A Brief History of the Virtual Magistrate Project: The Early Months


Robert Gellman

Executive Director

Virtual Magistrate Project

email: < rgellman@cais.com>

Prepared for

The On-Line Dispute Resolution Conference

Washington, DC

Sponsored by

The National Center for Automated Information Research

May 22, 1996

Beginnings The Virtual Magistrate Project had its origins in a meeting sponsored by the National Center for Automated Information Research (NCAIR) and the Cyberspace Law Institute (CLI) on October 25, 1995. The meeting took place in Washington, DC, at the offices of Counsel Connect. The participants were: Timothy C. Leixner, NCAIR; Ann Sloane, NCAIR; Ellen Kirsh, America Online; William Marmon, MCI; David R. Johnson, CLI & Counsel Connect; David Post, CLI & Georgetown Law School; Robert Gellman, CLI Fellow & Privacy and Information Policy Consultant; J. Beckwith Burr, CLI Fellow & Federal Trade Commission; and George H. Friedman, American Arbitration Association.

A two page document from the Cyberspace Law Institute formed the basis for discussion at the meeting. It described the purpose in these terms:

System operators in today's online environment face a difficult choice when their subscribers, or third parties, bring to their attention allegations of tortious communications appearing on their system (e.g., messages alleged to infringe the rights of a copyright holder, defamatory messages, or the like). Taking no action at all in the face of such an allegation would appear to be unreasonable; quite apart from questions of possible vicarious liability, most system operators, we may assume are not interested in allowing their systems to be used for illegal or otherwise harmful activity. At the same time, simply removing the allegedly tortious communication is equally unsatisfactory; the allegation may, of course, prove to be a false one, and the removal of the communications unfairly and unnecessarily impacts on the communication of third parties who have engaged in no wrongdoing. And determining whether the communication in question is, or is not, tortious may be extremely difficult; it may (and generally is not) clear from an examination of any particular message whether it contains infringing, or defamatory, material.

The "virtual magistrate" program will offer a means to deal with this problem. What we propose to convene, at the request of sysops to whom such allegations have been brought, [is] a neutral online adjudication panel that would be charged with making a rapid, real-time evaluation of the strength of the claim asserted, so as to provide the sysop with guidance as to the initial set of steps that should be taken.

A rough consensus emerged from the meeting about the role, jurisdiction, and procedures for the Virtual Magistrate Project. Between the October 23rd meeting and the public announcement on March 4, 1996, work continued on the structure, documentation, and network connections for the Project.

In November 1995, NCAIR committed $75,000 to finance a pilot project. The recipients of the funds and their roles are:

The American Arbitration Association serves as administrative clerk to the Project, receiving cases, reviewing them for suitability, and assigning them to magistrates. The AAA also agrees to defend and indemnify Virtual Magistrate Project participants in the event of litigation over their activities for the Virtual Magistrate Project. George Friedman and Kim McLaughlin are the responsible AAA officials.

The Villanova Center for Information Law and Practice operates the home page for the Project, maintains the discussion groups for cases. Professor Henry Perritt, Jr., and Kenneth Mortensen are responsible for the Villanova operations.

James Tierney serves as public prosecutor of consumer fraud cases and as an advisor to the Project.

Robert Gellman, Executive Director, prepares the documentation for the Project and coordinates the activities of other participants.

The Virtual Magistrate Project is, indeed, a virtual activity. For most of the participants, the October meeting was the only face-to-face meeting. Other contacts have been largely by electronic mail and telephone. Following the initial meeting, the task began of drafting the documentation to define the scope, rules, and procedures for the Virtual Magistrate Project. All documents circulated for comment electronically.

There were originally two basic documents: a concept paper and the rules. The concept paper offered an overview, explaining what the Project would accomplish, how it would proceed, and how it would be governed. The goals for the Project as defined in the concept paper are:

Establish the feasibility of using online dispute resolution for disputes that originate online.

Provide system operators with informed and neutral judgments on appropriate responses to complaints about allegedly wrongful postings.

Provide users and others with a rapid, low-cost, and readily accessible remedy for complaints about online postings.

Lay the groundwork for a self-sustaining, online dispute resolution system as a feature of contracts between system operators and users and content suppliers (and others concerned about wrongful postings).

Help to define the reasonable duties of a system operator confronted with a complaint.

Explore the possibility of using the Virtual Magistrate Project to resolve other disputes related to computer networks.

Develop a formal governing structure for an ongoing Virtual Magistrate operation.

As the two basic documents were circulated, revised and refined, it became clear that additional documentation would be needed. To track the style of the Internet, a FAQ (Frequently Asked Questions) was written for posting on the Virtual Magistrate Home Page. Also, to provide direction to magistrates, a Magistrates Handbook was prepared that described complaint and decision handling procedures; training requirements; and a suggested format for decisions. In addition to the Project documentation, the Villanova Center developed training materials for magistrates with the assistance of the AAA. All documents are available through the Virtual Magistrate Home Page.

While this was underway, CLI and the AAA worked together to select magistrates. Magistrates must be familiar with the law and with online systems. A geographically diverse group of eight magistrates agreed to serve. There is no requirement that magistrates be lawyers and, while most of the magistrates are lawyers, one is not. When the need arises to add more magistrates, there will be an effort to add more diversity to the pool, and magistrates from other countries will be sought.

The first class of magistrates includes: Professor Charles McClure School of Information Studies, Syracuse University; Professor Ethan Katsh, Department of Legal Studies; University of Massachusetts; N. M. Norton, Jr., Wright, Lindsey & Jennings, Little Rock, Arkansas; Paul Hoffman, Esq., Croton-on-Hudson, New York; Larry E. Meyer, Pohl, Bennett & Mathews, Houston, Texas; Alton W. Payne, Sroufe, Payne & Lundeen, Houston, Texas; Andrea W. Selvaggio, Esq., Washington, DC; and G. Gervaise Davis III, Davis & Schroeder P.C., Monterey, CA.

The hardest aspect of starting the Virtual Magistrate Project was searching for a model for its rules and procedures. There are useful models for dispute resolution, and the American Arbitration Association has a wealth of expertise in the arbitration process. One universally supported goal for the Virtual Magistrate Project was avoiding overly complex or legalistic rules or proceedings. It is important that the Virtual Magistrate be easy to understand and accessible to all. The main challenge has been developing a dispute processing system that reflects the methods and the culture of the Internet. For this, there is no readily identifiable precedent.

The Villanova Center proposed a complaint and message processing system using a "grist", a listserv/newsgroup message system. Participants in a case are registered in the grist and receive all messages posted to the grist. The address for the grist is the "reply-to" address in the initial notice sent to all parties informing them of the acceptance of a case. This simplifies contacting the parties and maintains fairness throughout the process. Submissions from the parties and communications to and from the magistrate go to this address. Every participant will automatically receive all messages about a case. The Villanova Center captures and saves all messages for future availability and analysis. When the magistrate reaches a decision, it will be mailed to the same address, and everyone will receive it at the same time. This system avoids the need for participants, magistrates, or Virtual Magistrate staff to forward messages.

Intervenors can be added to the grist for a case and can, if appropriate, have access to all stored messages about the case. The Docket System is password controlled, and password will be provided to participants. There is a procedure that allows for the filing of private messages with the magistrate if the need arises. Requests for confidential communications may be directed to a magistrate privately, and the magistrate will rule on whether confidentiality is appropriate.

The mechanics of the communication system reflect the methods and the culture of computer networks. To make sure that the system worked as designed, participants and volunteers from the Virtual Magistrate community took part in testing, and several test cases moved through the complaint system. This identified some glitches with the process, and adjustments followed.

The testing also revealed that the initial goal of resolving cases in 48 hours was unrealistic. Participants recognized from the beginning that this was an ambitious objective, but the immediacy of network communications and access makes it essential that decisions be reached quickly. Because of routine delays in electronic mail and the busy schedules of those involved in a case, the time limit was extended to 72 hours or three business days from the date of acceptance. In addition, revisions to the rules emphasized that the period for decision was a goal and not a binding commitment. Fairness is a more important objective than reaching decisions in a fixed period.

Going Public

The Virtual Magistrate Project was publicly announced on March 4, 1996, through a press release distributed exclusively through electronic mail, listservs, and other network information sharing devices. The announcement circulated widely and generated considerable attention and response. Inquiries immediately came from both traditional and network press, and many press stories resulted or are in process now. The Virtual Magistrates already received coverage in the New York Times, New York Post, San Jose Mercury-News, The Recorder, Chronicle of Higher Education, and Newsweek. International inquiries came from England, Germany, France, and Canada. More than twenty people volunteered to serve as magistrates or in another capacity in the Project. It is apparent from these responses that the Virtual Magistrate was largely successful in getting its message out to the Internet community.

The Virtual Magistrate Home Page < http://vmag.law.vill.edu:8080/> was selected as the Cool Site of the Day for April 16, 1996. This designation goes to Web innovative and interesting sites. The designation of the cool site is through a Web Page at < http://cool.infi.net> . Following selection, the number of hits on the Virtual Magistrate Home Page increased ten fold.

In the first six weeks of operation, the Project received three complaints, but none met the established criteria. In some way, each complaint, challenged the assumptions or procedures for the Project and produced some rethinking of the process or redrafting of the supporting documentation. Because the Virtual Magistrate Project is an experiment, this was not unexpected. Additional refinements and changes are certain as the Project continues.

The first complaint centered on a retail transaction involving a computer purchase. The complaint did not involve any computer network activity, and it did not fall within the jurisdiction of the Project. However, there was nothing express in the rules that defined the type of dispute that would be accepted. In response, the rules were amended so that they specified the jurisdictional limits clearly set out in the concept paper. This was just a clarification of the original intent and not a change in approach.

The two other complaints suggested that a considerable amount of effort would be required to contact all parties and convince them to submit to the jurisdiction of the Virtual Magistrate. For most complaints, an agreement from all major parties is necessary before a complaint is accepted. The original plan was for this pre- acceptance work to be conducted by the AAA staff.

Because both complaints suggested that there was a degree of personal animosity between the potential parties, it appeared that pre-acceptance negotiation would be required. If the complaints could be accepted, further negotiations between the parties were likely to be necessary. It is clearly preferable that all negotiations be handled by the same person and that the magistrate should undertake that responsibility.

The Project had not anticipated this circumstance. However, it was not difficult to alter the methods for these cases. A more complete review of the problem is underway, and an adjustment in procedures is under consideration. The obvious solution is to recognize that some complaints may be referred to magistrates before they are formally accepted. As the Project develops, more changes will occur as experience helps to identify the most effective way to process and decide complaints.

Key Features

There are three key features at the core of the Virtual Magistrate Project. The first is the commitment to a prompt resolution of disputes. The Net is a real time activity. Messages, files and postings become available worldwide at the touch of a key. Responses to problems must reflect the speed at which the Net operates. Remedies that take months or years to develop will not offer viable solutions.

The second key feature of the Virtual Magistrate Project is the limited nature of the decisions that it offers. The basic question for most disputes will be whether a message, file, or posting should remain publicly available on the Net. This type of decision is in the nature of a temporary restraining order. It is fast, rough justice. The Virtual Magistrate does not expect to resolve all legal disputes between parties, and it certainly will not be able to do that within the allotted time. If parties agree and the magistrates are willing, it may be possible to attempt resolution of related disputes. The basic decision offered, however, is narrow in scope.

Finally, the jurisdiction of the Virtual Magistrate is the same as that of the Net, and the Net does not map neatly onto the jurisdiction of any existing sovereign entity. This is a strength of the Virtual Magistrate as well as a problem. Disputes can be accepted and decided without the need for resolving potentially complex conflict of law questions for which there may be no existing legal precedent. Given the limited nature of the decision that the Virtual Magistrate will make, this is less of a barrier than it would be for a traditional, full-fledged legal dispute. On the other hand, the problem is that the Net does function within other jurisdictions and is not completely divorced from existing laws. The ultimate resolution of this broad and basic jurisdictional conflict between the Net and the rest of the world will likely continue for the rest of this century and beyond. But the Virtual Magistrate should be able to arbitrate individual cases in the interim. As arbitration decisions reached with the agreement of the parties, decisions should be enforceable in regular courts.

At this early stage in the Virtual Magistrate Project, it is difficult to assess its direction. The Virtual Magistrate represents an attempt by the Net community to police itself. Since there are no formal police powers, success must come through acceptance in the community. If the Net community sees the Virtual Magistrate as useful and fair, then it will find a place. But if the marketplace decides that the Virtual Magistrate is not acceptable or that other remedies are preferable, then the Project may not succeed.

It is apparent, however, from the enthusiastic response so far that the Virtual Magistrate has struck a chord in the Net community. There is a perceived need for a simple, rapid, and inexpensive dispute resolution mechanism for network disputes. Time will tell if the Virtual Magistrate is the best response to that need. The commitment from the Virtual Magistrate Project is to give the experiment every chance to be useful and to succeed.

VM Web Page: http://vmag.law.vill.edu:8080/

AAA Web Page http://www.adr.com

Complaints: vmag@mail.law.vill.edu

Help: vmag-question@mail.law.vill.edu


VM Operations: vmag-admin@mail.law.vill.edu


AAA Administrator: vmag-aaa@mail.law.vill.edu

VM Executive Director: rgellman@cais.com

Copies of the Virtual Magistrate's rules and other descriptive materials can also be obtained by sending an email message to

< vmag-docs@mail.law.vill.edu>

with anything in the subject line and the following in the body of the message:

index vmag-project

(for an index of available documents)


get vmag-project < filename>

(to retrieve the following documents:

concept -- Virtual Magistrate Concept paper

faq -- Frequently Asked Questions

goals -- Goals of the Virtual Magistrate Project

handbook -- Magistrate Handbook)