 16January 2006
213
Christopher T. Poggi, Electronic Commerce Legislation An Analysis of European and American Approaches to Contract Formation, 41Va. J. Intl L. (2000) 224. See Legal Aspects of Electronic Commerce Electronic Contracting: Provisions for a Draft Convention, note by the Secretariat (U.N. document A/CN.9/WG.IV/WP.95, 20 September 2001). See Legal Barriers to the Development of Electronic Commerce in International Instruments Relating to International Trade, note by the Secretariat (U.N. document A/CN.9/WG.IV/ WP.94, 11 February 2002). See Reports of the Working Group on Electronic Commerce on the Work of Its Thirty-eighth session (New York, 12-23 March 2001), U.N. document A/CN.9/484; id., Thirty-ninth session (New York, 14-15 March 2002), U.N. document A/CN.9/509; id., Fortieth session (Vienna, 14-18 October 2002), U.N. document A/CN.9/527; and id., Forty-first session (New York, 5-9 May 2003), U.N. document A/CN.9/528; id. Forty-second session (Vienna, 17-21 November 2003), U.N. document A/CN.9/546; id., forty-third session (New York, 15-19 March 2004), U.N. document A/CN.9/548; and id., Forty-fourth session (Vienna, 11-22 October 2004), U.N. document A/CN.9/571 (all reports and other document relating to the work of UNCITRAL in the area of electronic commerce are available in the six official languages of the United Nations at <http://www.uncitral.org>). The comments received by UNCITRAL are reproduced in documents A/CN.9/578 and addenda 1-17, which were circulated at the 38th annual session of UNCITRAL, in 2005. The deliberations of UNCITRAL are reflected in the report on the work of that session (See Official Records of the General Assembly, Fiftieth Session, and Supplement No. 17 (U.N. document A/60/17, paras. 12 to 167). General Assembly resolution 60/21, of 23 November 2005.
214
215
216
217
218
219
210
and has already attracted significant interest from governments.220 The following discussion of legal issues related to contract formation and performance through electronic focuses on the issues dealt with in the EEC and draws further on the international experience gained with the application of the United Nations Convention on Contracts for the International Sale of Goods (U.N Sales Convention),221 one of the most successful international instruments in the area of uniform commercial law.222
A. Applicable Law in International Electronic Contracting

The operation of rules of private international law nearly invariably requires that persons or their acts be associated with a geographic location. According to some systems of private international law, a contract is governed by the law of the place where it was concluded. The place of contract formation is generally either the place where the contracting parties where actually located at the time of reaching an agreement, or the place of their residence or place of business, if the contract was by correspondence or telecommunication. Other systems subject a contract to the law of the country that has the closest connection with a contract or its performance, which is in most cases the law of the country where the party required to perform the act that characterizes the contract (in
220
As of 3 December 2006, the Convention has been signed by the Central African Republic, China, Colombia, Iran (Islamic Republic of), Lebanon, Madagascar, Montenegro, Panama, Paraguay, Philippines, Russian Federation, Saudi Arabia, Senegal, Sierra Leone, Singapore, and Sri Lanka. At a special event held in connection with the 39th annual session of UNCITRAL, Mexico, Spain and the United States had also joined some of those States in declaring their intention to obtain the necessary domestic approvals to sign the Convention in the near future (see China, Singapore, Sri Lanka Sign United Nations Convention on Use of Electronic Communications in International Contracts, Press Release available at <www.uncitral.org/uncitral/en/about/news.html#news> (3 December 2007). 1489 UN.T.S. 25567, p. 3, also available at <www.uncitral.org/english/texts/sales/CISG. htm>. As of 28 November 2007, the U.N. Sales Convention had been ratified by the following 70 States: Argentina, Australia, Austria, Belarus, Belgium, Bosnia and Herzegovina, Bulgaria, Burundi, Canada, Chile, China, Colombia, Croatia, Cuba, Cyprus, Czech Republic, Denmark, Ecuador, Egypt, El Salvador, Estonia, Finland, France, Gabon, Georgia, Germany, Greece, Guinea, Honduras, Hungary, Iceland, Iraq, Israel, Italy, Kyrgyzstan, Latvia, Lesotho, Liberia, Lithuania, Luxembourg, Mauritania, Mexico, Moldova, Mongolia, Montenegro, Netherlands, New Zealand, Norway, Paraguay, Peru, Poland, Republic of Korea, Romania, Russian Federation, Saint Vincent and the Grenadines, Serbia, Singapore, Slovakia, Slovenia, Spain, Sweden, Switzerland, Syrian Arab Republic, The former Yugoslav Republic of Macedonia, Uganda, Ukraine, United States of America, Uruguay, Uzbekistan and Zambia (the list of Contracting States to the U.N. Sales Convention is available at <www.uncitral. org/uncitral/en/uncitral_texts/sale_goods/1980CISG_status.html>).
211
221
222
case of a sales contract, for example, the delivery of the goods) has its residence or place of business. A similar geographic focus can also be found in most international uniform law rules. Most international commercial law conventions have their field of application circumscribed to international transactions. The solutions adopted at both the national and international levels for defining an international contract range from general criteria, such as the contract having significant connections with more than one State or relating to international commerce, to more specific factors, such as the fact that the parties have their places of business or habitual residence in different countries.223 If a party has more than one place of business, those instruments refer to the place that has the closest relationship to the contract and its performance.224 Transposing traditional notions of conflicts of laws to a purely electronic environment may pose a number of problems,225 however, as discussed below. 1. Elements to Determine the Applicable Law When the parties to a contract concluded electronically clearly indicate where their relevant place of business is located, that indication is to be taken into account as an important criterion, if not the most important one, in determining the international character of a contract.226 However, this rule is of little help if no such indication has been made, since the location of the parties may not be evident from the electronic communications they exchange. The question then arises as to whether the technology used for electronic commerce requires an adjustment of concepts or suggests the need for taking additional factors into account for determining the parties location. This question
223
For example, U.N. Sales Convention, article 1, paragraph 1; Convention on the Limitation Period in the International Sale of Goods (U.N. Limitation Convention) (1511 U.N.T.S. 26119, p. 1), article 2, subparagraph (a); and article 1, subparagraph (a), of the United Nations Convention on Independent Guarantees and Stand-by Letters of Credit (U.N. Guarantees and Stand-by Convention) (Official Records of the General Assembly, Fiftieth Session, Supplement No. 17 and corrigendum (A/50/640 and Corr.1, annex). See also UNIDROIT Convention on International Financial Leasing, article 3, subparagraph 1 (a) (<www.unidroit.org/english/conventions/c-leas.htm>) and UNIDROIT Convention on International Factoring, article 2, subparagraph 1 (a) (<www.unidroit.org/english/ conventions/c-fact.htm>). E.g. U.N. Sales Convention, article 10 (a); U.N. Limitation Convention, article 2 (c); U.N. Guarantees and Stand-by Convention, article 4, paragraph 2 (a); UNIDROIT Convention on International Financial Leasing, article 3, paragraph 2; and UNIDROIT Convention on International Factoring, article 2, paragraph 2. See, generally, Sylvette Guillemard, Le Droit International Priv Face au Contrat de Vente Cyberspatial (Montral: ditions Yvon Blais, 2006). For instance, under the U.N. Limitation Convention or the U.N. Sales Convention.
224
225
226
212
was extensively debated within UNCITRAL during the negotiation of the ECC. Even if transmission protocols of electronic communications do not usually indicate where the parties are located, they often include a number of other types of apparently objective information, such as Internet Protocol (IP) addresses,227 domain names228 or information pertaining to intermediary information systems. The question then is what value, if any, could be attached to such information for the purpose of determining the physical location of the parties. (a) Location of Information Systems The location of the equipment and of its supporting technology are not adequate factors for determining the location of the parties, since they do not provide sufficient indication as to the ultimate parties to the contract. They may also change over time and may not be known or apparent to the parties during their communications. The management and operation of an information system may be entirely outsourced or run by a third party. For example, a contract made on behalf of the seller may be automatically concluded with the buyer by the computer of the Internet service provider (ISP) that hosts the sellers web site. Reliance on the location of equipment may thus lead to the undesirable result of linking a contract to a geographical location that, although related to the path followed by the electronic messages exchanged by the parties, bears perhaps little or no relationship to their actual location.229 Another undesirable result would be that a persons place of business, when negotiating a contract electronically, may end up being different from the same persons place of business when negotiating through other means.230
227
The Internet Protocol (IP) address is a 32-bit number (128 according to IP version 6) that identifies each sender or receiver of information that is sent in packets across the Internet. A domain name is a name assigned to a numerical IP functioning as part of a uniform resource locator (URL). The need to retain the same definitions that are used for off-line transactions is also mentioned in the EU Electronic Commerce Directive, supra note 200, where it is stated that: The place at which a service provider is established should be determined in conformity with the case-law of the Court of Justice according to which the concept of establishment involves the actual pursuit of an economic activity through a fixed establishment for an indefinite period; the place of establishment of a company providing services via an Internet website is not the place at which the technology supporting its website is located or the place at which its website is accessible but the place where it pursues its economic activity. The risks of establishing a dual regime for business, depending on the media being used, has been one of the main concerns expressed by the International Chamber of Commerce in connection with UNCITRALs current work on electronic contracting (see U.N. documents A/CN.9/WG.IV/WP.96 and A/CN.9/WG.IV/WP.101).
213
228
229
230
Nevertheless, it is conceivable that electronic commerce and the new economy may involve activities that are entirely or predominantly carried out through the use of information systems, without a fixed establishment231 or without any connection to a physical location other than, for instance, the registration of its articles of incorporation at a given registry. It has been argued that it would not be reasonable to apply to those so-called virtual companies the same criteria traditionally used to determine a persons place of business. In other words, the question is whether it is appropriate to give legal significance to the location of the equipment and technology supporting the information system or the places from which such a system may be accessed in order to establish where such a virtual company has its place of business. The issue of location of entities offering goods and services through electronic means was considered by the Organisation for Economic Cooperation and Development (OECD) in the context of its work on international aspects of taxation. On 22 December 2000, the OECD Committee on Fiscal Affairs adopted changes to the commentary on article 5 of the Model Tax Convention on Income and on Capital (the OECD Model Tax Convention) to deal with the issue of the application of the definition of permanent establishment, as understood in the context of the OECD Model Tax Convention, in connection with electronic commerce.232 The OECD Committee on Fiscal Affairs points out that, while a location where automated equipment is operated by an enterprise may constitute a permanent establishment in the country where it is situated, a distinction needs to be made between computer equipment, which may be set up at a location so as to constitute a permanent establishment under certain circumstances, and the data and software which is used by, or stored on, that equipment. According to that interpretation, an Internet web site, which is a combination of software and electronic data, does not in itself constitute tangible property and therefore does not have a location that can constitute a place of business as there is no facility such as premises or, in certain instances, machinery or equipment as far as the software and data constituting that web site is concerned. On the other hand, the OECD Committee on Fiscal Affairs points out that a server on which the web site is stored and through which it is accessible is a piece of equipment
231
Thibault Verbiest and Maxime Le Borne, Le Fonds de Commerce Virtuel: Une Ralit Juridique? <www.droit-technologie.org>, 24 May 2002 (30 November 2007). Organisation for Economic Cooperation and Development, Clarification on the Application of the Permanent Establishment Definition in E-Commerce: Changes to the Commentary on the Model Tax Convention on Article 5 (<www.oecd.org/dataoecd/46/32/1923380.pdf> (30 November 2007).
232
214
having a physical location and such location may thus constitute a fixed place of business of the enterprise that operates that server.233 OECD justifies the distinction between a web site and the server on which the web site is stored and used on the following grounds: the enterprise that operates the server may be different from the enterprise that carries on business through the web site. For example, it is common for the web site through which an enterprise carries on its business to be hosted on the server of an Internet Service Provider (ISP). Although the fees paid to the ISP under such arrangements may be based on the amount of disk space used to store the software and data required by the web site, these contracts typically do not result in the server and its location being at the disposal of the enterprise , even if the enterprise has been able to determine that its web site should be hosted on a particular server at a particular location. In such a case, the enterprise does not even have a physical presence at that location since the web site is not tangible. In these cases, the enterprise cannot be considered to have acquired a place of business by virtue of that hosting arrangement. However, if the enterprise carrying on business through a web site has the server at its own disposal, for example it owns (or leases) and operates the server on which the web site is stored and used, the place where that server is located could constitute a permanent establishment of the enterprise if the other requirements of [article 5 of the OECD Model Tax Convention]234 are met.
233 234
Id., para. 42.2. Article 5, paragraph 1 (Permanent establishment) of the OECD Model Tax Convention define permanent establishment as a fixed place of business through which the business of an enterprise is wholly or partly carried on. The term permanent establishment includes especially: a) a place of management; b) a branch; c) an office; d) a factory; e) a workshop, and f) mine, an oil or gas well, a quarry or any other place of extraction of natural resources (paragraph 2). However, it does not include: a) the use of facilities solely for the purpose of storage, display or delivery of goods or merchandise belonging to the enterprise; b) the maintenance of a stock of goods or merchandise belonging to the enterprise solely or the purpose of storage, display or delivery; c) the maintenance of a stock of goods or merchandise belonging to the enterprise solely for the purpose of processing by another enterprise; d) the maintenance of a fixed place of business solely for the purpose of purchasing goods or merchandise or of collecting information, for the enterprise; e) the maintenance of a fixed place of business solely for the purpose of carrying on, for the enterprise, any other activity of a preparatory or auxiliary character; f) the maintenance of a fixed place of business solely for any combination of activities mentioned in subparagraphs a) to e), provided that the overall activity of the fixed place of business resulting from this combination is of a preparatory or auxiliary character (paragraph 4).
215
For the purpose of distinguishing between a web site and the server on which it is stored, the OECD Committee on Fiscal Affairs stresses the importance of identifying the place of performance of the core functions of a business entity, as opposed to ancillary activities (e.g. provision of a communications link between suppliers and customers, advertising of goods or services, relaying information through a mirror server for security and efficiency purposes, gathering market data for the enterprise or supplying information). In that connection, OECD provides the following clarification: What constitutes core functions for a particular enterprise clearly depends on the nature of the business carried on by that enterprise. For instance, some ISPs are in the business of operating their own servers for the purpose of hosting web sites or other applications for other enterprises. For these ISPs, the operation of their servers in order to provide services to customers is an essential part of their commercial activity and cannot be considered preparatory or auxiliary. A different example is that of an enterprise (sometimes referred to as an e-tailer) that carries on the business of selling products through the Internet. In that case, the enterprise is not in the business of operating servers and the mere fact that it may do so at a given location is not enough to conclude that activities performed at that location are more than preparatory and auxiliary. What needs to be done in such a case is to examine the nature of the activities performed at that location in light of the business carried on by the enterprise. If these activities are merely preparatory or auxiliary to the business of selling products on the Internet (for example, the location is used to operate a server that hosts a web site which, as is often the case, is used exclusively for advertising, displaying a catalogue of products or providing information to potential customers), the location will not constitute a permanent establishment. If, however, the typical functions related to a sale are performed at that location (for example, the conclusion of the contract with the customer, the processing of the payment and the delivery of the products are performed automatically through the equipment located there), these activities cannot be considered to be merely preparatory or auxiliary. The above clarification shows the narrow conditions under which a server may be regarded as a permanent establishment for taxation purposes. While the term place of business, as generally defined in private law does not necessarily coincide with the notion of establishment under domestic and international tax law, the clarification provided by the OECD Committee on Fiscal Affairs offers elements that are also useful in a private law context. They confirm the perception
216
that by and large the location or information systems may be of limited use for determining where a company has its place of business.235 (b) Domain Names and Electronic Addresses Another related question is the extent to which the address from which the electronic messages are sent could be taken into account to determine a partys location. Thus, in the case of addresses linked to domain names connected to specific countries (such as addresses ending with .at for Austria, .nz for New Zealand, etc.) the party could be presumed to have its place of business in the corresponding country. In some countries, the assignment of domain names is only made after verification of the accuracy of the information provided by the applicant, including its location in the country to which the relevant domain name related. For those countries, it may be appropriate to rely, at least in part, on domain names for ascertaining a partys location. However, in countries where no such verification takes place, an electronic mail (e-mail) address or a domain name cannot automatically be regarded as the functional equivalent of the physical location of a partys place of business.236 Moreover, in certain branches of business it is common for companies to offer goods or services through various regional web sites bearing domain names linked to countries where such companies do not have a place of business in the traditional sense of the term. Furthermore, goods being ordered from any such web site may be delivered from warehouses maintained for the purpose of supplying a particular region, which may be physically located in a country other than those linked to the domain names involved. An additional level of uncertainty is created by the use of generic
235
Il serait erron de penser que le lieu o se trouve lordinateur o sont stocks les informations diffuss sur Internet puisse tre pris en considration aux fins de la dtermination de la rsidence, du domicile, de ltablissement ou des activits dun oprateur. En effet, dune part, les informations digitales ne sont pas des res dont on peut vrifier lemplacement physique, dautre part, les donnes peuvent tre facilement dplaces dun ordinateur un autre et lutilisateur ignore dans lequel dentre eux elles sont stockes. (Draetta, Internet et commerce lectronique, supra note 14, at 223). According to the Internet Corporation for Assigned Names and Numbers (ICANN), the assignment of top-level domain (TLDs) names including a country code (ccTLDs) is delegated to designated managers, who operate the ccTLDs according to local policies that are adapted to best meet the economic, cultural, linguistic, and legal circumstances of the country or territory involved (<www.icann.org/tlds/>). Needless to say, each country develops its own detailed rules for assigning domain names within its jurisdiction. The Swedish domain name registration system, for instance, seems to require proof of a companys claim to the domain name and its link to the country, whereas more liberal systems, such as that of Germany, only require the existence of a contact person in the country. See Frederik Roos, First Come, Not Served: Domain Name Regulation in Sweden, 17 Intl Rev. L. Comp. & Tech. (2003) 63, 70.
217
236
top-level domains237 such as .com or .net. These types of domain name and e-mail address do not show any link to a particular country, which is not surprising since the system of assigning domain names for Internet sites was not conceived in strictly geographical terms.238 Domain names and electronic addresses are, therefore, not capable of being reliably used as sufficient criteria for determining the location of contract parties in isolation of other evidence.239 (c) The Continued Usefulness of the Traditional Notion of Place of Business The above discussion has shown that peripheral information related to electronic messages, such as an IP address, domain names or the geographical location of information systems, have limited value for determining the physical location of the parties. Accordingly, the default rules on place of receipt and dispatch of an electronic communications in article 15 of the MLEC and article 10 of the ECC disregard elements such as domain names or location of information systems and refer back to the notion of place of business. The basic reason for these rules is that a party should not be regarded as being located in two different places depending on the media used for negotiating various contracts. Nevertheless, one of the central concerns of UNCITRAL has been the need for enhancing legal certainty and predictability for electronic commerce. UNCITRAL felt that it would be desirable to formulate uniform international provisions offering elements that allow the parties to ascertain beforehand the location of their counterparts.240 The ECC contains a set of rules for that purpose. They do not contemplate a duty for the parties to disclose their places of business, but establish a certain number of presumptions and default rules aimed at facilitating a determination of a partys location. The ECC attributes primary albeit not absolute importance to a partys indication of its relevant place of business. The ECC takes a cautious approach to peripheral information related to electronic communications, such as IP addresses, domain names, or the geographic
237
Generic TLDs are registered directly through ICANN-accredited registrars (for further information on the system, see <www.iana.org/cctld/cctld.htm>). seuls certains sites Internet sont rattachables un certain Etat, grce au suffixe utilis. Dautres utilisent des suffixes neutres du point de vue de la localisation, tels que com ou net. En second lieu, ladresse Internet est logique, non pas physique, et na rien voir aves lemplacement correspondant dune partie. (Draetta, Internet et commerce lectronique, supra note 14 at 222-223). It remains to be seen whether other technologies, such as geo-identification, if widely used, may offer a technical device for establishing location online (see Dan Svantesson, Geo-identification and the Internet A New Challenge for Australias Internet Regulation, 14 Murdoch U. eLaw J. (2007) 155. See U.N. document A/CN.9/484, para.103.
238
239
240
218
location of information systems that, despite their apparent objectivity, have little, if any, conclusive value for determining the physical location of the parties. Depending on the way that information systems are set up or how domain names are assigned, they may lack any reliable connection to a persons physical location.241 Moreover, UNCITRAL was mindful of the need to avoid situations where a persons place of business, when negotiating a contract electronically, might end up being different from the same persons place of business when negotiating through other means.242 (d) A Duty to Disclose the Place of Business? Some international instruments require the parties to electronic transactions to clearly indicate the location of their relevant places of business. However, during the negotiation of the ECC that proposition raised a number of questions, such as the extent to which such a duty, which does not exist for international paper-based transactions, might result in a duality of legal regimes.243 Another concern is what kind of legal consequence, if any, should be attached to the lack or inaccuracy of such information and how an international uniform instrument on electronic contracting could deal with that issue without unduly interfering with the underlying contract law.244 Accordingly, article 6 of the ECC merely creates a presumption in favour of a partys indication of its place of business, which is accompanied by conditions under which that indication can be rebutted and by default provisions that apply if no indication has been made. The article is not intended to allow parties to invent fictional places of business that do not meet the definition of place of business of its article 4, subparagraph (h).245 This presumption, therefore, is not absolute, and it was understood that the ECC does not uphold an indication
241
For an overview of this discussion, see Jose Angelo Estrella Faria, Harmonizing the Law of International Electronic Contracting: Adjust the Rules but Dont Rewrite Them, in Andrea Schulz (ed.), Legal Aspects of an E-Commerce Transaction, International Conference in The Hague, 26 and 27 October 2004 (Sellier European Law Publishers GmbH, 2006), 73, 77-79. The risks of establishing a dual regime for business, depending on the media being used, has been one of the main concerns expressed by the International Chamber of Commerce in connection with UNCITRALs current work on electronic contracting (see U.N. document A/CN.9/WG.IV/WP.96; see also the Note by the Secretariat of 25 February 2003 transmitting subsequent comments on UNCITRALs work by a task force established by the International Chamber of Commerce (U.N. document A/CN.9/WG.IV/WP.101). See U.N. document A/CN.9/509, para. 63. Id., paras. 4450 and 62-65; see also U.N. document A/CN.9/528, paras. 83-91. Article 4, subparagraph (h) defines place of business as any place where a party maintains a non-transitory establishment to pursue an economic activity other than the temporary provision of goods or services out of a specific location.
219
242
243 244
245
of a place of business by a party even where such an indication is inaccurate or intentionally false. 2. Substantive Applicable Law: Sales Law and Virtual Goods One particularly difficult question raised by electronic commerce concerns the appropriate legal treatment for contracts involving products that can be electronically delivered and how to treat such products for commercial law purposes. Software and other virtual goods have been identified as a good example of a situation where the intersection between goods and services has become increasingly blurred. In the past, these types of products crossed physical borders embodied on a tangible medium, and while new means of delivery have developed, the underlying characteristic of the product has not. Today, these products can flow across global networks and can be permanently retained on someones computer, and still retain the underlying function as if they were sent in physical form. (a) Computer Software under the UN Sales Convention Determining what substantive body of law applies to these transactions is particularly important because only international contracts for the sale of goods are subject to a truly universal uniform legal regime under the U.N. Sales Convention. However, it is generally accepted that the U.N. Sales Convention embodies a rather conservative concept of goods. According to most commentators, intangible rights, such as patent rights, trademarks, copyrights, a share of a limited liability company, as well as know-how, are not to be considered goods under the Convention.246 Nevertheless, considerable argument exists concerning the appropriate classification of software and other computer-related information. For a number of years now, courts in Germany have regarded software as goods247 and have applied German sales law to transactions involving standard software.248 The decisive test has been whether the buyer had the right to keep and use the software without limitation after having paid the price, irrespective of whether or not the software is incorporated in a tangible medium.249 Legal writings have followed suit and considered sales law at least mutatis mutandis
246
Franco Ferrari, 1 Vendita Internazionale di Beni Mobili, 52-53, in Scialoja/Branca (eds.) 1 Commentario del Codice Civile (Bologna, Zanichelli,1994). Bundesgerichtshof (BGH) 4 November 1987, BGHZ 102, p. 135 (at 144). Bundesgerichtshof (BGH) 22 December 1999, Neue Juristische Wochenschrift (2000) 1415. Id., pp. 99-100.
247
248
249
220
applicable to the purchase of standard software.250 Some authors expressly affirm the equivalence between software and goods (Ware),251 although some scholars regard the incorporation into a tangible medium (such as a disk) as an essential element for this analogy.252 Thus, it is not surprising that German courts have also taken the view that the sale of software can be covered by the U.N. Sales Convention as well. In obiter dictum, a German court of appeal stated that the sale of standard software could be considered a sale of goods, at least where the software was not custom-made.253 A German court of first instance reached the same result in a later case.254 Most German255 and Austrian256 commentators have also affirmed the applicability of the U.N. Sales Convention to software sales, at least as long as they are embodied on a tangible medium.257
For an overview of the discussion in Germany, see Abbo Junker and Martina Benecke, Computerrecht (Baden-Baden, Nomos Verlaggesellschaft, 3rd edn., 2003) pp. 120-124, Nos. 157-165 (who, however, underscore the difference between the intellectual property (software) and the tangible medium, sole element capable of being regarded as good and stress the applicability of sales law by analogy only (p. 122, No. 159). See also Heinrich Drner and Dirk Ehlers, Rechtsprobleme der EDV (Neuwied-Frankfurt, Metzner, 1989) 13 et seq.; Friederich Graf von Westphalen and Ulrich Seidel, Aktuelle Rechtsfragen der Software-Vertrags- und Rechtspraxis (Kln: RWS Verlag Kommunikationsforum, 3rd edn., 1992), p. 12. M. Michael Knig, Software (Computerprogramme) als Sache und deren Erwerb als Sachkauf , Neue Juristische Wochenschrift (1993) 3121-3124; Rigo Wenning, Die Behandlung von Standardsoftware im Wiener bereinkommen ber den Warenkauf von 1980 <archiv.jura.uni-saarland.de/projekte/online/evertrag.html> (30 November 2007), at II. 3. von Westphalen/Seidel, Aktuelle Rechtsfragen, supra note 250, 12. Oberlandesgericht (OLG) Koblenz, 17 September 1993, Recht der internationalen Wirtschaft 1993, p. 934 (English abstract published in Case Law on UNCITRAL Texts (CLOUT), case No. 281 (abstracts of this and all other cases compiled through the CLOUT system are published under at <www.uncitral.org> under Case Law-CLOUT). Landgericht (LG) Mnchen, 8 Feb. 1995, Case No. 8 HKO 24667/93 = CLOUT case No. 131. Burghard Piltz, Internationales Kaufrecht Das UN-Kaufrecht (Wiener bereinkommen von 1980), in Praxisorientierter Darstellung (Mnchen: C.H. Beck, 1993), p. 30 No. 48; Michael Bothe and Wolfgang Kilian, Rechtsfragen Grenzberschreitender Datenflsse (Kln: Dr. Otto Schmidt, 1992) 370; Ernst von Caemmerer and Peter Schlechtriem, Kommentar zum Einheitlichen UN-Kaufrecht (Mnchen: C. H. Beck, 2nd edn., 1995), Art. 1 No. 21; Ulrich Magnus, Wiener UN-Kaufrecht (CISG), Art 1 CISG No. 44, in Julius von Staudinger (ed.), II Kommentar zum Brgerlichen Gesetzbuch mit Einfhrungsgesetz und Nebengesetzen, (Berlin: Sellier, de Gruyter, 12th edn., 1994). Martin Karollus, UN-Kaufrecht, Eine systematische Darstellung fr Studium und Praxis (Wien-New York: Springer Verlag, 1991) 21. G. Beate Czerwenka, Rechtsanwendungsprobleme im Internationalen Kaufrecht: das Kollisionsrecht bei Grenzberschreitenden Kaufvertrgen und der Anwendungsbereich der Internationalen
221
250
251
252 253
254
255
256
257
The applicability of the U.N. Sales Convention to software sales, at least when embodied in a tangible medium, and for the same reasons that have influenced the debate in Germany, has also been affirmed in other countries with a similar tradition of private law, such as Italy.258 The same conclusion has also been proposed in Mexico, where, however, software transmitted electronically has been said to fall outside the scope of the Convention, since it can be assimilated to electricity, which is expressly excluded from the Convention.259 A slightly different approach seems to be taken by courts and scholars in France. The applicability of the U.N. Sales Convention to software transactions does not seem to have been put to test before French courts, and the position of French commentators is not clear. For purely domestic contracts, French courts seem to pay more attention to the parties intention in each concrete case rather than to abstract classifications. The result, however, accommodates both those who categorically reject260 the notion of software as goods to those who are willing to accept it, at least in respect of standard software incorporated on a tangible medium.261 As regards the U.N. Sales Convention, there is sufficient agreement concerning the exclusion of intangible goods (biens immatriels) from its scope of application.262 Even though the French text of the U.N. Sales
Kaufrechtsbereinkommen (Berlin: Duncker & Humblot, 1988) 148; Peter Schlechtriem, provides additional information the original understanding of the term goods in the U.N. Sales Convention: There are certainly good reasons to enlarge the sphere of applica tion of the Convention by understanding the concept of goods liberally not literally, but as far as I remember from the Vienna Conference, there was a strong conviction among many delegations that the sale and transfer problems of intellectual property and the like were not within the mandate of the Conference. This alone, of course, cannot answer the question whether computer software can be regarded as movables. It is a problem much dealt with in German literature not only in regard to the application of sales law provisions but also in regard to product liability. If the contract concerns so-called standard software, i.e. a program not designed especially to meet a specific customers demands, and if this program is recorded on a disk or tape, one could argue that the object of the sale falls under the Convention since it is movable and therefore goods (Uniform Sales Law The Experience with Uniform Sales Law in the Federal Republic of Germany, Juridisk Tidskrift (1991/92) 1, 8.
258 259
See, for example, Franco Ferrari, Vendita internazionale, supra note 246, 52-53. Diego Ricardo Galn Barrera, El mbito de aplicacin en la Convencin de las Naciones Unidas sobre los Contratos de Compraventa de Mercaderas, Part 2 <cisg.tij.uia.mx/ambito03.html> (30 November 2007). See the critical analysis of French case law by Michel Vivant et al., Droit de linformatique et des raiseaux (Paris: Lamy, 1999), Nos. 813- 833 and Nos. 969-994. For instance, Jrme Huet and Herbert Maisl, Droit de linformatique et des tlcommunications (Paris: Litec, 1989), No. 461. Vincent Heuz, La vente internationale de marchandises (Paris: L.G.D.J., 2000), No. 5, at pp. 5-6.
260
261
262
222
Convention uses the plain term marchandises, rather than the technical, and somewhat narrower, expression biens mobiliers corporels, which was used in the French text of the Conventions predecessor instrument, the Hague Uniform Sales Law,263 it is generally accepted that the U.N. Sales Convention applies to tangible movable goods. Nevertheless, there are also advocates of a broader interpretation of the notion of goods under the U.N. Sales Convention so as to encompass at least standard software.264 Although the United Kingdom is not a Contracting Party to the U.N. Sales Convention, the position taken by British courts on the matter may be indicative of the way in which other common law jurisdictions, including countries that have ratified the U.N. Sales Convention, might approach the problem. St Albans City and District Council v. International Computer Ltd. was the first case to address the issue of electronic software.265 In that case, an employee of the software vendor went to the premises of the other party and installed software on a computer system. The court considered whether electronic software constituted a good under the Sale of Goods Act of 1979. Focusing on the definition of a good, which is all personal chattels other than things in action and money,266 the court concluded that software without a disk was not a good. Since the employee of the software vendor personally installed the software rather than delivering it on a disk, the court concluded that the software contract is not the sale of a good. As could be expected in the country with the worlds largest software industry and the largest domestic market for electronic commerce, the discussion over the classification of virtual goods in the United States has been extensive, although not entirely settled. As was the case in other countries, the focus of the discussion in the United States has been on definition of software for purposes of domestic law. At the outset, the question was raised in connection with the application of tax law, when IBM announced in 1969 that it was separating the pricing of its software and services from the pricing of its hardware.267 For the taxing authorities
263
Convention portant loi uniforme sur la vente internationale des objets mobiliers corporels (Convention Relating to a Uniform Law on the International Sale of Goods (The Hague, 1 July 1964) <www.unidroit.org/french/conventions/c-ulis.htm>. For instance, Jean Thieffry and Chantal Granier, La vente internationale (Paris: Centre franais du commerce extrieur, 2nd edn., 1992) 39; and Vicent Heuz (La vente international, supra note 262, Nos. 22 and 84), who, however, emphasizes the need for an analysis of the adequacy of the substantive provisions of the Convention for dealing with software transactions, which he concedes. St Albans City and District Council v. International Computer Ltd., Court of Appeal [1996]4 All ER 481, 26 July 1996. Id. (quoting the Sale of Good Act 1979 and Supply of Goods and Services Act 1982). For an account of the evolution of the debate, see James Mogey, How Technology Has Affected the Legal System: Software as UCC Goods: a Critical Look, 34 Howard L.J. (1991) 299.
223
264
265
266 267
and ultimately the courts, the key question posed by IBMs action was whether such detached software was tangible personal property and, therefore, taxable, or intangible intellectual property not subject to taxation. In answering that question, two opposing lines of authority emerged: one asserting that software is essentially tangible, and the other declaring it to be intangible.268 When transposed to the context of private law the debate eventually led to the conclusion that, despite its inherently intangible nature, software should be regarded as goods for the purpose of Article 2 of the U.S. Uniform Commercial Code (UCC), dealing with sales of goods. At least initially, this was justified by the tangibility of the medium in which software was stored, much the same way as books or music records.269 But the more passionate advocates of extending UCC Article 2 treatment to computer software regarded the analogy between software and books or records as superficial and ultimately irrelevant, the important characteristics of a good for UCC application being movability, transferability, and identification at the time of sale.270 American commentators of the U.N. Sales Convention point out that the classification of computer software under the U.N. Sales Convention is a borderline case that has led to much controversy as software seemed difficult to distinguish from an exceedingly compact book or phonograph record.271 Other writers expressly affirm the applicability of the U.N. Sales Convention, on the basis of its legislative history, inner logic and purpose, and in light of comparative law, stressing also the benefits of the U.N. Sales Conventions uniform regime for international software transactions.272
268
By the early 1990s, the tension between the opposing camps was such that the cases did not even agree as to which was the majority view although a survey of available decisions indicated that the arguments for software as an intangible were prevailing (id., 300). Note: Computer Programs as Goods Under the U.C.C., 77 Mich. L.Rev. (1979) 1149, 1150-1151; Duncan M. Davidson, Protecting Computer Software: A Comprehensive Analysis, Ariz. St. L. J. (1983) 611. Bonna Lynn Horovitz, Computer Software as a Good under the Uniform Commercial Code: Taking a Byte out of the Intangibility Myth, 65 B.U.L. Rev. (1985) 129, 152. The author summarizes her position as follows: Whether computer programs are classified as tangible or intangible should ultimately be irrelevant to the decision of UCC applicability. While the UCC does distinguish between goods and things in action, excluding the latter from coverage, software, no matter how it is classified, should not fall within this exclusion. Software is unlike other intangibles. It is unlike intangible legal rights, intangible laws and principles, and intangible assets. A program is intangible in the sense that it cannot be touched or felt, but not in the sense that it cannot be moved and identified to a contract. (Id., at 162). John Honnold, Uniform Law for International Sales under the 1980 United Nations Convention (Deventer: Kluwer, 2nd edn., 1991) 56. L. Scott Primak, Computer Software: Should the U.N. Convention on Contracts for the International Sale of Goods Apply? A Contextual Approach to the Question, 11 Comp.
269
270
271
272
224
(b) A Special Legal Category for Virtual Goods? The discussion in the preceding section shows that, at a time when the Internet has created an unquestionable need for harmony in international contract law, the goal of uniformity as regards the legal treatment of software is but a distant dream for the international community.273 Most of the debate under domestic law took place before the exponential development of the Internet during the mid-to-late 1990s. Technological evolution showed, however, that software and other virtual goods could be transmitted entirely online, without the need for storage in any tangible medium. Continued application of the tangibility test would eventually mean that transactions involving the same software might be qualified differently depending on whether the software is downloaded from a web site or purchased over the counter. An indicator of this dichotomy, which may have practical consequences for domestic private law, but also for the applicability of international uniform instruments, is the definition of nearly all conceivable types of on-line transactions as services for the purposes of the EU internal market.274 In view of this controversy, it is not surprising that after having being proposed as a topic for future work that might be undertaken by UNCITRAL275 the issue of classification and legal treatment of virtual or digitalized goods from the perspective of the U.N. Sales Convention was discarded as a topic for international harmonization.276
L.J. (1991) 197.

273
Such is the conclusion of Trevor Cox, Chaos Versus Uniformity: the Divergent Views of Software in the International Community (<www.cisg.law.pace.edu/cisg/biblio/cox.html>). The rather complex system established by the EU Directives is based on the notion of Information Society service, which is understood as meaning any service normally provided for remuneration, at a distance, by electronic means and at the individual request of a recipient of services, except for a few exceptions such as radio and television broadcasting (see article 1, paragraph 2 and the indicative list of activities not covered by the definition in Appendix V, Directive 98/34/EC of the European Parliament and of the Council of 22 June 1998 laying down a procedure for the provision of information in the field of technical standards and regulations (Official Journal of the European Communities vol. L 204, 21 July 1998, pp. 37-48), as amended by Directive 98/48/EC of the European Parliament and of the Council of 20 July 1998, Official Journal of the European Communities vol. L 217, 5 August 1998, pp. 18-26.). See Report of the United Nations Commission on International Trade Law on the Work of its Thirty-third Session, (Official Records of the General Assembly, Fifty-fifth session, Supplement No. 17 (U.N. Document A/55/17), para. 384 The report of UNCITRAL on the work of its 36th annual session, refers to the Working Groups understanding that its work was not concerned with the question of whether and to what extent virtual goods were or should be covered by the United Nations Convention
225
274
275
276
One possible avenue to escape the nearly unsolvable quarrel over the classification of virtual goods could have been the development of a separate legal regime for software contracts and transactions involving other types of intangible information. This alternative had been proposed in view of the alleged inadequacy of sales law in general, and the U.N. Sales Convention in particular, to deal with contracts involving on-line databases, artificial intelligence systems, software, multimedia, and Internet trade in information, where the emphasis is not upon tangible goods, but upon intangibles and rights in those intangibles.277 Commentators criticize in fact the use of the expression software sales as being misleading, since transactions in software and virtual goods take the form of licenses, not sales.278 One of the proposals for future work by UNCITRAL that were made in 1999 was to consider the extent to which uniform rules could be extrapolated from the U.N. Sales Convention to govern dealings in services or virtual goods, that is, items (such as software) that might be purchased and delivered in cyberspace.279 If pursued to its ultimate consequences, that proposal might have involved the preparation of a new set of rules to deal with those transactions that might have obviated the need for their private law classification as either goods or services. Such was the course that the law in the United States seemed about to take at that time. Since the mid-1990s, the National Conference of Commissioners for Uniform State Laws (NCCUSL) and the American Law Institute (ALI) had been involved in drafting various proposals for a new Article 2B of the UCC. The motivation for the work was the perception that UCC Article 2, which deals with sales of goods, fits poorly with licensing of software and other computer information.
on Contracts for the International Sale of Goods (Id., Fifty-eighth session, Supplement No. 17 (U.N. Document A/58/17), para. 214.
277
Marcus G. Larson writes that [f]orcing an addendum to the CISG raises the specter of consumer goods being relevant to the Vienna Convention which is a highly unlikely prospect in light of its drafting history. Therefore, the logical course is the pursuit of an international convention on computer software transactions and information licensing which would be a separate entity from the CISG. This distinction will allow the drafters to include consumer transactions in this convention without tinkering with the existing CISG provisions (Comment: Applying Uniform Sales Law to International Software Transactions: the Use of the CISG, Its Shortcomings, and a Comparative Look at How the Proposed UCC Article 2b Would Remedy Them, 5 Tulane J. Intl & Compar. L. (1997) 445, 487. Michel Vivant et al., supra note 260. Report of the United Nations Commission on International Trade Law on the Work of its Thirty-third Session, Official Records of the General Assembly, Fifty-fifth session, Supplement No. 17 (U.N. Document A/.55/17), para. 384.
278 279
226
That project was not uncontroversial, though. From the beginning, the proposed new article 2B of the UCC was subject to attacks by academics, consumer groups and industry representatives.280 The criticism focused on the extent to which the proposed statute expanded the power and scope of contracts over existing law, especially with respect to its treatment of mass-market licenses. Critics also argued that the proposed uniform law interfered with the operation of numerous federal laws, including those relating to intellectual property, bankruptcy, and in particular, consumer protection.281 On 7 April 1999, the ALI and NCCUSL announced that they were abandoning the attempt to make the proposal part of the UCC. Instead, the NCCUSL announced that it was moving forward with a freestanding uniform act, the Uniform Computer Information Transactions Act (UCITA), which was eventually adopted by NCCUSL on 29 July 1999. Its advocates stress that UCITA went a long way to clarify the law governing computer information transactions and established a coherent basis of contract law tailored for the types of transactions that characterized the information industry. Furthermore, UCITA was said to address some long-standing legal issues related to software and computer information transactions, such as the debate over the validity of shrink-wrap licenses.282 UCITA further addressed issues of formation and terms of contracts, transfer of rights and interests, performance, warranties, and remedies. However, to protect the computer information owner, UCITA clearly established the relationship between selling a copy of computer information, and retaining the underlying intellectual property rights in the computer information. Despite some initial enthusiasm about UCITA, opposition to its implementation grew steadily, mainly out of a perception of UCITA as being an enforcement
280
For an overview of the debate, see Maureen A. ORourke, Progressing Towards a Uniform Commercial Code for Electronic Commerce or Racing Towards Nonuniformity?, 14 Berkeley Tech. L. J. (1999) 635. Michael Froomkin concludes his analysis of draft Article 2B as follows: Article 2Bs enormously ambitious strategy of providing a full regime for the sale and delivery of licenses in information resembles one of these self-installing software suites. While some of the rules regarding electronic contracting may be defensible, or even sensible, the total package makes a series of policy choices, especially those displacing consumer law for online transactions and enacting a national law on non-repudiation for digital signature-based e-commerce which do not seem to be required to achieve the end of rationalizing the law of information licenses (Article 2B as Legal Software for Electronic Contracting Operating System or Trojan Horse?, 13 Berkeley Tech. L.J. (1998) 1023, 1061. For some commentators, UCITAs recognition of shrink-wrap licenses alone would have sufficed as grounds for rejecting the act (for example, Roger E. Schechter, The Unfairness of Click-on Software Licenses, 46 Wayne L. Rev. (2000) 1735.
227
281
282
mechanism for software developers to the detriment of consumer interests.283 Over time, the arguments evolved into a discussion of whether a new body of contract law is at all necessary for computer information, the information industry, and the Internet.284 Leaving aside the controversial content of many of its provisions, UCITA had the merit of offering a number of ingenious solutions for handling novel issues. UCITA ended the debate on whether software and virtual goods are goods or services by using the conceptual category computer information, acknowledging the fact that there are similarities as well as differences between information and goods contracts.285 Even if one would disagree with the merits of its solutions, it is undeniable that UCITA could have offered a number of useful ideas to address electronic contracting issues on a global scale. Its failure to prosper in the United States, however, threw a negative light on UCITA as a whole and made it hard to promote even its less contentious provisions as a source of inspiration for international uniform law.286 Seen against this background, it is not difficult to understand the decision by UNCITRAL to put aside any proposals for developing a substantive law framework for transactions involving virtual goods.287
B. Contract Formation through Electronic Communications: Substantive Issues

The present section discusses how traditional notions of offer and acceptance may be applied to contract negotiation through electronic means. The discussion of these issues in the following paragraphs uses the U.N. Sales Convention as a paradigm of substantive transnational contract law.
283
For an overview of the criticism, see Bruce H. Kobayashi and Larry E. Ribstein Uniformity, Choice of Law and Software Sales, 8 George Mason L. Rev. (1999) 261. The authors attribute part of UCITAs failure to its drafting process. After analyzing costs and benefits of the uniform law process, offer the alternative of contractual choice of law as a better solution (at 294) Gregory E. Maggs, The Waning Importance of Revisions to U.C.C. Article 2, 78 Notre Dame L. Rev. (2003) 595. Amelia Boss, Taking UCITA on the Road: What Lessons Have We Learned?, 7 Roger Williams L. Rev. (2001) 167, 173. It is highly unlikely that UCITA would ever be found amenable to wholesale adoption on the international level.(Id. at 199). Report of the United Nations Commission on International Trade Law on the Work of its Thirty-sixth Session, Official Records of the General Assembly, Fifty-eighth session, Supplement No. 17 (U.N. Document A/58/17), para. 214.
284
285
286
287
228
1. Qualification of Parties Intent: Offers and Invitations to Make Offers Article 14, paragraph 1, of the U.N. Sales Convention provides that a proposal for concluding a contract that is addressed to one or more specific persons constitutes an offer if it is sufficiently definite and indicates the intention of the offeror to be bound in case of acceptance. Paragraph 2 of that article provides, however, that a proposal other than one that is addressed to one or more specific persons is to be considered merely as an invitation to make offers, unless the contrary is clearly indicated by the person making the proposal. Whether the parties negotiate by e-mail, electronic data interchange (EDI),288 or through more traditional means, the nature and legal effect of their communications will be established by their intention. In a paper-based environment, advertisements in newspapers, on the radio and television and in catalogues, brochures and price lists are generally regarded as invitations to submit offers (according to some legal writers, even in cases where they are directed to a specific group of customers), since in those cases the intention to be bound is considered to be lacking.289 (a) Offers and Advertisements in Electronic Commerce If the notion of offer of the U.N. Sales Convention is transposed to an electronic environment, a company that advertises its goods or services on the Internet or through other open networks should be considered to be merely inviting those who access the site to make offers. Thus, an offer of goods or services through the Internet would not prima facie constitute a binding offer.290
288
The Working Party on Facilitation of International Trade Procedures (WP.4) of the Economic Commission for Europe, which is the United Nations body responsible for the development of UN/EDIFACT technical standards has defined Electronic Data Interchange as the electronic transfer from computer to computer of information using an agreed standard to structure the information. This definition is also used in article 2 of the MLEC. Honnold, supra note 271, 195; von Caemmerer and Schlechtriem, supra note 255, art. 14, Nos. 13-15, pp. 144-146; Peter Schlechtriem, Commentary on the UN Convention on the International Sale of Goods (CISG) (Oxford: Clarendon Press, 1998) art. 14, Nos. 13-15, pp. 111-112; Heinrich Honsell, (ed.), Kommentar zum UN-Kaufrecht (Berlin/Heidelberg/New York: Springer, 1997), art. 14, Nos. 17-19, p. 121; Fritz Enderlein and Dietrich Maskow, International Sales Law (New York/London/Rome: Oceana, 1992) 83; Maria del Pilar Perales Viscasillas, La Formacin del Contrato de Compraventa Internacional de Mercaderas (Valencia: Tirant lo blanch, 1996) 289. A few commentators argue, however, that catalogue mailings addressed to named recipients might be regarded as binding offers, since such mailings could not be regarded as being to non-specified persons (Heuz, supra note 262, No. 175, p. 156; see also Bernard Audit, La Vente Internationale de Marchandises (Paris: L.G.D.J., 1990), No. 62, p. 58, and Jean Thieffry and Chantal Granier, supra note 264 at 89. Jens Werner, E-Commerce.co.uk Local Rules in a Global Net: Online Business Transactions and the Applicability of Traditional English Contract Law Rules, 6 Intl J. Communications
229
289
290
The difficulty that may arise in this context is how to strike a balance between a traders possible intention (or lack thereof) of being bound by an offer, on the one hand, and the protection of relying parties acting in good faith, on the other. The Internet makes it possible to address specific information to a virtually unlimited number of persons and current technology permits contracts to be concluded nearly instantaneously, or at least creates the impression that a contract has been so concluded. In legal literature, it has been suggested that the invitation-to-treat paradigm may not be suitable for uncritical transposition to an Internet environment. One possible criterion for distinguishing between a binding offer and an invitation to treat may be based on the nature of the applications used by the parties. Legal writings on electronic contracting have proposed a distinction between web sites offering goods or services through interactive applications and those which use non-interactive applications. If a web site only offers information about a company and its products and any contact with potential customers lies outside the electronic medium, there would be little difference from a conventional advertisement. However, an Internet web site that uses interactive applications may enable negotiation and immediate conclusion of a contract (in the case of virtual goods even immediate performance). Legal writings on electronic commerce have proposed that such interactive applications might be regarded, therefore, as an offer open for acceptance while stocks last, as opposed to an invitation to treat.291 This proposition is at least at first sight consistent with legal thinking for traditional transactions. Indeed, the notion of offers to the public that are binding upon the offeror while stocks last is recognized also for international sales transactions.292 However, the potentially unlimited reach of the Internet and the risk of errors in electronic communications, including in posting price and other product information on a web site, compounded by the use of automatic reply functions that do not provide an opportunity for review and correction of errors, seem to call for caution.293
L. & Policy (2000/2001) 1, 5.

291
Christoph Glatt, Comparative Issues in the Formation of Electronic Contracts, 6 Intl J. L. & Information Tech. (1998) 34, 50. Von Caemmerer and Schlechtriem, supra note 255, 144; Perales Viscasillas, supra note 289 at 295 and the example of Spanish legislation given in footnote 41 supra. Werner highlights the practical importance of the distinction between invitations to treat and offers with the following example: E-tailer Argos offered by mistake a Sony TV for 3.00 instead of 299.99. People who spotted the bargain placed numerous orders for TVs which would constitute an acceptance (and thus conclude a contract) if the webvertisement of Argos could be regarded as a genuine offer (E-Commerce.co.uk, supra note 290 at 5).
292
293
230
In support of this approach, it has been argued that parties acting upon offers of goods or services made through the use of interactive applications might be led to assume that offers made through such systems were firm offers and that by placing an order they were be validly concluding a binding contract at that point in time.294 Those parties, it has been said, should be able to rely on such a reasonable assumption in view of the potentially significant economic consequences of contract frustration, in particular in connection with purchase orders for commodities or other items with highly fluctuating prices. Attaching consequence to the use of interactive applications could help enhance transparency in trading practices by encouraging business entities to state clearly whether or not they accepted to be bound by acceptance of offers of goods or services or whether they were only extending invitations to make offers.295 UNCITRAL considered these arguments carefully. The final consensus was that attaching a presumption of binding intention to the use of interactive applications would be detrimental for sellers holding a limited stock of certain goods, if the seller were to be liable to fulfil all purchase orders received from a potentially unlimited number of buyers.296 In order to avert that risk, companies offering goods or services through websites that use interactive applications enabling negotiation and immediate processing of purchase orders for goods or services frequently indicate in their websites that they are not bound by those offers. UNCITRAL felt that, if this was already the case in practice, the ECC should not reverse it.297 In keeping with the principle of media neutrality, UNCITRAL took the view that the solution for online transactions should not be different
294
This seems to be the view taken by the High Court in Singapore, which has held that [i]n an Internet sale, a prospective purchaser is not able to view the physical stock available. The web merchant, unless he qualifies his offer appropriately, by making it subject to the availability of stock or some other condition precedent, could be seen as making an offer to sell an infinite supply of goods. A prospective purchaser is entitled to rely on the terms of the web advertisement. The law may not imply a condition precedent as to the availability of stock simply to bail out an Internet merchant from a bad bargain, a fortiori in the sale of information and probably services, as the same constraints as to availability and supply may not usually apply to such sales. Theoretically the supply of information is limitless. It would be illogical to have different approaches for different product sales over the Internet. It is therefore incumbent on the web merchant to protect himself, as he has both the means to do so and knowledge relating to the availability of any product that is being marketed. As most web merchants have automated software responses, they need to ensure that such automated responses correctly reflect their intentions from an objective perspective.(Chwee Kin Keong and others v. Digilandmall.com Pte Ltd, [2004] SGHC 71, 12 April 2004).
295 296 297
See U.N. document A/CN.9/509, para. 81. See U.N. document A/CN.9/546, para. 107. See U.N. document A/CN.9/509, para. 82; see also U.N. document A/CN.9/528, para. 116.
231
from the solution used for equivalent situations in a paper-based environment. Therefore, the ECC provides, as a general rule, that a company that advertises goods or services on the Internet or through other open networks should be considered as merely inviting those who accessed the site to make offers. Thus, an offer of goods or services through the Internet does not prima facie constitute a binding offer According to the ECC, the general principle that offers of goods or services that are accessible to an unlimited number of persons are not binding applies even when the offer is supported by an interactive application. Typically an interactive application is a combination of software and hardware for conveying offers of goods and services in a manner that allows for the parties to exchange information in a structured form with a view to concluding a contract automatically. The expression interactive applications focuses on what is apparent to the person accessing the system, namely that it is prompted to exchange information through that information system by means of immediate actions and responses having an appearance of automaticity.298 It is irrelevant how the system functions internally and to what extent it is really automated (e.g. whether other actions, by human intervention or through the use of other equipment, might be required in order to effectively conclude a contract or process an order).299 (b) Special Cases: Click-Wrap Agreements and Internet Auctions There are, however, special situations where electronic contracting models may well suggest that a person is making a binding offer. One such line of jurisprudence comprises cases dealing with so-called click-wrap agreements in the United States. Most if not all of those cases have involved contracts with Internet service providers or online purchases of software or other digitized information through web sites that allowed online download of software or immediate connection to a provider of Internet access services.300 Users were typically presented with messages on their computer screens requiring that they manifest their assent to the terms of the licence agreement by clicking on an icon. The products could not be obtained or used unless and until the icon was clicked. The main issue in such cases was the enforceability of contract terms purported to have been incorporated by reference and the conditions under which a consumer may be validly bound by such terms. While the cases did not
298 299
Id., para. 87. See U.N. document A/CN.9/546, para. 114. See Lawrence Groff v. America Online, Inc., LEXIS 46, WL 307001 (R.I. Super., 1998) <legal.web.aol.com/decisions/dlother/groff.html> (3 December 2007); Hotmail Corp. v. Van$ Money Pie, LEXIS 10729, WL 3888389 (N.D. Calif. 1998); Steven J. Caspi, et al. v. The Microsoft Network, L.L.C., et al., 323 N.J. Super. 118 (1999); and I. Lan Systems, Inc. v. Netscout Service Level Corp., 183 F. Supp. 2d 328 (D. Mass. 2002).
300
232
directly dealt with the nature of the sellers offer (i.e. whether it was a true offer or merely an invitation to treat), the reasoning used by the courts to deal with such cases implies a certain understanding of the nature of the communications from which a qualification of the offers may be inferred. Firstly, the courts that have thus far dealt with click-wrap cases, even those which have denied their enforceability as a whole or only of some of their terms against consumers,301 have not questioned the sellers intention to be bound by its Internet offer of a software or similar product. Furthermore, while some courts have questioned the effectiveness of clicking on an icon or I agree button for the purpose of indicating assent to the terms of the vendors software licence agreements, the courts have not required a subsequent act of the vendor as a condition for a contract to be concluded. Nor have the courts denied the existence of a contract on the ground that the consumers action represented a contract offer that needed to be accepted by the buyer. It is in fact implicit in the reasoning of the courts that at least in theory a valid contract could be formed once the consumer had validly indicated the intention to purchase the software. The courts have not regarded the customer as the actual offeror and have, albeit not expressly, clearly treated the web site offerings as a binding commitment on the vendor and not a mere invitation to treat. It may be argued that the fact that the products or services being offered allowed for immediate delivery by the vendor or immediate enjoyment by the customer was a decisive factor for the courts affirmation of contract formation through customer action without requiring subsequent acceptance by the vendor, even though no such mention is made by the courts in those cases. The second body of case law deals with Internet auctions and involves both business-to-consumer and business-to-business auctions. In an early case, a district court in Germany found that a person offering goods through an Internet auction platform had not made a binding offer, but had merely invited offers in respect of the goods during a set period of time.302 That decision was later reversed by the court of appeal, which found that the display of goods for auction purposes through an Internet auction platform constituted more than an invitation to treat and should be regarded as a binding contract offer.303 Such
301
For instance, Specht v. Netscape Communications Corp., 150 F. Supp. 2nd 585, affirmed in Specht v. Netscape Communications Corporation and America Online, Inc., (306 F. 3rd 17 (2d Cir. 2002), aff d 306 F. 3d 17 (2d Cir. 2002). Landgericht Mnster, Case No. 4 O 424/99, 21 January 2000, JurPC-Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 60/2000 <www.jurpc.de/rechtspr/20000060.htm> (30 November 2007). Oberlandesgericht Hamm, Case No. 2 U 58/00, 14 December 2000, JurPC-Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 255/2000 <www.jurpc.de/rechtspr/20000255. htm> (30 November 2007).
233
302
303
an offer did not represent an open-ended commitment to accept an unlimited number of offers, as it was limited to the acceptance of the highest bid remaining at the end of the auction period. The electronic auction process allowed for sufficient determination of the price, so that all essential elements of a binding offer to conclude a sales contract were present. That understanding was followed by other German courts304 and was also affirmed by the Federal Court (Bundesgerichtshof), which expressly affirmed the principle that an offer of goods for auction purposes through an Internet auction platform with the indication that the seller was committed to accepting the highest effective bid constituted a valid anticipatory acceptance of the highest bid and not only an invitation to treat.305 A court of appeals in the United States reached essentially the same conclusion in a case involving the auction of a domain name though the Internet. The court held that a partys ex post facto characterization of the contents of its web site as a mere advertisement did not by itself exclude the binding nature of a commitment to sell a certain item to the person offering the highest bid within a set period.306 UNCITRAL recognized that in some situations such as those described above it may be appropriate to regard a proposal to conclude a contract that was supported by interactive applications as evidencing the partys intent to be bound in case of acceptance. Some business models are indeed based on the rule that offers through interactive applications are binding offers. In those cases, possible concerns about the limited availability of the relevant product or service are often addressed by including disclaimers stating that the offers are for a limited quantity only and by the automatic placement of orders according to the time they were received.307 UNCITRAL also noted that some case law seemed to support the view that offers made by so-called click-wrap agreements and in Internet auctions may be interpreted as binding.308 However, the extent to which such intent indeed exists is a matter that, in the view of UNCITRAL, should be assessed in the light of all the circumstances (for example, disclaimers made by the vendor or the general terms and conditions of the auction platform). As a general rule, UNCITRAL considered that it would be unwise to presume
304
Amtsgericht Hannover, Case No. 501 C 1510/01, 7 September 2002, JurPC-Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 299/2002 <www.jurpc.de/rechtspr/20020299. htm> (30 November 2007). Bundesgerichtshof, Case No. VIII ZR 13/01, 7 November 2001, JurPC-Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 255/2001 <www.jurpc.de/rechtspr/20010255. htm> (30 November 2007). Je Ho Lim v. The TV Corporation International, 99 Cal. App. 4th 684 (2002). See U.N. document A/CN.9/546, para. 112. Id., para. 109.
305
306 307 308
234
that persons using interactive applications to make offers always intend to make binding offers, because that presumption would not reflect the prevailing practice in the marketplace.309 This general rule does not exclude the possibility, however, that on a given case the automated system is set up in such a way so as to suggest that there is such intention. 2. Time of Receipt and Dispatch of Electronic Communications and Contract Formation When business parties deal through more traditional means, the effectiveness of the communications they exchange depends on various factors, including the time of their receipt or dispatch, as appropriate. Although some legal systems have general rules on the effectiveness of communications in a contractual context, in many legal systems general rules are derived from the specific rules that govern the effectiveness of offer and acceptance for purposes of contract formation. From a legislative point of view, it is important to consider how to formulate rules on time of receipt and dispatch in a manner that adequately transposes the existing rules for traditional means of communication to an electronic environment. (a) Rules on Contract Formation Rules on contract formation often distinguish between instantaneous and non-instantaneous communications of offer and acceptance or between communications exchanged between parties present at the same place at the same time (inter praesentes) or communications exchanged at a distance (inter absentes). Typically, unless the parties engage in instantaneous communication or are negotiating face-to-face, a contract will be formed when an offer to conclude the contract has been expressly or tacitly accepted by the party or parties to whom it was addressed. Leaving aside the possibility of contract formation through performance or other actions implying acceptance,310 which usually involves a finding of facts, the controlling factor for contract formation where the communications are not instantaneous is the time when an acceptance of an offer becomes effective. There are currently four main theories for determining when an acceptance becomes effective under general contract law, although they are rarely applied in pure form or to all conceivable situations.311
309 310
Id., para. 112. See the commentary to article 2.6 of the UNIDROIT Principles of International Commercial Contracts (Unidroit, Rome, 1994). See the overview of the existing common law and civil law rules on contract formation in Perales Viscasillas, supra note 289 at 178 et seq.
235
311
Pursuant to the declaration theory,312 a contract is formed when the offeree produces some external manifestation of its intent to accept the offer, even though this may not yet be known to the offeror. According to the mailbox rule, which is traditionally applied in most common law jurisdictions,313 but also in some countries belonging to the civil law tradition,314 acceptance of an offer is effective upon dispatch by the offeree (for example, by placing a letter in a mailbox). In turn, under the reception theory, which has been adopted in several civil law jurisdictions,315 the acceptance becomes effective when it reaches the offeror. Lastly, the information theory requires knowledge of the acceptance for a contract to be formed.316 Of all these theories, the mailbox rule and the reception theory are the most commonly applied for business transactions. In some legal systems, both theories may be invoked, according to the context.317 The notion of receipt is sometimes understood not only as a question of time but also as a question of form or maybe even content of the communication of acceptance. Thus, for example, the rules of the German Civil Code318 on the legal effectiveness of legally relevant communications or declarations of will upon their receipt have been understood by German doctrine and case law to the effect that a communication has not only to reach the addressees sphere of control but it also has to be in such a form as to ensure the possibility for the addressee to
312
Which seems to be the general rule for contract formation in Switzerland, where the formation of a contract occurs lorsque les parties ont, rciproquement et dune manire concordante, manifest leur volont (Code des Obligations, art. 1). The mailbox rule was first adopted by the Kings Bench in 1818 in order to avoid the need for successive confirmations of receipt as that might continue ad infinitum (see Adams v. Lindsell, [1818] 160 ER 250). Despite some criticism, the mailbox rule has been nearly unanimously adopted in common law jurisdictions (see the references in Paul Fasciano, Internet Electronic Mail: A Last Bastion for the Mailbox Rule, 25 Hofstra Law Review (1997) 971, footnote 20. For instance, Argentina (Cdigo Civil, art. 1154) and Brazil (Cdigo Civil, art. 434). Such as in Austria (Allgemeines Brgerliches Gesetzbuch (ABGB), art. 862) and Germany (Brgerliches Gesetzbuch (BGB), sect. 130). For instance, Spain (Cdigo Civil, art. 1262) and Venezuela (Cdigo de Comercio, art. 120, para. 1). The information theory is the general rule for contract formation in Italy, where the contract is concluded when the offeror has knowledge of the acceptance by the offeree (Codice Civile, art. 1326). However, knowledge is presumed when the acceptance is received at the offerors address (Codice Civile, art. 1335), which in practice brings the Italian system closer to the reception theory.
313
314
315
316
317
This seems to be the case in France, where the Commercial Chamber of the Cour de cassation, in a judgment of 7 January 1981, affirmed the dispatch theory, but commentators continue to maintain the validity of the receipt theory (Franois Chabas, Note, Revue Trimestrielle de Droit Civil (1981) 849). BGB, sect. 130 (1).
318
236
become aware of it.319 The latter element has been further subdivided into various substantive requirements, such as, for example, accessibility of the language of the communication320 or delivery within normal working hours.321 The U.N. Sales Convention adopted the reception theory as a general rule.322 Under the U.N. Sales Convention, a contract is concluded at the moment when an acceptance of an offer becomes effective,323 which happens when the indication of assent reaches the offeror.324 For the purposes of the Conventions provisions on contract formation, an offer, declaration of acceptance or any other indication of intent reaches the addressee when it is made orally to him or delivered by any other means to him personally, to his place of business or mailing address or, if he does not have a place of business or mailing address, to his habitual residence.325 The notion of receipt has been understood by commentators to mean the time when the communication enters the sphere of control of the addressee. Until that time, the originator of the communication (in case of acceptance, the offeree) must ensure that the communication reaches the addressee and that it arrives within the required time. Where the notion of dispatch is relevant, the crucial moment is when the communication leaves the sphere of control of the originator. From that moment on, the originator would be relieved of the risk of loss or delay in the communication, with which instead the addressee would be concerned. Commentators of the U.N. Sales Convention have observed that the notion of reach in article 24 of the Convention was made dependent upon external, easily provable facts and was meant to relieve the originator of the risk of
319
Otto Palandt, Brgerliches Gesetzbuch (Mnchen C.H. Beck, 60th edn., 2001), 103, No. 3 (commentary on sect. 130 by H. Heinrichs); and Mnchener Kommentar zum Brgerlichen Gesetzbuch, vol. 1 (Mnchen: Becksche Verlagsbuchhandlung, 3rd edn., 1993), 1055, No. 10 (commentary on sect. 130 by H. Frschler). Transposed to the context of the U.N. Sales Convention, this requirement has led to the conclusion, for example, that standard contract conditions could not be relied upon if they have been sent in a language different from the one used during the negotiations (Amtsgericht Kehl, 6 October 1995 <cisgw3.law.pace.edu/cases/951006g1.html> (3 December 2007)). See the authorities cited elsewhere (Palandt, supra note 319, Brgerliches Gesetzbuch 103, No. 3 and Mnchener, supra note 319, Kommentar No. 12). However, dispatch is also relevant for the operation of a number of provisions of the Convention, such as articles 19, paragraph 2 (notice of objection to additional terms proposed by offeree); 20 (period of time for acceptance); and 21 (conditions for effectiveness of late acceptance). U.N. Sales Convention, art. 23. U.N. Sales Convention, art. 18, para. 2. U.N. Sales Convention, art. 24.
237
320
321
322
323 324 325
defective communications of a declaration within the recipients organizational sphere; circumstances that indicated that the provisions of article 24 contrary to the strict rules followed in some domestic laws should be interpreted to the effect that they generally do not require an opportunity for the recipient to gain awareness of the declaration.326 Other ways of applying the article, for example, by attempting to take national public holidays and customary working hours into consideration were said to lead to problems and to legal uncertainty in a law governing international situations.327 (b) Timing of Dispatch and Receipt of Electronic Communications The above considerations are equally important for the formation of contracts through electronic communications. Indeed, despite some early suggestions that contract negotiation through electronic means, in particular in an electronic data interchange (EDI) environment, replicates the pattern of face-to-face or instantaneous communications,328 the exchange of electronic messages, at least when electronic mail (e-mail) techniques are used, seems to be more analogous to exchange of postal correspondence.329 In any event, default rules on time and place of dispatch and receipt of electronic communications should supplement national rules on dispatch and receipt by transposing them to an electronic environment. Such provisions should be sufficiently flexible to cover both cases where electronic communication appears to be instantaneous and those where electronic messaging mirrors traditional mail. The following paragraphs analyse the way this has been done by the MLEC and in domestic legislation. (i) The Rule in Article 15 of the MLEC Article 15, paragraph 1, of the MLEC defines the time of dispatch of an electronic communication (data message in the MLEC terminology) as the time when
326
Peter Schlechtriem, Commentary on the UN Convention on the International Sale of Goods (CISG) (Oxford: Clarendon Press, 1998) art. 24, Nos. 13-14, pp. 167-168; see also von Caemmerer and Schlechtriem, supra note 255, art. 24, Nos. 13-14, pp. 202-203. Id. For example, Michael S. Baum and Henry H. Perritt, Jr., Electronic Contracting, Publishing and EDI Law (New York: Wiley Law Publications, 1991) 323, No. 6.8. The authors, however, recognize various factual circumstances that might lead to a different conclusion, such as a certain non-instantaneous characteristic of computerized offers and acceptances, regardless of whether mailboxes or store-and-forward techniques are used in the transmission. Despite common belief, [the transmission of Internet electronic mail] does not take place in a substantially instantaneous manner. Rather, it will typically take minutes, hours or in some cases days (Fasciano, supra note 313, 1000-1001).
327 328
329
238
the electronic communication enters an information system330 placed outside the control of the originator,331 which may be the information system of an intermediary or an information system of the addressee. Under that provision, an electronic communication should not be considered to have been dispatched if it merely reached the information system of the addressee but failed to enter it.332 For the time of receipt, paragraph 2 of the same article distinguishes between a few factual situations: (a) where the addressee designates a specific information system, which may or may not be his own, for the receipt of a message, the electronic communication is deemed to have been received when it enters the designated system;333 (b) if the electronic communication is sent to an information system of the addressee that is not the designated system, receipt occurs when the electronic communication is retrieved by the addressee; and (c) if the addressee has not designated an information system, receipt occurs when the electronic communication enters an information system of the addressee. The distinction between designated and non-designated information systems is intended to establish an appropriate allocation of risks and responsibilities between originator and addressee. The person who designates a specific information system for the receipt of electronic communications, even if it is a
330
Information system is a defined term under article 2, subparagraph (f), of the MLEC and means a system for generating, sending, receiving, storing or otherwise processing data messages. Depending on the factual situation, this may indicate a communications network, and in other instances could include an electronic mailbox or even a telecopier (Guide to Enactment of the MLEC, supra note 105, para. 40). The notion of control over an information system should not be understood as requiring the information system to be located on the premises of the addressee, since location of information systems is not an operative criterion under the MLEC (Guide to Enactment of the MLEC, supra note 105, para. 40). It should be noted that the MLEC, as pointed out in its Guide to Enactment of the MLEC, supra note 105, (para. 104): does not expressly address the question of possible malfunctioning of information systems as a basis for liability. In particular, where the information system of the addressee does not function at all or functions improperly or, while functioning properly, cannot be entered into by the data message (e.g. in the case of a fax that is constantly occupied), dispatch under the MLEC does not occur. It was felt during the preparation of the MLEC that the addressee should not be placed under the burdensome obligation to maintain its information system functioning at all times by way of a general provision. By designated information system the MLEC means a system that has been specifically chosen by a party, for instance in the case where an offer expressly specifies the address to which acceptance should be sent. Paragraph 102 of the Guide to Enactment of the MLEC, supra note 105, clarifies that a mere indication of an electronic mail or telecopy address on a letterhead or other document should not be regarded as express designation of one or more information systems.
239
331
332
333
system operated by a third party, should be expected to bear the risk of loss or delay of messages that effectively entered that system. However, if the originator chooses to ignore the addressees instructions and sends the message to an information system other than the designated system, it would not be reasonable to consider the message delivered to the addressee until the addressee has actually retrieved it. The rule in the event that no particular system was designated assumes that for the addressee it was indifferent to which information system the messages would be sent, in which case it would be reasonable to presume that it would accept messages through any of the information systems it uses. For both the definition of dispatch and that of receipt, an electronic communication enters an information system at the time when it becomes available for processing within that information system. It is not necessary for the recipient to know that the message has been received and there is no additional requirement that the recipient actually read or even access the message. If it reaches the recipients mailbox, receipt has occurred. Whether the electronic communication is intelligible or usable by the addressee is intentionally outside the purview of the MLEC, which does not set intend to overrule provisions of national law under which receipt of a message may occur at the time when the message enters the sphere of the addressee, irrespective of whether the message is intelligible or usable by the addressee.334 (ii) Electronic Communications in Domestic Enactments of the MLEC At the domestic level, there seems to be little disagreement with the proposition that, from a purely factual point of view, the time when an electronic communication enters an information system within the addressees control or enters an information system outside the senders control represent the evident electronic equivalents of the sphere of control tests used to define receipt and dispatch under both the reception and the mailbox rules. Except for France,335 the jurisdictions that have thus far adopted the MLEC have included provisions on time and place of dispatch and receipt of electronic
334
The Guide to Enactment of the MLEC, supra note 105, (para. 103) adds that the MLEC is also not intended to run counter to trade usage, under which certain encoded messages are deemed to be received even before they are usable by, or intelligible for, the addressee. It was felt that the MLEC should not create a more stringent requirement than currently exists in a paper-based environment, where a message can be considered to be received even if it is not intelligible for the addressee or not intended to be intelligible to the addressee (e.g. where encrypted data is transmitted to a depository for the sole purpose of retention in the context of intellectual property rights protection). The French enactment of the MLEC (Loi no. 2000-230 du 13 mars 2000 portant adaptation du droit de la preuve aux technologies de linformation et relative la signature lectronique) deals essentially with recognition and evidentiary legal value of electronic records, but does not deal with their communication.
335
240
communications. Without exception, all enactments of the MLEC have adopted the distinction between designated and non-designated systems.336 This is also the case in those countries in which uniform law has been prepared on the basis of the MLEC, such as Canada337 and the United States.338 That distinction, however, is not explicit in the U.S. Uniform Electronic Transactions Act (UETA), which contemplates, in addition to a designated information system, an information system that the recipient uses for the purpose of receiving electronic records or information of the type sent and from which the recipient is able to retrieve the electronic record.339 Notwithstanding that the language used in UETA differs from article 15 of the MLEC, both instruments distinguish between a system that was positively chosen by a party for the receipt of a particular message or type of message and other (non-designated) information systems merely used by the recipient. The latter category, in much the same way as the non-designated system under article 15 of the MLEC, was included in UETA out of a concern to [allow] the recipient of electronic records to retain control over where they would be sent and received.340 Court decisions in the United States suggest that
336
Australia (Electronic Transactions Act 1999, sect. 14, subsects. (3) and (4)); Colombia (Ley Nmero 527 de 1999: Ley de comercio electrnico, art. 24, subparas. (a) and (b)); Ecuador (Ley de comercio electrnico, firmas electrnicas y mensajes de datos of 2002, art. 11, subparas. (a) and (b)); India (Information Technology Act 2000, sect. 13); Ireland (Electronic Commerce Act, 2000, sect. 21, paras. (2) and (3)); Jordan (Electronic Transactions Law (No. 85) of 2001, art.17); Mauritius (Electronic Transactions Act 2000, sect. 14 (2)); Mexico (Decreto por el que se reforman y adicionan diversas disposiciones del Cdigo Civil para. el Distrito Federal of 26April 2000, art. 91); New Zealand (Electronic Transactions Act 2002, sect. 11, paras. (a) and (b)); Pakistan (Electronic Transactions Ordinance 2002, sect. 15, para. (2)); Philippines (Electronic Commerce Act 2000, sect. 22, paras. (a) and (b)); Republic of Korea (Framework Law on Electronic Commerce, 1999, art. 6, para. (2)); Singapore (Electronic Transactions Act 1998, sect. 15, subpara. (2) (a)); Slovenia (Electronic Commerce and Electronic Signature Act, 2000, art. 10, para. 2); Thailand (Electronic Transactions Act 2002, sect. 23); and Venezuela (Decreto no. 1024 de 10 de febrero de 2001 Ley sobre mensajes de datos y firmas electrnicas, art. 11). The same rules are also contained in the laws of the Bailiwick of Jersey (Electronic Communications (Jersey) Law 2000, art. 6), and the Isle of Man (Electronic Transactions Act 2000, sect. 2), both Dependencies of the British Crown; in the British overseas territories of Bermuda (Electronic Transactions Act 1999, sect. 18, para. 2) and Turks and Caicos (Electronic Transactions Ordinance 2000, sect. 16 (2) and (3)); and in the Hong Kong Special Administrative Region of China (Electronic Commerce Ordinance 2000), sect. 19 (2)). Uniform Electronic Commerce Act (UECA), supra note 131, sect. 23 (2). Uniform Electronic Transactions Act (UETA), supra note 132, sect. 15 (b). This formulation is also used in section 23 (b) of the Electronic Communications and Transactions Act 2002 of South Africa. The drafters of UETA recognized the fact that many people have multiple e-mail addresses for different purposes. [Subsection 15 (b) of UETA] assures that recipients can designate the e-mail address or system to be used in a particular transaction. For example, the
241
337 338 339
340
a designation of an electronic address may also be inferred from the fact that a company structured its business in such a way that it could be contacted only via its e-mail address and listed no easily discoverable street address.341 Domestic enactments of the MLEC are also remarkably uniform in defining the time of receipt of electronic communications sent to a designated system. Nearly all enactments reproduce the rule of paragraph 2 (a) (i) of article 15 of the MLEC, namely, that a message sent to a designated system is received when it enters that system. Minor domestic variations exist with regard to cases in which either the addressee has not designated a particular information system or the originator sends the message to a system other than the designated system. Most domestic enactments of the MLEC make that distinction.342 In those countries, the consequences are generally the same as in article 15 of the MLEC, that is, a message sent to an information system other than the designated one is only deemed to be received upon retrieval by the addressee,343 whereas a message sent in the absence of a designated system is deemed to be received upon entry
recipient retains the ability to designate a home e-mail for personal matters, work e-mail for official business, or a separate organizational e-mail solely for the business purposes of that organization. If A sends B a notice at his home which relates to business, it may not be deemed received if B designated his business address as the sole address for business purposes. Whether actual knowledge upon seeing it at home would qualify as receipt is determined under the otherwise applicable substantive law (Amelia H. Boss, supra note 114).
341
Rio Properties, Inc. v. Rio International Interlink, 284 F.3d 1007 (9th Cir. 2002). This case involved various trademark infringement claims by an American company against a foreign Internet business entity. After failed attempts to serve the defendant by conventional means in the United States of America, the claimant brought an emergency motion to effectuate alternative service of process by e-mail, which had been identified as being the defendants preferred means of communication. The Court of Appeals concluded that not only was service of process by e-mail proper that is, reasonably calculated to apprise the defendant of the pendent action and afford it an opportunity to respond but, in this particular case, it was the method of service most likely to reach the defendant. The Court noted in that connection that the defendant structured its business such that it could be contacted only via its e-mail address and that it listed no easily discoverable street address. Rather, on its web site and print media, the defendant designated its e-mail address as its preferred contact information. E.g. Bermuda, Colombia, Ecuador, India, Jordan, Mauritius, Mexico, Pakistan, the Philippines and the Republic of Korea. Some enactments, as in Bermuda, require, instead of retrieval, that the message come to the attention of the addressee. This does not, in practice, alter the substance of the rule.
342
343
242
into an information system of the addressee. However, in two countries in that group344 the law explicitly requires retrieval in both situations. The laws of other countries contemplate only cases where an addressee has not designated an information system.345 In those countries, receipt normally occurs once the message is retrieved or comes to the attention of the addressee, but in one country346 receipt occurs upon entry in a system regularly used by the addressee. Two countries contemplate only the hypothesis of a message being sent to an information system other than the designated system, in which case receipt occurs upon retrieval.347 It is not clear, for that group of countries, whether a message sent to one particular system despite an express designation of another system would follow the same rule. Arguably, both situations would be treated in the same manner, as is suggested by the law of one country that expressly provides that for all cases other than designated systems, the message is received when it comes to the addressees attention.348 The only apparently significant deviations from article 15 of the MLEC are found in UETA and the Uniform Electronic Commerce Act (UECA) of Canada. Both texts require that the electronic communication be capable of being retrieved and processed by the addressee in addition to entering the addressees system.349 It has been pointed out, in that connection, that the emphasis in the MLEC is on timing.350 Under the MLEC, the message enters a system when it is available for processing, whether or not it can in fact be processed. For UETA and UECA, on the other hand, proper receipt requires that the recipient should be able to retrieve the record from the system and that the message be sent in a form that the addressees system can process. Nonetheless, there is arguably no inconsistency between the UETA and the MLEC, as it may be understood that the MLEC defers to national law on the processability issue.351 Legal analysis and comparison of UETA and the MLEC has indicated, indeed, that, despite the different formulation, both instruments achieve the same result, as shown in the following example:352 Consider the situation where, because of a power failure or system failure, the system becomes inaccessible, precluding the recipient from ever
344 345 346 347 348 349 350 351 352
Mauritius and Mexico. E.g. Australia, Canada, Ireland and Venezuela. Venezuela. E.g. Slovenia and Thailand. New Zealand. UETA, sect. 15 (b) (1) and (2); UECA, sect. 23 (2) (a). Boss, supra note 114 at 328. Id. Id. at 330-331.
243
retrieving the record. The question would be (under both products): when did the failure occur? If it occurred before the electronic record entered the system, then no receipt has yet occurred under either formulation. If the message enters the system, the question initially under the UETA is whether the recipient is able to retrieve it. If the recipient is able to retrieve it, albeit for an instant, receipt has occurred. Subsequent failure of the system should not undo what has already occurred. The mere inability of the recipient to retrieve the electronic record at a later point in time is irrelevant once receipt has occurred. Another situation where both UETA and UECA seem at first sight to differ from the MLEC is when the recipient has designated an information system, but the sender sends the electronic record to another information system. Unlike the MLEC, UETA and UECA do not have specific rules for such a case, which would have to be solved in the light of their more general provisions. The result would probably not be substantially different from the result under the MLEC. If the information system, although not the designated system, was one used by the recipient for electronic records of this type, the record would be deemed to have been received (whether or not it was actually retrieved or received). If the system was not generally used for messages of this type, the presumption established by UETA would not apply. Arguably, that presumption would not be needed if the record was actually retrieved by the recipient, which would mean that in practice the result under the MLEC and UETA would be the same. If, however, the record entered an information system of the recipient that was neither the one designated by the recipient, nor the one used by the recipient for such messages, and the record was never retrieved by the recipient, it is claimed that the two laws would again produce the same result: there would be no receipt under the MLEC (because there was no retrieval), and none under the UETA (because it was not sent to the correct address).353 (iii) Electronic Communications in Other Domestic Laws The situation in countries that have not adopted the MLEC is not easily ascertainable in view of the scarcity of legal authorities. For the purposes of the present analysis, those countries may be placed in two broad groups: member States of the European Union (EU) and non-EU member States. Very few countries outside EU other than enacting States of the UNCITRAL MLEC have specific legislation on the types of issue related to electronic commerce that have been dealt with by UNCITRAL. Typically, where written laws exist, they deal only with digital signatures (sometimes also other forms of electronic
353
Id.
244
signatures) and certification services354 and rarely with issues of electronic contracting.355 The survey done by the Secretariat has not identified legislative provisions on time of dispatch and receipt of electronic communications in those countries. The situation is different within the EU. EU member States are bound to implement the principles set forth in the various Directives relevant for electronic commerce, in particular Directive 2000/31/EC (hereafter EU Electronic Commerce Directive).356 Article11 of the EU Electronic Commerce Directive provides that EU member States shall ensure, except when otherwise agreed by parties who are not consumers, that a customers order and the acknowledgement of receipt of the order by the merchant are deemed to be received when the parties to whom they are addressed are able to access them. Under the EU legislative system, member States are left with the choice of the means to achieve the result envisaged by the EU Electronic Commerce Directive. When, however, are the parties able to access electronic communications and which kind of ability is meant by the EU Electronic Commerce Directive? Is it sufficient that the parties have the abstract possibility of gaining access to the electronic communication, or is it necessary for the addressee to be actually in a position to retrieve the message? The preamble to the EU Electronic Commerce Directive does not explain the precise meaning of the words able to access. While a number of language versions favour a more general formulation,357
354
This is the case, for example, of the laws of Argentina (Ley No. 25.506 Ley de Firma Digital and Decreto No. 2628/2002 (Firma Digital), Reglamentacin de la Ley No. 25.506); Estonia (Digital Signatures Act, 2000); Israel (Electronic Signatures Act, 2000); Japan (Law concerning Electronic Signatures and Certification Services, 2001); Lithuania (Law on Electronic Signatures, 2000); Malaysia (Digital Signatures Act, 1997); Poland (Electronic Signatures Act, 2001); and Russian Federation (Law on Electronic Digital Signature (Federal Act No. 1-FZ) of 10 January 2002). One example is Tunisia, which enacted legislation on electronic commerce (Loi relative aux changes et au commerce lectroniques of 9 August 2000) that contains provisions on electronic contracting inspired by Directive 97/7/EC of the European Parliament and of the Council of 20May 1997 on the protection of consumers in respect of distance contracts (Official Journal of the European Communities, No. L 144, 4 June 1997, pp. 19-27).
355
356
Directive 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on certain legal aspects of information society services, in particular electronic commerce, in the internal market (Official Journal of the European Communities, No. L 17, 17 July 2000, p. 1). This seems to be the case for the French (lorsque les parties peuvent y avoir accs), Italian (quando le parte hanno la possibilit di acerdervi ), Portuguese (quando as partes tm possibilidade de aceder a estes) and Spanish (cuando las partes puedan tener acceso a los mismos) texts.
245
357
some of them seem to imply that the addressee must be actually able to retrieve the message.358 Arguably, the linguistic nuances in the various language versions of the EU Electronic Commerce Directive are not substantive. The main difficulty in fact seems to be that the formulation in article 11 of the EU Electronic Commerce Directive does not provide a presumption or indication of the time from which a party should be deemed to have been able to access a message. In legislation enacted to implement the EU Electronic Commerce Directive, Austria,359 Denmark,360 Germany,361 Ireland,362 Italy,363 Spain364 and the United Kingdom,365 for instance, have reproduced the formulation used in article 11 of the EU Electronic Commerce Directive with only slight changes.366 It is not entirely clear which rule applies in countries such as Ireland367 and Italy368 that already had statutory provisions on time of dispatch and receipt of electronic communications before the adoption of the EU Electronic Commerce Directive. The Irish law contains essentially the same rule as article 15 of the MLEC. The new law implementing the EU Electronic Commerce Directive provides that the specific rule on receipt of an order applies notwithstanding the earlier law. The rule in Italy is that an electronic document is deemed to have been
358
For instance, the German text (wenn die Parteien, fr die sie bestimmt sind, sie abrufen knnen). See Bundesgesetz mit dem bestimmte rechtliche Aspekte des elektronischen Geschfts- und Rechtsverkehrs geregelt (E-Commerce-Gesetz ECG) und nderung des Signaturgesetzes sowie der Zivilprozessordnung (Bundesgesetzblatt fr die Republik sterreich, 21 December 2001, p. 1977), sect. 12. See Lov om tjenester i informationssamfundet, herunder visse aspekter af elektronisk handel, sect. 12 (2). Article 11 of the EU Electronic Commerce Directive, supra note 200, has been incorporated in the new section 312e(1) of the German Civil Code (BGR). See European Communities (Directive 2000/31/EC) Regulations 2003, sect. 14 (1) (b). See Decreto legislativo 9 aprile 2003, n. 70, art. 13, para. 3. See Ley 34/2002, de 11 de julio, de servicios de la sociedad de la informacin y de comercio electrnico, art. 28, para. 2. See Electronic Commerce (EC Directive) Regulations 2002 (Statutory Instrument 2002 No.2013), sect. 11 (2). Section 312e(1) of the German Civil Code provides that an order and the acknowledgement of its receipt are deemed to have been received when the parties to whom they are addressed are able to retrieve them under normal circumstances (unter gewhnlichen Umstnden). The same formulation is used in the Austrian law. The Spanish law refers to the addressees ability to become aware (tener constancia) of the message, rather than access (tener acceso) the message. Electronic Commerce Act 2000, sect. 13, paras. (2) (a) and (b)). Decreto del Presidente della Repubblica 10 novembre 1997, n. 513, art. 12, para. 1.
359
360
361
362 363 364
365
366
367 368
246
dispatched by the originator and received by the addressee if it is transmitted to the electronic address indicated by the addressee.369 Although different in formulation, this rule would arguably lead in most cases to the same result as that of article 15 of the MLEC. Most of the countries that have implemented the EU Electronic Commerce Directive, however, did not have statutory rules on time of dispatch and receipt of electronic communications, although case law in some of them already provided criteria for the transposition to an electronic environment of traditional rules on dispatch and receipt. The result is typically consistent with article 15 of the MLEC. This is true even in countries which have not enacted the MLEC, such as Germany, where the courts have regarded the delivery of a message to a partys e-mail address, for instance, as the equivalent of receipt, regardless of whether or not the party had actually accessed the message.370 Incomplete delivery of the text of an electronic communication, owing, for instance to technical malfunctioning of the receiving equipment, does not exclude receipt if there is evidence that the message was transmitted entirely in electronic form.371 While proof of proper dispatch of an electronic communication may be regarded as
369
Il documento informatico trasmesso per via telematica si intende inviato e pervenuto al destinatario se trasmesso allindirizzo elettronico da questi dichiarato (Decreto del Presidente della Repubblica 10 novembre 1997, n. 513). See, for example, Landgericht Nrnberg-Frth, Case No. 2 HK O 9431/01, 7 May 2002, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 158/2003 <www.jurpc. de/rechtspr/20030158.htm> (30 November 2007) In this case, the claimants contract was terminated by the defendant through registered mail, which was later confirmed by an electronic message sent to the claimants e-mail address. The claimant challenged the effectiveness of the e-mail message, arguing that he had not been able to retrieve it, since the message was sent during his holiday and his e-mail account was not accessible through ordinary web browsers. The court held that the claimant had effectively received the message, as it had been delivered to his e-mail address. From that time on, the claimant bore the risk of loss of the message or delay in retrieving the message, for instance due to difficulties in accessing his e-mail account, as such a risk occurred within the claimants sphere of control. See Bundesgerichtshof, Case No. XII ZR 51/99, 14 March 2001, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 167/2001 <www.jurpc.de/rechtspr/20010167.htm> (30 November 2007)In this case, a court of appeals had rejected an appeal because the facsimile received did not contain counsels signature, which would have been contained in the fourth page of the statement of appeal, which the court did not receive. The Federal Court disagreed with the position taken by the court of appeals that only the pages received by it could be taken into account in determining whether the statement of appeal had been delivered within the deadline for its submission. The Federal Court held that when a document was completely (vollstndig) transmitted as a data message (durch elektrische Signale) from the appellants facsimile to the courts machine, but did not get printed completely and without errors, possibly as a result of technical malfunctioning at the destination, the document was deemed to have been received at the time of its transmission
247
370
371
prima facie evidence of its effective receipt by the other party,372 the courts have also stressed that the originator of an electronic communication is not entitled to rely on mere dispatch of the message, which creates no presumption that the message was actually received.373 The fact, however, that the EU Electronic Commerce Directive introduced the accessibility criterion to determine the time of receipt of electronic communications has caused some concern, as it was felt that the rule in the EU Electronic Commerce Directive should not be carried so far as to require actual retrieval of messages, a result that would conflict with existing case law. Indeed, it appears from the consultation process preceding the implementation of the EU Electronic Commerce Directive in some countries that some of the changes introduced in domestic legislation were intended to avoid the impression that receipt of a message required actual retrieval by the addressee. The final rule, it was said, should instead make it clear that only the technical possibility of retrieval was relevant, and not the addressees availability for retrieving the message.374 (iv) The Debate during the Negotiation of the EEC The original draft of the EEC375 followed closely the structure and formulation of article 15 of the MLEC. This was a natural choice in view of the wide acceptance of that provision of the MLEC. It also had the additional advantage of
as a facsimile, as long as the entire content of the document could be established through other means.
372
In particular in the case of highly reliable transmission methods, in view of the current stage of technological development, such as facsimile transmissions (see Oberlandesgericht Mnchen, Case No. 15 W 2631/98, 8 October 1998, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 153/1999 <www.jurpc.de/rechtspr/19990153.htm> (9September 2003). See Oberlandesgericht Dsseldorf, Case No. 23 U 92/02, 4 October 2002, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 167/2003 <www.jurpc.de/rechtspr/ 20030158.htm> (30 November 2007), in a case involving e-mail messages. This point was made expressly in the explanatory note to the draft bill introduced to implement the EU Directive in Austria. The Austrian Bar Association, in its comments on the draft bill, proposed that the law clearly provide that the only controlling factor was the technical retrievability (Abrufbarkeit) and that neither technical malfunctioning on the addressees part nor the addressees absence nor any other obstacle within the addressees sphere of control should hinder the effective receipt of the message (Rechtsanwaltskammer Wien, Stellungnahme zum Bundesgesetz mit dem bestimmte rechtliche Aspekte des elektronischen Geschfts- und Rechtsverkehrs geregelt werden (E-Commerce-Gesetz ECG), 31 August 2001 <www.rakwien.at/import/documents/stellungnahme _ecommerce_fuer_homepage.pdf> (30 November 2007). U.N. document A/CN.9/WG.IV/WP.101, annex.
373
374
375
248
its compatibility with article 24 of the U.N. Sales Convention.376 Surprisingly enough, no other provision of the EEC caused grater debate than the definition of the time of receipt of electronic communications.377 One criticism was that the provision is overly complex and that there may be no practical need for distinguishing between designated and non-designated information systems. That criticism was formulated in the following terms in a statement by the German Bar Association on the preliminary draft of the ECC:378 A point requiring clarification seems to be the [distinction] between an information system designated by the addressee for the receipt of data messages and a system other than the designated system. This distinction is more relevant for Electronic Data Interchange (EDI), but not for e-mail communications. The consequence is that, in the context of e-mail communications, the decisive factor should be the actual entry of the data message in the recipients computer station . On the basis of [article 4, subparagraph (f)], an information system means a system for generating, sending, receiving, storing or otherwise processing data messages. This wide definition encompasses not only a providers web server, but also the computer stations of the providers clients, from which they retrieve their messages or through which they forward their messages to the provider for transmission to the addressees. It is important in this connection to clarify whether the entry in the providers server is sufficient to establish the receipt of the message, or whether the data message needs to be actually retrieved by the addressee at its computer station. This distinction depends on whether the addressee has designated a particular information system for the receipt of the data message and which is the designated system. Normally, the user does not make use of system-specification (Systembenennung) as a receipt-address (Zustelladresse), but rather uses an e-mail address, from which no specified system is recognizable. A specification should not in fact be necessary, since, in protocol-based Internet data communication, messages are
376
Sieg Eiselen, E-Commerce and the CISG: Formation, Formalities and Validity, 6 Vindobona J. Intl Comm L. & Arb. (2002) 310. See U.N. documents A/CN.9/509, paras. 93-98, and A/CN.9/528, paras. 141-151. Stellungnahme der Bundesrechtsanwaltskammer: UNCITRAL-bereinkommensentwurf ber internationale Vertrge, die mit elektronischen Mitteln geschlossen oder nachgewiesen werden, submitted in March 2002 by the Committee on Private International Law and International Procedural Law (Ausschuss Internationales Privat- und Prozessrecht) <www.brak.de/seiten/ pdf/EndfUNCITRAL-Uebereinkentwurf.pdf> (30 November 2007).
249
377 378
transmitted without empty signs through special computers that use appropriate destination tables. On this technical basis, [the rules on receipt in the ECC], would become idle, because the originator is not provided with a specific system of destination, and only requires an e-mail address, which is independent from a computer station. The transmission of the message to another, non-designated system does not occur, since to that end there should have been a designated system in the first place. The problems identified in that analysis might in fact be significant if the notion of information system was understood to refer to the telecommunication channels and infrastructure used to transport messages to their final destination, rather than to the electronic address designated by a party for the purpose of receiving messages. As understood by UNCITRAL, however, the notion of information system was intended to cover the entire range of technical means used for transmitting, receiving and storing information, which, depending on the factual situation, may be a communications network, and in other instances could include an electronic mailbox or even a telecopier.379 Another criticism was that the rule of article 15 of the MLEC might be excessively rigid because the entry of a message in the addressees system or another system designated by the addressee did not always allow the conclusion that the addressee is capable of accessing the message. It was proposed that the notion of entry should be rendered more flexible by adding the notion of accessibility of the electronic communication, which would be given when the communication is capable of being processed and retrieved by the addressee.380 One proposal would have linked the receipt to the time when the retrieval of that electronic communication by the addressee could normally be expected.381 However, there were also objections to that proposal, as the reference to the time when the addressee could normally be expected to retrieve might deviate from the accepted notion of availability of the message for processing within an information system, as an objective test, towards a more subjective approach.382 In considering those issues, UNCITRAL noted that there was no disagreement with the general objective of developing default rules on dispatch and receipt of messages that aim at establishing a fair allocation of risks and responsibilities
379 380 381 382
Guide to Enactment of the MLEC, supra note 105, para. 40. See U.N. document A/CN.9/509, paras. 94 and 96. See U.N. document A/CN.9/528, para. 148. Id., para. 149.
250
between the originator and the addressee.383 It should not be difficult to reach international consensus on the principle that a person who manages an information system, or designates a specific information system for the receipt of electronic communications, even if it is a system operated by a third party, should bear the risk of loss or delay of messages that have effectively entered that system. Where no specific system has been designated, the rule to be ultimately adopted should be such that it would allow a judge or arbitrator called to decide upon a dispute on the time of receipt of an electronic communication to apply a test of reasonableness to the choice of an information system by the originator in the absence of a clear designation by the addressee. (v) The Rule in the ECC UNCITRAL considered at length various options to bridge the gap between the opposing views on that aspect of the ECC. One possibility, which had been proposed even in connection with systems that follow the information theory for the purposes of contract formation, might be to attach a presumption of knowledge (in the sense of accessibility or possibility of knowledge) of an electronic communication to the effective delivery of a communication to the addressees information system. It would thus be for the addressee to adduce evidence that, through no fault of its own or of any intermediary of its choosing, it could not access the communication.384 As eventually agreed, the time of receipt of an electronic communication under the ECC was defined in article 10, paragraph 2, as the time when the electronic communication becomes capable of being retrieved by the addressee at an electronic address designated by the addressee. This is presumed to occur when the electronic communication reaches the addressees electronic address. The rule in this article, too, is based on a similar rule of the MLEC.385 The differences in wording between the ECC and the MLEC are more evident in this provision than in the definition of dispatch. In essence, however, both texts should achieve the same result.
Id., para. 145. Giovanni Comand and Salvatore Sica, Il commercio elettronico (Turin: G. Giappichelli, 2001) 57. The authors propose this approach as a combined interpretation of articles 1136 (which requires the offerors knowledge of the acceptance for contract formation) and 1135 (which provides that the partys knowledge is presumed when the acceptance was communicated to an appropriate address), both of the Italian Civil Code, and article 12, subparagraph (l), of Decree No. 513/1997 (which provides that an electronic document is deemed to have been received by the addressee when it has been transmitted to the electronic address indicated by it). The authors point out that such an interpretation would also be in line with the notion of accessibility of a data message for the purposes of the EU Directive.
383 384
385
Article 15, paragraph 2.

251
Indeed, the requirement that an electronic communication should be capable of being retrieved should not be seen as adding an extraneous subjective element to the rule contained in article 15 of the MLEC. In fact, entry in an information system is understood under article 15 of the MLEC as the time when an electronic communication becomes available for processing within that information system,386 which is arguably also the time when the communication becomes capable of being retrieved by the addressee. Whether or not an electronic communication is indeed capable of being retrieved is a factual matter outside the ECC. UNCITRAL took note of the increasing use of security filters (such as spam filters) and other technologies restricting the receipt of unwanted or potentially harmful communications (such as communications suspected of containing computer viruses). The presumption that an electronic communication becomes capable of being retrieved by the addressee when it reaches the addressees electronic address may be rebutted by evidence showing that the addressee had in fact no means of retrieving the communication.387 Following the example of a number of domestic laws, the ECC uses the term electronic address instead of information system, which was the expression used in the MLEC. In practice, the new terminology, which appears in other international instruments such as the Uniform Customs and Practices for Documentary Credits (UCP 500) Supplement for Electronic Presentation (eUCP),388 should not lead to any substantive difference. Indeed, the term electronic address may, depending on the technology used, refer to a communications network and, in other instances, could include an electronic mailbox, a telecopy device, or another specific portion or location in an information system that a person uses for receiving electronic messages.389 The notion of electronic address, like the notion of information system, should not be confused with information service providers or telecommunications carriers that might offer intermediary services or technical support infrastructure for the exchange of electronic communications.390 The ECC retained the distinction made in article 15 of the MLEC between delivery of messages to specifically designated electronic addresses and delivery of messages to an address not specifically designated. In the first case, the rule of receipt is essentially the same as under article 15, paragraph (2)(a)(i), of the MLEC, that is, a message is received when it reaches the addressees electronic
386 387
See Guide to Enactment of the MLEC, supra note 105, para. 103. See Official Records of the General Assembly Sixtieth Session, Supplement No. 17 (UN. document A/60/17), para. 80. See also U.N. document A/CN.9/571, paras. 149 and 160. See James E. Byrne and Dan Taylor, ICC Guide to the eUCP (Paris: ICC, 2002) 54. See U.N. document A/CN.9/571, para. 157. See U.N. document A/CN.9/528, para. 149.
388 389 390
252
address (or enters the addressees information system, in the terminology of the MLEC). The ECC does not contain specific provisions as to how the designation of an information system should be made or whether the addressee could make a change after such a designation. In distinguishing between designated and non-designated electronic addresses, paragraph 2 aims at establishing a fair allocation of risks and responsibilities between originator and addressee. In normal business dealings, parties who own more than one electronic address could be expected to take the additional care of designating a particular one for the receipt of messages of a certain nature and to refrain from disseminating electronic addresses that they rarely use for business purposes. By the same token, however, parties should be expected not to address electronic communications containing information of a particular business nature (e.g. acceptance of a contract offer) to an electronic address that they knew or ought to have known would not be used to process communications of such a nature (e.g. an e-mail address used to handle consumer complaints). It would not be reasonable to expect that the addressee, in particular large business entities, should pay the same level of attention to all of the electronic addresses that it uses.391 One noticeable difference between the ECC and the MLEC, however, concerns the rules for receipt of electronic communications sent to a non-designated address. The MLEC distinguishes between communications sent to an information system other than the designated one and communications sent to any information system of the addressee in the absence of any particular designation. In the first case, the MLEC does not regard the message as being received until the addressee actually retrieves it. The rationale behind this rule is that if the originator chose to ignore the addressees instructions and sent the electronic communication to an information system other than the designated system, it would not be reasonable to consider the communication as having been delivered to the addressee until the addressee has actually retrieved it. In the second situation, however, the underlying assumption of the MLEC was that, for the addressee, it was irrelevant to know in advance to which information system the electronic communication would be sent. In such a case, it would be reasonable to presume that it would accept electronic communications through any of its information systems. In this particular situation, the ECC follows the approach taken in a number of domestic enactments of the MLEC and treats both situations in the same manner. Thus for all cases where the message is not delivered to a designated electronic address, receipt under the ECC only occurs when (a) the electronic communication becomes capable of being retrieved by the addressee (by reaching
391
See U.N. document A/CN.9/528, para. 145.

253
an electronic address of the addressee) and (b) the addressee actually becomes aware that the communication was sent to that particular address. In cases where the addressee has designated an electronic address but the communication is sent elsewhere, the rule in the ECC is not different in result from article 15, paragraph (2)(a)(ii), of the MLEC, which itself requires, in those cases, that the addressee retrieves the message (which in most cases would be the immediate evidence that the addressee has become aware that the electronic communication has been sent to that address). The only substantive difference between the ECC and the MLEC, therefore, concerns the receipt of communications in the absence of any designation. In this particular case, UNCITRAL agreed that practical developments since the adoption of the MLEC justified a departure from the original rule. It also considered, for instance, that many persons nowadays have more than one electronic address and could not be reasonably expected to anticipate receiving legally binding communications at all of the addresses that they maintain.392 The addressees awareness that the electronic communication has been sent to a particular non-designated address is a factual manner that could be proven by objective evidence, such as a record of notice given otherwise to the addressee or a transmission protocol or other automatic delivery message stating that the electronic communication had been retrieved or displayed at the addressees computer. The rules on dispatch of electronic communications under the ECC were far less contentious than the definition of time of receipt. Article 10, paragraph 1, of the ECC defines the dispatch of an electronic communication as the time when it: leaves an information system under the control of the originator or of the party who sent it on behalf of the originator or, if the electronic communication has not left an information system under the control of the originator or of the party who sent it on behalf of the originator, the time when the electronic communication is received. This provision follows, in principle, the rule on dispatch set out in article 15, paragraph 1, of the MLEC.393 One minor difference is that the formulation finally adopted in the EEC refers to the time when the electronic communication leaves an information system under the control of the originator, rather than the time when the electronic communication enters an information system outside the control of the originator, as had been used in the MLEC. This shift in focus was done so as to more closely mirror the notion of dispatch in a non-electronic environment, which is understood in most legal systems as
392
Official Records of the General Assembly Sixtieth Session, Supplement No. 17 (U.N. document A/60/17), para. 82. Unless otherwise agreed between the originator and the addressee, the dispatch of a data message occurs when it enters an information system outside the control of the originator or of the person who sent the data message on behalf of the originator.
393
254
the time when a communication leaves the originators sphere of control.394 In practice, the result under the EEC should be the same as under article 15, paragraph 1, of the MLEC, since the most easily accessible evidence to prove that a communication has left an information system under the control of the originator is the indication, in the relevant transmission protocol, of the time when the communication was delivered to the destination information system or to intermediary transmission systems. Another minor difference is that article 10 of the ECC also covers situations where an electronic communication has not left an information system under the control of the originator. This hypothesis, which was not covered in article 12 of the MLEC, may happen, for example, when the parties exchange communications through the same information system or network so that the electronic communication never really enters a system under the control of another party. In such cases, dispatch and receipt of the electronic communication coincide. 3. Automated Information Systems Automated computer systems, sometimes called electronic agents, are being increasingly used in electronic commerce and have caused extensive theoretical discussion.395 They have made some scholars revisit traditional common law theories of contract formation to assess their adequacy to contracts that come into being without human intervention.396 Existing uniform law conventions do not seem in any way to preclude the use of automated systems, for example, for issuing purchase orders or processing purchase applications. This seems to be the case in connection with the U.N. Sales Convention, which allows the parties to create their own rules,397 for example, in an EDI trading partner agreement regulating the use of electronic agents. The MLEC, too, lacks a specific rule on the matter. While nothing in the MLEC seems to create obstacles to the use of fully automated systems, it does not deal specifically with those systems, except for the general rule on attribution in article13, paragraph 2 (b).398
394
See U.N. document A/CN.9/571, para. 142. See generally Fabio Bravo, Contrattazione telematica e contrattazione cibernetica (Milan, Giuffr, 2007) 170-347, and the literature cited in p. 182, note 18.
395
396
See Anthony J. Bellia, Jr., Contracting with Electronic Agents, 50 Emory L. J. (2001) 1047. U.N. Sales Convention, art. 9. Article 13, paragraph 2 (b), of the MLEC provides that, as between the originator and the addressee, a data message is deemed to be that of the originator if it was sent by an information system programmed by, or on behalf of, the originator to operate automatically.
255
397 398
(a) Responsibility for Automated Information Systems In an early discussion of the matter, UNCITRAL was of the view that, while the expression electronic agent had been used for purposes of convenience, the analogy between an automated system and a sales agent was not appropriate. General principles of agency law (for example, principles involving limitation of liability as a result of the faulty behaviour of the agent) could not be used in connection with the operation of such systems. UNCITRAL was also of the view that, as a general principle, the person (whether a natural person or a legal entity) on whose behalf a computer was programmed should ultimately be responsible for any message generated by the machine.399 At present, the attribution of actions of automated information systems to a person or legal entity is based on the paradigm that an automated information system is capable of performing only within the technical structures of its preset programming. However, at least in theory it is conceivable that future generations of automated information systems may be created with the ability to act autonomously, and not just automatically. That is, through developments in artificial intelligence, a computer may be able to learn through experience, modify the instructions in their own programs, and even devise new instructions.400 That possibility has led some commentators to go as far as to advocate the attribution of at least some elements of legal personality to automated computer systems401 or to a transposition of the general theory of agency to computer transactions.402 Other commentators, however, seem less inclined to impose liability upon machines and prefer to apply general principles of law, such as reliance and good faith, to establish the link between the computer and the person on whose behalf it functions.403 Even if no modification appears to be needed in general rules of contract law, it was felt that it would be useful for a new international instrument to make it clear that the actions of automated systems programmed and used by people will
399
See Report of the Working Group on the work of its thirty-eighth session (New York, 12-23 March 2001) (U.N. document A/CN.9/484), paras. 106 and 107. Tom. Allen and Robin. Widdison, Can Computers Make Contracts?, 9 Harv. J. L. & Tech. (1996) 25. For instance, Lawrence B. Solum, Legal Personhood for Artificial Intelligences, 70 N.C.L. Rev. (1992) 1231; and Leon E. Wein, The Responsibility of Intelligent Artifacts: Toward an Automated Jurisprudence, 6 Harv. J. L. & Tech. (1992) 103. David D. Wong, The Emerging Law of Electronic Agents: E-commerce and Beyond, 33 Suffolk L. Rev. (1999) 83. See Jean-Franois Lerouge, The Use of Electronic Agents Questioned Under Contractual Law: Suggested Solutions on a European and American Level, 18 John Marshall J. Comp. & Info. L. (1999) 403. Similarly, from a common law perspective, see C. C. Nicoll, Can Computers Make Contracts?, 1998 J. Bus. L. (1998) 42.
400
401
402
403
256
bind the user of the system, regardless of whether human review of a particular transaction has occurred. Accordingly, article 12 of the ECC recognizes that contracts may be formed as a result of actions by automated message systems (electronic agents), even if no natural person reviewed each of the individual actions carried out by the systems or the resulting contract. (b) Errors in Messages and Communications Closely related to the use of automated computer systems is the question of mistakes and errors in electronic commerce. Such errors may be either the result of human actions (for example, typing errors) or the consequence of malfunctioning of the information system used. (i) Human Errors The MLEC was not concerned with substantive issues that arise in contract formation. Therefore, it did not deal with the consequences of mistake and error in electronic contracting. However, uniform legislation enacting the MLEC, such as the UECA in Canada and UETA in the United States, contain provisions dealing with errors made by natural persons when dealing with an automated computer system of another person. The relevant provisions in UECA (sect. 22) and in UETA (sect. 10) set out the conditions under which a natural person is not bound by a contract in the event that the person made a material error. The rationale for provisions such as those contained in UECA and in UETA seems to be the relatively higher risk that an error made in transactions involving a natural person, on the one hand, and an automated computer system, on the other, might not be noticed as easily in transactions that involve only natural persons. Errors made by the natural person in an entirely automated environment may become irreversible once acceptance is dispatched. In favour of formulating a substantive rule on the consequences of computer errors, it could be said that other international texts deal with the consequences of errors for the validity of the contract, albeit restrictively.404 However, a counter-argument could be that a provision of that type would interfere with well-established notions of contract law and might not be appropriate in the context of an instrument specifically concerned with electronic commerce, in view of the risk of duplication of legal regimes. A slightly different approach might be to envisage only provisions that promote best business practices, such as provisions that would induce businesses to make available procedures for detecting and correcting errors in electronic
404
Such as the UNIDROIT Principles of International Commercial Contracts (see arts. 3.5 and 3.6).
257
contract negotiation, without dealing with the consequences of errors for the validity of the contract. For example, article 11, paragraph 2, of the EU Electronic Commerce Directive creates such an obligation for providers of information society services. It is recognized, however, that, in implementing the EU Electronic Commerce Directive, States have added various consequences for a partys failure to provide procedures for detecting and correcting errors in electronic contract negotiation. For example, in Austria,405 Ireland,406 Italy407 and Spain,408 such failure constitutes an administrative offence and subjects the infringer to payment of a fine.409 In Germany,410 the consequence is an extension of the period within which a consumer may avoid a contract, which only begins to run from the time when the merchant has fulfilled its obligations. A similar consequence is provided in the United Kingdom, where the customer is entitled to rescind the contract unless any court having jurisdiction in relation to the contract in question orders otherwise on the application of the service provider.411 UNCITRAL considered carefully the various possibilities. The view that eventually prevailed was that, while the ECC should avoid interfering with the law of mistake, it was important to deal with the a type of error specific to electronic commerce, in view of the relatively higher risk of human errors being made in communications exchanged with automated message systems.412 Article 14 of the ECC does not oblige the operators of automated message systems to make available procedures for detecting and correcting errors in electronic contract negotiation. Nevertheless, it authorizes a party who makes an error to withdraw the portion of the electronic communication where the error was made if the automated message system did not provide the person with an opportunity to correct errors. This right is subject to two general conditions: the person must notify the other party as soon as possible and must not have used or received any material benefit or value from the goods or services received from the other party.
405 406 407 408
E-Commerce-Gesetz ECG, sect. 26, para. 4. European Communities (Directive 2000/31/EC) Regulations 2003, regulation 13 (5). Decreto legislativo 9 aprile 2003, n. 70, art. 21, para. 1. Ley 34/2002, de 11 de julio, de servicios de la sociedad de la informacin y de comercio electrnico , arts. 38 and 39. E-Commerce-Gesetz ECG, sect. 26, para. 4. Brgerliches Gesetzbuch, sect. 312e, para. 1, first sentence. Electronic Commerce (EC Directive) Regulations 2002 (Statutory Instrument 2002 No. 2013), regulation 15. See U.N. documents A/CN.9/509, para. 105 and A/CN.9/548, para. 17.
409 410 411
412
258
(ii) Errors Generated by Information Systems Another issue considered by UNCITRAL was whether the ECC should deal with errors made by the automated system itself. At its initial discussion of the issue, UNCITRAL was of the view that errors made by any such system should ultimately be attributable to the persons on whose behalf they operated. Nevertheless, UNCITRAL recognized that there might be circumstances that justify a mitigation of that principle, such as when an automated system generates erroneous messages in a manner that could not have reasonably be anticipated by the person on whose behalf the messages were sent. It was suggested that elements to be taken into account when considering possible limitations for the responsibility of the party on whose behalf the automated system was operated include the extent to which the party had control over the software or other technical aspects used in programming the system.413 The complexity of the issues involved can be illustrated by three very similar cases where German courts arrived at opposing results.414 The cases related to sales of goods erroneously offered over the Internet for a price below the price intended by the seller. They all involved interactive applications that generated automatic replies from the seller stating that the customers order (Auftrag) would be immediately carried out (ausgefhrt). It was surmised that the errors were computer-made and had occurred during processing and posting of the sellers information on web sites maintained by independent Internet service providers. The courts affirmed the principle that automated communications were attributable to the persons on whose behalf the system had been programmed and in whose names the messages were sent. The courts consistently regarded the advertisement of goods via the Internet as a mere invitation to treat (invitatio ad offerendum) and considered that a binding contract would only come into being once the seller had accepted the buyers bid (offer). The courts further affirmed the legal value of the messages sent by the automatic reply function as binding expressions of intention (Willenserklrung) and valid acceptances for purposes of contract formation. Nevertheless, one court of appeals found that the pricing error in the Internet advertisement vitiated the sellers acceptance and rendered it invalid.415 Two
413 414
See U.N. document A/CN.9/484, paras. 108. Oberlandesgericht Frankfurt, 20 November 2002, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 91/2003 <www.jurpc.de/rechtspr/20030091.htm>; Landgericht Kln, 16 April 2003, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok138/2003 <www.jurpc.de/rechtspr/20030138.htm> and Amtsgericht Westerburg, Case No. 21 C 26/03, 14 March 2003, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 184/2003 <www.jurpc.de/rechtspr/20030184.htm> (30 November 2007). Die unrichtige bermittlung der invitatio ad offerendum wirkte bei der infolge der entsprechenden Programmierung automatisch erstellten und dann an den Rechner des Klgers elektronisch
259
415
district courts, in turn, regarded the invitation to treat expressed through the Internet advertisement as a separate legal act from the eventual acceptance of the buyers offer, so that the error in the first instance did not affect the validity of the sellers acceptance.416 While some factual differences between the cases might have influenced their outcome,417 the discrepancy between the judgements seems to result from conflicting views regarding the allocation of risks for malfunctioning of commercial web sites. Courts in Singapore, in turn, have tended to consider a number of factors before deciding on an appropriate allocation of the risk of errors in electronic commerce, including (a) the need to observe the principle of upholding rather than destroying contracts, (b) the need to facilitate the transacting of electronic commerce, and (c) the need to reach commercially sensible solutions while respecting traditional principles applicable to instances of genuine error or mistake.418 Court hold that the party who selects the means of communication should bear the consequences of any unexpected events. However, the general law on errors still applied and the risk may be shifted to the other party. In doing so, the courts also take into account the conduct of the beneficiary party and consider whether such party knew or ought to have known that an error had been made by the system.419
bermittelten Annahmeerklrung der Beklagten noch fort (Oberlandesgericht Frankfurt).

416
Eine auf diesen Irrtum gesttzte Anfechtung kommt gleichwohl nicht in Betracht, weil der Irrtum nach dem klgerischen Sachvortrag allenfalls bei der Einstellung der Preisangaben ins Internet, nicht aber zum massgeblichen Zeitpunkt der Abgabe der Willenserklrung vorgelegen hat (Landgericht Kln; similarly, Amtsgericht Westerburg). Such as the fact that in the Frankfurt case the erroneously advertised price represented 1percent of the ordinary value of the product, whereas in the Cologne case the court found that the price, which arguably fell some 50 per cent below the ordinary market price, was not extraordinary (keine Seltenheit) for an Internet sale. Chwee Kin Keong and others v. Digilandmall.com Pte Ltd, High Court, [2004] SGHC 71, 12 April 2004, confirmed by the Court of Appeals ([2005] SGCA 2, 13 January 2005). Id. In this case, a website operated in Singapore by Digiland, a Singapore company (the respondent), advertised on sale a laser printer worth Singapore dollars (S$) 3,854 for only S$ 66. This pricing error was due to the uploading on the website of a set of figures prepared for a training template. By the time the mistake was discovered a few days later 784 individuals (six of whom were the appellants for this case) had already placed 1008 purchase orders via the Internet for 4086 laser printers. In total, they ordered 1,606 printers for a total price of S$105,996 against a market value of S$ 6,189,524. After the discovery of the pricing error on the website, Digiland refused to honour the contracts on the basis that there was a mistake in the posted price. The courts found that each of the six buyers had actual knowledge that there was a mistake in the pricing on the websites, and that the contracts were void on the ground of unilateral mistake. VK Jajah, JC noted that [t]he stark gaping difference between the price posting and the market price of the
417
418
419
260
4. Incorporation and Availability of Contract Terms One additional question concerning contract formation through the intervention, in whole or in part, of automated information systems is the legal effect of the incorporation by reference of contractual clauses accessible through a hypertext link. Another issue relates to the availability and retention or reproduction of contract terms. (a) Incorporation of Terms and Conflicting Contract Terms The question of incorporation of contract terms is dealt with in article 5 bis of the MLEC. That provision sets out the general rule that information shall not be denied validity or enforceability solely because it has been incorporated by reference. Domestic laws typically go beyond that general rule and set down the substantive conditions for the enforceability of terms incorporated by reference. In doing so, it seems that courts make a distinction between terms formulated by one party, which seeks to enforce them against the other party, and terms established by a third party and intended to apply to all transactions being negotiated in a particular market or through a particular facility offered by such third party. In the first situation, courts in many legal systems seem not to automatically assume a partys acceptance of the terms incorporated by reference. Courts have in fact required a specific act of incorporation and held that the mere existence of such terms in an easily accessible resource (such as a partys web site) was not sufficient to effectively incorporate those terms into a contract in which they were not otherwise referred to.420 The courts do not seem to have categorically excluded the possibility of incorporating terms by the mere clicking of an I agree button on a computer screen.421 Yet, courts have often required unambiguous demonstration that the accepting party either had an opportunity to actually access and read those terms or that the party was adequately alerted, through a conspicuously placed notice or otherwise, of the existence of those terms and their relevance for the transaction in question.422
laser printer would have made it obvious to any objective person that something was seriously amiss.
420
Hanseatisches Oberlandesgericht Hamburg, Case No. 3 U 168/00, 13 June 2002, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 288/2002 <www.jurpc.de/ rechtspr/20020288.htm> (30 November 2007). See supra note 300. For instance, Specht v. Netscape Communications Corp., 150 F. Supp. 2d. 585, affirmed in Specht v. Netscape Communications Corporation and America Online, Inc., 306 F. 3d 17 (2d Cir. 2002).
261
421 422
In some legal systems, courts seem to establish a distinction between contractual terms developed by one of the parties and contract terms developed by another entity (a third party) that offers the electronic platform for the parties to conduct their negotiations. This question has arisen, for instance, in connection with Internet auctions in Germany. In an early case, a district court in Germany found that a person offering goods through an Internet auction platform had not made a binding offer, but had merely invited offers in respect of the goods during a set period of time.423 The fact that the general conditions of the operator of the auction platform qualified the offer of goods for auction as binding and irrevocable was not regarded as being controlling. That decision was later reversed by the court of appeal, which found that there was no need for the parties to specifically refer to or otherwise incorporate into their communications the general conditions of the operator of the auction platform, which highlighted the binding character of offers of goods for auction. Both parties should be deemed to have accepted those general conditions beforehand.424 This understanding was followed by other courts425 and was also affirmed by the Federal Court (Bundesgerichtshof), which held that the seller could have avoided, if it had wished, the impression of being bound by its offer by introducing an appropriate statement in its automatic reply messages. However, a reservation to the general conditions that was not recognizable as such by the addressees of the offer could not be held against them.426 Another question concerning contract formation through automated information systems is the legal effect of contract terms displayed on a video screen but not necessarily expected by a party. Directly related to this question is the issue of the battle of the forms, which may be a serious problem in the context of electronic transactions, in particular where fully automated systems are used and no means are provided for reconciling conflicting contractual terms. Neither of these issues is dealt with in article 5 bis of the MLEC, which only contains a general provision intended to uphold the legal effect of information
423
Landgericht Mnster, 21 January 2000, Case No. 4 O 424/99, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 60/2000 <www.jurpc.de/rechtspr/20000060.htm> (30 November 2007). Oberlandesgericht Hamm, 14 December 2000, Case No. 2 U 58/00, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 255/2000 <www.jurpc.de/rechtspr/20000255. htm> (30 November 2007). Amtsgericht Hannover, 7 September 2002, Case No. 501 C 1510/01, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 299/2002 <www.jurpc.de/rechtspr/20020299. htm> (30 November 2007). Bundesgerichtshof, 7 November 2001, Case No. VIII ZR 13/01, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 255/2001 <www.jurpc.de/rechtspr/20010255. htm> (30 November 2007).
424
425
426
262
incorporated by reference. Furthermore, neither the MLEC nor the U.N. Sales Convention expressly provides a solution for the well-known problem of battle of the forms.427 The magnitude of the problem and the profound differences, both in policy and approach, in the manner in which those issues are addressed under domestic laws428 suggest that there would be significant obstacles for international harmonization. UNCITRAL therefore decided to leave these questions outside the scope of the ECC. (b) Availability of Contract Terms Except for purely oral transactions, most contracts negotiated through traditional means result in some tangible record of the transaction to which the parties can refer in case of doubt or dispute. In electronic contracting, such a record, which may exist as an electronic communication, may only be temporarily retained or may be available only to the party through whose information system the contract was concluded. Thus, some recent legislation on electronic commerce, such as the EU Electronic Commerce Directive, requires that a person offering goods or services through information systems accessible to the public should provide means for storage or printing of the contract terms. The rationale for creating such specific obligations seems to be an interest in enhancing legal certainty, transparency and predictability in international transactions concluded by electronic means. Thus, it may not be unreasonable to require certain information to be provided or technical means to be offered in order to make available contract terms in a way that allows for their storage and reproduction, in the absence of a prior agreement between the parties, such as a trading partner agreement or other type of agreement. No similar obligations exist under the U.N. Sales Convention or most international instruments dealing with commercial contracts. During its deliberations on the ECC, UNCITRAL was therefore faced with the question of whether, as a matter of principle, it should propose specific obligations for parties conducting business electronically that may not exist when they contract through more traditional means. One objection to the inclusion of disclosure obligations in a new international uniform law instrument was that the consequences of a
427
The U.N. Sales Convention offers an implicit solution for the question in article 19, paragraph 2. Specific rules on the matter can be found in the UNIDROIT Principles of International Commercial Contracts, Unidroit, Rome, 1994. An overview of the differences between American and European law can be found in James R. Maxeiner, Standard Terms Contracting in the Global Electronic Age: European Alternatives, 28 Yale J. Intl L. (2003) 109.
263
428
party failing to comply with any such obligation would have to be considered and well defined.429 The EU Electronic Commerce Directive does not prescribe what the consequences are if information society services fail to comply with its provisions on this point. In the absence of uniform sanctions, EU member States have provided a variety of different consequences in their national laws.430 The laws of Austria,431 Ireland,432 Italy433 and Spain,434 for example, provide that failure to make the contract terms available constitutes an administrative offence and subject the infringer to payment of a fine.435 In the United Kingdom, the law distinguishes between disclosure of information and availability of contract terms. In the first case, those duties shall be enforceable, at the suit of any recipient of a service, by an action against the service provider for damages for breach of statutory duty.436 In the second case, the customer may seek an order from any court having jurisdiction in relation to the contract requiring that service provider to comply with that requirement.437 In Germany, the consequence is an extension of the period within which a consumer may avoid the contract, which does not begin to run until the time when the merchant has complied with its obligations.438 In most cases, these sanctions do not exclude other consequences that may be provided in law, such as sanctions under fair competition laws.439
429
See Legal aspects of electronic commerce Electronic contracting: provisions for a draft convention Comments by the International Chamber of Commerce (U.N. document A/CN.9/WG.IV/ WP.96), annex, p. 6. This has been one of the arguments put forward by the International Chamber of Commerce in its criticism of the corresponding provision in an early draft of the convention (U.N. document A/CN.9/WG.IV/WP.101, p. 6). E-Commerce-Gesetz ECG, sect. 26, para. 4. European Communities (Directive 2000/31/EC) Regulations 2003, regulation 7 (2). Decreto legislativo 9 aprile 2003, n. 70, art. 21, para. 1. Ley 34/2002, de 11 de julio, de servicios de la sociedad de la informacin y de comercio electrnico , arts. 38 and 39. E-Commerce-Gesetz ECG, sect. 26, para. 4. Electronic Commerce (EC Directive) Regulations 2002 (Statutory Instrument 2002 No. 2013), sect. 11 (2), regulation 13. Electronic Commerce (EC Directive) Regulations 2002 (Statutory Instrument 2002 No. 2013), sect. 11 (2), regulation 14. Brgerliches Gesetzbuch, sect. 312e, para. 1, first sentence. German courts have decided, for example, that failure by a company to disclose its name and address, as required under the German Distance Sales Law (Fernabsatzgesetz), which is based largely on another EU directive, represented an act of unfair competition against which the violators competitors could seek an injunction (Oberlandesgericht Frankfurt, Case No. 6 W 37/01, 17 April 2001, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 135/2001 <www.jurpc.de/rechtspr/20010135.htm> (30 November 2007).
430
431 432 433 434
435 436
437
438 439
264
But what kind of sanctions could be contemplated in a uniform commercial law instrument? On the one hand, rendering commercial contracts invalid or unenforceable for failure to comply with disclosure obligations may be an undesirable and unreasonably intrusive solution. On the other hand, providing for other types of sanctions, such as tort liability or administrative sanctions, would probably be outside the scope of the work that UNCITRAL had done so far. Moreover, the operation of sanctions of this type presupposes administrative structures and enforcement measures that cannot be provided in an international uniform law instrument. The views within UNCITRAL were for some time divided between two groups. On the one side were those who believed that obligations to disclose certain information should be left for international industry standards or guidelines, or, at the national level, for regulatory regimes governing the provision of online services, especially under consumer protection regulations, but should not be included in an international convention dealing with electronic contracting.440 On the other side were those who believed that disclosure obligations of certain basic information about a business entity would promote good business practices and enhance confidence in electronic commerce.441 The first group eventually prevailed within UNCITRAL, and the final text of the ECC defers the matter to domestic law by providing, in its article 13, that nothing in the ECC affects the application of any rule of law that may require a party that negotiates some or all of the terms of a contract through the exchange of electronic communications to make available to the other party those electronic communications that contain the contractual terms in a particular manner or relieves a party from the legal consequences of its failure to do so.
Chapter II. Particular Issues in Government Contracting

Electronic communications are being increasingly used in public procurement, and it is generally believed that they help to increase competition and streamline public purchasing, particularly in terms of time and costs savings.442
440 441 442
See U.N. document A/CN.9/509, para. 63. See U.N. document A/CN.9/509, para. 63. See, Directive 2004/17/EC of the European Parliament and of the Council of 31 March 2004 coordinating the procurement procedures of entities operating in the water, energy, transport and postal services sectors (Official Journal of the European Union, No. L 134, 30 April 2004, p. 1), Preamble paragraph 20; see also Directive 2004/18/EC of the European
265
Many procurement laws were adopted long before the development of new information and communications technologies or at a time when they were not yet very widely used. Thus, procurement laws are often not primarily concerned with legal issues related to the use of new technologies, and the wording of a number of domestic procurement rules indicates that they were conceived against the background of communications, record-keeping and evidentiary systems that were largely based on information recorded on tangible media (essentially, written on paper).443 Like in other areas of the law, legislators and policymakers had to adapt existing rules and procedures, which to a very large extend rely on, or presuppose, paper-based communications, to an electronic environment. General legislation on electronic commerce, such as legislation based on the MLEC may provide some guidance in respect of some of those issues. However, more specific solutions may be needed to enable electronic procurement. In practice, countries do not seem to rely exclusively on the general legislation on electronic transactions to enable electronic communications in the procurement process. Indeed, in some countries, general rules on electronic communications may be excluded in connection with procurement activities of public bodies,444 or have been incorporated into the existing framework for private law in such a way that they do not seem to apply automatically to government functions.445 The reason for this particular treatment is that procuring entities in many countries have an interest in establishing conditions for the use of electronic communications taking into account their respective levels of sophistication, security concerns and other relevant factors. Some countries have enacted both rules governing the use of electronic communications in Government (including procurement), and general legislation on electronic commerce, some of which is declared to apply to the public sector, while other aspects do not seem to have been conceived for Government use. In a number of other countries, however, general legislation on electronic commerce and electronic transactions is expressly intended to bind Government, except for a number of specifically excluded
Parliament and of the Council of 31 March 2004 coordinating the procurement procedures of entities operating in the water, energy, transport and postal services sectors (Official Journal of the European Union, No. L 134, 30 April 2004, p. 114), Preamble para. 12.
443
Examples include references to documentary evidence and similar concepts, or rules on preparation, modification, withdrawal, submission and opening of tenders, particularly in view of common requirements, such as that tenders be submitted in a sealed envelope. United States (Electronic Signatures in Global and National Commerce Act, Public Law 106-229, June 30, 2000, sect. 102(b)). This is the case, for example, in France (see Loi no. 2000-230, of 13 March 2000, Journal officiel, 14 March 2000) and Mexico (see Decreto por el que se reforman y adicionan diversas disposiciones del Cdigo Civil para. el Distrito Federal of 26 April 2000).
444
445
266
areas,446 but even in countries that follow this approach electronic commerce legislation often contains specific rules for the use of electronic communications in governmental functions,447 or contemplate the enactment of specific regulations for that purpose.448 Lastly, some countries not all of which have adopted a general framework for electronic commerce and electronic transactions have enacted detailed provisions on electronic communications in the procurement process.449 The use of electronic applications in the procurement process may fall generally under two broad categories: electronic tendering and electronic purchasing systems.450 Electronic tendering systems can be defined as systems developed to support carefully regulated competitive bidding processes based on detailed bidding documents and technical specifications.451 Electronic tendering systems are said to be particularly suitable for procurement of large public works, of production capabilities such as a power plant, of performance capabilities such as large
446
Australia (Electronic Transactions Act 1999); Ireland (Electronic Commerce Act, 2000); and New Zealand (Electronic Transactions Act 2002). India (Information Technology Act, 2000, sect. 4-10); Ireland (Electronic Commerce Act, 2000, sect. 12); Mauritius (Electronic Transactions Act 2000, sect. 40); Philippines (Electronic Commerce Act 2000, sect. 27-29); and Singapore (Electronic Transactions Act 1998, sect. 47). Republic of Korea (Framework Law on Electronic Commerce 1999, art. 27); Thailand (Electronic Transactions Act 2001, sect. 35); and Venezuela (Decreto no. 1024 de 10 de febrero de 2001 Ley sobre mensajes de datos y firmas electrnicas, art. 3). In the Philippines, in addition to general legislation on electronic commerce (Electronic Commerce Act 2000), there are specific rules concerning the use of electronic communications in the procurement process under Republic Act No. 9184 (known as The Government Procurement Reform Act <www.procurementservice.net/English/ AboutEPS/ RepublicAct9184-GPRA.pdf> and its Implementing Rules and Regulations <www.ato.gov. ph/Downloads/RA9184_rules.pdf> (12 November 2007)). Brazil, however, does not have general legislation on electronic commerce or the legal value of electronic communications, but has enacted specific legislation on certain procurement application of information technologies, such as electronic reverse auctions and on electronic catalogues under Lei no. 10.520, of 17 July 2002 <www.planalto.gov.br/ ccivil_03/Leis/2002/L10520.htm>) and Decreto no. 3.697, of 21December 2000 <www.planalto.gov.br/ccivil_03/decreto/ D3697.htm>). See Eduardo Talero, Electronic Government Procurement: Concepts and Country Experiences, World Bank Discussion Paper (September 2001), paras. 30-40 <wbln0018.worldbank. org/OCS/egovforum.nsf/c3c9b2819079a45d852569bc007722a0/e5596442988c ccfd85256af5006af56a/$FILE/ATTUQ5LL/egpdiscpaperdraft16.pdf> (12 November 2007). See also Elaine Curran, Andrea Bernert, Anke Wiegand, Electronic Procurement in the Public Sector: Factsheet on Latest Developments in E-procurement in the EU and its Member States <www.eic.ie/downloads/e_procurement.pdf> (12 November 2007). Talero, supra note 450, para. 31.
267
447
448
449
450
451
information systems, or of sophisticated services such as design and management of virtual private communication networks. Electronic tendering systems may provide various types of support functions for the conduct of procurement proceedings. Depending on the extent of use of information technology in a country, the systems may evolve through the following stages: (a) First stage: In this stage, the use of electronic communications is essentially limited to making procurement-related information available through electronic means, such as Internet web sites. Such a system is not very complex technologically and requires minimum or no legislative change; (b) Second stage: In this stage, invitations to prequalify and solicitation documents are made available electronically and may be either downloaded by suppliers from a designated website or are transmitted by e-mail upon request. In addition, a number of other actions may be carried out electronically, such as the online registration of suppliers and contractors and notices of impending business opportunities through electronic mail based on supplier profiles; (c) Third stage: This stage involves conversion to full electronic processing and requires substantially more complex technology, operating capabilities and legal and regulatory infrastructure. In this stage, all pre-bidding steps are accomplished electronically invitation to participate in the procurement, registration, supply of solicitation documents, clarifications, modifications to process or substance of the procurement. Furthermore, submission of bids, opening of bids, filing of minutes of the bidding session, recording of the award decision, reception and filing of complaints, and notice of disposition of complaints, may all de done electronically; (d) Fourth stage: The last stage involves, in addition to the capabilities covered by the third stage, highly developed support and oversight functions, including functions such as settlement of transactions made through the procurement platform; advanced demand aggregation services (whereby the procurement platform operator identifies aggregation possibilities for public sector demand of particular goods or services and actively markets electronic auctions designed to capture associated economies of scale); or advanced buyer support services (whereby the procurement platform operator develops procurement profiles for individual government agencies, particularly for recurrent purchases, and custom tailors market research and transaction facilities that improve the efficiency and economy of those purchases).452
452
Id., supra, para. 106.
268
Countries might be at varying stages in the use of electronic communications. Even within the same country different procuring entities may not be at an equal level of sophistication as regards the use of information technology in the procurement process, although this situation may rapidly change as more experience is gained and technology becomes more widely used. Those disparities affect uniformity and may cause a number of practical problems as a result, for instance, of lack of interoperability between systems set independently by various procuring entities at different levels of technological advancement. For electronic procurement systems to function properly, complex technology, operating capabilities, legal and regulatory infrastructure, and systems that allow for the submission and opening of bids electronically should be in place. These systems should also ensure security, reliability, and accessibility of the process. Implementation costs, in particular in connection with designing appropriate software or adapting generic software to local conditions, may be significant and of concern, especially if the costs are not commensurate with the value of procurement, or if the use of electronic procurement is not so extensive to ensure that the system will pay for itself in the long term. Another concern is that, in the countries where Internet penetration is low, electronic procurement systems may have a potentially discriminatory effect on suppliers depending on their access to new technology and on quality of the connection The following sections discuss general issues related to the use of electronic communications in the procurement process and particular issues arising out of the use of certain electronic procurement methods.
A. General Issues in Connection with the Use of Electronic Communications in the Procurement Process
There are two broad categories of general issues related to the use of electronic communications in the procurement process. The first category concerns the publication of procurement related information. This covers both general information as well as information relating to specific procurement opportunities. The main issue here is safeguards that are needed to avoid discrimination among potential bidders. The second category relates to the actual use of electronic communications during the procurement process. In this respect, concerns about possible discrimination among bidders are coupled with concerns to ensure transparency, data protection and integrity of communications. These matters are discussed in the following sections. 1. Publication of Procurement-Related Information Electronic publication of procurement-related information may provide wider dissemination of such information than would be achieved through traditional
269
paper means by making it more accessible to a potentially larger group of suppliers. This helps promote transparency and public authorities should be encouraged to disseminate electronically the information they are required by law to publish. A significant issue in this regard is the extent to which electronic publication should be mandatory or optional, that is, in a particular case effected by electronic means alone, or by electronic means as an addition to traditional paper-based means. Publications relating to contract opportunities may take the form of general information on forthcoming procurement opportunities, or may extend to invitations to participate in specific procurement proceedings. Information of the first category is typically of a non-binding nature and serves general purposes such as to promote better planning of Government procurement or to allow potential suppliers to make advance arrangements for participation in forthcoming procurement processes. Not many laws currently require the publication of such information. Typically, procuring entities would issue periodically (e.g. once every year) general information on their forecasted procurement needs for the relevant period, without any commitment on their part to actually procure the goods or services indicated. This information is being increasingly disseminated through electronic publication, and may appear on both procurement entities individual websites, or in centralized electronic systems covering many entities. Furthermore, many countries and entities now use electronic means to publish invitations for suppliers to participate in specific procurements (including those required to be published by law).453 Electronic publications may be valuable means to enhance transparency and competition, and in domestic practice, there seems to be an interest in replacing paper publications entirely with electronic publications.454 Procuring entities are also typically required to publish notices
453
Argentina (<onc.mecon.gov.ar>), Australia (<www.tenders.gov.au/federal/index.shtml>>), Brazil (<www.comprasnet.gov.br>), Canada (<www.merx.com>), Chile (<www. chilecompra.cl>), the European Union (<ted.europa.eu>), France (<www.journal-officiel. gouv.fr>), Mexico (<web.compranet.gob.mx>), the Philippines (<www.procurementservice. net/Default.asp>), Singapore (<www.gebiz.gov.sg>), United States (<www.fedbizopps. gov>). Permitting electronic notice of business opportunities as a substitute for the currently required paper publication is key to agencies ability to realize the efficiencies in electronic processes that justify agency investments in these processes (United States, Interim Rule of 16May 2001, Federal Register, vol. 66, No. 95 (66 FR 27407). In Chile, article 24 of the Reglamento de Ley no. 19.886 de Bases sobre Contratos Administrativos de Suministro y Prestacin de Servicios (Decreto no. 250, of 9 March 2004, modified by Decreto no. 1.562, of 25 December 2005 and Decreto no. 20, of 12 January 2007) already requires all procuring entities to publish invitations to participate in procurement through the electronic tendering system. Article 62 of the Decreto admits tendering in paper form only in exceptional circumstances (available under Orientacin normativa at <www.
454
270
of contract awards above a certain threshold in the manner specified by the law. Some countries already publish this information electronically, which is believed to greatly enhance transparency. Some go even further and publish in electronic form other information, such as information on the status of ongoing procurement proceedings. The essential legal question that arises in connection with either type of publication is the extent to which a procuring entity may use electronic publication of contract opportunities and contract awards without discriminating among suppliers and members of the general public. Publications of this type were traditionally made by some official channel, such as a governments gazette or even local or national newspapers. These publications typically enjoyed a presumption of general knowledge, that is, no one could claim not having received notice when a procurement notice was published in the governments gazette, for example. Transposing the same degree of authority to electronic publications is a policy matter that ultimately depends on satisfaction that electronic means are sufficiently widespread so as to enjoy a presumption of general accessibility. In order to avoid discrimination, domestic laws usually required that the means of publication should not compromise the general principle of accessibility.455 2. Use of Electronic Communications in the Procurement Process Electronic communications may be used in any procurement process conducted electronically, even if the particular procurement method used was not specifically designed for electronic application. They may also be used in special procurement
chilecompra.cl> (12 November 2007)). (The Reglamento defines Information System as an information system for public procurement and electronic contracting which is composed as software, hardware and electronic communications and support infrastructure that allows to conduct procurement).
455
Domestic laws on electronic publications in procurement often provide such a requirement. In France, for example, Dcret no. 2006-975 du 1er aot 2006, article 56, I, authorizes the publication of invitations to tender in an information network (raiseau informatique) from which any interested person should be able to download the invitation and related documents (Journal officiel, No. 179, 4 August 2006, p. 11627). A similar requirement exists in Austria, where 3(1) of the Verordnung der Bundesregierung betreffend die Erstellung und bermittlung von elektronischen Angeboten in Vergabeverfahren (E-Procurement-Verordnung) requires the procuring entity to choose the means of communication for the transmission of electronic offers and the electronic address to which they shall be transmitted in a non-discriminatory manner (Bundesgesetzblatt fr die Republik sterreich, 28 April 2004, part II). In the United States, procuring entities must ensure that any notice of agency requirements or agency solicitation for contract opportunities is provided in a form that allows convenient and universal user access through a single, Government-wide point of entry (United States Code Service, title 41, chapter 7, section 426(c)(4) (41 U.S.C.S., 426 2004)).
271
techniques that are primarily conceived for use in electronic form. The discussion in the following paragraphs applies to both situations. (a) Electronic Qualification and Suppliers Lists Suppliers lists are established for use as a pool of readily available information about listed suppliers in more than one procurement process. In paper or electronic form, they are employed around the globe for various purposes. Electronic procurement systems and techniques have expanded and diversified the recourse to them for procurement and non-procurement purposes, and made their maintenance and operation more efficient and less costly for both procuring entities and suppliers. Especially in electronic procurement systems, the use of some type of registration lists is to some extent indispensable for the operation, safety and security of electronic systems.456 Depending on how registration on the list affects the eligibility of suppliers to participate in a procurement process, lists may be mandatory or optional. Lists are mandatory when registration on them is required for participation in procurement covered by the list. Where the absence of registration on the list does not affect the right of suppliers to participate in procurement proceedings covered by the list, lists are optional. At the one end of the spectrum are lists that often operate as mailing lists, where prior registration does not involve any assessment of eligibility of suppliers to participate in procurement: all those suppliers with an interest in the contracts covered by the list are included in the list and qualification are checked in the context of specific procurements. At the other end of spectrum are suppliers lists the main purpose of which is to screen potential suppliers for future procurements. The degree of screening may vary from an initial assessment of minimum information on the eligibility of suppliers to participate in procurement generally,457 to prequalification of all or some criteria that potential suppliers have to meet for participation in
456
For instance, so that the system can identify and register potential suppliers, provide them with access rights to the electronic procurement portal, differentiate those rights, communicate with suppliers by sending information to and/or validating the information received from them. E.g. Compranet registration requirements in Mexico (<www.compranet.gob.mx/), and CCR registration requirements in the United States (<www.ccr.gov/handbook.asp#info). Requirements to provide minimum information for listing are also found, for example, in Argentina and in article 92 of the Public Procurement Act of the Slovak Republic (Act No. 523/2003 of 24 October 2003 on Public Procurement and on Amendment of Act. No. 575/2001 Coll. on the Organisation of Activities of the Government and on the Organization of Central State Administration, as amended). The information in these systems is usually limited to basic data about suppliers (e.g.identification information, legal form, goods supplied, contact information).
457
272
procurements covered by the list.458 These types of lists are often referred to as qualified suppliers lists or lists of approved or qualified suppliers. In electronic procurement systems, the distinction between qualified suppliers lists and simple registration lists may not necessarily be clear. For instance, some electronic registration lists, initially operating as a yellow pages business catalogue, where suppliers were listed and identified by their basic data (such as name, address and types of goods and services offered) have evolved into more complex systems. Some of them are being linked to trade, professional or other registries and systems (in particular, tax and social security payment systems), which allows registration on the list with simultaneous automatic assessment of at least basic data. With more technology possibilities, other more complex functionalities may be integrated into electronic suppliers lists, allowing for automatic evaluation of other qualification elements, such as checking performance history and ranking suppliers accordingly.459 One problem in the operation of suppliers lists is that they may substantially restrict access to procurement and reduce competition by excluding from the procurement suppliers who are not registered. The greatest risks for transparency and competition in procurement arise with lists that operate in a disguised nontransparent manner. Concerns are often expressed that not all essential elements in the operation of suppliers lists are disclosed to the public in general or to the suppliers concerned. For instance, conditions for excluding suppliers from a list or blocking suppliers admittance into a list are rarely set out in detail. In addition, under procurement regimes that authorize some but not all suppliers on the list to be solicited in any given procurement proceeding, systems employed for the selection from the lists (rotation, chronological order of registration on the list or other systems) are also rarely made public, leading to uneven distribution of procurement opportunities contradictory to the principles of transparency, equality and non-discrimination among potential suppliers.
458
Requirements to provide more detailed information are found, for example, in Brazil (Law No. 8.666 of 21 June 1993, article 35), China (Interim Measures of the Public Procurement Centre for Central Government Authorities Regarding Registration of Suppliers Qualifications, article 5), Chile (Law 19.886, article 94), Costa Rica (Decreto No. 25038-H, Reglamento General de Contratacin Administrativa of 6 March 1996, article 59.2). Criteria for listing in those systems may extend to experience, technical, managerial and financial capacity, organization and availability of equipment, staff and skills. E.g. the Government Electronic Procurement System in the Philippines (GEPS, <www. procurementservice.net), which operates on the basis of a registry of suppliers, includes a performance tracking mechanism (IRR-A, section 9.1.5). In Brazil, as well, under article 36 2 of Law No. 8.666, the suppliers conduct in the performance of its obligations shall be mentioned in the respective registry. Similarly, in Hong Kong, the Environment, Transport and Works Bureau, which maintains a list of approved contractors, also maintains a contractor performance report system.
273
459
Concerns have also been raised that qualified suppliers lists are difficult to maintain in practice, which may lead to obsolescence or the information.460 Lists could also involve administrative costs, which in some cases may be high, for both suppliers and procuring entities when suppliers that are not likely to win contracts register or seek to register. For example, combining open access to lists with screening that requires maintaining an ongoing status review for a long list of suppliers when only a few will be qualified for a specific procurement may be expensive. However, if such an ongoing review is not in place, the value of information on the list, or submitted for registration on the list, would be questionable as it would not reflect changes in capacity achieved by potential suppliers and in other data on which registration had relied. As a result, contracts could be awarded to bidders without adequate qualifications or qualified bidders could be excluded, particularly in the context of market segmentation.461 Domestic regulations on suppliers lists aim at mitigating concerns arising from the use of the lists, in particular, their potential anti-competitive effect and risks of corruption. Commonly found controls specify conditions for the use of suppliers lists and require: (i) open approach to market for the establishment of a list; (ii) continuous publicity of lists, rules regulating their operation, criteria for listing and delisting and amendments thereto; (iii) objective, non-discriminate, transparent and proportionate criteria for listing that are assessed in objective manner; (iv) open access to lists at any time; (v) regular updating, including by limiting validity of entries on the list; and (vi) due process (proper notifications of decisions related to listing, debriefing, reasonable timeframe for taking decisions and availability of appropriate challenge mechanisms).462
460
See Strengthening Procurement Capacities in Developing Countries. International Benchmarks and Standards for Public Procurement Systems (Paris: OECD/DAC World Bank Roundtable, 22-23 January 2003) 3. Id. Some or all of these requirements are set forth, for instance, in Australia (Financial Management Guidance No. 13, pp. 56-57); Brazil (Law No. 8.666, article 34 1); Chile (Executive Guidelines to Law 19.886, article 97 <www.chilecompra.cl/portal/centro_informaciones/ fr_ley_compras.html>); China (Interim Measures of the Public Procurement Centre for Central Government Authorities Regarding Registration of Suppliers Qualifications, article12); Colombia (Law 598); Costa Rica (Decreto No. 25038-H, Reglamento General de Contratacin Administrativa, article 59, section 3); the European Union (EU directive
461 462
274
Provided that these safeguards are in place, the enhanced transparency and speed of communications that result from the use of information technology for procurement purposes helps procuring entities benefit from the practical advantages of a continuously available qualification system while at the same time avoiding or reducing the discriminatory or improper impact that the operation of suppliers has had in the past. (b) Electronic Supply of Solicitations and Other Documents Some countries expressly authorize procuring entities to transmit solicitation documents, including specifications, project description, draft contracts and other related information by electronic means, subject to a number of controls, such as that there must be a record of the date and time of transmission and receipt of the content of the transmission and that proper identification of originator and addressee be provided.463 Another way of supplying solicitation documents that, depending on the technology supporting electronic procurement, may become widely used is the posting of documents on an accessible database or information system such as a special web site from which suppliers can download them. The invitation to tender may even incorporate those documents by reference, similarly to what commercial entities do in respect of general conditions of contract made available through the Internet. For the avoidance of doubt, the laws of some countries clearly state that a procuring entitys duty to provide the solicitation documents may be met by making those documents available through a publicly accessible electronic information system from which they can be downloaded or printed by the suppliers.464 (c) Form of Other Communications during the Procurement Process Once electronic procurement is authorized, it should be possible for the procuring entity to deliver notices and other communications that it may be required
2004/17/EC, articles 41(3), 49 (5) and 53 and annex XIV, and EU directive 2004/18/ EC, article 52 (6)); Hong Kong (Tender Procedures for Government Procurement (Chapter III of the Stores and Procurement Regulations, section 320(c)); Mongolia (article 37 of the Public Procurement Law); Serbia (Public Procurement Law, article 53); and in the United States (FAR 9.202(a)(2), 9.204(a), (c) and (d), and 205).
463
Spain, Real Decreto 1098/2001 (Reglamento general de la Ley de Contratos de las Administraciones Pblicas), of 12 October 2001, article 80, paragraph 2 (Boletn oficial del Estado, No. 257, 26 October 2001, p. 39252). In Mexico, for example, article 31 of the Ley de Adquisiciones, Arrendamientos y Servicios del Sector Pblico provides that solicitation documents are made available at the address indicated by the procuring entity as well as by electronic means of publication established by the Government.
275
464
to send to all bidders (such as, for example, an addendum to the solicitation documents, invitations to meetings of suppliers, and notices of the extension of deadlines for submission of tenders) to the electronic addresses provided by the suppliers. Depending on the technology used by the procuring entity, it may however appear more expeditious to post those notices and documents on an accessible database or information system such as a special web site from which suppliers can download them. Recognizing that possibility, as some countries already do,465 helps avoiding bid protests by excluded suppliers on grounds of insufficient notice. A more delicate issue relates to the conduct of meetings with suppliers or contractors and the manner of handling requests for clarifications of solicitation documents and responses thereto. The principle of non-discrimination requires that a procuring entity must communicate clarification and modifications to bidding documents to all suppliers or contractors to which the procuring entity has provided those documents. The situation may be more complex in connection with meetings with suppliers or contractors, insofar as the word meeting usually suggests the physical presence of persons at the same place and time. Some countries have therefore, introduced enabling provisions that authorize procuring entities to dispense with the requirement of an actual meeting, as long as it is possible for the procuring entity and the suppliers to establish some other form of simultaneous communication, such as by using teleconferencing facilities.466 (d) Electronic Submission of Tenders, Proposals and Quotations In some countries, procuring entities have the right to choose when tenders may be submitted electronically467 and, if so, whether or not tenders may be submitted
465
For example, sect. 8.4 of the Implementing Rules and Regulations of Republic Act No. 9184 of the Philippines, which provides that requests for clarification from bidders may be sent electronically. Supplemental Bid Bulletins as well as all other notices to be made to the bidders or prospective bidders shall be posted in the [government electronic procurement] bulletin board and sent electronically to the e-mail address indicated in the bidders registration. For example, sect. 8.4 of the Implementing Rules and Regulations of Republic Act No. 9184 of the Philippines, which provides that the requirement for face-to-face bidding conferences may be replaced once videoconferencing or similar technology becomes the norm in business transactions in the country. Procuring entities with videoconferencing capabilities that have manufacturers, suppliers, distributors, contractors and/or consultants that also have videoconferencing capabilities may conduct their pre-bidding conferences electronically. In Germany, for example, public procurement regulations require procuring entities to inform potential suppliers of the means of communication chosen, which may include mail, faxes, direct communications, electronic means or any combinaiton thereof (Verdingungsordnung
466
467
276
in paper form as well, which in some countries is generally admitted unless the invitation to tender states otherwise.468 However, in those countries suppliers are not allowed to switch from one medium to the other or to use both media to submit tenders or parts thereof.469 A somewhat different approach is taken by countries in which procuring entities are given the right to choose the means of communications with suppliers, including electronic communications, but the procuring entity may not refuse offers and tenders that the suppliers choose to submit through other means.470 Lastly, some countries require procuring entities to accept tenders and other documents submitted electronically, as long as they are authenticated with methods prescribed by the law.471 Certain methods of procurement (such as electronic reverse auctions or dynamic purchasing systems) are nearly always conducted by electronic means only.472 It is indeed an essential element of those procurement methods that all suppliers are required to submit their bids by electronic means only. Accordingly, once the conditions for use of any such special procurement method are
fr Leistungen (VOL/A), 16(4); Verdingungsordnung fr freiberufliche Leistungen (VOF), 4(6); Vergabe- und Vertragsordnung fuer Bauleistungen (VOB/A), 1(1) (2006 versions available at <www.bmwi.de/BMWi/Navigation/Service/gesetze,did=191324.html>).
468
In Austria, 43, paragraph 3 of the Bundesgesetz ber die Vergabe von Auftrgen (Bundesvergabegesetz 2006 BvergG) provides that electronic submission of tender is possible where the means chosen are compatible with generally used information and communications means. If the procuring entity has made no declaration on the possibility to deliver offers electronically, the delivery of offers by electronic means is not permitted (91, paragraph 1). Where the submission of tenders by electronic means is admitted, 91, paragraph 2 requires the invitation to tender to state whether tenders can be delivered only electronically or whether they can be submitted either electronically or in paper form. If the procuring entity has made no declaration on this, the delivery of offers is allowed either by electronic means or in paper form (Bundesgesetzblatt fr die Republik sterreich, Teil I. Nr. 17/2006, 31 January 2006, p. 1). Austria, Bundesvergabegesetz 2002, 113, paragraph 3: If the delivery of offers by electronic means is permitted, the bidders that have delivered an electronic offer may not make an offer or parts of an offer in paper form. The foregoing does not apply to parts of offers such as [documentary evidence require by the law] as far as these parts of the offer are not available electronically. This is the case, for example, in France until 1 January 2010, when procuring entities will generally have the right to requir submission of tender in electronic form only (see Dcret no. 2006-975, article 56, II) This is the case, for example, in Argentina (see Decreto delegado No. 1023/2001 con las modificaciones introducidas por los Decreto Nros. 666/2003 y 204/2004 y por la Ley 25.563. Rgimen General. Contrataciones Pblicas Electrnicas. Contrataciones de Bienes y Servicios. Obras Pblicas. Disposiciones Finales y Transitorias <onc.mecon.gov.ar>, under Normativa(12 November 2007), article 21. Only a few countries admit reverse auctions outside the context of electronic procurement. One of them is Brazil, where the matter is regulated in Lei no. 10.520, of 17 July 2002.
277
469
470
471
472
met, the procuring entity must have the right to refuse to accept bids submitted by other means. Even for procurement methods which do not by their nature require the use of electronic communications, the procuring entity might have a legitimate interest, for purposes of economy or efficiency, to take advantage of fully or partly automated devices for receiving and processing tenders, such as a specially designated portal or Internet web site. Another problem is that procurement laws often require that tenders must be submitted in writing, signed and in a sealed envelope.473 Requirements of this nature are intended to guard the confidentiality of tenders and prevent opening of the tenders prior to the deadline for submission of tenders. This question is closely related to the issue of control over the use of electronic communications, in particular as regards security, confidentiality and authenticity of submissions, and integrity of data. Efficient and reliable electronic procurement systems indeed require appropriate controls as regards security, confidentiality and authenticity of submissions, and integrity of data, for which special rules and standards might need to be formulated. Current regulations and practices allow identifying a number of general principles and goods practices in the area. UNCITRAL is currently working on a revision of its Model Law on Procurement of Goods, Construction and Services, inter alia, to accommodate electronic procurement methods. Already at an early stage of this work, there was agreement on a number of principles that provide a good basis for the formulation of specific rules, standards or guidance on the matter.474 The main such principles are the following: (a) The procuring entity should apply procedures to make it possible to establish the origin and authenticity of communications. The means and mechanisms used should also be such as to ensure that the integrity of data is preserved;475
473 474
E.g. UNCITRAL Model Procurement Law, article 30(5)(a). See UNCITRAL, Report of Working Group I (Procurement) on the work of its sixth session (Vienna, 30 August-3 September 2004) (U.N. document A/CN.9/568), paras. 41 and 42. Directive 2004/17/EC, in its Annex XXIV provides that devices for the electronic receipt of tenders, proposals or quotations must at least guarantee, through technical means and appropriate procedures, inter alia, that electronic signatures relating to tenders, requests to participate and the forwarding of plans and projects comply with national provisions adopted pursuant to the Electronic Signatures Directive.
475
278
(b) There should be technical means in place to determine the time of receipt of documents, when the time of receipt is significant in applying the rules of the procurement process;476 (c) The means and mechanisms used by the procuring entity should also ensure that tenders and other significant documents are not accessed by the procuring entity or other persons prior to any applicable deadline.477 (d) Lastly the confidentiality of information submitted by, or relating to, other suppliers should be maintained throughout the procurement process.478 In a number of countries recourse to electronic procurement is only allowed after the interested procuring entities have certified that they possess the capability to ensure that their electronic communications provide the same level of reliability that is generally assumed to exist in the case of paper-based communications. 479
476
EU Directive 2004/17/EC, Annex XXIV (requiring that devices for the electronic receipt of tenders, proposals or quotations must, inter alia, offer means to determine precisely the exact time and date of the receipt of tenders, requests to participate and the submission of plans and projects).
477
See, for example, EU Directive 2004/17/EC, Annex XXIV (requiring that devices for the electronic receipt of tenders, proposals or quotations must at least guarantee, through technical means and appropriate procedures, inter alia, that no one must have access to data before the applicable deadlines and that any infringement of access restriction must be clearly detectable; that only authorized persons may set or change the dates for opening data received; and that simultaneous action by authorized persons must give access to data transmitted only after the prescribed date). In Mexico, article 27 of Ley de Adquisiciones, Arrendamientos y Servicios del Sector Pblico provides that tenders submitted electronically shall use technology that ensures the confidentiality and inviolability of the information and that an agency of the Government shall provide certification services to support the electronic identification methods used by suppliers and contractors. See, for example, EU Directive 2004/17/EC, Annex XXIV, requiring that devices for the electronic receipt of tenders, proposals or quotations must guarantee that during the different stages of the contract award procedure access to all data submitted must be possible only through simultaneous action by authorized persons and that data received and opened in accordance with these requirements must remain accessible only to authorized persons. In the United States, sect. 4.502 of the Federal Acquisition Regulations requires the heads of procuring agencies before using electronic commerce to ensure that the agency systems are capable of ensuring authentication and confidentiality commensurate with the risk and magnitude of the harm from loss, misuse, or unauthorized access to or modification of the information. In the Philippines, government electronic procurement systems (G-EPS) are subject to a number of general requirements set forth in sect. 8.1.2 of the Implementing Rules and Regulations of Republic Act No. 9184, including that the Bid Awards Committee shall have complete control of the bidding process and sole authority to open bids, that the systems must be virus-resilient and must provide sufficient
279
478
479
Another matter relates to the conditions for functional equivalence between written tenders submitted in a sealed envelope and electronic tenders is the manner of opening tenders. Some countries have introduced enabling provisions that contemplate opening of tenders through an electronic information system that would automatically transmit the information that is usually announced at the opening of tenders.480 (e) Procurement Contracts and Electronic Signatures In domestic practice, some countries authorize the notice of acceptance of a tender to be sent electronically.481 In principle, it should be possible for a procuring entity in a country where the law does not create obstacles to the legal recognition of contacts negotiated through electronic means to accept electronically executed procurement contracts. However, countries may also wish to prescribe the manner in which the parties will sign or otherwise authenticate a procurement contract concluded electronically.482
security such as firewall and encryption devices, that they must provide for the use of electronic signatures and other current electronic authentication devices and have sufficient redundant back-up facilities.
480
In Chile, for example, article 33 of Decreto no. 250, of 9 March 2004 provides that the opening of tenders shall be effected through an Information System, which shall automatically release and open the tenders at the date and time provided in the solicitation documents. It provides further that the Information System shall ensure certainty as to the date and time of opening and shall allow the tenderers to know at least the following conditions of the remaining tenders: (a) identity of tenderer; (b) basic description of good or service tendered; (c) initial and global price of tender; (d) identification of tender security, if any. United States Code Service, title 41, section 253 b (c) (41 U.S.C.S. 253b (2004)): The award of a contract shall be made by transmitting, in writing or by electronic means, notice of the award to the successful bidder. Within 3 days after the date of contract award, the executive agency shall notify, in writing or by electronic means, each bidder not awarded the contract that the contract has been awarded. In Austria, 131 of the Bundesvergabegesetz 2006 provides in paragraph 2 that notice of award can be sent to suppliers and contractors electronically. However, 134, paragraphs 1, 2, and 3 contemplate the execution of the procurement contract through the exchange of paper documents by registered mail tenders, while paragraph 3 of that provision only authorizes the Federal Government to issue regulations on contract execution (Vertragsabschluss) by electronic means, including regulations to guarantee the confidentiality, authenticity and integrity of data transmitted electronically by means of secure electronic signatures, as well as their confidentiality. Rules on the authenticity and integrity of electronic tenders are contained in regulations issued recently (the aforementioned E-Procurement-Verordnung 2004). Although the regulations do not expressly refer to the execution of the procurement contract, the same requirements would arguably apply.
481
482
280
To the extent that a procurement contract would be formed by the exchange of electronic communications between the procuring entity and the selected supplier, the time of contract formation should follow the general rules on the matter. By the same token, at least in theory, there is no reason why procurement laws should make specific signature requirements beyond and above, or at variance with, the general rules on electronic authentication and signature in force in the country. However, the type of authentication methods that a procuring entity is capable of accepting may be limited for various reasons, including concerns over the appropriate level of reliability and availability of supporting technology. Lastly, issues related to the interoperability of information systems, both at the national level and as within a given region, suggest that countries should have broad latitude in determining which methods of authentication they would accept in the procurement process.483
B. Legal Issues arising out of Specific Electronic Procurement Techniques

By and large, the legal issues that have arisen in connection with the conduct of government procurement through electronic means, as described in the following sections, are not intrinsically related to the electronic medium. Indeed, most of the same issues would also arise if the procurement process was conducted through traditional means. Nevertheless, the fact electronic communications and automated devices make it easier or more economical to carry out certain procurement methods than would have been the case with traditional means, possibly leading to their overuse, may exacerbate the risks and problems that these procurement methods already have. This may call for caution in their use and stress the need for safeguards to ensure that the overall objectives of sound government procurement (transparency, non-discrimination, integrity and value-for-money) are preserved. 1. Electronic Reverse Auctions An electronic reverse auction (ERA) can be defined as an online, real-time dynamic auction between a buying organization and a number of suppliers who compete against each other to win the contract by submitting successively
483
See Christine Kirchberger and Jon Ramn y Olano, Issues of Security and Interoperabiity in Electronic Public Procurement, 47 Scandinavian Studies of Law (2004) 51.
281
lower priced bids during a scheduled time period.484 ERAs are used in both the private and public sectors. A traditional selling auction involves a single seller and many buyers, the latter bidding for the right to purchase and the former using market forces to drive buyers to raise the price of purchase, in a reverse auction. In an ERA, by contrast, there is a single buyer and many suppliers: the buyer indicates its requirement, and suppliers progressively bid downwards to win the right to supply. In this instance, the buyer uses market forces to drive suppliers to lower prices. (a) The Extent of Use of ERAs The extent of the use of ERAs is determined to a large degree by the extent of electronic commerce activity in the overall economy. In countries where electronic commerce is widespread, the trend towards the use of ERAs in public procurement is strong. Countries in which application of ERAs to public procurement has been pioneered include in particular Australia,485 Brazil,486 Canada, France, Singapore, Thailand, the United Kingdom487 and the United States.488
484
See Stewart Beall et al., Role of Reverse Auctions in Strategic Sourcing (Center for Advanced Purchasing Studies (CAPS Research), 2003) <www.capsresearch.org/publications/ pdfs-public/beall2003es.pdf> (3 December 2007). Australia has had an integrated national electronic procurement framework since May 1999 (see the Framework for National Cooperation on Electronic Commerce in Government Procurement <www.apcc.gov.au/docs/NationalECFramework.pdf>). This framework consists of Commonwealth initiatives developed to promote electronic procurement. State governments have also established business centres to encourage acceptance of online procurement and have developed their own online portals for e-procurement, including New South Wales (<www.cpsc.nsw.gov.au/e-procurement/links.htm), Victoria (<www. ec4p.dtf.vic.gov.au/domino/web_notes/ec4p/ec4p.nsf/ frameset/EC4P?OpenDocument) and Queensland (<www.qgm.qld.gov.au/prc/English/ prc_intro.htm).
485
486
In Brazil, ERAs were introduced to public procurement by Decree 3.697 of 21 December 2000. In 2001, 3.2 per cent of the total volume of goods and services procured by the Federal Government were procured through ERAs, growing to 12 per cent, in 2003, and to approximately 20 per cent, in 2004. COMPRASNET (<www.comprasnet.gov.br>) is the web platform for e-Government Procurement of the Federal Government in Brazil. See the website of the Office of Government Commerce in the United Kingdom (OGC) (<www.ogc.gov.uk>). For news on experience in the United States and elsewhere in governments use of ERAs, see <www.egov.vic.gov.au/Research/OnlineAuctions/auctions.htm>. For a discussion group on successes and failures in the use of ERAs in U.S. procurement, see <www.wifcon. com/arc/forum62.htm>.
487
488
282
Regulations on ERAs may be found in statutes (such as legislation implementing the EU Directives and in Brazil489) or in provincial regulations (such as in China490). In other countries, guidance on the conduct of ERAS is given in internal documents of procuring agencies (such as in Singapore491), or in centralized regulatory guidelines (such as in Australia492 and the United Kingdom493). In the United States,494 the procedures are largely determined on an agency-by-agency basis, and sometimes on a procurement-by-procurement basis. Regulations provide for two ways of using ERAs: either as a stand-alone method of procurement or as an optional phase in other methods of procure-
489
Federal Law No. 10.520/2002 of 17 July 2002, complemented by Law 8.666 to the extent it does not conflict; Decrees 1.094/94, 3.555/00, 3.693/00, 3.697/01 and 3.784/01; and Internal Instructions (Portarias) SAF/PR 2.050 and MARE 5. See the Nanning City Interim Measures for the Management of Online Public Procurement Bidding of 18 June 1999; the Zhejiang Province Interim Measures for the Management of Online Public Procurement Bidding of 1 September 2000; the Hefei City Interim Measures for the Management of Online Public Procurement Bidding of 13 March 2001; the Wuxi City Interim Measures for the Management of Online Public Procurement Bidding of 1 April 2001; the Zhuhai City Interim Measures for the Management of Online Public Procurement Bidding of 27 June 2002; the Shenzhen City Interim Measures for the Management of Online Public Procurement Bidding of 15 October 2003; and the Shanghai Interim Measures for the Management of Online Public Procurement Bidding of December 2004. The text of the latter in Chinese and its unofficial translation in English are available with the UNCITRAL secretariat. See, for example, the Administrative guidelines for assisted reverse auction event of the Ministry of Defense (the Singapore Guidelines). In Australia, regulation is currently limited to policy documents, non-statutory procurement guidelines and broad statutory provisions about electronic procurement. New South Wales remains the only State to provide any specific guidance on topic. See the NSW Government Procurement Guidelines on Reverse Auctions of March 2001 <www.dpws. nsw.gov.au/ NR/rdonlyres/ezac4yppqkqqzaj5qdjgerv3aj62n4ishpa3xhofh4fdl3cqut4m7l4ibv3a2w67sslw5zuhmjpois43joel4ees4xe/Reverse+Auctions.pdf>) (the Australian Guidelines). In the United Kingdom, rules on public procurement are mainly limited to those of the EU law. The British Government has considered that the EU current directives allow scope for ERAs in public procurement and has endorsed their use. OGC, in promoting the use of ERAs in government procurement, has issued the on-line guidance <www.ogc.gov.uk/ index.asp?docid=1001034>. Attempts to formulate centralized binding rules have not yet been successful, reportedly because of industry opposition and because there is no consensus on when reverse auctions should be used. See, for example, Susan L. Turley, Wielding the Virtual Gavel DOD Moves Forward with Reverse Auctions, 173 Military L. Rev. (2002) 1, 25-31, (discussing sources of industry opposition to ERAs).
283
490
491
492
493
494
ment.495 The latter approach is taken by the EU Directives,496 in Australia,497 Singapore498 and the United States.499 In most of those cases, an ERA is a final stage preceding the award of a public contract, while in some jurisdictions, like in Brazil and China, an ERA is a distinct procurement method. In the latter case, ERAs can be conducted in an open market to all suppliers or to a limited number of pre-selected or pre-qualified suppliers. However, it is generally recognized that not all types of procurement are appropriate for ERAs. The primary factor to consider in deciding whether a certain type of procurement is appropriate for ERA is the level of product or service complexity for the procurement and with what level of accuracy the procurement can be specified (i.e. whether suppliers can easily understand the requirement or the requirement can only be defined superficially and needs early supplier intervention). Other factors considered are: (a) predicted value of procurement to determine whether procurement would be attractive to suppliers; (b) market competition (whether it is high enough to ensure the participation of sufficient number of suppliers in the ERA);500 and (c) award procedure (to what extent the procurement award criteria are quantifiable). There has been a general tendency in international practice to confine the use of this procurement technique to standardized goods and some simple types of services. In Australia, for example, the use of ERAs is restricted to products
495
Although this means that ERAs can be used in open tendering proceedings, it has been observed that, in practice, the restricted procedure will normally be used when an ERA is involved. ERAs are likely to be used only rarely in negotiated procedures since many of the grounds permitting recourse to such procedures are concerned with situations in which specifications and other conditions cannot be easily set in advance, something which is generally essential for an auction (See Sue Arrowsmith, Electronic Reverse Auctions Under the EC Public Procurement Rules: Current Possibilities and Future Prospects, 11 Pub. Procurement L. Rev. (2002) 299. EU Directive 2004/18/EC, article 54 (2). Under the Australian Guidelines, ERAs could be used as part of the tender process, as a means of obtaining quotes from suppliers, and as the second stage of a two-stage tender process where price is the remaining selection criteria. Section 1.1 of the Singapore Guidelines. In the United States, in the absence of explicit prohibition, ERAs could be used in combination with any available procurement methods and is also used in the context of frameworks and dynamic purchasing systems. Some systems specifically address a minimum number of participants in the ERA while in other systems general provisions of procurement law apply. The requirement of at least three participants in an ERA is commonly found in the regulations. See, for example, article 22 of the Shanghai Interim Measures for the Management of Online Public Procurement Bidding. It has been observed that a higher number of participants effectively prevents the risk of collusion. In Austria, participation in ERA of minimum ten participants is required (see Purchase Contract Awards Act 2002, para. 116.7).
496 497
498 499
500
284
or commodities with no or little value-added or service component, capable of description in homogeneous terms, available in a competitive market, with price as a primary selection criterion.501In some countries, for example in Brazil,502 the use of ERAs is also allowed for the procurement of simple services. Works, however, are usually excluded from ERAs.503 In some countries, such as Canada and the United States, grave concerns have indeed been expressed particularly over the use of ERAs for the procurement of construction.504 In some states of the United States, such as Pennsylvania and Kansas, state procurement regulation explicitly prohibits procurement of construction contracts through ERAs.505 It has been observed, however, that some construction works and services (e.g. road maintenance) may be appropriately procured through ERAs. Under the EU Directives, any purchases can be procured through ERAs provided that specifications can be established with precision, such as recurring supplies. Intellectual works or services, such as the design of works, are explicitly excluded.506 (b) Procedural Aspects of ERAs (i) Auction Stage At the auction stage the participating bidders access a screen by logging in to the auction address provided in the notice of auction or invitation to the auction, as applicable, using their respective identification and personal password that permits them to participate in the auction. The object of the ERA and the auction
501 502
The Australian Guidelines. Under article 1 of Law No. 10.520 of 17 July 2002, auctions can be used only for common goods and services, defined as those for which quality and performance standards can be objectively and precisely defined according to standard specifications used in the market. Brazilian regulations do not mention works in the list of eligible items for ERAs. Under the Australian Guidelines as well, the use of ERAs is to be restricted to the procurement of products or commodities only. Special Bulletin of the Canadian Construction Association, December 2001 <www. cca-acc.com/news/committee/rag/rag-owner.pdf> (28 November 2007). See AGCs white paper, Reverse Auctions Over the Internet: Efficiency At What Cost?, 2003. See article 1(7) of EU Directive 2004/18/EC that defines reverse auction. It explicitly provides that certain service contracts and certain works contracts having as their subjectmatter intellectual performances, such as the design of works, may not be the object of electronic auctions. Recital paragraph 14 explains that only elements suitable for automatic evaluation by electronic means, without any intervention and/or appreciation by the contracting authority, may be the object of electronic auctions, that is the elements which are quantifiable so that they can be expressed in figures or percentages.
285
503
504
505
506
rules (i.e., start time, duration, minimum bid, the method of termination, etc.) are announced and the call for bids is the communicated simultaneously to all bidders. A minimum period of time is usually required to elapse between the issuance of the notice of auction or invitation to the auction, as the case may be, and the opening of the auction.507 In practice, the more complicated the procurement, the longer such a period usually is. Usually, regulations provide flexibility to procuring entities with respect to the closure of the auction. For example, under the EU Directives, auctions may be closed in one or more ways: (a) at the date and time fixed in advance as communicated to tenderers in the invitation; (b) when contracting authorities receive no more new prices or new values which meet the requirements concerning minimum differences; or (c) when the number of phases fixed in the invitation has been completed.508 The end of the bidding session may be set electronically or, as in Brazil, if expressly provided in the notice of auction, announced by the auctioneer.509 The Brazilian system permits bidders to challenge the time fixed for the auction and request extensions of the auction. However, it is within the discretion of auctioneer to satisfy such a request.510 In practice, the greater the value and complexity of the procurement, the longer the duration of the ERA. ERAs rarely close after a fixed duration of time has expired (what is called hard close time). Usually, the closing time of the ERA is automatically extended for a specified period of time (e.g. 5 minutes) if a new lowest bid or a bid that changed top bid rankings (usually one of the top threeranked bids) is received in the last few minutes (e.g. within 2 minutes of the closing time). Such extensions may be continuous for an indefinite period of time (known as unlimited soft close) or limited in the amount of overtimes (e.g. maximum of three 5-minute extensions). This process continues until there are no longer any lower bids being submitted within the stated period prior to closing. (ii) Bidding Requirements When the price only is subject to ERA, it is usually required that the value of each bid has to be necessarily lower than the value of the last bid registered by
507
See Law No.10.520 of Brazil (requiring at least 8 days after the notice of auction is given); EU Directive 2004/18/EC, article54(4) (providing that ERAs may not start sooner than twoworking days after the date on which invitations are sent out). EU Directive 2004/18/EC, article 54(7) Decree 3.697. Auctioneers discretion in this regard has been criticized by some analysts, as well as by multilateral development banks, which prefer fully automated systems without little, if at all, human intervention.
508 509 510
286
the system. In some countries, participants who did not bid at all or did not vary their bids within the fixed increment are excluded after each round of bids.511 In Brazil, no provisions on exclusion of bidders by the auctioneer are found but bidders are permitted to withdraw their bids at any time and, if they are not interested in continuing, disconnect at any time. The time of disconnection is automatically registered and this comes out in the records of the ERA. (iii) Disclosure of Information during the Auction One of the inherent features of an ERA is that it enables current information about the status of the auction to be provided to bidders automatically and instantaneously as an auction unfolds. It has been observed that, unless properly regulated, this feature of ERAs gives rise to concerns, especially from the standpoint of competition law.512 Most regulations are flexible with respect to the extent of information disclosure. Some systems, generally require that participating bidders be provided certain basic information about the progress of an auction on an on-going basis, such as information that allows bidders to ascertain their current relative rankings,513 the number of bidders participating in the proceeding,514 and the time remaining for the reverse auction. To determine relative ranking, some systems require real-time disclosure of the lowest price to be bid515 while in other systems, only ranking but not prices are disclosed.516 Under the EU Directives, information about other prices or values submitted may be disclosed if the specifications or auctions rules, as applicable, provide for the disclosure of that information.517
511 512
See Austria, Bundesvergabegesetz 2006, 147, paragraph 4(5). See Ciara Kennedy-Loest and Ruth Kelly, The EC Competition Law Rules and Electronic Reverse Auctions: A Case for Concern?, 12 Pub. Procurement L. Rev. (2003) 27. Article 54(6) of EU Directive 2004/18/EC requires contracting authorities to instantaneously communicate to all tenderers at least sufficient information to enable them to ascertain their relative rankings at any moment. See, for example, EU Directive 2004/18/EC, article54(6). See, for example, Austria, Bundesvergabegesetz 2006, 148, paragraph 2. See, for example, the Shanghai Interim Measures for the Management of Online Public Procurement Bidding. According to the analysts of the EU competition law rules, showing during the auction overall ranking instead of prices is preferable from competition law prospective. See Kennedy-Loest and Kelly, supra note 512 at 29. EU Directive 2004/18/EC, article 54(6).
287
513
514 515 516
517
(iv) Contract Award Depending on the permitted criteria for the award of a contract procured through ERAs, two systems are found: those based on the lowest price alone and those that permit additional criteria. In the systems where the price is the only permitted criterion for the award, as in Brazil and China,518 quality requirements are limited and factored in the bidding documents as minimum qualification requirements, which, if met, put suppliers in an equal footing. In addition, in Brazil, quality requirements of the goods being procured are established in advance when the goods and services were included in the appropriate catalogues, but the bidder offering the lowest bid may be later disqualified in the course of evaluation of its qualifications, which takes place after the auction.519 In China, some quality requirements, such as ability to provide quality after-sale service and complete technical maintenance, are evaluated upon the application for the membership in the online public procurement bidding system, without which no participation in ERAs is possible.520 Quality aspects are also taken into account in the event of a price tie, when a supplier with a higher credibility is selected.521 By contrast, the EU Directives differentiate two types of auctions: simple ERAs, in which the price is the only award criterion; and other types of ERAs where the technically and financially most advantageous offer is given the award on the basis of evaluation of all award criteria fixed in the tender documents.522 The values of only those features that are quantifiable and can be expressed in figures or percentages can be the subject of ERAs.523 However, all features of the tender, auctionable and non-auctionable, are to be evaluated prior to the auction in accordance with their relative weightings.524 The outcome of the full evaluation of each bidder is made known before the ERA in the invitation to
518 519 520
This approach has also been advocated by the multilateral development banks. Law 10.520 and Decree 3.697. See, for example, the Shanghai Interim Measures for the Management of Online Public Procurement Bidding, article 6, and the Zhejiang Province Interim Measures for the Management of Online Public Procurement Bidding, article 8. See the Shanghai Interim Measures for the Management of Online Public Procurement Bidding, article 19; and the Zhejiang Province Interim Measures for the Management of Online Public Procurement Bidding, article 22. Factors considered to assess suppliers credibility are inter alia a good record of legal compliance, past performance, business integrity, strong credit standing, considerable capital strength and sound financial status. EU Directive 2004/18/EC, article 54(3). EU Directive 2004/18/EC, article 54(3). Id., article 54(4).
521
522 523 524
288
the auction. The invitation also states the mathematical formula525 to be used in ERA to determine automatic reranking on the basis of the new prices or new values submitted. Except in those auction models that admit an evaluation of the non-auction elements of the bid after closing of the auction phase a possibility which is the object of some controversy the winning bidder should be known in the end of the auction and would normally be the bidder offering the lowest bid. However, even in auction without subsequent evaluation of non-auction elements of bids, there may be situations in which the lowest bid is not necessarily the winning bid. Some systems allow the notice on ERA or the terms and conditions of auctions to state that the final decision to accept the offer rests with the procuring entity allowing it to ascertain if the product meets the standards required and whether the supplier is able to proceed with the supply.526 Existing regulations usually allow the procuring entity to accept the second best offer, if for legitimate reasons it rejected the lowest bid, provided that such a possibility was disclosed in advance to the bidders.527 In other systems, however, if the lowest bid is invalidated, the procuring entity has to reconduct ERA or adopt other methods of procurement.528
525
Id., article 54 (5). That formula incorporates the weightings of all criteria fixed to determine the most economically advantageous tender, as indicated in the contract notice or in the specifications; for that purpose, any ranges shall be reduced beforehand to a specified value. Where variants are authorized, a separate formula is provided for each variant. See, for example, in China, the Shanghai Interim Measures for the Management of Online Public Procurement Bidding, articles 22 and 27; the Shenzhen City Interim Measures for the Management of Online Public Procurement Bidding, article 31; the Zhuhai City Interim Measures for the Management of Online Public Procurement Bidding, article 29; and the Hefei City Interim Measures for the Management of Online Public Procurement Bidding, article 25. In China, for example, the lowest bid may be invalidated by the procuring entity, for instance, if it is higher than the market price or abnormally low or in case of a misconduct of the winning bidder during the bidding process or the registration (See the Liuzhou City Rules of Implementation for Public Procurement through Online Procurement, article 19). The Brazilian system permits an auctioneer to approach the bidder who submitted the second lowest bid if the winning bidder is disqualified or the bid is deemed to be unacceptable or non-responsive See, in China, the Shanghai Interim Measures for the Management of Online Public Procurement Bidding, article 22; the Shunyi District of Beijing Interim Measures for the Management of Online Public Procurement Bidding, article 20; and the Liuzhou City Rules of Implementation for Public Procurement through Online Procurement, article 22.
289
526
527
528
(c) Benefits and Concerns In the view of some analysts, if used properly, ERAs have the potential to improve value for money,529 efficient allocation of resources,530 and transparency in the process of awarding contracts.531 It has also been observed that they can make governmental systems more accessible and user-friendly, allow governments to keep up with changes in technology, business practices and prices found in the private sector, gain better knowledge of the market and open government bidding markets to suppliers who had not enjoyed access to them previously.532 The potential of ERAs to exert a positive effect on competition, in particular by dismantling preferential purchasing patterns in some member States of the European Union, has been recognized in the 2004 EU Directives.533 Most analysts agree that ERAs are successful for goods and services that can clearly be specified, whose non-price criteria can be quantified, for which switching costs (e.g. replacement of suppliers) are acceptable, and for which a competitive market exists. In contrast, it is generally considered that for one-of products where quality is more important than price, and for strategic items, for which alliance level supplier relationships are critical, they are not suitable.534 In some countries where ERAs have been introduced, there have been concerns that, at least for some types of procurement, ERAs seldom provide benefits comparable to currently-recognized selection procedures. For instance, it has been suggested that ERAs:
529
See Andrew Stein and Paul Hawking, Reverse Auction E-procurement: A Suppliers Viewpoint, <ausweb.scu.edu.au/aw02/papers/refereed/stein/paper.html> (28 November 2007). See also David C. Wyld, Auction Model: How the Public Sector Can Leverage the Power of E-commerce Through Dynamic Pricing, Price Waterhouse Cooper Endowment for the Business of Government, 2000 <www.businessofgovernment.org/pdfs/WyldReport. pdf> (28 November 2007); and Curran, Bernert and Wiegand, supra note 450. Also OGCs E-procurement: Cutting Through the Hype <www.ogc.gov.uk>. See CAPS Research Focus Study, Role of Reverse Auctions in Strategic Sourcing (2003). Shortened time frames for actions in the context of ERAs have already been reflected in some legislative texts. On the other hand, it is observed that such a pattern may impact other methods of procurement where, however, reduction of time frames for certain acts may not be justifiable. See Ohad Soudry, Promoting Economy: Electronic Reverse Auctions Under the EC Directives on Public Procurement, 4 J. Pub. Procurement (2004) 340, 354. See also Wyld, supra note 529. See CAPS Research Focus Study, supra note 530 and Wyld, supra note 529. See Ohad Soudry, supra note 531 at 340-342. See, for example, the OGC guidance <www.ogc.gov.uk/index.asp?docid=1001034> (28 November 2007). Also, CAPS Research Focus Study, supra note 530.
530
531
532 533 534
290
(a) do not guarantee the lowest responsible and responsive price535 and continued savings in subsequent ERAs;536 (b) may encourage imprudent bidding and thus create a higher risk of abnormally low bids;537 (c) do not adequately handle non-price factors, such as quality of performance and buyer-supplier relationships;538 (d) are more vulnerable than traditional bidding processes to collusive behaviour by bidders, especially in projects characterized by a small number of bidders, or in repeated bidding in which the same group of bidders participate;539 and (e) have negative effects on the market, including an anti-competitive impact.540 In addition, some analysts question the legality of ERAs on the grounds that their inherent features conflict with traditional procurement principles
535
Id. It is suggested that in ERAs, each bidder recognizes that it will have the option to provide successive bids and therefore has a little incentive to offer its best price and subsequently may never offer its best price. Consequently, the winning bid may be simply an established increment below the second lowest bid rather than the lowest responsible and responsive bid. Id. For the analysis of existing approaches for handling the risk of abnormally low prices, including in ERAs, see U.N. document A/CN.9/WG.I/WP.36. Jeffrey K. Liker and Thomas Y. Choi, Building Deep Supplier Relationships, Harv. Bus. Rev. (2004) 104, 106. Collusion can occur in the ERA when two or more bidders work in tandem to manipulate the price of an auction, or, alternatively, when a seller uses shells to enter fake bids and drive up the asking price. As a result, contracting authorities might face higher prices and the members of the cartel will enjoy profits above the competitive prices. See Soudry, supra note 531, 360-66. See generally, Peter Trepte, Electronic Procurement Marketplaces: The Competition Law Implications, 10 Pub. Procurement L. Rev. (2001) 260, (discussing anti-competitive concerns in the context of an electronic government procurement market). See also Ciara Kennedy-Loest and Ruth Kelly, supra note 512: (the authors discuss three main areas of concerns identified by the European Commission in relation to the compatibility of electronic marketplaces (and, by analogy, ERAs) with the EU competition rules: information exchange, in particular because the auction marketplace provides a forum for competitors to exchange commercially sensitive information; access and foreclosure issues (has a marketplace or auction been set up to exclude certain competitors or to require them to participate on an exclusive basis?); and the aggregation of purchasing power (does the auction or marketplace facilitate joint purchasing or joint selling by participants in an auction?)). The use of ERAs in the private sector has also been questioned, in particular as it carries the risk of abuses of monopoly or dominant positions (See Draetta, supra note 14 at 169-173).
291
536 537
538
539
540
and practices, such as rules forbidding the disclosure of information on other bids, pre-closing negotiations or bid-shopping.541 The risk of collusion is probably the main reason why auctions are not used in traditional procurement methods. One of the reasons why the use of ERAs has been advocated is because technology allows for the conduct of auctions in a situation of nearly absolute anonymity, and only an adequate assurance of anonymity can justify the use of a method where so much sensitive information is shared during the bidding process. Therefore, it is general practice, often required by law, that the identity of those submitting particular tenders in the auction phase not be disclosed to other bidders.542 In some systems, the identity is not disclosed to the procuring entity.543 Usually anonymity requirements apply until the closure of ERA and are ensured through computerized or automatic means. In some systems, anonymity of bidders is preserved also after the auction. 2. Dynamic Purchasing Systems Dynamic purchasing systems can be described as arrangements where one or more suppliers who were selected on the basis of their responses to request for proposals undertake to supply certain goods or services, or range of goods or services for a certain period of time pursuant to periodic orders that are subsequently placed with the supplier or suppliers chosen as particular requirements arise. Dynamic purchasing systems are a type of framework agreement, but differ in that they are conducted exclusively by electronic means.544 Dynamic
541
The extension of bid closing times and the ability to resubmit prices as allowed by ERAs can be interpreted as a form of pre-closing negotiation or bid-shopping which may compromise a fair and open competitive process. See Pierre Boucher, Technology Versus Industry Practices, February 2003 <www.findarticles.com/p/articles/mi_qa4088/is_200302/ ai_n9176581>(28 November 2007). See, for example, EU Directive 2004/18/EC, article 54(6). In Brazil, the system only provides an identification number of the bidders, so that the reverse auctioneer may control the receiving of bids from the different bidders, but not be able to identify them physically. The bidders themselves have sufficient information to perceive only which is the lowest bid and if it is theirs or not. A framework agreement is defined in EU Directive 2004/17/EC, article 1(4) as an agreement between one or more contracting entities and one or more economic operators, the purpose of which is to establish the terms governing contracts to be awarded during a given period, in particular with regard to price and, where appropriate, the quantities envisaged. A dynamic purchasing system, in term, is defined in article 19(5) of the directive as a completely electronic process for making commonly used purchases, the characteristics of which, as generally available on the market, meet the requirements of the contracting entity, which is limited in duration and open throughout its validity to any economic operator which satisfies the selection criteria and has submitted an indicative
542 543
544
292
purchasing systems further differ from classical framework agreements under EU Directive 2004/18/EC in that they permit a system that is ongoing, open to all qualified suppliers, and to which new suppliers can be added (which means that the systems is not binding as between the procuring entity and the initial suppliers). Dynamic purchasing systems are also related to suppliers lists drawn up in anticipation of procurements. Both identify suppliers for future awards of procurement contracts, and although framework agreements (of which DPS are a species) and suppliers lists have been described from a commercial point of view as shades on a single spectrum,545 there are significant differences between the two.546 They can be distinguished in that in the case of a dynamic purchasing system, like generally in framework agreements, there is an initial invitation to tender or other invitation to participate in a procurement that contains a specification of the goods to be procured and other requirements for the procurement; and the terms and conditions upon which the various suppliers will supply the goods (such as price and delivery charges and times). Dynamic purchasing systems are most commonly used for goods, services or construction for which a procuring entity has a repeated need, in particular standardized goods (such as stationery or spare parts) or simple works and services (bulk shipment, transportation, cleaning services), but for which delivery times and quantities are not known at the time of the initial invitation. Other uses include the purchase of items from more than one source, such as electricity and medicines, and centralized purchasing for several procuring entities. Under the system introduced by EU Directive 2004/18/EC, for example, dynamic purchasing systems may be used for the purchase of commonly used goods, works or services, the characteristics of which, as generally available on the market, meet the requirements of the contracting authority.547 (a) Scope and Duration of Dynamic Purchasing Systems Dynamic purchasing systems are too new a phenomenon to have already given rise to established practices. To the extent that they are a special type of framework agreement, some of the same principles may apply. Under a framework agreement a procuring entity may or may not be bound to place any orders at
tender that complies with the specification. Identical definitions are contained article 1 (4) and 1(5) of EU Directive 2004/18/EC.
545
Sue Arrowsmith, Framework Purchasing and QualificationLlists under the European Procurement Directives: Part I, 8 Pub. Procurement L. Rev. (1999) 115. It is true, however, that some types of dynamic purchasing system and some types of suppliers lists can be difficult to classify. As with dynamic purchasing systems, its use is not confined to specific goods or services.
293
546
547
all, depending on the legal system.548 Where the terms of dynamic purchasing systems do not oblige the procuring entity to make purchases under the dynamic purchasing system, the entity could purchase outside the agreement if more favourable terms are available elsewhere. Some systems require framework agreements to set minimum and maximum quantities or values of purchases under the framework, and to this extent, the dynamic purchasing systems are binding on the procuring entity. In the United States, for example, the Federal Acquisition Regulations require all taskanddelivery order contracts, which are a form of framework agreement, to specify the period of the contract, the maximum quantity of goods or services to be purchased, and although the agreement must also stipulate a minimum monetary value that will be purchased under the agreement, the amount is typically low and of little real significance. Regulations specific to dynamic purchasing systems, such as those introduced by the 2004 EU Directives, do not appear to establish a similar obligation, which is understandable in view of the non-specialized nature of the supplies they typically cover. Like framework agreements, dynamic purchasing systems are also usually subject to a maximum duration, the most common being one year, or ranging from three to five years.549 (b) The Operation of Dynamic Purchasing Systems It is likely that the types of recurrent purchases for which dynamic purchasing systems are commonly used will dictate the use of an open and competitive procurement method. Following the conclusion of the procurement proceedings, the procuring entity includes the selected suppliers into the dynamic purchasing system with one or more suppliers. It is common under more complex dynamic purchasing systems with several components or variables that the identity of the supplier whose offer will turn out to be the lowest-priced or lowest evaluated when a purchase order is placed not to be known at the time the dynamic purchasing system is concluded. For example, where the dynamic purchasing system covers more than one product (for example, a range of computer equipment), not all suppliers are able to offer all products, and the best price for each product may be offered by different suppliers. For practical reasons, dynamic purchasing systems therefore allow suppliers to
548
A European Commission guidance paper on the operation of framework agreements under EU Directive 2004/18/EC notes indeed that whether or not the procuring entity is bound to place orders under a framework agreement, and the supplier to fulfill them, is a matter of national law (European Commission, Directorate General Internal Market and Services Public, Procurement Policy, CC/2005/03_rev 1, of 14 July 2005, page 3). EU Directive 2004/18/EC, in article 33 (7), provides that a dynamic purchasing system may not last for more than four years, except in duly justified exceptional cases.
549
294
revise their prices and other terms of their tenders, without necessarily calling for a formal tender phase. Typically, the procuring entity chooses the best offer existing at the time of a particular order, possibly refining the specification as it does so. Observers have noted the advantages of this mechanism, including that the costs of full re-tendering in such circumstances might be disproportionate and the use of a dynamic purchasing system consequently could be seen to be cost-effective. It is also common that this type of system is open, such that new suppliers can be admitted to the dynamic purchasing system at any time. This type of system may take the form of an electronic catalogue in which procuring entities can search for suppliers current prices. Such facilities enable prices to be changed regularly, and their increasing use (which also helps reduce the transaction costs involved in changing suppliers) has provided more impetus for the use of dynamic purchasing systems generally. Electronic catalogues therefore allow procuring entities to select goods and services swiftly, while still exerting competition. Dynamic purchasing systems differ from classical framework agreements under EU Directive 2004/18/EC in that they permit a system that is ongoing, open to all qualified suppliers, and to which new suppliers can be added (which means that the systems is not binding as between the procuring entity and the initial suppliers). The rules provide that tenders can be altered at any time, and that there must be a second phase competition for each specific contract. Also, unlike in framework agreements, before issuing the invitation to tender, a procuring entity must publish a simplified contract notice inviting all interested suppliers to submit an indicative tender and a procuring entity may not proceed with tendering until it has completed evaluation of all the indicative tenders received within a fixed time limit. Only then may a procuring entity invite all tenderers admitted to the system to submit a tender. (c) Benefits and Concerns The main purposes of dynamic purchasing systems include the reduction of transaction costs and transaction time, and assuring the security of supply. As in dynamic purchasing systems the suppliers are identified, their qualifications assessed and the specification, terms and conditions of the future procurement established before an order is placed, recurrent costs can be avoided and purchases can be made with lower overall transaction costs and shorter delivery times than would be the case were each purchase procured separately. The types of dynamic purchasing systems in which all competition takes place in the first phase of the award process are straightforward to operate in the second award phase, and thus the potential for savings in transaction costs and times is significant, particularly where individual purchase orders under the dynamic purchasing system are made electronically. In addition, observers have commented that
295
dynamic purchasing systems, as is generally the case with framework agreements, can also lower inventory costs (as supplies are ordered only when needed), and allow the procuring entity greater flexibility in scheduling requirements, both in terms of timing and quantity.550 Dynamic purchasing systems may be useful to ensure rapid and secure supply of items to be procured (for example, the agreement may require the supplier to fulfil all orders placed and to keep a permanent stock of a product available). It has also been observed that one of the benefits of the use of dynamic purchasing systems in the procurement process is that, similarly to framework agreements, they allow a procuring entity to realize the benefits of an ongoing competitive environment throughout the duration of the contract,551 and to seek price reductions through the anticipated volume of orders. Whether as a single or multi-supplier system, whenever a range of products is covered, which is often the case, it is unlikely that the suppliers admitted to the dynamic purchasing system will during the entire lifetime of the system be able to supply the whole range of products and that they will maintain their cost advantage as compared to supplies from outsiders. This is a dilemma for the procuring entity, which is known from experience with framework agreements. Allowing procurement from outsiders removes the incentive for participating in the system. Prohibiting outside competition locks the procuring entity into commercial terms that may, over time, lo longer reflect the best offers available on the market. The system established by EU Directive 2004/18/EC aims at ensuring the widest possible choice for the procuring entity, therefore requiring a dynamic purchasing system to remain open to any potentially interested supplier throughout its duration.552 However, the short duration of the system, compounded with the absence of any purchase obligations on the side of the procuring entity has caused fears that the system may be of limited practical use.553
550
Dynamic purchasing systems are considered by some commentators to be better than suppliers lists for urgent procurement in cases such as utilities (which might otherwise be conducted using the request for quotations method envisaged by the UNCITRAL Model Law on Procurement of Goods, Construction and Services, also known as shopping). However, particularly so far as utilities are concerned, other commentators stress the quality risks that arise in such outsourcing. U.S. Office of Management & Budget, Office of Federal Procurement Policy, Best Practices for Multiple Task and Delivery Contracting 7 (Washington, D.C., July 1997). EU Directive 2004/18/EC, article 33(4). Roger Bickerstaff and Helen Kingston, The New Procurement Law Regime: Implementation in the United Kingdom, (8 March 2006) <www.twobirds.com/english/publications/ articles/The_New_ Procurement_Law_Regime.cfm?RenderForPrint=1> (3 December 2007).
551
552 553
296
Dynamic purchasing systems may also share some other shortcomings of framework agreements. A procuring entity would usually be under a general duty to afford the suppliers a fair opportunity to compete.554 Practical experience with framework agreements has shown, however, that procuring entities often fail to make use of a second phase competition, demonstrating a certain propensity for single source awards. Observers of framework agreements have cited various reasons for such non-competitive second-phase awards, including continuity of supplier (initial low-value awards being followed by others of greater value), practical considerations such as timing and lack of adequate notice favouring incumbent contractors, collusion, biased or inadequate technical specifications and inadequate assessment of prices submitted.555 EU Directive 2004/18/EC aims at ensuring maximum competition by requiring the procuring entity to issue invitations to tender for each specific contract to be awarded under a dynamic purchasing system.556 Dynamic purchasing systems may also enable a central procuring entity or an external purchasing body to undertake procurement on behalf of or for a number of entities. Such aggregated purchasing can lead to bulk purchase discounts, enhancing value for money, and offering freedom of choice for endusers where contracts are entered into with several suppliers with differing products. However, experience with framework agreements shows that there is a risk that a central or external purchasing body may accommodate customer agencies by reducing competition, and that may, in turn, mean using competitive techniques, or technical requirements, or prequalification requirements that favour a specific firm and which unreasonably restrict competition, and may lead to suppliers gaining effective monopolies.557 Further, a dynamic purchasing system may be of long duration and wide in coverage, closing off markets from the periodic competition contemplated by the procurement regulations (for example, effectively securing the market for a certain group of suppliers). To this extent, dynamic purchasing systems could compromise the common procurement objectives of fair and equitable treatment, integrity and public
554
FAR 16.505 (b)(1). See, further, paragraph 30. U.S. General Accounting Office, Contract Management: Not Following Procedures Undermines Best Pricing Under GSAs Schedule Report No. GAO-01-125 (28 November 2000) <www.ogc.doc.gov/ogc/contracts/cld/papers/GAO1-125.pdf> (4 December 2007); Benjamin, Multiple Award Task and Delivery Order Contracts: Expanding Protest Grounds and Other Heresies, 31 Pub. Contract L. J. (2002) 429.
555
556 557
EU Directive 2004/18/EC, article 33(4). It should be noted that certain systems, such as the EU under Directive 2004/18/EC, do not permit contracts between entities other than parties to the initial dynamic purchasing system (article 32 (2)), and therefore ad hoc centralized purchasing is not possible. However, in the United States, for example, many government agencies permit other organizations to place orders on their multiple-award contracts.
297
confidence in the procurement system if their operation is not appropriately regulated and overseen.
Chapter III. Contract Performance through Electronic Means

Contracts are performed by the delivery of the goods sold or the services promised. There is in principle no difference between contracts negotiated by electronic means or otherwise. In either situation, contract performance means that a new right over the object of performance is created to the benefit of the purchasing or hiring party, or an existing right is assigned to that party. Generally the rule is the same, whether contract performance involves rights to a tangible thing (such as such as moveable or immovable property), or an immaterial good (such as technical or professional advice or other form of information). Differences exist, however, as regards the manner in which such rights must be created or transferred in order to produce the intended legal effect. Perfection and transfer of rights is one of the areas of contract performance where existing law may pose obstacles to the use of electronic means. Typically, such problems will be the result of the possessory nature of some security interests or of the requirement of a tangible medium to effect the assignment. As used in this chapter, the expression rights in tangible goods refers to property rights or security interests in corporeal moveable property, including particularly commodities and manufactured goods, other than the money in which the price (in case of a sales contract) is to be paid. The expression other rights refers to intangible assets (other than property rights in tangible goods or intellectual property rights558), which have an economic value that makes them capable of being negotiated in the course of business, including in particular trade or financial receivables, investment and other securities.
A. Traditional Methods for Transfer and Perfection of Rights

The challenge for the establishment of workable systems for electronic transfer of rights in tangible goods and other rights consists in how to devise adequate electronic methods that are functionally equivalent to the transfer methods recognized by the law. For a better understanding of the problems raised by the
558
This chapter does not deal with electronic delivery of digitized goods or on-line software licensing.
298
use of electronic communications in connection with creation and perfection of rights, it is useful, therefore, to briefly recapitulate the essential rules for the transfer of rights by traditional means. 1. Transfer of Rights in Tangible Goods and Other Rights Methods for transfer of property interests in tangible property are generally based on two legal concepts, namely, the principle of consent559 and the principle of delivery.560 Additional methods include registration and symbolic delivery. Although these additional methods are usually regarded as conceptual variants of either the principle of consent or the principle of delivery, they are presented hereafter separately for ease of reading. (a) Transfer by Consent According to the principle of consent, the property passes from the transferor to the transferee by means of a contract between them implying the transfer of property.561 In legal systems that follow the principle of consent, all that is required for the transfer of property under a validly concluded sales contract is the parties agreement about sale of the goods and their status as buyer and seller. However, some legal systems give special emphasis to the intention of the parties with respect to the transfer of property.562 Those legal systems require clear evidence of the parties agreement upon the ownership of the transferee. Such intention may be expressed in the underlying contract (such as a sales contract) but is to be understood individually. It may even be concluded without
559
Transfer by consent is the prevailing method in common law legal systems and, within the civil law tradition, in jurisdictions influenced by French law. For a comparative overview of methods of transferring moveable property, see Rodolfo Sacco, Le transfert de la proprit des choses mobilires dtrmines par acte entre vifs, in Akadmiai Kiad, Pteri and Lamm (eds.), General Reports to the 10th International Congress of Comparative Law (Budapest, 1981) 247; see also Ulrich Drobnig, Transfer of Property, in Hartkamp et al. (eds.), Towards a European Civil Code (The Hague/London/Boston: Kluwer, 2nd ed. 1998) 495; information on various legal systems may further be found in Alexander von Ziegler et al. (eds.), Transfer of Ownership in International Trade (Paris/New York: Kluwer, 1999). Transfer by delivery is the general rule in civil law jurisdictions that follow the Roman law distinction between title (titulus) and form (modus) of transfer, such as most Ibero-American legal systems, and in jurisdictions influenced by the German law or in codifications inspired by the German Civil Code. E.g. France (Code Civil, Articles. 1138, 1583, 938); Italy (Codice Civile, Art. 1376), Japan (Civil Code, Art. 176). This is particularly the case in common law jurisdictions, such as Australia (Alexander von Ziegler, supra note 559 at 12), the common law provinces of Canada (id., at 83), England (id., at. 135).
299
560
561
562
a contract of sale. However, in some of those legal systems transfers of property in general, or in respect of specific goods, while valid and effective as among the transferor and the transferee, may not be enforceable against third parties until the transfer is registered in a registry system, or until the goods are actually delivered to the transferee. Apart from tangible goods, the consent of the parties is in many legal systems sufficient for the transfer of other property (intangibles) as well. Special rules are often found, however, in respect of assignments of payment claims (receivables).563 Indeed, while an assignment may be valid and binding on the assignor and the assignee, it has no effects on the debtor, unless the debtor has acquired knowledge of the assignment. In this respect, legal systems differ as to whether a notice to the debtor is required or whether any other act results in the debtor acquiring knowledge of the assignment. (b) Transfer by Registration Based also on consent is the principle of registration, which requires consent of the parties and registration by an office with statutory rights to take records.564 The transfer is completed with the inclusion of a corresponding record of the transaction in the registry system. Registration serves to ensure legal certainty especially when the achieved ownership cannot be primarily shown by physically shifting possession (e.g. with immovables). Transfer by registration is sometimes needed in respect of certain forms of intangible property. For example, transfer of shares or other securities issued by companies may need to be effected through appropriate records in the companys books, at least for the purpose of becoming effective vis--vis the company or third parties. Some jurisdictions have also established a system of filing information about assignments of trade receivables for the purpose of providing evidence of title to the receivables, notice about assignment to interested third parties or a method for determining priorities.565
563
Legal Aspects of Receivables Financing: report of the Secretary-General (U.N. document A/CN.9/397), para. 30 (Yearbook of the United Nations Commission on International Trade Law, Volume XXV: 1994 (United Nations publication, Sales No. E.95.V.20), part two, chap. V, sect. A). E.g. Germany (Brgerliches Gesetzbuch (BGB), 873 sect. 1) for immovable property. U.N. document A/CN.9/397, para. 30.
564 565
300
(c) Transfer by Delivery The principle of delivery is also based on consent but requires in addition the physical delivery of the asset to the transferee.566 Countries take different approaches as to the relationship between the underlying consent expressed in the contract and an additional consent about the transfer of goods itself (real agreement) which comes with the delivery. As far as the underlying contractual consent is the basis for the transfer by delivery, the validity of the transfer is affected by the validity of the contract itself.567 On the other hand, an independent real agreement to transfer is not affected by the contract, the validity of the transfer is in this case determined independently (doctrine of abstraction).568 Transfer by delivery is the norm for the effective transfer of certain types of intangible property. Negotiable instruments, such as bills of exchange and promissory notes, are typically negotiated by transfer of possession, whether voluntary or involuntary, of the instrument by a person other than the issuer to a person who thereby becomes its holder. Except for negotiation by a remitter, if an instrument is payable to an identified person, negotiation requires transfer of possession of the instrument and its endorsement by the transferor. If an instrument is payable to bearer, it may be negotiated by transfer of possession alone.569 (d) Transfer by Symbolic Delivery Even in countries that build on the principle of delivery, physical delivery of the goods is not always necessary. Possession of the goods can be left with the transferor or an agent of the transferor, where the parties agree on a legal relationship that assigns indirect possession to the transferee.570 Property rights in goods may also be deemed to have been transferred when the transferee is given the means for exercising or claiming control over the goods. For example,
566
E.g. Austria (Allgemeines brgerliches Gesetzbuch (ABGB), 426, Germany (BGB, 929 sect. 1), Greece (Civil Code Art. 1034), Netherlands (Civil Code book 3, Art. 3:84 para. 1), Russian Federation (Civil Code, Art. 223, sect. 1), Spain (Cdigo Civil, Art. 609), Switzerland (Civil Code, Art. 714, sect.1). Netherlands. Germany. Article 13 of the U.N. Convention on International Bills of Exchange and International Promissory Notes, for example, reflects this principle by providing that an instrument is transferred by endorsement and delivery of the instrument by the endorser to the endorsee; or by mere delivery of the instrument if the last endorsement is in blank. The same principle can be found in articles 11 and 16 of Annex I to the Convention Providing a Uniform Law on Bills of Exchange and Promissory Notes (Geneva, 7 June 1930) (League of Nations, Treaty Series, vol. CXLIII, p. 259, No. 3313 (1933-1934). E.g. Austria (ABGB, 427), Germany (BGB, 930).
301
567 568 569
570
in many jurisdictions, the transfer of the bill of lading operates as a transfer of the constructive possession of the goods, and may operate as a transfer of title to the goods.571 Other examples include surrendering to the transferee the keys of a warehouse where the goods are stored or surrendering to the transferee the documents (such as a warehouse receipt) necessary to claim delivery of the goods from a bailee holding such goods to the order of the holder. 2. Security Interests in Tangible Goods and in Intangible Property This section briefly describes the main methods for creating and perfecting security interests.572 For that purpose, it is important to distinguish between formal requirements, if any, for a security agreement to be binding as between the parties and those requirements that need to be met in order for the security creditor to be able to enforce the security against third parties. Except for a few jurisdictions that dispense altogether with form requirements for all or at least for certain kinds of security interests, such as purchase money, security agreements are in most cases subject to certain form requirements and typically need to be in writing.573 In most legal systems, a formal contract, whilst necessary, does not exhaust the legal requirements; it must be supplemented by other means of publication. If the secured party does nothing more than to enter into the security agreement with the debtor, that security interest is unperfected. An unperfected security interest may be completely valid and enforceable against the debtor, but may not be effective against third parties or may be subordinate to the rights of certain third parties, such as the trustee in a bankruptcy proceeding or creditors of the debtor. The ways in which a security
571
In Sanders Bros v. Maclean & Co. (1883) 11 QBD 327, at 341, the court stated: A cargo at sea while in the hands of the carrier is necessarily incapable of physical delivery. During this period of transit and voyage, the bill of lading by the law merchant is universally recognised as its symbol, and the indorsement and delivery of the bill of lading operates as a symbolical delivery of the cargo. Property in the goods passes by such indorsement and delivery of the bill of lading, whenever it is the intention of the parties that the property should pass, just as under similar circumstances the property would pass by an actual delivery of the goods. The information provided in this section draws on conclusions reached at an earlier study by the Secretariat on security interests (U.N. document A/CN.9/131, Yearbook of the United Nations Commission on International Trade Law, Volume VIII: 1977 (United Nations publication Sales No. E.78.V.7), part two, chap. II, sect. A) and on an earlier note by the Secretariat on article 9 of the Uniform Commercial Code of the United States of America (Id., part two, chap. II, sect. B). Although some details of the information contained in those documents may be dated, the basic principles and concepts set out in those documents are still relevant. U.N. document A/CN.9/131, (UNCITRAL Yearbook 1977), supra note 572 at 180.
572
573
302
interest can be perfected typically depend on the nature of the collateral and the underlying transaction. (a) Perfection by Possession Transfer of possession used to be (and in some legal systems still is) the main method for perfecting security interests in tangible goods. The secured party usually has possession from the moment the collateral is in its physical pos session or in the physical possession of a third person who holds it for the secured partys account. Perfection by possession serves two important purposes. Firstly, possession by the secured creditor serves as a notice to third parties that the creditor has a security interest in the goods in its possession. Secondly, because no two persons can physically possess the same goods at the same time, perfection by possession effectively avoids the creation of conflicting security interests in the same goods, thus guaranteeing the singularity of the creditors security interest. However, with a view to alleviating the limitation to the debtors ability to trade the goods pledged as security, in many legal systems, perfection by possession has been increasingly replaced with other methods, and has become of reduced commercial significance. Nevertheless, even in such legal systems, transfer of possession remains essential for the creation of security interests in respect of negotiable instruments, bills of lading, warehouse receipts and other nego tiable documents of title. In each case possession of the paper document creates a security interest in the claim, the rights or the goods represented by that document. (b) Perfection by Registration Another method for perfecting security interests is registration. Generally, a security agreement that is otherwise in accor dance with the appropriate requirements has the effect of giving rise to a legal relationship between the contracting parties even before registration. However, registration, where required, is typically a condition precedent for giving effect to a security interest vis--vis third persons.574 As the need for credit for businesses in particular expanded, and businesses came to generate and make use of intangible assets, countries were obliged to develop other mechanisms for recording security rights. Many countries ultimately came to rely on the concept of registration as the primary mechanism for recording security rights in movable property.575 One reason for this preference is that registration facilitates searches by third persons. It also avoids, on the part of the
574 575
Id., at 182. Security Interests: Draft Legislative Guide on Secured Transactions Note by the Secretariat (U.N. document A/CN.9/631, add.3, of 11 June 2007), para. 3.
303
creditor, all doubts about the proper place of registration, and avoids refiling in case of removal of the debtors domicile or the location of the goods. (c) Other Methods Formalities other than a contract or registration mainly take the form of marking the encumbered goods or of advertising the security interest. Marking of the encumbered goods with the secured creditors name is prescribed for certain goods in some jurisdictions either in addition to, or in place of, registration; rarely is it the exclusive method of publication. Much the same way as the registration of security interests, marking of en cumbered goods is intended to warn third parties against the existence of security interests; it may also help to prevent unauthorized dispositions by the debtor. In some countries private systems of collecting and publishing informa tion on security interests seem, in effect, to combine registration with advertisement. Indeed, in some countries the registration of security interests is published in private trade journals. Advertisement of security interests may serve as the basis of private registers kept by credit agencies.
B. Transfer or Creation of Rights by Electronic Means of Communication

1. General Legal Obstacles Legal obstacles to the electronic transfer of property rights in tangible goods and intangible property or to the creation of security interests in either type of property may result from form requirements for the validity, effectiveness or proof of the agreements to transfer or create the rights in question. Additional obstacles may relate to difficulties in establishing the functional equivalence between the transfer or creation method in a paper-based environment and its electronic analogous. All methods for transferring both property rights in tangible goods and intangible property or for creating security interests in either type of property presuppose at least the agreement of the parties on the transfer of such property or creation of the security interest. The spectrum of form requirements for the validity of the agreement may range from a written document signed by the parties, which in some jurisdictions may be made by a stamp or mechanical means as well as by hand, to a public deed drawn by a notary public. Intermediate requirements include other formalities, such as a certain number of witnesses or authentication of signatures by a notary public. In some legal systems, a statutory contract form is required.
304
The replacement of paper-based methods for transferring rights in tangible goods, transferring intangible property or creating security interests in tangible goods or intangible property with electronic equivalents presupposes therefore the resolution of the following legal issues: the satisfaction of writing and signing requirements; the evidential value of electronic communications; the determination of the place of contract formation. These issues were discussed in detail in chapter I above. The following sections will therefore focus on particular challenges posed by electronic negotiability. 2. Electronic Equivalents of Negotiable Instruments and Documents of Title Surmounting the issues of writing and signature in an electronic context does not solve the issue of negotiability which has been said to be perhaps the most challenging aspect of implementing electronic commerce in international trade practices.576 Where the law requires physical delivery of goods for the purpose of transferring property or perfecting security interests in such goods, a mere exchange of electronic messages between the parties would not be sufficient for effectively transferring property or perfecting a security interest, however evident the parties intention to transfer the property or perfect the security interest might have been. Therefore, even in jurisdictions where the law recognizes the legal value and effectiveness of electronic communication or records, no such communication or record could alone effectively transfer property or perfect a security interest without an amendment of the law governing transfer of property or perfection of security interests. (a) Basic Conditions for Electronic Negotiability The prospects for developing electronic equivalents of acts of transfer or perfection might be more positive where the law has at least in part dispensed with the strict requirement of physical delivery, for instance, by attributing to certain symbolic acts the same effect as the physical delivery of certain goods. One such example may be where the law attributes to the transferee or secured creditor the constructive possession of the goods transferred or pledged by virtue of an act of the parties that confers on the transferee the means for claiming control over the goods. Conceivably, the law could attribute the same effect to the entry of the transfer agreement into a registry system administered by a trusted third party or to an acknowledgement sent by the party in physical possession of the goods that these are held to order of the transferee or the secured creditor. Rights in goods represented by documents of title are typically conditioned by the physical
576
See Jeffrey B. Ritter and Judith Y. Gliniecki, International Electronic Commerce and Administrative Law: The Need for Harmonized National Reforms, 6 Harv. J. L. & Tech. (1993) 263, 279.
305
possession of an original paper document (the bill of lading, warehouse receipt, or other similar document). Early analysis of the legal basis for the negotiability of documents of title had indicated that [t]here is generally no statutory means in place by which commercial parties, through the exchange of electronic messages, can validly transfer legal rights in the same manner possible with paper documents.577 This conclusion is also essentially valid for rights represented by negotiable instruments. Moreover, the legal regime of negotiable instruments is in essence based on the technique of a tangible original paper document, susceptible to immediate visual verification on the spot. In the present state of legislation, negotiability cannot be divorced from the physical possession of the original paper document.578 Thus, the main obstacle to accommodating electronically transmitted documents of title is to generate them in such a way that holders who claim due negotiation will feel assured that there is a document of title in existence, that it has no defects upon its face, that the signature, or some substitute therefore is genuine, that it is negotiable, and that there is a means to take control of the electronic document equivalent in law to physical possession.579 The practical significance of requiring transfer of actual or constructive possession of tangible goods, or of the physical surrender of a document that represents goods, or embodies other rights is to prevent conflicts between parties claiming entitlement to the same goods or rights. As no two persons can possess certain goods or hold a given document or negotiable instrument at the same time, the requirement of possession ensures the regularity of the chain of negotiation. Ultimately, the law aims at ensuring that only one person can effectively exercise control over the goods, the document that represents them or the negotiable instrument. Based on that premise, one could conceive the formulation of rules that establish the conditions under which electronic systems can replicate that function. One set of general rules to that effect is contained in the United States Uniform Electronic Transactions Act (UETA). The rationale for including such a provision in UETA is explained in the official commentary to UETA as follows:
577 578
Id. See K. Bernauw, Current Developments Concerning the Form of Bills of Lading Belgium, in A.N. Yannopoulos (ed.),Ocean Bills of Lading: Traditional Forms, Substitutes and EDI Systems (The Hague: Kluwer Law International, 1995) 87, 114. Donald B. Pedersen, Electronic Data Interchange as Documents of Title for Fungible Agricultural Commodities, Idaho L. Rev. (1995) 719, 726.
579
306
Paper negotiable instruments and documents are unique in the fact that a tangible token a piece of paper actually embodies intangible rights and obligations. The extreme difficulty of creating a unique electronic token which embodies the singular attributes of a paper negotiable document or instrument dictates that the rules relating to negotiable documents and instruments not be simply amended to allow the use of an electronic record for the requisite paper writing. However, the desirability of establishing rules by which business parties might be able to acquire some of the benefits of negotiability in an electronic environment is recognized by the inclusion of this section on Transferable Records. Section 16, Transferable records of UETA establishes the criteria for the legal equivalence of electronic records to notes or records under Articles 3 and 7, respectively, of the Uniform Commercial Code (UCC). The essential criterion for such equivalence is that the electronic record needs to be of such nature that a person may exercise control over the record. Under Section 16 acquisition of control over an electronic record serves as a substitute for possession of an analogous paper negotiable instrument. More precisely, control under Section 16 serves as the substitute for delivery, endorsement and possession of a negotiable promissory note or negotiable document of title. Section 16(b) allows control to be found so long as a system employed for evidencing the transfer of interests in the transferable record reliably establishes [the person claiming control] as the person to whom the transferable record was issued or transferred. The key point, as indicated in the official commentary, is that a system, whether involving third party registry or technological safeguards, must be shown to reliably establish the identity of the person entitled to payment. A person is considered to have control of a transferable record under UETA if a system employed for evidencing the transfer of interests in the transferable record reliably establishes that person as the person to whom the transferable record was issued or transferred.580 A person having control of a transferable record acquires the status of holder of the record, for the purposes of Section
A system satisfies this requirement, and a person is deemed to have control of a transferable record, if the transferable record is created, stored, and assigned in such a manner that: (1) a single authoritative copy of the transferable record exists which is unique, identifiable, and, except as otherwise provided in paragraphs (4), (5), and (6), unalterable; (2) the authoritative copy identifies the person asserting control as the person to which the transferable record was issued; or the authoritative copy indicates that the transferable record has been transferred, the person to which the transferable record was most recently transferred; (3) the authoritative copy is communicated to and maintained by the person asserting control or its designated custodian;
580
307
1-201(20) of the UCC, and has the same rights and defences as a holder of an equivalent record or writing under the UCC, including the rights and defences of a holder in due course, a holder to which a negotiable document of title has been duly negotiated, or a purchaser, respectively. Delivery, possession, and endorsement are not required to obtain or exercise any of the rights under this subsection. The official commentary to UETA suggests that control requirements may be satisfied through the use of a trusted third party registry system, but a technological system which met such exacting standards would also be permitted under Section 16. According to the official commentary, section 16 provides legal support for the creation, transferability and enforceability of electronic note and document equivalents, as against the issuer/obligor. A similar provision is contained in the draft convention on the carriage of goods, which is being considered by UNCITRAL since 2001 (hereafter the UNCITRAL Draft Carriage of Goods Convention).581The UNCITRAL Draft Carriage of Goods Convention includes provision to the effect that the issuance, control, or transfer of an electronic transport record shall have the same effect as the issuance, possession, or transfer of a paper document. A condition for such same effect is that the use of a negotiable electronic transport record must be subject to procedures that, amongst others, provide for (a) a method to effect the exclusive transfer of that record to an intended holder; (b) an assurance that the negotiable electronic transport record retains its integrity; and (c) the manner in which the holder is able to demonstrate that he is the holder. Furthermore, the draft makes any provision that applies to a transport document, equally applicable to the corresponding electronic version of it. These rules constitute an electronic supplement to the negotiability system envisaged by the UNCITRAL Draft Carriage of Goods Convention. To support the operation of transport documents both in paper and in electronic form, the UNCITRAL Draft Carriage of Goods Convention deals with a number of questions thus far not dealt with by any existing international maritime convention, with a

581
(4) copies or revisions that add or change an identified assignee of the authoritative copy can be made only with the consent of the person asserting control; (5) each copy of the authoritative copy and any copy of a copy is readily identifiable as a copy that is not the authoritative copy; and (6) any revision of the authoritative copy is readily identifiable as authorized or unauthorized. The latest version of the draft convention on the carriage of goods [wholly or partly] [by sea] is contained in U.N. document A/CN.9/WG.III/WP.101. The latest summary of the preparatory work is contained in the provisional agenda for the twenty-first session of UNCITRAL Working Group III (Transport Law) (U.N. document A/CN.9/WG.III/ WP.100). Both documents are available <www.uncitral.org/uncitral/en/commission/ working_groups/3Transport.html>.
308
view to clarifying who is in control of the goods during their carriage. Therefore, the UNCITRAL Draft Carriage of Goods Convention includes a comprehensive set of rules about the right of control under the contract of carriage. The right of control is defined in such a way that it should satisfy the requirements that may exist under national law for the validity of a pledge on the goods. It includes, among other attributes, the right of the controlling party to change the name of the consignee and to demand delivery of the goods at an intermediate place en route before their arrival at destination. Second, the UNCITRAL Draft Carriage of Goods Convention clarifies who is the controlling party, provides for rules on the transferability of the right of control as well as for the qualified obligation of the carrier to follow the instructions given by the controlling party. Both UETA and the UNCITRAL Draft Carriage of Goods Convention were formulated at a level of generality that should accommodate both electronic registries and a technical device such as an electronic token or unique electronic record. The certainty created by rules of this nature should provide the requisite incentive for industry to develop the systems and processes, which involve significant expenditures of time and resources, to enable the use of such electronic documents.582 This is the point where the imagination of the law-maker meets the limits of science. Modern technology makes it possible to transmit information in electronic form satisfactorily down a chain of parties. The same process could conceivably be used by any of the parties to transmit the information that it renounces its title in favour of another person, thus amounting to an endorsement of the instrument. However, if a person is to receive an exclusive benefit, such as possessory title, by receiving a particular electronic communication or record, the addressee will need to be satisfied that no identical communication or record could have been sent to any other person by any preceding party in the chain, creating the possibility of other claimants to the title. It is true that no electronic communication or record can be actually the very same as another electronic communication or record; but as long as it is technically possible, with no possibility of detection, to replicate an electronic communication or record exactly and send it to someone else, there cannot be a guarantee of singularity.
582
R. David Whitaker, Rules under the Uniform Electronic Transactions Act for an Electronic Equivalent to a Negotiable Promissory Note, 55 Bus. Law. (1999) 437, 449. The author notes, however, that meeting the standards of UETA section 16 will not be an easy task, and will require a carefully designed and supervised set of systems and practices. Meeting those standards will not be an easy task, and will require a carefully designed and supervised set of systems and practices. The key element will be data integrity. Courts evaluating the control of a transferable record may be expected to focus on the systemic protections e.g. division of labour, complexity of backup systems, activity logs, and security of copies stored offsite to verify content which make it difficult to tamper with the record without detection.
309
As we have seen, the main difficulty in developing electronic negotiability systems is how to create an electronic equivalent of possession of goods, negotiable instruments or documents of title. At least in theory, these problems could be solved if computer technology were able to create a unique electronic record that could be exclusively held by a holder and transferred to another without of replication at some point down the negotiating chain.583 Unfortunately, computer technology has not yet been able to create such a unique electronic record. This means that, for the time being, all electronic negotiability systems rely on electronic registries. Electronic registries have been in place for some time now, with more or less success, depending on the industry affected. The fact that they have not yet entirely replaced traditional negotiation methods, despite the sophistication of some of them, is a result of some of their limitations, which are discussed below. (b) Practical Experience: Electronic Registry Systems Recognizing that electronic communications lack the tangibility that characterizes traditional negotiable instruments and documents of title, registry systems attempt to replace both paper documents and their physical transfer with electronic records. Typically, there will be one or more electronic records that contain essentially the same information that would otherwise be contained in the paper document. This information would be kept in permanent records maintained by the registry. Furthermore, there would be a series of electronic communications recording transactions in the goods, rights or receivables represented by the electronic record. In essence, an electronic registry system is not different from traditional registry systems, such as a land registry system. The difference is the intended purpose of substituting entries in a registry for the physical delivery of a negotiable instrument or document of title. A few examples of such systems are given in the following paragraphs to illustrate their functioning. (i) Dematerialized Securities The system of using dematerialized securities essentially seeks to enable transactions of securities to be conducted and completed electronically using a system of account transfers without any physical exchange of documents, such as share certificates and transfer deeds.584 These securities systems are
583
See a theoretic discussion in Robert E. Kahn and Patrice A. Lyons, Representing Value as Digital Objects: A Discussion of Transferability and Anonymity, 5 J. Telecom. & High Tech. L. (2006) 189. Dematerialization has become an essential feature of modern trade in securities by settlement systems such as Euroclear in Brussels, Cedel in Luxembourg, the Depository
584
310
designed to reduce the paperwork, expense and risks associated with physical documents, which are replaced with records in electronic form. Besides the basic system of using dematerialized securities, some securities systems also offer an immobilization system, that is, they retain the physical security in a vault and give the holder dematerialized rights to the security by virtue of the holder being the account-holder. Although there are variations between jurisdictions, the key participants in a dematerialized securities system are the depository (sometimes also referred to as custodian), the issuer, trading intermediaries and the investor. The depository is an organization whose primary function is to maintain an electronic system of accounts in a central registry. This central registry contains a record of the holdings of dematerialized securities and the rights and restrictions arising therefrom, which are held by depository participants on behalf of investors at any time. Trading intermediaries are normally financial institutions, brokers and other entities authorized to be members of the depository and who hold accounts with the depository. In any securities transaction, investors who trade in dematerialized securities through their trading intermediaries will do so in a recognized securities market, such as the stock exchange. Details of these transactions on any given day will normally be transmitted automatically to the depository for clearing and settlement585 and, if not, trading intermediaries will inform the depository on their own accord. Once details of these transactions are transmitted to the depository, the clearing and settlement process begins and trading intermediaries will begin to deal directly with the depository. The clearing and settlement process is normally based on a delivery versus payment principle, which means that by the announced settlement day (usually a number of days after the trading day) payments will have to be made by the buying investors, and securities will have to be transferred from the selling investors to the buyers. In the interim period, the depository will have informed all trading intermediaries of every buy and sell order made on the day of trading
Trust Corporation in the United States, CREST and the Central Gilts Office in the United Kingdom, SICOVAM in France, Monte Titoli in Italy and numerous comparable systems elsewhere.
585
Apart from maintaining the central registry and any immobilization of securities, the depository may also undertake the function of clearing and settlement where this is not undertaken by another organization. Clearing refers specifically to the processing of a trade and establishing what the investors owe each other as a result of that trade. Settlement refers to the transfer of value between the investors so as to complete the transaction. If a separate organization is used for clearing and settlement, the role of the depository is limited to the maintenance of the central register of information. In the United States, for example, a separate organization known as the National Securities Clearing Corporation has been set up to undertake this function.
311
in accordance with electronic reports received from the securities market and the trading intermediary. The purpose of this is to allow the depository participants to confirm the transactions and make corrections to the records. The depository will then proceed to process the net securities traded and the net amounts due by each trading intermediary, or rather the balance of securities and payments due or owed by each trading intermediary on behalf of their respective investors.586 On settlement day, the depository causes the accounts of each trading intermediary to reflect the net settlement of securities by re-allocating securities from the accounts of the net sellers to the net buyers by electronic means. Trading intermediaries also meet the net financial obligations of each investor by wiring funds between designated settling banks. The intended transfer of dematerialized securities is completed when the latest securities holding information is entered into the central register at the depository but the transfer of ownership is recognized as of the date of the transaction. A legal arrangement created for entirely dematerialized securities may take one of several approaches. The fungible nature of book-entry securities may be explicitly recognized, leading to a new characterization of the investors property interest. The investor may be treated as a co-owner of all the securities of the type it has purchased that are held by the intermediary. The investor then retains a specific property interest in the securities but can only claim it on a proportional basis. However, where a different model is used, the legal arrangement may instead deprive the investor of its property interest in the securities and place it in a debtor/creditor relationship with the intermediary. In that case, the deposit of securities becomes analogous to a bank deposit with special characteristics. In such an arrangement, the investors interest may be further refined. The investors claim may be secured with the specific assets held for the investor serving as collateral for the claim. Alternatively, the investor may become part of a preferred class of creditors, with a claim that is secured generally by all securities held by the intermediary for customers. A study by the Bank for International Settlements587 indicates that market participants have made considerable efforts to simplify the flow of securities across borders through the development of global custody networks, international central securities depositories and links between national central securities depositories.
586
The transmission of instructions and information during the clearing and settlement process is conducted through various secure communication networks such as S.W.I.F.T. or Cedcom. These instructions may be checked against validation rules such as the International Securities Identification Number (ISIN) to ensure its accuracy. ISIN is a code that uniquely identifies a specific securities issue. The organization that allocates ISINs in any particular country is the National Numbering Agency (NNA), which is typically a recognized stock exchange. Cross-border Securities Settlement (Bank for International Settlements, March 1995) at 46.
587
312
The availability of book-entry settlements makes it possible for settlement systems, securities depositories and custodians to offer comparable settlement services in a wide range of national markets. However, the comparability of settlement services masks important distinctions between the legal frameworks that may be applied to the same securities in different countries. (ii) Electronic Registries as a Replacement of Paper Bills of Lading A bill of lading performs a number of functions in connection with the contract of carriage, the underlying sales contract and the documentary credit issued to facilitate payment. In its basic function, the bill of lading is an acknowledgement by the carrier of the receipt of the goods indicated in the bill of lading and their loading on board a vessel. The bill of lading also evidences the contract of carriage and embodies its essential terms. These two functions are also common to non-negotiable transport documents, such as the seaway bill. The bill of lading performs however one additional and unique function, in that it allows for negotiation of goods represented by the bill of lading while there are in transit. It is this function that explains the central role played by the bill of lading in the documentary credit system: by embodying rights to the goods, the bill of lading is capable of being pledged as collateral for loans or letters of credit issued to secure payment to the seller. Replicating the first two functions of the bill of lading in an electronic environment is a relatively simple task, and in many trades seaway bills have already been effectively replaced with electronic communications. It is the negotiability function of the bill of lading that gives rise to most difficulties, mainly because it is impossible to physically hold, endorse or deliver an electronic record. These functions must therefore be fulfilled by other means capable of establishing, at any time, who is the rightful holder of the electronic equivalent of the bill of lading, that is, who has the right to give instructions to the carrier for the deliver of the goods, who may transfer that right to another person and who is entitled to claim delivery of the goods when they arrive at destination. During the past decades, many attempts have been made by a number of international organizations, whether intergovernmental or non-governmental, and by various groups of users of electronic communication techniques to reproduce the functions of a traditional paper-based bill of lading in an electronic environment.588 The following paragraphs describe one such system, the Bill of
588
For an overview of various developments, see Marek Dubovec, The Problems and Possibilities for Using Electronic Bills of Lading as Collateral, 23 Ariz. J. Intl & Comp. L. (2006) 437, 466. Information on earlier initiatives, such as the Sea Docs experiment and the CMI Rules for Electronic Bills of Lading can be found in earlier studies done by UNCITRAL (see U.N. document A/CN.9/WG.IV, WP.69) (UNCITRAL Yearbook 1996, part two, chap. II, sect. B).
313
Lading for Europe (Bolero).589 The Bolero system, which became operational in September 1999, relies on a messaging system, where users communicate with each other through a central registry application with standard EDI messages. It is a closed system in that only subscribers are permitted to use it.590 One of the key components of the Bolero system is a registry for Bolero Bills of Lading, which will store data on behalf of the users of the Bolero system. A Bolero Bill of Lading is designed to replicate the essential business functions of a bill of lading via electronic communications recorded in a central database operated by a trustworthy third party. The Bolero Bill of Lading is an electronic record that contains essentially the same information as a conventional bill of lading issued by a carrier to a shipper, and is designed to replicate the functions of a physical bill of lading as evidence of a contract of carriage, a receipt for the goods, and a document representing the entitlement to possession of the goods. Transactions involving a Bolero Bill of Lading are effected through the Bolero Title Registry Record. The Title Registry holds a record for each consignment, which is updated when instructions are received from the holders of rights to the consignments represented by a Bolero Bill of Lading. The registry requires the holder to demonstrate its rights by means of the identification and authentication procedures set forth in the Bolero Rulebook591 in order for the instructions to be registered. These instructions may establish a right in the favour of a third party (for example a pledge to the benefit of a bank issuing a documentary credit), communicate a transfer of ownership to the consignment represented by the Bolero Bill of Lading, or confirm the delivery of the cargo and surrender of the Bolero Bill of Lading. The designation of a new holder becomes effective by means of an acknowledgement, by the carrier, that from that time on it holds the goods described in the Bolero Bill of Lading to the order of such new holder. The new holder (either owner or secured creditor) then replaces the previous holder and is issued the authentication devices (cryptographic keys) for communicating
589
For an extensive discussion of the Bolero system, see Tobias Eckardt, The Bolero Bill of Lading Under German and English Law (Frankfurt: Quadis/Sellier, 2004).
590
Users of the Bolero System are required to accept the terms of the Bolero Rulebook. Appended to the Rulebook are the Operating Procedures, which is a detailed descrip tion of Bolero System operations with a few specific and technical rules to ensure that the technology and legal infrastructure mesh together without gaps or inconsistencies. Operational Service Contracts provide for the services that Bolero Inter na tion al supplies, as well as for system security, information dissemination and retention, and similar rights and duties involved in a central information service. A service contract governs the rights and obligations of the Bolero Association and its members and participants (for further information, see <www.boleroassociation.org/dow_docs.htm>). <www.boleroassociation.org/downloads/rulebook1.pdf>.
591
314
with the registry. Taken together, these messages and instructions replicate the physical transfer of a paper bill of lading. The legal relationships among all parties involved are set forth in the Bolero Rulebook,592 which deals, inter alia, with the validity of electronic transac tions and the legal effect of Bolero Bills of Lading. The Bolero Rulebook establishes security procedures to ensure that the entitlements are generated, authenticated and transferred only by the authorized holder.593 As bailor, the shipper is declared to have constructive possession of the goods represented by the Bolero Bill of Lading, which may be transferred by the designation of a new holder (either as a new holder-to-order, pledgee holder, bearer holder, or consignee holder). The designation of a new holder becomes effective by means of an acknowledgement, by the carrier, that from that time on it holds the goods described in the Bolero Bill of Lading to the order of such new holder. As regards the contract of carriage, the system envisages its transfer through novation: each carrier in the Bolero System appoints Bolero International to act as its agent, and Bolero International re-makes each contract of carriage on behalf of the carrier with each new transferee.594
592 593
Id. For instance, Section 2.2.1 of the Rulebook requires all users of the Bolero system to digitally sign their messages, which is done by using private keys duly certified for use within the system. By adhering to the terms of the Rulebook, Bolero users agree to accept the evidential admissibility of electronic data and messages and are estopped from repudiating Bolero messages they send. The Rulebook makes it possible to incorporate, directly or by reference, the provisions of underlying contracts, notably the carriage contract and the letters of credit, so as to bind parties who are liable and to benefit those intended to receive the rights. Thus, an example of a sale of goods financed with a documentary credit using the Bolero system may be as follows: Upon receipt of the cargo from the seller, the carrier creates a Bolero Bill of Lading and designates the seller as the shipper and holder of the Bolero Bill of Lading and the importer as the to order party. The seller sends a message to the registry designating the confirming bank of the documentary credit as the pledgee holder of the Bolero Bill of Lading and sends on the required documents via digitally signed Bolero messages. The confirming bank examines the Bolero Bill of Lading, finds it in order, credits the sellers account, and designates a bank that issued the documentary credit as the new pledgee holder. The issuing bank performs any additional checking of the documents that it requires and charges the importers account. The issuing bank then relinquishes its pledge and, by message to the registry, designates the importer as the holder of the Bolero Bill of Lading. The importer is already to order party for the bill, and now, as holder also, can transfer the bill. On behalf of the carrier, Bolero International notifies the importer that the carrier holds the goods to its order. The importer sells the goods in transit. Accordingly, the importer designates the buyer as the holder-to-order (i.e. both holder and to order party) of the Bolero Bill of Lading. On behalf of the carrier, Bolero International notifies the holder-to-order that the carrier holds the goods to its order. The goods arrive at the destination port and the buyer surrenders the Bolero Bill of Lading. No
315
594
(iii) The International Registry of Security Interests on Mobile Equipment A more recent example of an electronic registry system is the International Registry for the recording of international interests in aircraft objects, which was established in Shannon, Ireland, in June 2004, to carry out the registry functions under the Convention on International Interests in Mobile Equipment (also known as the Cape Town Convention),595 and the Protocol to the Convention on International Interests in Mobile Equipment on Matters specific to Aircraft Equipment (the Aircraft Protocol).596 The Cape Town Convention was adopted to facilitate the efficient financing of mobile equipment.597 Traditionally, many countries would subject the constitution and effects of consensual and non-consensual proprietary interests in movables to the law of the place where they are located. In a practical sense, however, mobile equipment does not have a specific location. Therefore, the traditional rule is considered to be inadequate when applied to objects that are continuously moving from country to country in the ordinary course of business. In the case of space assets, such as telecommunication satellites, these objects are not connected to a country at all. Consequently, an international framework providing for the recognition, enforcement and priority of proprietary rights in mobile equipment was needed. The Cape Town Convention establishes a system for the perfection of international interests in mobile equipment by means of registration in the
further Bolero-based transactions are now possible for the Bolero Bill of Lading. Bolero International gives notice of surrender to the carrier and confirms surrender to the buyer. The buyers representative appears at the port with the proof of identification required by the carrier or port. The carrier delivers the goods to the buyers representative.
595
The Convention was adopted at a diplomatic conference convened under the joint auspices of the International Institute for the Unification of Private Law (UNIDROIT) and the International Civil Aviation Organization (ICAO) at Cape Town, South Africa on 16 November 2001 (see the text at <www.unidroit.org/english/conventions/mobileequipment/main.htm>). The Cape Town Convention entered into force on 1 April 200 and had, as of 6 December 2007, 20 Contracting States (see <www.unidroit.org/english/ implement/i-2001-convention.pdf>). See also Roy M. Goode, The Official Commentary on the Convention on International Interests in Mobile Equipment and Protocol Thereto on Matters Specific to Aircraft Equipment (Rome: UNIDROIT, 2001). Protocol to the Convention on International Interests in Mobile Equipment on Matters specific to Aircraft Equipment (Cape Town, 16 November 2001) (see the text at <www. unidroit.org/english/conventions/mobile-equipment/main.htm>). The Aircraft Protocol entered into force on 1 March 2006 and had, as of 6 December 2007, 19 Contracting States (see http://www.unidroit.org/english/implement/i-2001-aircraftprotocol.pdf>). In general, the term mobile equipment encompasses objects that by their very nature are used internationally. Examples of such objects are aircraft, satellites, trains, movable oil-rigs and containers.
596
597
316
international registry system. In this manner, international creditors will be provided with a secure system to record their interests in aircraft objects, and to establish the priority of their interests vis--vis competing creditors within and outside the insolvency of the debtor. The Convention/Protocol provides the holders of international interests with a range of default and insolvency-related remedies. The international registry operates under a supervisory authority, a body with legal personality under public international law and enjoying the privileges and immunities usually granted to international organizations, such as immunity from legal process and enforcement measures. The registration system is an electronic, internet-based service. It will be available on-line without interruption. According to article 16 of the Cape Town Convention, access to the international registry is open to all parties that comply with the registration requirements, regardless of whether they are located in Contracting States or non-Contracting States. The system is designed, however, to prevent the recording of interests without the required prior electronic consent. The checking of the registration applications, the registrations themselves, and the responses to searches will be processed automatically, without requiring human involvement. The registration of an interest is complete upon entry of the required information into the international registry data base so as to be searchable.598 Thus, any risk associated with interruption of services provided by the registry system will be allocated depending upon the point in the process of registration when the services became unavailable. If that happened before the registration data have been entered into the database, the risk is borne by the party submitting the data for registration. If it occurs after that point, the risk is borne by a person searching the registry.599 (c) Limitations and Problems of Registry Systems The establishment of electronic equivalents to paper-based registration systems raises a number of particular problems. They include the satisfaction of legal requirements on record-keeping, the adequacy of certification and authentication methods, possible need of specific legislative authority to operate electronic registration systems, the allocation of liability for erroneous messages, communication failures, and system breakdowns; the incorporation of general terms and conditions; and the safeguarding of privacy.
598
Cape Town Convention, article 19, paragraph 1.Pursuant to article XX of the Aircraft Protocol, the search criteria for an aircraft object include at least the name of its manufacturer, its manufacturers serial number and its model designation, supplemented as necessary to ensure uniqueness. Ronald C.C. Cuming, The International Registry for Interests in Aircraft: An Overview of Its Structure, 11 Uniform L. Rev. (2006) 18, 38.
317
599
Record-keeping, written form and signature requirements have been discussed previously. The following analysis will focus, therefore, on two main aspects: liability and functionality of electronic registries. (i) Liability for System Failure A registry system may be exposed to a number of risks. The registry may delay the recording or notification of entries in the system, which may deprive a secured creditor or holder from the protection it hoped to enjoy in the event, for example, of the debtors insolvency prior to the perfection of its security interest. There may also be errors in entering information into the system (for example, by attributing the right to a person other than the legitimate holder, or by misrepresenting the amount of the debt or the value of the goods). Also, the registry may fail to apply appropriate identification procedures, thus making it possible for an unauthorized person to impersonate a bona fide holder. Information and records kept in the registry may be compromised by unauthorized access by parties not entitled to deal with the registry, or may even be completely lost, for example, as a result of a system breakdown if appropriate recovery procedures are not in place. To some extent these risks are not specific to the electronic environment, and might occur in connection with the operation of any system whereby a person or entity is entrusted with safekeeping of legally relevant record and other information and with recording transactions concerning the assets covered by the registry. Thus for example, a land registry is exposed to the risk that the records are lost by fire or looting, and a notary public may be held liable for effecting transactions done by unauthorized person due to negligence in confirming their identity or powers. What makes electronic registries different is the reliance on information technology, which due to its inherent capability of connectivity, increases their exposure to outside attacks.600 Attacks from outsiders may aim both at defrauding the system and effecting unauthorized transactions, as well as at causing the system to collapse. The potential damage that can be caused by loss of information, transactions delay or interruption caused by deliberate attacks can be hardly overestimated. The risk of security threats is compounded by the risk of unintended system failure, such as for instance, a crash down in the
600
In connection with electronic real estate registry systems, for example, it has been said that one indirect cost is the risk of different types of fraud and theft. While individual fraud may be less likely with electronic than with paper documents, the new electronic databases of official real estate records will create an attractive new target for sophisticated cyber-criminals and will require additional security to prevent attacks (Sam Stonefield, Electronic Real Estate Documents Context: Unresolved Cost-Benefit Issues and a Recommended Decisional Process, 24 W. N. Eng. L. Rev. (2002) 205, 233-234.
318
supporting computer systems, or unavailability of telecommunication by no fault of the registry.601 It is not surprising, therefore, that liability becomes an important issue in the development of registry systems. Under the Cape Town Convention, for example, the registrar is liable for compensatory damages for loss suffered by a person directly resulting from an error or omission of the registrar and its officers and employees or from a malfunction of the international registration system except where the malfunction is caused by an event of an inevitable and irresistible nature, which could not be prevented by using the best practices in current use in the field of electronic registry design and operation, including those related to back-up and systems security and networking.602 However, the registrar is not liable for factual inaccuracy of registration information received by the registrar or transmitted by the registrar in the form in which it received that information nor for acts or circumstances for which the registrar and its officers and employees are not responsible and arising prior to receipt of registration information at the international registry. The Cape Town Convention requires the registrar to procure insurance or a financial guarantee covering its liability.603 The instruments governing the functioning of the international registry do not expressly establish a monetary ceiling for the registrars liability. In the case of the Bolero system, the liability of Bolero International Ltd., under whose authority the system is run, is subject to the limitations and conditions set forth in the Operational Service Contract, which is entered into between individual users and Bolero International Ltd. Liability in connection with misdirection or loss of messages, delay in sending messages, alteration, incorrect identification, false creation, breach of confidentiality or other errors in connection with messages processed by Bolero International Ltd. is generally subject to a limit of US$ 100,000 per user per occurrence. The same limit applies to errors and service failures in connection with certificates issued by Bolero International Ltd. However, in the event that all certificates issued by Bolero International become unreliable or unsuitable for usage as stated in their documentary forms, and, as a direct result, the user suffers loss, Bolero International undertakes to pay damages to the user up to the limit of US$1,000,000. The aggregate limit of loss per calendar year is US$ 10,000,000 irrespective of the number of claims or of the number of users entitled to claim in any calendar year.
601
For a few practical examples of business interruption in the on-line securities market, see Anna Lee, Insuring Cyberspace: Why Traditional Insurance Policies Are Not Enough: The Nature of Potential E-Commerce Losses and Liabilities, 3 Vand. J. Ent. L. & Pract. (2001) 84. Cape Town Convention, article 28, paragraph 1. Cape Town Convention, article 28, paragraph 4.
319
602 603
(ii) Effectiveness of Electronic Communications for Transfer of Rights From a legal point of view, perhaps the most interesting issue in connection with electronic registry systems is the interplay between the medium used for registration and the nature of the rights being registered. Experience shows that, if the starting point of most initiatives was merely to render a negotiating mechanism more efficient by replicating electronically the transactions that used to be done by other means, the result sometimes was that the very nature of the rights being registered ended up being affected by the disappearance of the tangible medium to represent those rights. This interplay between medium and substantive law became particularly evident as a result of the expanding trade in dematerialized securities, which raised various questions regarding the nature of the securities and the relationships between the parties involved. In some instances, the new medium led to a redefinition of legal concepts traditionally applied in securities transactions, including, in some cases by legislative action. A study conducted some time ago by the French National Council of Credit and Securities,604 for example, identified the following main issues, and the answers, as appropriate, that were given to them in practice: (a) Legal nature of securities. Investment securities issued in paper form had traditionally been regarded as corporeal moveable goods that incorporated or represented certain rights (e.g. a credit against the issuing company or shareholders rights). Without the paper support, it became necessary to reclassify investment securities as intangible property;605 (b) Nature of rights established by a book entry. As long as investment securities were regarded as tangible property, the rights of the holder in the securities were typically regarded as property rights. That understanding was questioned in the case of dematerialized securities, which often
604
Conseil National du Crdit et du Titre, Problmes juridiques lis la dmatralisation des moyens de payement et des titres (Paris, Banque de France, 1997), p. 122. This, in term, affects the remedies available to the holder under the laws of the forum. For instance, where securities have the nature of rights in rem, the holder can easily escape the consequences of the intermediarys insolvency proceedings by claiming restitution of the securities. If, however, the securities are only treated as creating rights ad personam, the holder would be forced to file a claim as any other unprivileged creditors of the insolvent intermediary (See generally, Steven L. Schwarcz, Intermediary Risk in a Global Economy, 50 Duke L. J. (2001) 1541; Kon Sik Kim, A Civil Law Jurists Perspective on Intermediary Risk in the Indirect Holding System For Securities: A Comment on Schwarcz & Benjamin, 12 Duke J. Comp. & Intl L. (2002) 335; for a discussion of conflicts of law issues in connection with capital markets, see Herbert Kronke, Capital Markets and the Conflict of Laws, 286 Recueil des Cours de lAcadmide de Droit International de La Haye (2000) 245.
605
320
are not individualized, being sometimes not even susceptible of being individualized; (c) Effect of book entry. The introduction of an intermediary between the issuer of the securities and their holder has raised the question as to whether the record of issuance or transfer of the securities in the depositorys accounts (book entry) was simply a means of evidencing the rights of the holder or whether it was constitutive of such rights; (d) Nature of contract between depository and investor. As long as investment securities were represented by paper documents, it had been held that the relationship between the holder and the depository of the certificates was assimilated to the relationship between a bailor and a bailee. The absence of a tangible instrument capable of being physically or constructively possessed by either party has given rise to doubts as to the nature of the contract between depository and investors and the extent of the latters remedies in case of breach by the depository. A study on issues of cross-border securities settlements prepared by the Bank for International Settlements in 1995606 points out that there are considerable differences among countries with regard to the legal framework applying to the ownership, transfer and pledging of securities. The legal framework for multi-tiered systems falls into one of two general types: one applies the conventional legal framework for securities to book-entry systems by presuming the existence of physical securities; the other builds a new legal framework for dematerialized securities that are issued solely in electronic form. The first type of arrangement relies on a legal fiction to fit book-entry securities into a paper-based legal theory. The law pretends that the securities exist in physical form. Ownership rights and the transfer and pledging of book-entry securities are then explained in terms of possession and delivery through the mechanisms of immobilization or global certificates, in which physical securities are deemed to be deposited and kept in fungible (interchangeable) form. An investor shown on the books of the intermediary is regarded as having physical possession of the respective securities and, as a consequence, acquires a property interest in them. The completion of book entries is deemed to have the same effect as physical delivery of the relevant securities. Domestic variations in the legal underpinnings of the systems for dematerialized securities lead to inevitable conflicts in connection with cross-border securities settlements.607 In addition to legal issues arising out of the involvement of the
606
Cross-border Securities Settlement (Bank for International Settlements, March 1995), at 50. Id., 47-57.
321
607
intermediaries608 or relating the accounting practices and safekeeping procedures employed by the custodian and sub-custodians609 the Bank for International Settlements noted a wide variation among countries in their legal treatment of securities, which raises significant issues for cross-border securities transactions.610 Other issues identified by the Bank for International Settlements include systemic risks, conflict-of-laws problems, difficulties in establishing finality of delivery and payment and problems related to the bankruptcy of participants in the system. Although these problems are not specifically related to or caused by the use of electronic records or messages, they are aggravated by the complexities of dematerialization. Recent international efforts have aimed at eliminating or at least reducing some of these legal problems by harmonizing conflicts of law rules and introducing uniform substantive rules for securities held by intermediaries. The Hague Conference on Private International Law has a prepared an international convention that determines the law applicable to a number of issues in respect of securities held with an intermediary (hereafter The Hague Securities Convention).611 The Hague Securities Convention was negotiated in view of the practical need in a large and growing global financial market to provide legal certainty and predictability as to the law applicable to securities
608
E.g. that the intermediaries may become insolvent, act negligently or commit fraud. The issuer seeks discharge of its obligations, but risks performing to the wrong party. E.g. the risk that custodians and sub-custodians, although making appropriate debits and credits to the investors accounts, may not have sufficient securities to support the total number of accounting entries they make. Shortfalls in custodial holdings may develop for a number of reasons: inefficiencies in the settlement process, poor accounting controls, or intentional fraud. For example, dematerialized securities issued in one country may be handled in the book-entry system of a second country that relies on an immobilization scheme and the legal fiction of possession. In that case, it may be unclear whether the dematerialized securities qualify as securities in the second country. If they do not, the transferee of the dematerialized security may acquire a legal interest, which is significantly different from the one it expected. The question of a securitys status under the law becomes critical if an intermediary becomes insolvent. Further difficulties arise in connection with depository receipts, which are issued in one country to establish entitlement to a security held in custody in another country. Depository receipts are then traded and settled in the domestic market in place of the foreign securities that they represent. However, the legal status of these quasi-securities is not always clear. For example, a depository receipt may not entitle the investor to make a claim on the issuer of the original securities; it may only symbolize a claim on the intermediary or serve as evidence of a debtor/creditor relationship between the intermediary and the investor. Moreover, it is not clear what happens to depository receipts if the underlying securities are invalid, or if depository receipts are over-issued relative to the amount of the underlying securities. Convention on the Law Applicable to Certain Rights in respect of Securities held with an Intermediary (The Hague, 5 July 2006), not yet in force.
609
610
611
322
that are held through clearing and settlement systems or other intermediaries. The Hague Securities Convention established uniform rules for determining the law applicable in respect of the following issues: (a) the legal nature and effects against the intermediary and third parties of the rights resulting from a credit of securities to a securities account; (b) the legal nature and effects against the intermediary and third parties of a disposition of securities held with an intermediary; (c) the requirements, if any, for perfection of a disposition of securities held with an intermediary; (d) whether a persons interest in securities held with an intermediary extinguishes or has priority over another persons interest; (e) the duties, if any, of an intermediary to a person other than the account holder who asserts in competition with the account holder or another person an interest in securities held with that intermediary; (f) the requirements, if any, for the realization of an interest in securities held with an intermediary; and (g) whether a disposition of securities held with an intermediary extends to entitlements to dividends, income, or other distributions, or to redemption, sale or other proceeds. By harmonizing conflicts of law rules in those areas, the Hague Securities Convention helps reduce legal risk, systemic risk and associated costs in relation to cross-border transactions involving securities held with an intermediary so as to facilitate the international flow of capital and access to capital markets. The new conflicts of law rules provided by the Hague Securities Convention will be supplemented in the future by a discrete set of substantive rules that are currently at the final stages of negotiation at the International Institute for the Unification of Private Law (UNIDROIT).612 The new international convention being prepared by UNIDROIT aims to promote internal soundness and cross-border system compatibility by providing the basic legal framework for intermediated securities
612
Preliminary Draft Convention on Substantive Rules Regarding Intermediated Securities (hereafter the Preliminary Draft Convention) was prepared by the UNIDROIT Study Group on Transactions on Transnational and Connected Capital Markets (UNIDROIT Study Group LXXVIII) and is currently being considered by UNIDROIT Committee of Governmental Experts for the Preparation of a Draft Convention on Substantive Rules Regarding Intermediated Securities The latest version of the Preliminary Draft Convention was adopted by the Committee oat its fourth session (Rome, 21-25 May 2007) and is available at <www.unidroit.org/english/publications/proceedings/2007/study/78/s78-94-e.pdf> (10 December 2007).
323
holding systems. Within the context of holding securities with an intermediary, the draft UNIDROIT convention sets forth the rights resulting from the credit of securities to a securities account and establishes the minimum characteristics for the acquisition and disposition of securities and for the provision of securities as security or collateral. The new draft UNIDROIT convention also clarifies the rules regarding finality of book entry transfers and irrevocability of instructions by declaring that a book entry is effective when made and precludes upper-tier attachment. Furthermore, the draft UNIDROIT convention: establishes a priority ranking among competing interests with respect to securities; protects the good faith holder of securities from adverse claims; declares the rights of the account holder and the responsibilities of the intermediary in the event of insolvency; establishes a regime for loss allocation; and clarifies the legal relationship between the collateral taker, the intermediary and the account holder in situations where the latter provides its securities as collateral. No such general legal framework exists for the electronic transfer of rights in tangible goods as a substitute for the documents of title currently used. This has been recognized as a serious limitation to international negotiation of goods in transit through electronic means whenever an equivalent of a traditional document of title (such as a bill of lading) is needed for trading or financing purposes. Transfer of title to tangible goods, or creation of security interests in tangible goods often require transfer of physical or symbolic possession of such goods. The development of documents that represent such goods has greatly facilitated the movement of goods in international trade. That result was legally possible by legislative recognition of the function of transport and warehousing documents as substitutes for physical delivery of the goods. A similar conclusion may be reached in connection with the function of the endorsement of negotiable instruments such as letters of exchange and promissory notes. Systems whereby title to goods and receivables might be transferred by means of electronic messages, without creation and circulation of paper documents, might result in significant savings in the overall cost of trade transactions. To a large extent, practical solutions may be crafted by contractual arrangements binding upon the users of any such systems. However, voluntary rules, upon which some systems may be based, give way when they conflict with a States laws and may not be enforceable against or binding upon third parties.613 This
The lack of success of some registry systems developed thus far can be explained by a reluctance of the industry, in particular the banking industry, to accept electronic equivalents of paper bills of lading as adequate collateral under a letter of credit. Developers of technological solutions take the view that the acceptance of electronic documents is not a matter of changing transportation law to enable electronic documentation, but is predominantly a matter of gaining the trust and security of the customers who use shipping documents in their trade relations. (Dubovec, supra note 588, at 465). This may be partly true, but the trust and security of those customers would be strengthened by
613
324
becomes even more of a problem where documents of title and other negotiable instruments are used to support trading and financing transactions that take place outside a regulated exchange accessible only to pre-registered participants. The experience with trade in dematerialized securities shows that, although electronic trading systems can be established at a national or regional level, their smooth operation in a globalized economy sooner or later demands the adoption of international rules on conflicts of laws, as well as some uniform substantive law rules. Indeed courts may be inclined to accept electronic book entries as the functional equivalent of endorsement and physical transfer of a document of title as between persons and entities that participate in a closed negotiation system where all parties agree in advance to be bound by the rules of that system. The result may differ, however, in situations where the courts are asked to enforce rules of an electronic system against a third party that did not expressly agree to be bound by those rules. For example, the administrator in insolvency proceedings of a company that sold goods in transit to another participant in an electronic system shortly before becoming insolvent would be expected to try to obtain delivery of the goods to the benefit of the insolvency estate. When faced with a concurring claim by a buyer admitted to the electronic system, the insolvency administrator would very likely refuse being bound by rules that he or she has not accepted and that are not statutorily enforceable against him or her. The insolvency administrator would likely insist that the court should apply the laws of the place where the goods are located to determine whether a transfer of property had validly occurred. In such a case, unless the applicable law recognizes the validity of the negotiability rules of the electronic system, the court would likely hold that only the actual endorsement of a bill of lading, a nearly universally accepted means of transferring the right to claim delivery of the goods, could have granted that right to the alleged holder. The difficulty in such a situation is that most general cargo in international shipping is not negotiated in established exchanges and, therefore, is not subject to trade-specific rules. The introduction of a general statutory recognition of electronic systems for the transfer of rights in goods in transit, as contemplated by the UNCITRAL Draft Carriage of Goods Convention, will possibly ease some of the legal concerns that currently exist. It remains to be seen, however, how
clear legal rules enacted by state legislators, and not with the confusing legal structure created by some registry systems. It has been said that one of the reasons why the use of electronic commerce is not developing in line with technological capability is that there is little law governing its use. The exchange of data electronically does not itself pose a problem. However, when the data represents negotiable documents that cover valuable assets, an established legal structure is needed.(Id.).
325
these systems will be set up in practice, how commodity-specific systems will interplay with carriers and bankers practices, how trading systems can be developed for cargo not ordinarily negotiated in established exchanges, and whether communications and information technology will be able to offer an alternative to electronic registry systems. The same considerations apply, mutatis mutandis, to the negotiation of traditional negotiable instruments, such as bills of exchange and promissory notes.
326
CONcLUSION
Information and communications technology have revolutionized the way business is conducted both at the national and international level. The extraordinary growth in the worlds economic output and the expansion of international trade that we have witnessed in the past decades would have been inconceivable without the supporting role of technology. This has led not only to increased efficiency in business, both in terms of time and cost savings, but also to the emergence of new industries and business models. Called upon to develop a legal and regulatory environment to facilitate electronic commerce, legislators and policymakers around the world soon discovered the limitations of the legal concepts and principles inherited from times when paper documents represented the paradigm for legally relevant communications and records. They soon realized that the intrinsically borderless nature of electronic communications challenged the territorial nature of State law. Switching from a paper-based to an electronic communications system has therefore demanded an adaptation of legal concepts so as to ensure the legal recognition of contracts negotiated electronically and the evidentiary function of electronic records. UNCITRAL made a significant contribution to assist countries in that transition, by offering principles and rules that permitted the application of existing laws to the new electronic environment. Technology has its limits, however. The advantages brought by the ease and speed of communication are balanced by the difficulty in replicating all the features of traditional documents with a comparable level of simplicity. This is particularly the case when it comes to enabling contract performance through electronic means. Here new structures were needed to provide legal certainty for the sometimes complex procedures that must be in place to satisfy the security needs of the international trading and financing systems. Achieving that result in a manner that does not tie the law to the technology of the day bound to become obsolete faster than the law could be changed has been one of the greatest legislative challenges of the past years. The fact that not all obstacles have yet been removed means that the law like the technology whose operation it intends to support will remain a work in progress for some time.
327
BIBLIOGRAPHy
I. Books
AUDIT, Bernard, La vente internationale de marchandises (Paris, L.G.D.J., 1990). BOTHE, Michael and KILIAN, Wolfgang, Rechtsfragen grenzberschreitender Datenflsse, (Kln, Dr. Otto Schmidt, 1992). BAUM, Michael S. and PERRITT JR., Henry H., Electronic Contracting, Publishing and EDI Law (New York, Wiley Law Publications, 1991). BRAVO, Fabio, Contrattazione telematica e contrattazione cibernetica (Milan, Giuffr, 2007). BYRNE, James E. and TAYLOR, Dan, ICC Guide to the eUCP (Paris, ICC, 2002). CZERWENkA, G. Beate, Rechtsanwendungsprobleme im internationalen Kaufrecht: das Kollisionsrecht bei grenzberschreitenden Kaufvertrgen und der Anwendungsbereich der internationalen Kaufrechtsbereinkommen (Berlin, Duncker & Humblot, 1988). COMAND, Giovanni and SIcA, Salvatore, Il commercio elettronico, Turin, G. Giappichelli, 2001 Drner Heinrich and Elhers Dirk, Rechtsprobleme der EDV (Neuwied/ Frankfurt, Metzner, 1989). DRNER, Heinrich and EHLERS, Dirk, Rechtsprobleme der EDV (Neuwied Frankfurt, Metzner, 1989). DRObNIG, Ulrich, Transfer of Property, in Hartkamp and others (eds.), Towards a European Civil Code (2nd edn, The Hague/London/Boston, Kluwer, 1998). EckARDT, Tobias. The Bolero Bill of Lading under German and English Law (Frankfurt, Quadis/Sellier, 2004). ENDERLEIN, Fritz and MASkOW, Dietrich. International Sales Law (New York/London/ Rome, Oceana, 1992). FERRARI, Franco, Vendita internazionale di beni mobili, vol. 1, pp. 52-53, in Scialoja/ Branca (eds) Commentario del Codice Civile (Bologna, Zanichelli,1994). GOODE, Roy M., The Official Commentary on the Convention on International Interests in Mobile Equipment and Protocol Thereto on Matters Specific to Aircraft Equipment (Rome, UNIDROIT, 2001). GRAf VON WESTPHALEN, Friederich and SEIDEL, Ulrich, Aktuelle Rechtsfragen der Software-Vertrags- und Rechtspraxis, (3rd edn, Kln, RWS Verlag Kommunikationsforum 1992). GUILLEMARD, Sylvette, Le droit international priv face au contrat de vente cyberspatial (Montreal, ditions Yvon Blais, 2006).
328
HONNOLD, John, Uniform Law for International Sales under the 1980 United Nations Convention (2nd edn, Deventer, Kluwer, 1991). HONSELL, Heinrich (ed.), Kommentar zum UN-Kaufrecht (Springer, Berlin/Heidelberg/ New York, 1997). HUET, Jrme and MAISL, Herbert, Droit de linformatique et des tlcommunications (Paris, Litec, 1989). JUNkER, Abbo and BENEckE, Martina, Computerrecht (3rd edn, Baden-Baden, Nomos Verlaggesellschaft, 2003). KAROLLUS, Martin, UN-Kaufrecht, Eine systematische Darstellung fr Studium und Praxis (Wien/New York, Springer Verlag, 1991). MAGNUS, Ulrich, Wiener UN-Kaufrecht (CISG), Art 1 CISG No. 44, in Julius von Staudinger (ed.), Kommentar zum Brgerlichen Gesetzbuch mit Einfhrungsgesetz und Nebengesetzen, Band II (12th edn, Berlin, Sellier/de Gruyter, 1994). PERALES VIScASILLAS, Maria del Pilar. La formacin del contrato de compraventa internacional de mercaderas (Valencia, Tirant lo blanch, 1996). PILTZ Burghard, Internationales Kaufrecht Das UN-Kaufrecht (Wiener bereinkommen von 1980) in praxisorientierter Darstellung) (Mnchen, C.H. Beck, 1993). PALANDT, Otto, Brgerliches Gesetzbuch (60th edn, Mnchen, C.H. Beck, 2001). ScHLEcHTRIEM, Peter, Commentary on the UN Convention on the International Sale of Goods (Oxford, Clarendon Press, 1998). THIEffRY, Jean and GRANIER, Chantal, La vente internationale (2nd edn, Paris, Centre franais du commerce extrieur, 1992). VIVANT, Michel, et allii, Droit de linformatique et des raiseaux (Paris, Lamy, 1999). VON CAEMMERER, Ernst and ScHLEcHTRIEM, Peter, Kommentar zum einheitlichen UNKaufrecht (2nd edn, Mnchen, C. H. Beck, 1995).
II. Articles
AALbERTS, Babette and VAN DER HOf, Simone, Digital Signature Blindness: Analysis of Legislative Approaches toward Electronic Authentication, November 1999, p. 8 <rechten.uvt.nl/simone/Digsigbl.pdf> (3 December 2007). ALLEN, Tom and WIDDISON, Robin, Can Computers Make Contracts?, 9 Harvard Journal of Law and Technology (Winter 1996) 25-52. ARROWSMITH, Sue, Electronic Reverse Auctions under the EC Public Procurement Rules: Current Possibilities and Future Prospects, 11-6 Public Procurement Law Review (2002) 299-330. ARROWSMITH, Sue, Framework Purchasing and Qualification Lists under the European Procurement Directives: Part I, 8 Public Procurement Law Review (1999) 115. BAkER, Stewart A., LIcHTENbAUM, Peter, SHENk, Maury D. and YEO, Matthew S. E-Products and the WTO, 35-1 The International Lawyer (Spring 2001) 5-21.
329
BAkER, Stewart A., and YEO, Matthew S. (in collaboration with the secretariat of the International Telecommunication Union), Background and Issues Concerning Authentication and the ITU, briefing paper presented to the Experts Meeting on Electronic Signatures and Certification Authorities: Issues for Telecommunications, Geneva, 9 and 10 December 1999, Document No.2 <www.itu.int/osg/ spu/ni/esca/meetingdec9-101999/briefingpaper.html> (30 November 2007). BEALL, Stewart et allii, Role of Reverse Auctions in Strategic Sourcing, Center for Advanced Purchasing Studies (CAPS Research), 2003 <www.capsresearch.org/ publications/pdfs-public/beall2003es.pdf> (3 December 2007). BELLIA, JR. Anthony J.. Contracting with Electronic Agents, 50 Emory Law Journal (2001) 1047-1092. BENjAMIN, Michael, Multiple Award Task and Delivery Order Contracts: Expanding Protest Grounds and Other Heresies, 31 Public Contract Law Journal (2002) 429-477. BERNAUW, K, Current Developments Concerning the Form of Bills of Lading Belgium, in A.N. Yannopoulos (ed.), Ocean Bills of Lading: Traditional Forms, Substitutes and EDI Systems (The Hague, Kluwer Law International, 1995) 87-125. BLANcHETTE, Jean-Franois, Defining electronic authenticity: an interdisciplinary journey, <polaris.gseis.ucla.edu/blanchette/papers/dsn.pdf> (26 November 2007) (paper published in a supplemental volume of the 2004 International Conference on Dependable Systems and Networks (DSN 2004), Florence, Italy, 28 June-1 July 2004), 228-232. BOSS, Amelia, Electronic Data Interchange Agreements: Private Contracting Toward a Global Environment, 13 Northwestern School of Law Journal of International Law & Business (Spring/Summer 1992) 31-70. BOSS, Amelia, Taking UCITA on the Road: What Lessons Have We Learned?, 7 Roger Williams University Law Review (2001) 167-213. BOSS, Amelia H., Electronic Commerce and the Symbiotic Relationship between International and Domestic Law Reform, 72 Tulane Law Review (1998) 1931-1984. BOUcHER, Pierre, Technology versus Industry Practices, February 2003 <www. findarticles.com/p/articles/mi_qa4088/is_200302/ai_n9176581>(28 November 2007). CASEY, Timothy D. and MAGENAU, Jeff, A Hybrid Model of Self-regulation and Governmental Regulation of Electronic Commerce, 19 Santa Clara Computer and High Technology Law Journal (December 2002) 1-36. COX, Trevor, Chaos Versus Uniformity: the Divergent Views of Software in the International Community (<www.cisg.law.pace.edu/cisg/biblio/cox.html>). CUMING, Ronald C.C. The International Registry for Interests in Aircraft: An Overview of its Structure, 11 Uniform Law Review (2006) 18-59.
330
CURRAN, Elaine, BERNERT, Andrea and WIEGAND Anke, Electronic Procurement in the Public Sector: Factsheet on Latest Developments in E-procurement in the EU and its Member States <www.eic.ie/downloads/e_procurement.pdf> (28 November 2007). DRAETTA, Ugo, Internet et commerce lectronique en droit international des affaires, 314 Recueil des Cours de lAcadmie de Droit International de La Haye (2005) 99-232. DUbOVEc, Marek, The Problems and Possibilities for Using Electronic Bills of Lading as Collateral, 23 Arizona Journal of International and Comparative Law (2006) 437-466. DUMORTIER, Jos and VAN DEN EYNDE, Sofie, Electronic Signatures and Trusted Archival Services, <www.law.kuleuven.ac.be/icri/publications/172DLM2002.pdf?where> (25 November 2007). DAVIDSON, Duncan M, Protecting Computer Software: A Comprehensive Analysis, Arizona State Law Journal (1983) 611-784. DRObNIG, Ulrich, Transfer of Property, in Hartkamp and others (eds.), Towards a European Civil Code (2nd edn, The Hague/London/Boston, Kluwer, 1998) 495-510. ESTRELLA FARIA, Jose Angelo. Harmonizing the Law of International Electronic Contracting: Adjust the Rules but Dont Rewrite Them, in Andrea Schulz (ed.), Legal Aspects of an E-Commerce Transaction, International Conference in The Hague, 26 and 27 October 2004 (Sellier European Law Publishers GmbH, 2006) 73-98. EISELEN. Sieg, E-Commerce and the CISG: Formation, Formalities and Validity, 6-2 Vindobona Journal of International Commercial Law and Arbitration (2002) 310-311. FAScIANO, Paul, Internet electronic mail: a last bastion for the mailbox rule, 25-3 Hofstra Law Review (1997) 971-1003. FIScHER, Susanna F, Saving Rosencrantz and Guildenstern in a Virtual World? A Comparative Look at Recent Global Electronic Signature Legislation, 7-2 Boston University Journal of Science and Technology Law (2001) 229-238. FROOMkIN, Michael,Article 2B as Legal Software for Electronic Contracting - Operating System or Trojan Horse?, 13 Berkeley Technology Law Jour (1998) 1023-1062. GREENbERG, Marc H, A Return to Lilliput: The LICRA v. Yahoo! Case and the Regulation of Online Content in the World Market, 18 Berkeley Technology Law Journal (2003) 1191-1258. GLATT Christoph, Comparative Issues in the Formation of Electronic Contracts, 6 International Journal of Law and Information Technology (1998) 34-68. HERT, Paul de, Biometrics: Legal Issues and Implications, background paper for the Institute for Prospective Technological Studies of the European Commission (European Communities, Directorate General Joint Research Centre, 2005) 13 <http://cybersecurity.jrc.es/pages/ProjectlibestudyBiometrics.htm> (30 November 2007).
331
HOROVITZ, Bonna Lynn,Computer Software as a Good under the Uniform Commercial Code: Taking a Byte out of the Intangibility Myth, 65 Boston University Law Review (1985) 129-164. KAHN, Robert E. and LYONS, Patrice A., Representing Value as Digital Objects: A Discussion of Transferability and Anonymity, 5 Journal on Telecommunications & High Technology Law (2006) 189-198. KENNEDY-LOEST, Ciara and KELLY, Ruth. EC competition law rules and electronic reverse auctions: a case for concern?, 12-1 Public Procurement Law Review (2003) 27-30. KIM, Kon Sik. A Civil Law Jurists Perspective on Intermediary Risk in the Indirect Holding System For Securities: A Comment on Schwarcz & Benjamin, 12 Duke Journal of Comparative & International Law (2002) 335-338. KIRcHbERGER, Christine and RAMN Y OLANO, Jon, Issues of Security and Interoperabiity in Electronic Public Procurement, 47 Scandinavian Studies of Law (2004) 51-77. KObAYASHI, Bruce H. and RIbSTEIN, Larry E. Uniformity, Choice of Law and Software Sales, 8 George Mason Law Review (1999) 261-306. KObRIN, Stephen J., Economic Governance in an Electronically Networked Global Economy, in R. Hall and T. Biersteker (eds), The Emergence of Private Authority: Forms of Private Authority and Their Implications for Global Governance (Cambridge University Press) <www-management.wharton.upenn.edu/kobrin/Research/ revision%201.pdf> (30November 2007). KNIG, M. Michael, Software (Computerprogramme) als Sache und deren Erwerb als Sachkauf , Neue Juristische Wochenschrift (1993) 3121-3124. KRONkE, Herbert, Capital Markets and the Conflict of Laws, 286 Recueil des Cours de lAcadmide de Droit International de La Haye (2000) 245-386. LARSON, Marcus G., Comment: Applying Uniform Sales Law to International Software Transactions: the Use of the CISG, Its Shortcomings, and a Comparative Look at How the Proposed UCC Article 2b Would Remedy Them, 5 Tulane Journal of International and Comparative Law (1997) 445-488. LEE, Anna, Insuring Cyberspace: Why Traditional Insurance Policies are not Enough: The Nature of Potential E-Commerce Losses & Liabilities, 3 Vanderbilt Journal of Entertainment Law & Practice (2001) 84-93. LEROUGE, Jean-Franois, The Use of Electronic Agents Questioned under Contractual Law: Suggested Solutions on a European and American Level, 18 John Marshall Journal of Computer and Information Law (1999) 403-433. LIkER, Jeffrey K. and CHOI, Thomas Y.,Building Deep Supplier Relationships, Harvard Business Review (December 2004) 104. McCULLAGH, Adrian, LITTLE, Peter and CAELLI William, Electronic Signatures: Understand the Past to Develop the Future, 21-2 University of New South Wales Law Journal (1998).
332
MASON, Stephen, Electronic Signatures in Practice, VI-2 Journal of High Technology Law(2006) 148-164. MOGEY, James, How Technology Has Affected the Legal System: Software as UCC Goods: a Critical Look, 34 Howard Law Journal (1991) 299-312. MAGGS, Gregory E., The Waning Importance of Revisions to U.C.C. Article 2, 78 Notre Dame Law Review (2003) 595-628. MAXEINER, James R., Standard Terms Contracting in the Global Electronic Age: European Alternatives, 28-1 Yale Journal of International Law (2003) 109-182. NIcOLL, C. C., Can Computers Make Contracts?, The Journal of Business Law (January 1998) 35-49. OVERbY, Brooke A., Model Law on Electronic Commerce:Will Cyber Law Be Uniform? An Introduction to the UNCITRAL Model Law on Electronic Commerce, 7 Tulane Journal of International and Comparative Law (1999) 219-235. OROURkE, Maureen A.,Progressing Towards a Uniform Commercial Code for Electronic Commerce or Racing Towards Nonuniformity?, 14 Berkeley Technology Law Journal (1999) 635-658. POLk, William T. and HASTINGS, Nelson E., Bridge Certification Authorities: Connecting B2B Public Key Infrastructures, National Institute of Standards and Technology, September2000 <csrc.nist.gov/groups/ST/crypto_apps_infra/documents/ B2B-article.pdf > (3 December 2007). POGGI, Christopher T., Electronic Commerce Legislation An Analysis of European and American Approaches to Contract Formation, 41 Virginia Journal of International Law (2000) 224-277. PRIMAk, Scott L., Computer Software: Should the U.N. Convention on Contracts for the International Sale of Goods Apply? A Contextual Approach to the Question, XI-2 Computer Law Journal (1991) 197-231. PEDERSEN, Donald B., Electronic data interchange as documents of title for fungible agricultural commodities, 31 Idaho Law Review (1995) 719-746. RObERTSON, R. J., Jr, Electronic Commerce on the Internet and the Statute of Frauds, 49 South Carolina Law Review (1998) 787-846. RAZZANO, Robert T, Error 404 Jurisdiction Not Found: The Ninth Circuit Frustrates The Efforts Of Yahoo! Inc. To Declare A Speech-Restrictive Foreign Judgment Unenforceable, 73 University of Cincinnati Law Review (2005) 1743-1764. REED, Chris, What is a Signature?, 3 Journal of Information, Law and Technology (2000) <www2.warwick.ac.uk/fac/soc/law/elj/jilt/2000_3/reed/#fn6> (3 December 2007). RITTER, Jeffrey B. and GLINIEckI, Judith Y., International Electronic Commerce and Administrative Law: The Need for Harmonized National Reforms, 6 Harvard Journal of Law and Technology (1993) 263-285. ROOS, Frederik, First Come, Not Served: Domain Name Regulation in Sweden, 17-1 International Review of Law Computers and Technology 63-72.
333
SAccO, Rodolfo, Le transfert de la proprit des choses mobilires dtrmines par acte entre vifs, Pteri and Lamm (eds), General Reports to the 10th International Congress of Comparative Law (Budapest, Akadmiai Kiad, 1981) 247-268. ScHEcHTER, Roger E., The Unfairness of Click-on Software Licenses, 46 Wayne Law Review (2000) 1735-1803. ScHLEcHTRIEM, Peter, Uniform Sales Law The Experience with Uniform Sales Law in the Federal Republic of Germany, Juridisk Tidskrift (1991/92) 1-28. ScHWARcZ, Steven L., Intermediary Risk in a Global Economy, 50 Duke Law Journal (2001) 1541-1607. SNEDDON, Mark, Legislating to Facilitate Electronic Signatures and Records: Exceptions, Standards and the Impact of the Statute Book, 21-2 University of New South Wales Law Journal (1998) <www.austlii.edu.au/au/journals/ UNSWLJ/1998/59.html> (28 November 2007). SOLUM, Lawrence B. Legal Personhood for Artificial Intelligences, 70 North Carolina Law Review (1992) 1231-1287. SORIEUL, Renaud, CLIfT Jennifer R. and ESTRELLA-FARIA, Jos Angelo, Establishing a Legal Framework for Electronic Commerce: The Work of the United Nations Commission on International Trade Law, 35-1 The International Lawyer (2001) 107-122. SOUDRY, Ohad, Promoting Economy: Electronic Reverse Auctions under the EC Directives on Public Procurement, 4-3 Journal of Public Procurement (2004) 340-374. STEIN, Andrew and HAWkING, Paul, Reverse Auction e-Procurement: A Suppliers Viewpoint <ausweb.scu.edu.au/aw02/papers/refereed/stein/paper.html> (28 November 2007). STONEfIELD, Sam, Electronic Real Estate Documents Context: Unresolved Cost-Benefit Issues and a Recommended Decisional Process, 24 Western New England Law Review (2002) 205-243. SVANTESSON, Dan. Geo-identification and the Internet A New Challenge for Australias Internet Regulation, 14-2 Murdoch University eLaw Journal (2007) 155-177. TREPTE, Peter. Electronic Procurement Marketplaces: The Competition Law Implications, 10 Public Procurement Law Review (2001) 260-280. TALERO, Eduardo, Electronic Government Procurement: Concepts and Country Experiences, World Bank Discussion Paper (September 2001), paras 30-40 <wbln0018. worldbank.org/OCS/egovforum.nsf/c3c9b2819079a45d852569bc007722a0/ e5596442988cccfd85256af5006af56a/$FILE/ATTUQ5LL/egpdiscpaperdraft16. pdf> (12 November 2007). TURLEY, Susan L., Wielding the Virtual Gavel DOD Moves Forward with Reverse Auctions, 173 Military Law Review (2002) 1-66.
334
VERbIEST, Thibault and LE BORNE, Maxime, Le fonds de commerce virtuel: une ralit juridique? <www.droit-technologie.org>, 24 May 2002 (30 November 2007). WERNER, Jens, E-Commerce.CO.UK: Local Rules in a Global Net: Online Business Transactions and the Applicability of Traditional English Contract Law Rules, 6 International Journal of Communications Law and Policy (2000/2001) 1-10. WEIN, Leon E., The responsibility of intelligent artifacts: toward an automated jurisprudence, 6 Harvard Journal of Law and Technology (1992) 103-154. WHITAkER, R. David, Rules under the Uniform Electronic Transactions Act for an Electronic Equivalent to a Negotiable Promissory Note, 55 The Business Lawyer (1999) 437-453. WONG, David D., The Emerging Law of Electronic Agents: e-Commerce and Beyond, 33 Suffolk University Law Review (1999) 83-106. WYLD, David C., Auction Model: How the Public Sector can Leverage the Power of eCommerce through Dynamic Pricing, Price Waterhouse Cooper Endowment for the Business of Government, 2000 <www.businessofgovernment.org/pdfs/ WyldReport.pdf> (28 November 2007).
III.Cases
Canada Turner v. TELUS Communications Inc., Federal Court of Canada (2005 FC 1601),29 November 2005. Colombia Juzgado Segundo Promiscuo Municipal Rovira Tolima, Juan Carlos Samper v. Jaime Tapias, 21 julio 2003, Rad. 73-624-40-89-002-2003-053-00. France Court dappel de Paris, 11th Chamber, Section A, 10 November 1999, Presdok et Siline Gmbh D.J. v. F.C.O. fiduciaire SA <www.juriscom.net/documents/caparis19991110.pdf> (3 December 2007). Tribunal de Grande Instance de Paris, Ordonnence de rfer, 22 May 2000, Union des Etudiants Juifs de France (UEJF) et La Ligue contre le racisme et lantismitisme (LICRA) c/. Yahoo! Inc. et Yahoo France <www.juriscom.net/txt/jurisfr/cti/ tgiparis20000522.htm>. Germany Amtsgericht (District Court) Bonn, Case No. 3 C 193/01, 25 October 2001, JurPC Internet-Zeitschrift fr Rechtsinformatik und Informationsrecht, JurPC Web-Dok. No. 332/2002 <www.jurpc.de/rechtspr/20020332.htm> (11 September 2003).
335
Amtsgericht Erfurt, Case No. 28 C 2354/01, 14 September 2001, JurPC InternetZeitschrift fr Rechtsinformatik und Informationsrecht, JurPC Web-Dok. No.71/2002 <www.jurpc.de/rechtspr/20020071.htm> (25August2003). Amtsgericht Hannover, Case No. 501 C 1510/01, 7 September 2002, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 299/2002 <www.jurpc.de/ rechtspr/20020299.htm> (30 November 2007). Amtsgericht Westerburg, Case No. 21 C 26/03, 14 March 2003, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 184/2003 <www.jurpc.de/ rechtspr/20030184.htm> (30 November 2007). Bundesgerichtshof (Federal Court of Justice), Case No. XI ZR 367/97, 29 September 1998, JurPC Internet-Zeitschrift fr Rechtsinformatik und Informationsrecht , JurPC Web-Dok. No. 05/1999 <www.jurpc.de/rechtspr/19990005.htm (29 November 2007). Bundesgerichtshof, Case No. XII ZR 51/99, 14 March 2001, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 167/2001 <www.jurpc.de/ rechtspr/20010167.htm> (30 November 2007). Bundesgerichtshof, Case No. VIII ZR 13/01, 7 November 2001, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 255/2001 <www.jurpc.de/ rechtspr/20010255.htm> (30 November 2007). Hanseatisches Oberlandesgericht Hamburg, Case No. 3 U 168/00, 13 June 2002, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 288/2002 <www. jurpc.de/rechtspr/20020288.htm> (30 November 2007). Landgericht Kln, 16 April 2003, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok138/2003 <www.jurpc.de/rechtspr/20030138.htm>. Landgericht (Land Court) Konstanz, Case No. 2 O 141/01 A, 19 April 2002, JurPC Internet Zeitschrift fr Rechtsinformatik und Informationsrecht, JurPC Web-Dok. No. 291/2002 <www.jurpc.de/rechtspr/20020291.htm> (25 August 2003). Landgericht Mnster, Case No. 4 O 424/99, 21 January 2000, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 60/2000 <www.jurpc.de/ rechtspr/20000060.htm> (30 November 2007). Landgericht Nrnberg-Frth, Case No. 2 HK O 9431/01, 7 May 2002, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 158/2003 <www.jurpc. de/rechtspr/20030158.htm> (30 November 2007). Oberlandesgericht (Court of Appeal) Frankfurt, 20 November 2002, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 91/2003 <www.jurpc.de/ rechtspr/20030091.htm>. Oberlandesgericht Frankfurt, Case No. 6 W 37/01, 17 April 2001, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 135/2001 <www.jurpc.de/ rechtspr/20010135.htm> (30 November 2007).
336
Oberlandesgericht Hamm, Case No. 2 U 58/00, 14 December 2000, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 255/2000 <www.jurpc.de/ rechtspr/20000255.htm> (30 November 2007). Oberlandesgericht Karlsruhe, Case No. 14 U 202/96, 14November 1997, JurPC Internet Zeitschrift fr Rechtsinformatik und Informationsrecht, JurPC Web-Dok. No. 09/1998 <www.jurpc.de/rechtspr/19980009.htm> (29 November 2007). Oberlandesgericht Mnchen, Case No. 15 W 2631/98, 8 October 1998, JurPC Internet Zeitschrift fr Rechtsinformatik, JurPC WebDok 153/1999 <www.jurpc. de/rechtspr/19990153.htm> (9September 2003).
Singapore
Chwee Kin Keong and others v. Digilandmall.com Pte Ltd, High Court, [2004] SGHC 71, 12 April 2004. Chwee Kin Keong & Others v. Digilandmall.com Pte Ltd., Court of Appeal, [2005] SGCA 2 13 January 2005.
Sm Integrated Transware Pte Ltd v. Schenker Singapore (Pte) Ltd, High Court, [2005] 2 SLR 651, 30March 2005.
United Kingdom Bonnier Media Ltd v. Greg Lloyd Smith and Kestrel Trading Corporation, Scotland, Court of Session, Outer House, 2002 SCLR 977, 1 July 2002. Brennan v. Kinjella Pty Ltd., Supreme Court of New South Wales, 24 June 1993, 1993 NSW LEXIS 7543, 10. Hall v. Cognos Limited (Hull Industrial Tribunal, Case No. 1803325/97) (unreported) Hill v. Hill, Court of Appeal, [1947] Ch 231, 5 December 1946. Hill v. Regem, UK, Kings Bench Division [1945] KB 329, 1 March 1945. Lazarus Estates, Ltd. v. Beasley, Court of Appeal [1956] 1 QB 702, 24 January 1956. Lobb v. Stanley, UK, Queens Bench, (1844) 5 Q.B. 574, 114 E.R. 1366. London County Council v. Vitamins Ltd, Court of Appeal [1955] 2 QB 218, [1955] 2 All ER 229, 31March 1955. Lyell v. Kennedy (No 3), UK, Court of Appeal [1881-1885] All ER Rep 814, 8 April 1884. Mehta v. J. Pereira Fernandes S.A., Chancery Division,[2006] EWHC 813 (Ch), [2006] 2 Lloyds Rep 244,7 April 2006. Pretty Pictures Sarl v. Quixote Films Ltd., Queens Bench [2003] EWHC 311 (QB), 30 January 2003. Sanders Bros v. Maclean & Co (1883) 11 QBD 327, at 341. St Albans City and District Council v. International Computer Ltd., Court of Appeal [1996]4 All ER 481, 26 July 1996. The Queen v. Joseph Marijancevic, 1991 VIC LEXIS 452.
337
Tucker & Co., Ltd. v. Board of Trade, UK, Chancery Division [1955] 2 All ER 522, 19 May 1955.
United States Bensusan Restaurant Corp. v. King, U. S. District Court Southern District of New York, (937 F. Supp. 295), 9 September 1996. Caspi et al. v. The Microsoft Network, L.L.C., et al, Superior Court of New Jersey, Appellate Division, 2 July 1999 (323 N.J. Super. 118). Central Illinois Light Company (CILCO) v. Consolidation Coal Company (Consol), United States District Court for the Central District of Illinois, 30 December 2002 (235 F. Supp. 2d 916), 30 December 2002. Cloud Corporation v. Hasbro, Inc., US Court of Appeals for the Seventh Circuit (314 F. 3d 296) 26December 2002. Commonwealth Aluminum Corporation v. Stanley Metal Associates, United States District Court for the Western District of Kentucky, 9 August 2001 (186 F. Supp. 2d 770). Compuserve, Inc. v. Patterson, United States Court of Appeals for the Sixth Circuit (89 F.3d 1257, 1261), 22 July 1996. Edwards, LLC v. Fiddes & Son, Ltd., US District Court for the District of Maine (245 F. Supp. 2d 245), 14 February 2003. Farm Credit Bank of St. Paul v. William G. Huether, US, Supreme Court of North Dakota (454 N.W.2d 710, 713), 12 April 1990. Fix My PC, LLC d/b/a Fixx My PC v. N.F.N. Associates, Inc., United States District Court For The Northern District Of Texas, Dallas Division (48 F. Supp. 2d 640), 26 March 1999. Gorman v. Ameritrade Holding Corp., United States Court of Appeals for the District of Columbia Circuit (293 F.3d 506, 5112-513). Groff Lawrence v. America Online, Inc., Superior Court of Rhode Island, 27 May 1998, LEXIS 46 (R.I. Super., 1998) <legal.web.aol.com/decisions/dlother/groff.html> (30 November 2007). Gutnick v. Dow Jones & Co Inc [2001], Supreme Court of Victoria, VSC 305, 28 August 2001. Hotmail Corp. v. Van$ Money Pie, United States District Court for the Northern District of California, 16 April 1998, U.S. Dist. LEXIS 10729 (U.S. Dist., 1998). I. Lan Systems, Inc. v. Netscout Service Level Corp., United States District Court, District of Massachusetts, 2 January 2002 (183 F. Supp. 2nd 328). Inset Systems Inc. v. Instruction Set, Inc., United States District Court for the District of Connecticut (937 F. Supp. 161), 17 April 1996 (D. Conn. 1996). Lim v. The TV Corporation International, (State) Court of Appeal of California, 24 June 2002 (99 Cal. App. 4th 684). Machulsky v. Hall, United States District Court for the District of New Jersey (210 F. Supp. 2d 531), 9 July 2002.
338
Metcalf v. Lawson, Supreme Court of New Hampshire (148 N. H. 35). Mieczkowski v. Masco Corp., United States District Court for the Eastern District of Texas, Texarkana Division, 997 F. Supp. 782, March 18, 1998. Millennium Enterprises, Inc. v. Millennium Music, LP, United States District Court for the District of Oregon, 33 F. Supp. 2d 907, January 4, 1999. Rio Properties, Inc. v. Rio International Interlink, United States Court of Appeals for the Ninth Circuit (284 F.3d 1007), 20 March 2002. Sea-Land Service, Inc. v. Lozen International, LLC, United States Court of Appeals for the Ninth Circuit, 3 April 2002 (285 F.3d 808). Shattuck Jonathan P. v. Klotzbach David K., Superior Court of Massachusetts, 11 December 2001 (14 Mass. L. Rep. 360). Specht v. Netscape Communications Corporation and America Online, Inc., United States Court of Appeals for the Second Circuit, 1 October 2002 (306 F. 3rd p. 17). Superhighway Consulting, Inc. v. Techwave, Inc., United States District Court for the Northern District of Illinois, Eastern Division, 16 November 1999 (1999 U.S. Dist. LEXIS 17910). Weber v. De Cecco Dante, United States, New Jersey Superior Court (1 N.J. Super. 353, 358), 14 October 1948. Winfield Collection Ltd. v. Mccauley, United States District Court for the Eastern District of Michigan, Southern Division (105 F. Supp. 2d 746, 751), 24 July 2000. Yahoo! Inc. v. La Ligue contre le Racisme et lantismitisme, United States District Court for the Northern District of California, San Jose Division (169 F. Supp. 2d 1181, 1192), 7 November 2001. Yahoo! Inc. v. La Ligue contre le Racisme et lantisemitisme, United States Court of Appeals for the Ninth Circuit (379 F.3d 1120), 23 August 2004. Yahoo! Inc. v. La Ligue Contre le Racisme et lAntisemitisme,United States Court of Appeals for the Ninth Circuit (399 F.3d 1010),10 February 2005. Zippo Manufacturing Company v. Zippo Dot Com, Inc. United States District Court for the Western District of Pennsylvania (952 F. Supp. 1119), 16 January 1997.
339

Legal Aspects of Electronic Commerce - Rules of Evidence, Contract Formation and Online Performance

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Legal Aspects of Electronic Commerce - Rules of Evidence, Contract Formation and Online Performance

Hochgeladen von

Copyright:

Verfügbare Formate

LEGAL ASPEctS OF ELEctRONIc COMMERcE: RULES OF EVIDENcE, CONtRAct FORMAtION AND ONLINE PERFORMANcE

JOSE ANGELO EStRELLA FARIA

Part One Legal Value of Electronic Communications

Collected Courses 2009, Volume 2

Part Two Electronic Communications in Contract Formation and Performance 208

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

LISt OF PRINcIPAL PUBLIcAtIONS

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Commerce, 30 June 1999 <www.uscib.org/index.asp?documentID=1318> (29 November 2007).

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

PARt ONE LEGAL VALUE OF ELEctRONIc COMMUNIcAtIONS

Chapter I. Rules of Evidence, Form Requirements and Electronic Communications

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

127 128 129

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

141 142 143

Collected Courses 2009, Volume 2

151 152 153

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Chapter II. Electronic Signature and Authentication Methods

A. Notion of Electronic Authentication and Signature

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

B. Legal Treatment of Electronic Authentication and Signatures

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

PARt TwO ELEctRONIc COMMUNIcAtIONS IN CONtRAct FORMAtION AND PERFORMANcE

Chapter I. Contract Formation through Electronic Communications

Collected Courses 2009, Volume 2

A. Applicable Law in International Electronic Contracting

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2

Collected Courses 2009, Volume 2