Citation: Ethan Katsh, Bringing Online Dispute Resolution to Virtual Worlds: Creating Processes through Code, 49 N. Y. L. Sch. L. Rev. 271 (2004) Content downloaded/printed from HeinOnline Tue Nov 27 03:16:37 2018 -- Your use of this HeinOnline PDF indicates your acceptance of HeinOnline's Terms and Conditions of the license agreement available at https://heinonline.org/HOL/License -- The search text of this PDF is generated from uncorrected OCR text. -- To obtain permission to use this article beyond the scope of your HeinOnline license, please use: Copyright Information Use QR Code reader to send PDF to your smartphone or tablet device BRINGING ONLINE DISPUTE RESOLUTION TO VIRTUAL WORLDS: CREATING PROCESSES THROUGH CODEt ETHAN KATSH* "Conflict is a growth industry" - Fisher & Ury, Getting to Yes (1982) I. INTRODUCTION Conflict is present in all environments. On the Internet, there are many new and interesting environments, among the most novel and most unusual being Multi-player Massive Online Role Playing Games or MMORPGs. 1 In these virtual "places," where transactions, interactions, and relationships can begin and end quickly, where barriers of time and space are not constraints on communication, where identities can be transformed on the screen, and where potentially valuable new forms of intellectual property can be created at the keyboard, it is not surprising that disputes are occurring. Since dispute resolution and dispute prevention were probably not thought about and were certainly not high priorities when these worlds were designed, the questions of whether these disputes are a problem, how big a problem they are, and what to do about them are, significant.2 While disputes and dispute resolution may be a growing concern for game developers and participants, they present an opport This article is adapted from a presentation made at the "State of Play" conference, New York Law School, November 14, 2003. * Professor of Legal Studies and Director of the Center for Information Technology and Dispute Resolution at the University of Massachusetts at Amherst; J.D. Yale Law School (1970); B.A. New York University, 1967. 1. Bren Lynne, What Is A MMORPG?, at http://www3.sympatico.ca/iambren/ MMORPG.html (last visited Aug. 30, 2004); see alsoLastowka & Hunter, The Law of Virtual Worlds, 92 CALIF. L. REv. 1 (2004). 2. "MMORPGs have brought some unique new challenges to developers. I think that all of the social and moral aspects caught a lot of developers by surprise. The initial concept of a CRPG with lots of participants seemed simple enough. In reality, however, human behavior is a lot more unpredictable than most developers imagined." Paul E. Schwanz II, Morality in Massively Multi-PlayerOnlineRole-PlayingGames, at http:// www.mud.co.uk/dvw/moralityinmmorpgs.html (last visited Oct. 20, 2004). NEW YORK LAW SCHOOL LAW REVIEW tunity for the field of dispute resolution. Virtual worlds are environments created out of informational tools. Dispute resolution processes are also created out of informational tools. The nature of a virtual world depends upon the tools provided and the manner in which the tools can be used. Much the same thing can actually be said about dispute resolution. A MMORPG or a virtual world may be more fun to engage in than a dispute resolution process but the fact that both are informational activities opens opportunities to experiment with new approaches to dispute resolution and to integrate new methods for resolving disputes into the fabric of the virtual world. The most common forms of dispute resolution, litigation, mediation, arbitration and negotiation, involve participants working with information, communicating it, storing it, organizing it, evaluating it, and shaping how, when, where and to whom it flows. This was noticed by an early computer pioneer who defined law as "ethical control applied to communication." 3 What differentiates one dispute resolution model from another is largely how information is managed and regulated. One process, for example, may make a body of information called rules the focal point of decision-making while in other processes the parties may decide that other types of information are more important than rules. Very generally, mediators, arbitrators and judges all have information processing roles but use different methods for different reasons and with different goals. The growing interest in and need for dispute resolution and dispute prevention in virtual worlds is paralleling a growing interest in and use of online dispute resolution ("ODR"), a new approach to dispute resolution that employs Internet-based resources. 4 ODR recognizes and tries to build upon the centrality of communication and information processing to dispute resolution. It uses high speed computer networks to allow parties to communicate in new ways, at a distance, and powerful information processing capabili- 3. NORBERT WIENER, THE HuMAN USE OF HuMAN BEINGS 143 (1950). 4. See ETHAN KATSH & JANET RIFEIN, ONLINE DISPUTE RESOLUTION: RESOLVING CONFLICTS IN CYBERSPACE (2001); COLIN RULE, ONLINE DISPUTE RESOLUTION FOR BUSINESS (2002). [Vol. 49 DISPUTIE RESOLUTION TO VIRTUAL WORLDS ties to manage the flow and use of information that is at the heart of dispute resolution. ODR's initial efforts were focused on disputes arising out of online activities. Cyberspace itself is a virtual world, or perhaps many virtual worlds, and many of the same factors that are now creating an interest in using ODR in MMPORGs led to the initial interest in ODR.5 At the time, the most active virtual entities were ecommerce marketplaces. The largest of these, eBay, was the first to recognize that it would benefit from an ODR process. In the last four years, as will be described in more detail below, an Internet start-up, SquareTrade.com, has handled over one and a half-million disputes, most linked to eBay transactions. In this short period of time, SquareTrade has become the world's largest dispute resolution provider. Disputes occur for many reasons and some level of disputing in any environment is inevitable. 6 Disputes can, at times, be a sign of trouble and poor planning, and can interfere with the accomplishment of goals. At other times, however, disputes may be a sign of growth, of creative and competitive energies, even of progress. In virtual worlds, disputes occur as a consequence of being able to use informational tools in new ways, and to engage in activities and establish relationships that could not occur in physical space. In such a context, it is not the presence of disputes but the presence of too many disputes, or the absence of any means to respond to disputes, that is cause for concern. If dispute resolution processes consist of strings of informational activities engaged in by the disputants (negotiation) or administered by the third party (mediation or arbitration), new technologies allow new interactions between the parties and new opportunities to work with information. Virtual worlds are an attractive arena into which to introduce ODR because they have participants who have demonstrated a willingness to engage in environments that are themselves an experiment in online interactions. Thus, these environments not only have a need for ODR but 5. See National Center for Automated Research Conference on Dispute Resolution (May 22, 1996), Conference Papers, at http://www.odr.info/ncair (last visited Oct. 5, 2004). 6. STUART HAMPSHIRE, JUSTICE Is CONFLICT (2000). 2004] NEW YORK LAW SCHOOL LAW REVIEW have a population that should not be resistant to new approaches to resolving disputes at a distance. While ODR in virtual worlds may be most instructive in how to respond to disputes that arise out of online activities, such environments can also be development and testing environments for approaches that may be of value in the physical world. A virtual world is an environment in which there are only informational activities. What are physical activities in the physical world, such as constructing buildings or moving from place to place, are informational activities in a virtual world. When everything is data and when all data is collected and resides in a database, opportunities should exist for innovation in both dispute prevention and resolution. While some informational activities in the virtual world cannot be duplicated offline, managing environments through the use of information is also routine in the physical world. Much of what is learned in a virtual environment, therefore, should be transferable to a physical one. As Professor Marc Galanter has observed, law: usually works not by exercise of force but by information transfer, by communication of what's expected, what forbidden, what allowable, what are the consequences of acting in certain ways. That is, law entails information about what the rules are, how they are applied, with what costs, consequences, etc. For example, when we speak of deterrence, we are talking about the effect of information about what the law is and how it is administered. Similarly, when we describe 'bargaining in the shadow of the law,' we refer to regulation accomplished by the flow of information rather than directly by authoritative decision. Again, 'legal socialization' is accomplished by the transmission of information. In a vast number of instances the application of law is, so to speak, self administered - people regulate their conduct (and judge the conduct of others) on the basis of their knowledge about legal standards, possibilities and constraints. 7 As will be described in more detail below, ODR has grown greatly in use during the past few years, in both online and offline 7. Marc Galanter, The Legal Malaise: Or,Justice Observed, 19 LAW AND Soc'y. REV. 537, 545 (1985). [Vol. 49 DISPUTE RESOLUTION TO VIRTUAL WORLDS environments. In most early experiments with ODR, the goal was to use the network to connect disputants with a human third party and allow them to communicate at a distance. This Article suggests that virtual worlds allow ODR to be extended and enhanced by developing online tools that exploit the information processing capabilities of machines. As the number and variety of these tools grow, the result is likely to be a process in which there is, in addition to the disputants and the third party, a kind of "fourth party." This metaphor suggests not that the third party will be replaced, but that, as software applications are developed, new roles will be created, different patterns of interaction will be possible, and traditional approaches will be questioned. Virtual worlds are a particularly appropriate environment for exploring and experimenting with technological resources that will someday be of value in responding to all kinds of disputes. II. A BRIEF HISTORY OF ONLINE DISPUTE RESOLUTION ODR began as an offshoot of the Alternative Dispute Resolution ("ADR") movement. During the past twenty-five years, there has, at least in the United States, been a significant effort to employ ADR processes such as mediation and arbitration in lieu of litigation.9 Such processes are seen as being quicker, cheaper, and more flexible than litigation. 0 The increase in use of such processes has led many to refer to ADR as an acronym for appropriatedispute resolution rather than alternativedispute resolution. 1 8. KATSH & RIFKIN, supra note 5, at 93-116. 9. Deborah R. Hensler, OurCourts, Ourselves: How the Alternative DisputeResolution Movement IsRe-Shaping OurLegalSystem, 108 PENN. ST. L. REv 165 (2003); Carrie MenkelMeadow, Mothers and FathersofInvention: The IntellectualFoundersofADR, 16 OHIO ST. J. ON Disp. RESOL. 1 (2000); Richard Birke & Louise Ellen Teitz, Mediation in 2001: The Path thatBrought America to Uniform Laws and Mediation in Cyberspace,50 AM. J. CoMP. L. 181 (2002); Carrie Menkel-Meadow, Symposium On AlternativeDisputeResolution: Introduction: What Will We Do When AdjudicationEnds?A BriefIntellectualHistory ofADR, 44 UCLA L. REv. 1613 (1997). 10. Frank Sander, VarietiesofDispute Processing,Address Before the NationalConference on the Causes of PopularDissatisfactionwith the AdministrationofJustice, 70 F.R.D. 79, 114 (1976). 11. Jeffrey Scott Wolfe, Symposium: Alternative Dispute Resolution In the Twenty-First Century: Across the Ripple Of Time: The FutureOfAlternative (Or,Is It "Appropriate?")Dispute Resolution, 36 TUlSA L.J. 785 (2001). 2004] NEW YORK LAW SCHOOL LAW REVIEW The Internet has a thirty-five-year history, 12 but it was not until the early 1990s that online disputes became enough of a concern to receive attention. 13 Disputes undoubtedly occurred early in the Internet's history but no one suggested the need for formal dispute resolution institutions. During its first two decades, the Internet was used by a limited number of people in a very limited number of ways. Those with Internet access were associated with either the military or with academic institutions and even in those environments relatively few computers had Internet access. What is common today, such as screens with images and email with advertisements, was unknown at that time. The World Wide Web was not invented until 198914 and, perhaps even more significantly, until 1992 the National Science Foundation banned commercial activity from the Internet.15 In the early 1990s, listservs began to be employed for group communication and this form of many-to-many discussion soon generated disputes categorized as "flaming" and violations of "netiquette." 16 Disputes also occurred in MOOs and MUDs, the ancestors of today's virtual worlds. 17 MOOs and MUDs allowed people to come together and interact, as their imaginations directed, facilitated by a machine that could be located anywhere. Various online mechanisms were employed to deal with conflicts that arose but there were no organized dispute resolution institutions that were devoted specifically to ODR. Indeed, the acronym ODR had not yet been invented. The decision by the National Science Foundation in 1992 to lift its ban on commercial activity was a highly controversial deci- 12. Barry M. Leiner et al., A Brief Histoiy of the Internet, at http://www.isoc.org/ internet-history/brief.html (last visited Aug. 30, 2004). 13. Ian MacDuff, Flames on the Wire: Mediatingfrom an ElectronicCottage, 10 NEGOTIATIONJ. 5 (1994). 14. TIM BERNERS-LEE & MARK FISCHETrI, WEAVING THE WEB: THE ORIGINAL DESIGN AND ULTIMATE DESTINY OF THE WORLD WIDE WEB (1999). 15. Jay P. Kesan & Rajiv C. Shah, Fool Us Once Shame on You - Fool Us Twice Shame on Us: What We Can Learnfrom the Privatizationsof the Internet Backbone Network and the Domain Name System, 79 WASH. U. L. Q. 113 (2001). 16. VIRGINIA SHEA, NETIQUETrE (1994). 17. Julian Dibbell, A Rape in Cyberspace, THE VILLAGE VOICE, Dec. 21, 1993, at 36- 42; Jennifer Mnookin, Virtual(ly) Law: The Emergence of Law in LambdaMOO, at http:// www.ascusc.org/jcmc/vol2/issuel/lambda.html (last visited Aug. 30, 2004). [Vol. 49 DISPUTE RESOLUTION TO VIRTUAL WORLDS sion and an enormously significant one. Shortly after the ban on commercial activity was removed, disputes related to commerce began to surface. In April, 1994, for example, the first commercial spain occurred when two lawyers tried to recruit clients to participate in an immigration scam.18 A few months later, in September, 1994, the U.S. Federal Trade Commission filed its first case of online fraud. The case involved an AmericaOnline subscriber who advertised that for $99.00 he could show people how to create new credit files at all the major credit bureaus. He advertised that the process was both legal and guaranteed. The FTC did not consider the process to be legal or guaranteed. As a result of the FTC action, the subscriber agreed to stop advertising credit repair programs and to provide compensation to consumers. 1 9 The idea for online dispute resolution emerged out of a recognition that disputes would grow as the range of online activities grew. The origins of ODR, therefore, are traceable to a very simple insight, namely that the more transactions and interactions there are online, the more disputes there will be. In addition, it was understood that the Internet, since it was an information resource, should be able to support information-dependent activities such as dispute resolution. In other words, the Internet was part of the problem, in that it generated disputes, but it was also part of the solution in that resources might be found online to respond to disputes. III. Two EXAMPLES OF ODR A. EBay: Assisted Negotiation, then Mediation Ebay is an online auction site with over 125 million registered users and where over 24 million items are offered for sale each day. Ebay makes it possible for sellers to sell to buyers who may be located anywhere. Ebay itself is not a party to any transaction and, in general, assumes no responsibility for problems that arise between 18. Sharael Feist, The Fatherof Modern Spain Speaks Out, at http://news.com.com/ 2008-1082-868483.html (last visited Aug. 29, 2004). 19. Information SuperhighwayPromoterof "CreditRepair"Planto Pay FullRedress Under Settlement with FTC, at http://www.ftc.gov/opa/predawn/F95/chaseconsult2.htm. (last visited Sept. 9, 2004). 2004] NEW YORK LAW SCHOOL LAW REVIEW buyers and sellers. In 1999, eBay decided that having a dispute resolution process might further enhance trust between buyers and sellers. It therefore authorized the Center for Information Technology and Dispute Resolution 20 at the University of Massachusetts to conduct a pilot project to test the viability and value of a dispute resolution process that would allow parties who could not resolve some problem by themselves to receive expert assistance from a mediator. In what was the first large scale use of ODR, the Center handled over two hundred disputes during a two week period.21 Several months after the completion of the University of Massachusetts pilot project, eBay selected SquareTrade.com to be its dispute resolution provider. 22 SquareTrade's approach to ODR differed from the University of Massachusetts approach in two ways, each of which represented an important advance in ODR. First, before providing a human mediator, SquareTrade added a technology-supported negotiation process in which parties could try to resolve the dispute themselves before requesting a mediator. Second, SquareTrade employed the Web rather than email as the means for communicating with and working with the disputants. The manner in which SquareTrade employs the Web illustrates how even relatively small changes in how communication occurs can have large consequences. Most of the people who file complaints with SquareTrade have already tried to negotiate via email and have reached impasse. Not only do parties seem more willing to negotiate via the Web than via email but the negotiations are more frequently successful. The Web site designed by SquareTrade provides a more structured set of exchanges between the parties than occurs with email. SquareTrade knows that almost all eBay disputes fall into about ten categories. This allows it to create forms which the parties fill out and these forms clarify and highlight both what is dividing the parties and what solutions are desired. While parties do have an opportunity to describe concerns in their own 20. The Center for Information Technology and Dispute Resolution, at http:// www.umass.edu/dispute (last visited Aug. 29, 2004). 21. Ethan Katsh et al., E-Commerce, E-Disputes, andFDisputeResolution:In the Shadow of eBay Law, 15 OHIO ST.J. ON Disp. RESOL. 705, 710 (2000). 22. SquareTradecertifiesthe authenticity ofsellerson eBay, Yahoo (via Overture), and other online marketplaces, at http://www.squaretrade.com/cnt/jsp/prs/npost.052204 (last visited Aug. 29, 2004). [Vol. 49 DISPUTE RESOLUTION iTO VIRTUAL WORLDS words, the forms and the form summaries that parties receive inevitably reduce the amount of free text complaining and demanding that is done, a result that appears to have the effect of lowering the amount of anger and hostility between the parties. Negotiation is, by definition, between the disputants with no third party present. Using the Web in the SquareTrade manner adds a novel element to traditional negotiation, a kind of "virtual presence." The Web site, particularly the forms that are employed, frames the back and forth communication and provides some of the value that might otherwise be provided by a mediator. There are no algorithms at work that analyze responses and thus this is only a first step toward a more sophisticated online negotiation process. The more that technology works with the parties in negotiation, however, the less clear the classic distinction between negotiation and mediation will be. When Web-based negotiation fails, SquareTrade provides a human mediator for a fee of twenty dollars. The Web interface is still employed but the conversation is facilitated by a human third party neutral. Use of the Web provides a structure and format that allows parties to participate whenever they wish and with a mediator who may be located anywhere. B. ODR and Arbitration:ICANN and Domain Name Disputes For the Internet to function, every computer connected to it must have a unique identifying number or Internet address. Such addresses typically look something like: 128.119.28.27. Because humans find it difficult to remember strings of numbers, a system was developed that allowed a domain name, such as "adr.org." to be typed in instead of the number string. What occurred when someone typed in the domain name was that a machine somewhere translated it into the number string, something the computer could process to find a particular machine. The demand for domain names grew as commercial activity on the Internet grew and as businesses wanted potential customers to have an easy way to find them. The domain name system had been designed before commercial activity was permitted on the Internet and it had not been anticipated that many businesses with similar names might want the same domain name, or that owners of trade- 2004] NEW YORK LAW SCHOOL LAW REVIEW marks would be upset if someone registered a domain name that was similar to a trademark. The combination of domain name scarcity and the concerns of trademark holders led to disputes over domain names. In 1998, the United States government agreed to allow a new organization, the Internet Corporation for Assigned Names and Numbers ("ICANN") to manage the domain name system. One of the first things ICANN did was enact the Uniform Dispute Resolution Policy ("UDRP") establishing both a process and a set of rules for deciding domain name disputes. 23 Both the approach ICANN chose, a modified arbitration process, and the systems which have implemented this approach, represent another choice in moving dispute resolution online. The process employed to resolve domain name disputes is interesting in a number of ways. For example, UDRP dispute resolution occurs without face-to-face meetings and, except in rare instances, without telephone communication. It is, therefore, dispute resolution at a distance. The process employed by the current dispute resolution providers, however, involves limited use of the Internet. One of the original dispute resolution providers, eResolution.com, employed a completely online system but it stopped handling cases in 2001. The two main current providers, the World Intellectual Property Organization ("WIPO") and the National Arbitration Forum ("NAF") have online systems that could be employed and probably will be employed in the future. Currently, online filings are occurring with increasing frequency and email is employed sometimes. Unlike the eBay mediations, however, the Web is not employed and if there is added value that could be provided by Web-based processes, such value is not yet present. Furthermore, the UDRP is not classic arbitration in that the decisions are not binding and enforceable in court.24 UDRP arbitrators are referred to as panelists since the word arbitrator denotes someone who can make a decision that is enforceable in court. UDRP panelists are empowered by terms in the contract agreed to 23. Elizabeth G. Thornburg, Going Private: Technology, DueProcess, and InternetDispute Resolution, 34 U.C. DAVIs L. REv. 151, 160 (2000); see Uniform Domain Name Dispute Resolution Policy, available at http://www.icann.org/udrp. (last visited Aug. 29, 2004). 24. See Thornburg, supra note 23. [Vol. 49 DISPUTE RESOLUTION TO VIRTUAL WORLDS when a domain name is registered. Decisions of arbitrators are enforced by making necessary changes in the domain name registry. The UDRP created an efficient, although somewhat unorthodox process, and not without controversy.25 IV. NEW APPROACHES TO DISPUTE RESOLUTION: THE "FOURTH PARTV' AND VIRTUAL WORLDS The MOOs and MUDs of the early 1990s, the scene of some of the earliest online disputes, were extremely creative efforts.26 They were also, compared to today's online environments, starkly different. Unlike current applications, for example, they were wholly textual. Participants engaged in activities not by clicking a mouse or manipulating images but by typing in textual commands that directed characters to do something or to go somewhere. There were no images, no graphics, no video, no sound, nothing except sets of instructions that told a participant what might be happening. Someone who wished to cause a problem might be able to do so by typing in a command such as "hit X." Disputes, of course, could not be resolved simply by typing in "settle dispute." One lesson that was learned from these environments was that the generation of online disputes does not require sophisticated communication capabilities. Simply providing a method for communication and interaction is sufficient. With simple communications, the range of disputes may be limited but the intensity of the disputes that occur might still be very high. Disputes can, as a result, be disruptive, and if the context is competitive and value or values are contested, the need for dispute resolution will only increase over time. While simple communication tools provide the opportunity for dispute creation, they do not enable very effective dispute resolution. The wizards of LambdaMOO created roles and responsibilities for human arbitrators but there were few tools that were available other than sending and accepting messages. Such arbitra- 25. MILTON MUELLER, RULING THE ROOT: INTERNET GOVERNANCE AND THE TAMING OF CYBERSPACE (2002); Michael Geist, Fair.com?: An Examinationof the Allegations of Systemic Unfairness in ICANN UDRP, at http://www.udrpinfo.com (last visited Aug. 30, 2004). 26. Pavel Curtis, Mudding: SocialPhenomenain Text-Based Virtual Realities, athttp:// www.scara.com/-ole/literatur/mudding.html (last visited Aug. 30, 2004). 2004] NEW YORK LAW SCHOOL LAW REVIEW tors could issue rulings but no software was available to allow the third party to interact efficiently with the parties. The reason for having arbitrators was to allow the arbitrator's expertise to be delivered in some way over the network. This could work if expertise could be represented in words, such as in documents delivered and decisions rendered. It would not work, or not work very well where, as in mediation, more was required than merely delivering words. Stated another way, mediation requires regulating who hears and says what when, something that requires software to facilitate organizing, managing or processing informational exchanges with parties. Such software was not available at the time. The most frequently heard concern about ODR has been that online processes and interactions cannot match the richness of the face-to-face sessions that are at the heart of offline mediation. Faceto-face sessions enable a mediator to regulate who says and hears what simply by physically including or excluding parties from the room. In addition, the mediator gets feedback from the parties both by hearing what is said and by seeing how it is said. Other elements of the mediation process, such as building trust and maintaining a non-hostile environment, are also assisted by behavioral interactions. For those unfamiliar with it, mediation can be a process that is somewhat hard to analyze and explain because it is so tied up in all the options for verbal and non-verbal expression that are present in a face to face setting. Mediators, by definition, do not have authority to impose solutions or make decisions, and, for various reasons, tend not even to propose solutions. The manner in which a mediator leads parties to a successful agreement may seem obscure and even the mediator, relying on the nuances of language, the rephrasing and refraining what one party has said, and the behavioral cues picked up from the parties, may find it hard to explain how consensus came about. Mediation has grown rapidly in use offline because litigation often did not serve the needs of parties to a dispute. Litigation was costly, slow, formal, and inflexible. ADR was used to avoid court and was also used by judges to divert cases out of court to save money and reduce backlogs of pending cases. In addition to promising cost and time savings, it also provided greater opportunities to [Vol. 49 DISPUTE RESOLUION TO VIRTUAL WORLDS repair and build relationships, and to be more flexible and creative in designing outcomes. As disputes generated by online activities began to grow, what were recognized to be problems with litigation offline seemed to be even bigger problems online. There was a need for speed, flexibility and the other benefits usually claimed for ADR and there was also the very large problem that one could probably not sue in court even if one wanted to. Even if one found an online dispute that seemed appropriate for a court to consider, the parties were likely to be in different jurisdictions. If courts were costly to users who lived close by, they were even less accessible to parties who lived at a distance. As attention began to be directed to online disputes, therefore, an online version of ADR, rather than litigation, seemed to be the only realistic means of dealing with conflicts that arose online. ADR also was attractive because the network allowed easy communication between disputants and a human third party with expertise in mediation or arbitration. The first efforts at ODR assumed that the network alone was a sufficiently rich conduit for effective intervention by a mediator or decision-making by an arbitrator. These efforts had some success and, for online disputes where there were no realistic alternatives for using land-based systems, had clear benefits. Yet, it was also clear that fostering communication and exchanging messages was only part of the offline expertise of mediators. Face-to-face mediations also involved managing communication, knowing when parties should be talking with each other, and knowing when to stop and start communication. Simply allowing disputants to exchange messages, in other words, permitted only some of the expertise of a third party to be employed online. One noted authority has written that mediation is essentially negotiation that includes a third party.27 The most widely known book about conflict resolution, Gettingto Yes, also focuses on negotiation, not on mediation or arbitration. Negotiation is, almost always, the first process employed to resolve a dispute and it is, as these works suggest, appropriate to view mediation and arbitration not as parallel processes to negotiation but as extensions of negotia- 27. See CHRISTOPHER MOORE, THE MEDIATION PROCESS (1996). 2004] NEW YORK LAW SCHOOL LAW REVIEW tion. For ODR, this is a very helpful perspective to have and one that early ODR efforts did not have. Early efforts, my own included, tended to ignore negotiation because few tools were available at the time to help parties resolve disputes themselves and parties who requested assistance already seemed to have negotiated and reached impasse. These early efforts recognized that there were novel capabilities for linking to experts who might be anywhere and hoped that simply making an expert available online to parties who were in conflict could be beneficial. The metaphor of the "fourth party" suggests that the novel element in ODR may be less the online connection to a human third party than the connection to software tools that can help parties negotiate and, if that fails, assist interactions with a mediator or arbitrator. Looked at in this way, ODR can develop in several different ways. It can be perceived of as a new process or set of processes when the network is the exclusive mode of interaction. Or, ODR over time may transform all three ADR approaches by using online tools even when face to face sessions occur. In the latter case, all negotiation in the future will be online negotiation in that online tools will be recognized to be of value to disputants and will be routinely employed, and all mediation and arbitration will be online mediation or online arbitration in a similar way. ODR has already moved out of the stage it was in when disputants simply conversed online with a mediator. The SquareTrade approach indicates that software that only organizes and focuses a conversation on a Web page can be a powerful improvement over communication via email that is less structured. In addition, in ways that parties may not be aware of, SquareTrade forms can be modified on the basis of data provided previously and opportunities for using graphics in lieu of text can be exploited. These techniques do not aim to duplicate the richness of face-to-face sessions online but to take advantage of some of the capabilities that information processing and communication together provide that are not present offline. ADR off-line emphasizes face-to-face meetings and ADR training stresses techniques that can be employed in working with parties face to face. A concern about ODR has been that techniques that are part of traditional mediator training may not be effective in [Vol. 49 DISPUTERESOLUTION TO VIRTUAL WORLDS a non-face-to-face environment using current communication software. Many of the goals that such techniques try to accomplish, however, might still be achieved through the development of online tools that allow parties to brainstorm, to identify priorities, and to identify interests and needs. These tools recognize the value not simply of transmitting data but of collecting, storing and using it. Third parties online, in other words, can try to use techniques they were taught but they should recognize that various new means will be appearing that can be employed at a distance and that can be effective in leading to the same ends. ADR was the original model for ODR and many goals of ADR will remain goals of ODR. The process for reaching these goals, however, need not mimic processes employed by offiline mediators. In one respect, ODR is in the process of diverging from offline ADR by employing the language of tools in lieu of or in addition to techniques. In another respect, offline mediation is becoming ODR, as the value of using online tools in addition to traditional techniques and face-to-face meetings is understood. What the offline ADR process instructs us is not that face to face meetings are indispensable, but that a mediation process consists of a very large number of managed informational exchanges, some of which can be managed very effectively through software. While the goal of the "fourth party" is to make information processing and information management resources available, this does not minimize the importance of any improvements that may occur in the future as to how efficiently data moves online from one party to another. When video conferencing is more cost effective and improved technologically, mediation may even occur online without losing very much of what is present when parties meet face to face. Yet, the maturing of ODR, something that can be expected from its use in virtual worlds, will consist as much in designing new online tools to process data as in tools to transmit data. In the absence of video conferencing, and probably in most disputes even when video conferencing is available, ODR will have to have tools that may have the same overall goals as offline mediation and arbitration but lead there in very different ways. Virtual worlds, where the resistance to experimentation may be less than in the physical world, may be the environment in which 20041 NEW YORK LAW SCHOOL LAW REVIEW such applications first appear and are first used. In a world where avatars are commonplace, a virtual "fourth party" with an array of tools and approaches to conflict resolution should seem almost natural, and certainly not as alien a concept as it might offline. There should also not be anything startling about the "fourth party" avatar becoming more skilled and intelligent over time. At first, the avatar might be of most value in simple disputes, disputes that involve two parties and one or two issues. Experience with version one of such an avatar might lead to more skilled avatars or avatars that can interact with human third parties to provide users with a combination of human and machine expertise. ODR has made an impact already because the first two largescale uses of ODR have occurred in two high visibility situations, eBay and ICANN. Yet, ODR is still in an early stage of its development because software applications and tools are emerging slowly. ODR, as the eBay experience has demonstrated, can respond to large numbers of disputes if appropriate software is available. It will also be able to respond to more complex disputes, such as multiparty or multi-issue disputes, when appropriate software is available. ODR is also being adopted by government agencies involved in dispute resolution and this can be expected to accelerate use of ODR. 28 Generic communication and information processing applications such as threaded discussion software, project management software or collaboration software of various kinds provide some support to ODR efforts in the absence of software that is both powerful and targeted to specific tasks. For ODR to mature, however, what is needed are tools that are linked more to processing information than to transmitting data. Virtual worlds are an exciting development for ODR because not only is the need present and not only are the game designers in a position to make some tools available, but users themselves might see opportunities in creating dispute resolution tools and applications. The "fourth party," in some instances, will encroach upon and displace techniques employed by third parties and in other situa- 28. Alternative Dispute Resolution (ADR) and Online Dispute Resolution (ODR), at http://www.nmb.gov/adrservices/adrmenu.html (last visited Aug. 29, 2004). [Vol. 49 DISPUTERESOLUTION TO VIRTUAL WORLDS tions will enhance third party capabilities. Here are only a few examples: 1. Computational. We already have a tool with a computational focus that manages the flow of information and employs the network. It is called "blind bidding" and virtual worlds provide an opportunity to make the tool more accessible and make its use commonplace. Blind bidding is used when parties are negotiating about money or something else that is quantitative. In its most generic form, each party submits an offer over the network, a computer compares the offers and if the difference is 30 percent or less, the parties have agreed to split the difference. 29 If the offers are far apart, the process can be repeated through several rounds or it can be ended with neither side having lost anything since nothing was revealed to the other party. Blind bidding can be accomplished with a human third party who accepts the offers and makes the calculations but the process is cumbersome. Yet, when done online, blind bidding appears to be almost too simple to be considered any more of a dispute resolution process than flipping a coin or drawing a card from a deck. Unlike these other techniques, however, in which the outcome has nothing to do with the needs or interests of the parties, blind bidding allows parties to reach an agreement, not simply an outcome or result. It does so through a calculation but rules on disclosure are as much a part of blind bidding as is the calculation. Blind bidding may be simpler than tools for other tasks because the rules governing it are very clear and understandable and the calculation is elementary. These are also its virtues. 2. Building Trust. This is an important component of any dispute resolution process and one of the first tasks undertaken by mediators. By achieving trust, it is felt that parties will be more forthcoming and willing to work with the mediator. Trust, it has been written, "is earned through a mediator's behavior during the mediation process. °30 Yet, building trust is a process that can be redesigned to be more informational than behavioral. Questions implicit in the trust building process, such as how it can be guaranteed that communications will be kept confidential, and what expe- 29. What is Cyrbersettle?, at http://www.cybersettle.com (last visited Aug. 29, 2004). 30. Richard Salem, Trust in Mediation, at http://www.beyondintractability.org/m/ trustmediation.jsp (last visited Aug. 29, 2004). 2004] NEW YORK LAW SCHOOL LAW REVIEW rience and expertise does the mediator have, can be responded to online not by verbal assurances but by software that can provide information, and monitor and secure messages. 3. Identfying Interests. Identifying interests and distinguishing interests from positions is a fundamental part of mediation sessions.31 Parties may not themselves understand clearly what they really want and mediators may have to meet with parties more than once to identify what their needs are as opposed to what they claim they want as an end result. This can be a difficult challenge for a mediator and we do not yet have interactive applications that can achieve this or assist the mediator in this task and that are as easy to use as blind bidding.3 2 There is no reason, however, that such applications cannot be developed. 4. Ascertaining Facts. Many disputes arising out of online transactions involve facts, such as whether something was delivered or paid for. Experience thus far has been that ascertaining what occurred in a transaction, because it occurred at a distance, is more difficult than when someone can be "looked in the eye." Disagreements about facts, however, can frequently be prevented by systematic record keeping about such facts as when something was shipped or whether a payment was made. The delivery of virtual goods in virtual worlds is today a problem that will be solved better by a software application than by an arbitration proceeding. 5. Designing Solutions. Many solutions to problems in the physical world are built around the qualities of physical space. Parties who have been fighting with each other might agree to visit a place at different times or to erect a fence between their properties. These can be effective solutions but a virtual world has the potential to be more creative. A property dispute offline might result in the placement of a physical barrier. Virtual world technology could be applied in a way that would make an ugly neighboring property invisible to the person who is bothered by it. 6. MonitoringAgreements. In the physical environment, there are very limited tools for monitoring performance. When record 31. See ROGER FISHER & SILLIAM URY, GETTING TO YES (1982). 32. But see http://www.smartsettle.com (last visited Aug. 30, 2004) (online "eNegotiation system" that claims to assist parties in negotiating dispute resolutions through "patented optimization algorithms"). [Vol. 49 DISPUTE RESOLUTION TO VIRTUAL WORLDS keeping and data collection is part of performance, however, notification of problems and defaults can occur as soon as the problem occurs. In addition, criteria can be established for early warning signs, measures that would alert parties if something occurred late or out of sequence. The importance attached to written agreements is a consequence of the ability of words on paper to be both a record of what happened at a particular time and place and to be a reminder as to what was agreed would happen in the future when the parties may not be in contact. In a pre-electronic world, paper contracts may have been the best alternative for structuring performance but it will be increasingly open to question in the future whether such documents are the best of all existing alternatives. 33 Written agreements will not go away in the near future but tools to monitor performance, particularly using graphical means in lieu of text, such as the use of a "red flag" icon appearing on screen when an item has been agreed upon, will be increasingly employed. 7. Buildingon Experience. Virtual worlds present opportunities to obtain feedback from users and to learn more about disputes that have arisen. Data collection can facilitate the resolving of disputes but, in many situations, can be even more valuable when used proactively to understand why disputes occur and then to prevent them from occurring. In their fascinating article "The Laws of the Virtual Worlds," Gregory Lastowka and Dan Hunter ask about what might be done in the following dispute: A twelve-year-old castle owner alleges that his giant avatar's virtual goose, which laid golden eggs (worth U.S. $ 100 daily), was stolen by a virtual juvenile (and real-world adult) who climbed a giant beanstalk to get it. The plaintiff alleges trespass, invasion of privacy, and conversion. The defendant, in turn, counterclaims against the virtual world's wizards who made the goose in question capable of being stolen, and throws in an additional charge against a third party of fraud involving a virtual cow.3 4 33. See M. ETHAN KATSH, LAW IN A DIGITAL WORLD (1995). 34. F. Gregory Lastowka & Dan Hunter, The Laws of Virtual Worlds, 92 CALIF. L. REv. 1, 71 (2004). 20041 NEW YORK LAW SCHOOL LAW REVIEW ADR and ODR are normally not of value in cases of fraud or theft, in situations where one party wants the other punished, or where there is a need for a legal standard or principle to be established or clarified. On the other hand, ODR could be employed if the goal is restitution or if what appears to be theft turns out to be something different. In this particular case, an individual claims to have suffered harm but there is also a need for understanding how such actions can be prevented, or at least discouraged. While one party considers the other a thief, it would not be impossible, at the end of a successful dispute resolution process, for the "thief' to be transformed into a buyer/seller of the golden eggs. This dispute is brought by an individual who claims to have been wronged. This is also a dispute that has implications for the game designers, since it is their design that has allowed such a problem to occur. It may not be possible to anticipate every dispute that might arise in an environment as novel and as creative as the one described. The suggestion in this article, however, is that an environment that can create a market in golden geese should also be an environment in which there are effective alternatives other than appealing to an outside court. V. CONCLUSION This article was not intended to be a precise blueprint for dispute resolution in virtual worlds but an exploration of how and why such worlds might employ and enhance relatively new online dispute resolution processes. Every dispute arises in a setting or context, and the setting from which it arises may shape the expectations of the parties, the timing of settlement, the perceived urgency of resolution, the consequences of and available alternatives to failure, the role of the third party, and even the form of dispute resolution.3 5 Professor Frank Sander has suggested that in determining what dispute resolution system to employ, one should 35. Context can influence the approach of the neutral, the choice of process, and the behavior and attitudes of disputants. In any environment, context can affect the kinds of disputes that are likely to arise and also affect who the parties are who are likely to be involved in the dispute. Context implicitly feeds us information about the extent or nature of the injury as well as how the injury or dispute is perceived by those involved. Context situates a dispute (Vol. 49 DISPUTE RESOLUTION TO VIRTUAL WORLDS try to match the "forum to the fuss."3 6 Virtual worlds are intriguing because they present us with new forums and new fusses. They also present us with opportunities to develop new tools that might be employed in both online and non-online environments. Marshall McLuhan wrote that "when cultures change, games change."37 McLuhan took games seriously and, indeed, devoted a whole chapter in UnderstandingMedia to the subject of games. He did so because he considered games, along with art and literature, to be early warning systems, artifacts that react to and inform us about the nature and direction of early shifts in a culture. The theme of this article has been that network-based online games and virtual worlds may be more than early warning systems. The games or virtual worlds that are the subject of this article are not simply contests but new environments. As such, they may not only suggest something about the future direction of dispute resolution but allow us to design and experience new models, systems and approaches. in a particular time and place, and we react and adjust accordingly as the parameters of the environment become clear to us. Ethan Katsh, The OnlineOmbuds Office: AdaptingDisputeResolutionto Cyberspace, at http:// www.umass.edu/dispute/ncair/katsh.htm (last visited Aug. 30, 2004). 36. Frank E. A. Sander & Stephen B. Goldberg, Fittingthe Forum to the Fuss:A UserFriendly Guide to Selecting an ADR Procedure,10 NEGOT. J. 49, 50 (1994). 37. MARSHALL McLUHAN, UNDERSTANDING MEDIA: THE EXTENSIONS OF MAN 221 (1964). 2004]