Electronic Dispute Resolution

BACKGROUND

On September 30, 1993, President Clinton signed Executive Order 12866 establishing a Regulatory Working Group (Working Group) by the Administrator of the Office of Information and Regulatory Affairs (OIRA) in the Office of Management and Budget. The Working Group was a forum to assist agencies in identifying and analyzing important regulatory issues including (1) the development of innovative regulatory techniques, (2) the methods, efficacy, and utility of comparative risk assessment in regulatory decision-making, and (3) the development of short forms and other streamlined regulatory approaches for small businesses and other entities. The Executive Order provided that agencies should design regulations in the most cost-effective manner to achieve the regulatory objective. Agencies were directed to consider incentives for innovation, consistency, predictability, the costs of enforcement and compliance (to the government, regulated entities, and the public), flexibility, distributive impacts, and equity. While the Executive Order was not mandatory for independent agencies, these concepts have been recently codified as part of the Small Business Regulatory Fairness Act of 1996, Pub. Law. 104-121, 5 U.S.C. Chapter 8.

On December 10, 1993, the Nuclear Regulatory Commission addressed several recommendations made by the National Performance Review as well as the National Science Foundation High Performance Computation, Communications, and Information Technology (HPCCIT) Initiatives. Staff was to recommend appropriate applications and facilitate the transfer of this technology to the NRC environment. In addition, the Commission directed the staff to investigate and report to the Commission on the feasibility of formal NRC participation in the National Performance Review's REGNET.

REGNET was a proposal submitted by Construction Systems Associates, Inc. to Advanced Research Projects Agency (ARPA) in the Department of Defense in July 1993. Its technology reinvestment project (TRP) was jointly sponsored by Lawrence Livermore National

Laboratory (LLNL). The proposal included an idea for an Accelerated Regulatory Information System (ARIS) which was referenced by Andy Campbell in an August 26, 1993 report to NPR.

The ARIS report provided an overview describing the problem being addressed as follows:

"Over the past twenty years, the United States has developed agencies and regulations to govern industry in this country. The regulatory intent is to protect the health and safety of U.S. citizens and the physical and social environment in which they live. The reality is that an industry usually answers to several regulatory agencies (at the Local, State and Federal level); and a myriad of regulations which they must understand, and which sometimes are redundant or inconsistent. Further, different agencies often have regulations which accomplish the same goal, and therefore only act to create more reporting requirements. The cost and delays associated with the regulatory process in this country, has [sic] been a primary factor in the decision by many U.S. industries to build their facilities outside the United States. This proposal describes implementation of an electronic information system which will lower the cost and accelerate the regulatory process for industry.

"ARIS describes a regulatory information system whose purpose is to ease and speed up the regulatory process in this country. When fully developed, it will electronically contain regulatory information (regulatory databases and regulatory support computing analysis codes) at the Local, State, and Federal Level. These regulatory databases and regulatory support computing codes will be electronically integrated so that an industry can test a potential submittal to a regulatory agency and evaluate whether the submittal meets the regulatory requirements, whether the submittal must have more analysis, or provide more supporting information. ARIS will comprise the same tools that regulatory agencies will use to evaluate an applicant's submittal for regulatory approval. ARIS will have specifications, criteria and standards developed to guide regulatory agencies in connecting and integrating into the regulatory information system.

"ARIS will have a regulatory based component, an analytical based component and an economic based component which will be integrated together through a graphical user interface (GUI) to make it ‘industry user friendly.' This integrated system will contain the functional and environmental criteria required to design, construct or produce, and operate a specific type of system, product, or facility currently approved for defense, research, commercial or private operation in the United States. ARIS will provide rapid, concise information regarding if and where the design model fails the evaluation and give industry rapid feedback on the types of changes needed to meet regulatory requirements. As envisioned, when ARIS is fully developed, it will determine for a given design, subject to regulator's approval:

"1. Compliance with local, state, and federal development, operating, and maintenance requirements for the proposed area of application or construction.

"2. If any, what type of additional design changes are required. Where appropriate the regulatory support analytical codes will challenge the design model to confirm submitted design assumptions; along with intelligent access to previously certified designs.

"3. The impact the design will have on social economic issues and visa versa. "From a regulator's perspective, ARIS will be used to evaluate the impact of a regulatory change on the existing regulated entities. It will also indicate where existing regulations already accomplish what an impending regulatory change is meant to provide (thereby acting to eliminate redundant reporting requirements). ARIS will also be used to identify conflicting regulatory requirements."

As a result of interactions with the NPR, the Administrative Conference of the United States, and Lawrence Livermore National Laboratory, the Commission decided to pilot an electronically facilitated rulemaking using the High Performance Computing Mosaic interfaces as described in the ARIS proposal while at the same time incorporating some of the dispute resolution procedures recommended over the years by ACUS and developed in the course of the Commission's experience with enhanced participatory rulemakings. Each Commissioner expressed willingness to move forward to develop the concept. OMB's OIRA and NPR staff agreed that the concept was consistent with government-wide initiatives described in "Reg Net."

go to table of contents

THE DESIGN AND EXPERIENCE WITH RULENET

NRC's project was designed to demonstrate and use rapidly developing computer based collaborative tools that would support electronic conferences and neutral facilitation as a online implementation of regulatory negotiation. Anticipated advantages were thought to include: improvements in the quality of the developed rule, reduced contentiousness resulting from addressing disagreements as they occur, and reduced drafting and implementation time.

The conceptual agreement worked out with Lawrence Livermore where the systems development would occur included an electronically networked facilitated rulemaking that had all the components of the traditional notice and comment process. The internet component, however, had enhancements. The first set of enhancements involved electronic forum technology. This is a medium that allows for group interaction in a structured environment. Participation can be at one's convenience and accessible from any designated electronic address. Access and participation can be controlled with several parameters. Participation in an electronic forum can be limited to a particular group. The ability to read or write to the forum can be controlled. The entire public can be allowed to see a dialogue but participation in the dialogue can be limited. Even postings by those allowed to participate can be held until after editing or an offline dialogue for clarification purposes is conducted. More than one forum can be conducted at a time and forums can be organized by topic created specifically to address particular issues or to allow particular persons to interact. Organization of subjects, issues, and participants can be done by a designated facilitator or can be self- selected by participants. Participants can choose to respond in several forums or in only one.

Another set of enhancements that the Commission wanted to try might be called "virtual meetings" which are held at a specified time and place. This was a video conference involving multiple sites with a virtual transcript threaded to an Internet address. The transcript was also corrected and posted for later use of other participants. The meeting was structured to allow for public questions concerning matters presented in the meeting. The questions can be virtual or can be posed after the meeting allowing time for more thoughtful questions with answers to be supplied at some later date certain.

It was also decided to enhance the electronic participation with trained facilitators who would assist with the on-line negotiations much in the same way that a consensus builder /facilitator would operate in a meeting. Ground rules would be established and order would be maintained, issues identified, and interests explored. Several different types of facilitation and consensus building techniques were developed and tried.

Once the project was underway, it became evident that it would not be feasible to try to demonstrate all the "distributed" and "dynamic" database capabilities that the Internet and web browsers present an agency. In discussing government data (and government access to data) on the internet, it has become common to refer to distributed data as that information that is not maintained on a person's own computer by which that person routinely accesses from another location. Dynamic data refers to that information that is constantly changing and may be different at different points in time when two or more individuals look at it. The most familiar example may be stock market quotes when an exchange is in session.

The ARIS "RegNet" proposal's complexity can be envisioned by a situation where a downstream pollution monitor triggers a signal to a satellite uplink. The data is simultaneously down linked to a local plant's computer control room, the local emergency planning office, the state designated environmental control location, and the designated federal agencies with jurisdiction. Each entity then takes appropriate actions within seconds of the notification based on artificial intelligence built into the programs. This would include necessary consultations, corrective actions, public notifications, and other enforcement activities. Because the NRC's pilot effort was going to be restricted to a single agency in a rulemaking context, the name given to the pilot was "RuleNet" rather than "RegNet" to indicate its more limited definitions.

Addressing the issue of fire protection at nuclear power plants, the rulemaking design hoped to optimize the flow of information and expert opinion between participants and the NRC, and also among the participants themselves.

The Commission's overall approach to safety issues in recent years has been to move in the direction of performance-based regulations and away from prescriptive regulations. In RuleNet, the Commission wished consider whether this approach should be applied to fire protection requirements as well, and if so, how it should be implemented. In addition, the NRC had received an industry petition for rulemaking in February, 1995, seeking a performance-based alternative to the existing prescriptive regulations. Since public comment had been received on this petition in the traditional way, this topic presented two advantages: 1) there was a clearly defined topic of interest to the technical regulated community so that significant useful participation could be anticipated; and 2) there was a baseline in the traditional paper notice and comment process with which to compare results.

The NRC had experience in two previous innovations in the rulemaking arena on which it wished to draw. It had developed the concept of "enhanced participatory rulemaking," designed to promote early public comment and interaction on rulemaking issues before a proposed rule. This built on its early experience with negotiated rulemaking but had broader applicability. The second innovation was the use of electronic bulletin boards, which allowed comments on a proposed rule to be submitted electronically. These BBS' were maintained on "Fed World" but had revealed a significant drawback: participants wanted until the comment due date to file anything. Thus, there was little interaction among commentators. The procedures developed for RuleNet a merged these two approaches.

Communications technology, developed specially for this purpose by the Lawrence Livermore National Laboratory, permitted participants to deal with one another and with the NRC using Graphical User Interfaces (GUI) and the internet. (Participants were not restricted to communication by computer. Written comments could be submitted in place of or in addition to electronic communications. (Written comments would have been scanned and placed at the appropriate locations had it been necessary.)

Traditionally, NRC publishes either an advance notice of proposed rulemaking or, eliminating that step, issues a proposed rule. In the former case, the process is generally extremely time-consuming; in the latter, there is a risk that the agency is too wedded to the proposed approach to be able to rethink the issue from the ground up if a wholly new proposal is submitted by a commenter. In either case, the structure is not conducive to working toward any common ground. Rather, all comments are submitted to the agency, which then reviews conflicting views and approaches, synthesizes them as best it can, and decides on the best way to proceed. When meetings on a proposal are held, they are generally in the Washington, D.C. area, which can mean heavy burdens for faraway participants and an advantage for those who are able to attend. Even the "enhanced" rulemakings, where facilitated workshops were noticed and held in locations other than Washington, posed geographic, time, money, and other practical restrictions that may prevented further effective participation by some interested persons who filed written comments.

Rulemaking using computer-based communications technology, on the other hand, has several advantages. One is rapidity. Proposals can be submitted, analyzed, responded to, revised, and resubmitted, on tight schedules that keep participants focused on the task at hand. Another is the capacity for interaction between participants and the agency and among participants.

In traditional rulemaking, if a particular matter raises questions, participants can only point out the issue in their written comments. If the rulemaking is on a proposed rule, the commenter may not learn how the agency resolves the question until the final rule is issued.

In an internet based computer forum, however, the agency can analyze the comments and questions as received, ascertain which questions arise most often, and then post electronically a list of "Frequently Asked Questions" and their answers. In this way, doubtful points can be clarified before, not after, comments are filed.

(1) The dialogue through the RuleNet computer network does not supplant formal comments (submitted in writing or electronically). Rather, RuleNet is intended to provide additional opportunities for commenters to interact with agency personnel before the agency has developed text on which formal written comments are required to be filed.

This can mean better informed, focused, and influential comments. Likewise, the ability of commenters to interact among themselves before comments are filed means that misunderstandings and miscommunications can be corrected in a timely way. However, because the electronic communications will be considered by the agency in the rulemaking process, a copy of these communications will be placed in the rulemaking record.

For facilitating exchanges of views, a tool was developed for the rulemaking called the "caucus". It was designed to allow discussions among subgroups of participants. These caucuses could be of two kinds. First, participants with similar views or interests could request a caucuse to focus solely on an issue or issues to maximize the effectiveness of their participation. Second, caucuses could be used to allow a specific issue to be placed before all participants for highly focused consideration. In this way, a particular topic could be considered in detail, the strengths and weaknesses of conflicting positions could be analyzed, and the possibilities of a compromise resolution could be explored. Caucusing was designed to take place either separate from the rulemaking, by the private interaction of participants, or through the rulemaking's electronic communications, either with or without the use of the agency's professional facilitator.

Several types of facilitators and/or moderators were supplied. The facilitators served a variety of functions, both substantive and procedural. An expert faciliator who had fire protection training and experience was available to help process and synthesize substantive comments on fire protection issues, and to formulate alternatives for consideration, thus helping to maximize the usefulness of the electronic communications process. Two types of process facilitation was provided. One was to review comments in the forum before posting and to interact with the posting party to ensure effectiveness of communication. In addition, these form of facilitation was responsible for the "bozo" monitor which automatically screened postings for profanity and abuse. Such postings were to be returned to the sender with a request to excise the offending language. If the participant refused, the posting was rejected for electronic transmission. The author could reduce the comment to paper and mail it to the agency in the normal course where it would be place in the public document room. The second process faciliator was a dispute resolution professional. This person was a part of the RuleNet design team. Two consensus tools were developed based on collaborative software principles. The first was a simple forms-based voting device which permitted the facilitator to draft a proposition and call for a vote of registered participants.

(2) The second was a more sophisticated table tool that was called "consensus evaluation." It was developed in consultation with John Helie, of ConflictNet who was retained as a facilitator. It was also patterned on a collaborative tool developed at the University of Arizona called Ventana Software. The idea is to provide participants the ability to register the strength of their views on a scale 1-5. The facilitator can require additional postings to explain the vote. Thus if scale numbers 4 and 5 are "opposed" and "strongly opposed", the facilitator can block the automatic registration of the vote unless the person voting provides an explanation. The facilitator does not have to personally review each vote but can instruct the computer to block the vote until a certain number of responses are received.

Computer-based technology does not substitute altogether for the actual reading of comments submitted by participants (except where the computer identifies a comment as identical to one previously filed and analyzed). Computer technology can, however, facilitate greatly the process of analyzing and tabulating comments. Using these tools, rather than the agency characterizing the positions of the participants, the participants can do it themselves. In addition, computer technology, searching for specific words and phrases, can make it easier to find where if at all a participant is addressing a particular issue in his or her comment.

The electronic forum provides for a democratization of the rulemaking process. The individual person with expertise and good ideas to offer has as much access to the forum as any governmental unit, corporation, or law firm, and may be just as influential or more so.

With access to the forum already available in millions of homes nationwide (and at terminals in public libraries, for those who do not already have access elsewhere), there is the potential to level the playing field to an unprecedented degree.

go to table of contents

Phases of the Rulemaking

As a preliminary step, the NRC made relevant information on fire protection available to all who can use it, that is, both potential participants and those who want only to observe the process. This reference material was provided on line in the "reference library."

The first phase of the rulemaking itself began with a "virtual kickoff" in which all participants will be able to communicate in a simultaneous discussion in cyberspace. A combination of five video-conference centers around the country were used. In addition the ASCII text was extracted from the court reporter's equipment and fed to the internet a page at a time.

Consequently, any internet user with web access could view the actual transcript with less than a four minute delay. In addition a forms based "mailto" document was provided on the screen that enabled Internet viewers to submit questions to the Commission at the Washington video-conference site.

The kickoff was followed by a 10 day period for review of the background documents by the participants and any caucuses; for the posting of questions and requests for clarification, directed either to the NRC or to other participants, and for the posting of answers to those questions; and for the identification of the particular issues to be addressed, or challenges to be met, in the rulemaking. This "issue identification" phase was denominated "phase 1" and was designed to be an electronic analogy to meeting "brainstorming" when no comment is considered off-limits.

In the second phase of the pilot, which also was 10 days, the participants were asked to propose solutions to the challenges and issues identified in the first phase. This also was the opportunity for participants to respond to comments and suggestions made by others during the first phase.

After the second phase, the NRC technical staff, acting in conjunction with staff supplied by the Lawrence Livermore Laboratories, consolidated and synthesized the challenges and the proposed solutions, using them to develop more concrete proposals, which were posted electronically.

In the third phase, the participants were asked to respond to the proposals identified. While there was the opportunity for participants to caucus either within the electronic rulemaking or outside of it and to continue to interact as they had in the first two phases, in reality most participants waited until the last day and posted more formal comments. In this regard, phase 3 was most like the agency's prior experience with electronic bulletin boards.

Several Issues raised by Professor Perritt in his convening paper were matters that were considered as part of the development of the RuleNet project and so some comment upon them seems appropriate.

go to table of contents

CREDIBILITY, NEUTRALITY, AND LEGAL EFFECT

The RuleNet pilot was closed in February and NRC is currently evaluating its experience. As a part of the evaluation, surveys were distributed to the registered participants. The survey, of course, has an immediately obvious built-in bias: only people who have computers and are inclined to use them probably registered in the first place. Consequently, one intuitively would expect to be sampling people who are receptive to technology in the first place. There were exceptions, however. Some industry participants who were hostile to the technological approach participated because their stake in the outcome was high. The survey revealed a high level of satisfaction from those users with access to the key agency staff although they were less satisfied that the process was without legal pitfall. In the latter respect, the major concern was whether the agency could define a proper record in the event of a court challenge. Participants expressed satisfaction with the neutrality of the facilitators and perhaps more surprising with the neutrality of the agency personnel who participated. It was made clear that agency personnel were participating in their individual capacities and that no one had the right to speak for the agency. Although Commissioners were told by the General Counsel that they could also post comments without fear since no decisional document was yet developed and before them, none chose to do so.

Nevertheless they did state their views at the initial video conference kick-off and did monitor the activity as it progressed.

go to table of contents

PARTICIPANT ACCESS TO TECHNOLOGY

Efforts were made before the start of the pilot to contact organizations known to have an interest in the issues but limited resources. In one case, Ohio Citizens for Responsible Energy, it was known that there was not a convenient local dial up access point for the representative. The reservations expressed were not based on access to necessary technology. Rather, the time commitment was cited as a more significant barrier. Reservations were expressed by some registered participants about whether RuleNet unfairly excluded citizens from access to the process. In every case, however, the participant expressing the concern had high speed access to the Internet at either office or home.

go to table of contents

FULL MOTION VIDEO

Even if full motion video were available, the NRC's experience with the video conference raises a question concerning whether it is the most useful technology in rulemaking consensus building. As previously indicated, participants could either go to a video conference site or participate from a computer in their home or office. More interactive questions were received from the Internet participants than from the five video conference sites. Fewer complaints were registered from the Internet than from the video conference sites. In several conferences this author has been on panels were discussion has centered on the hybrid nature of Internet communications in a virtual situation such as chat rooms or forums. It seems to combine aspects of both written and spoken communications.

Thoughts are not as well polished as one would expect in a carefully edited paper. Nevertheless, they are more refined and cogent than one would expect in a meeting where people tend to talk in stream of conscience. It is also true, that interaction taking place over the course of days takes longer than interactions that are face to face. More experimentation with methods to use a variety of these techniques is necessary before an optimal solution is self-evident.

go to table of contents

PRIVACY, AUTHENTICATION, and FREE SPEECH

Professor Perritt raises the issues of privacy and authentication. To these two should be added another, free speech. These topics torch more flaming on the Internet than any others. People in dialogue on the net about these issues tend to have their minds made up before the discussion starts. Whether the debate is about the clipper chip, encryption, export restrictions, the public square, pornography and children, or virtual banking the rhetorical heat is high. When congress recently passed legislation in the area, many sites drapped themselves in black. Unfortunately, the issues are complicated and the answers are not as easy as the debators would have us believe. Since it is purpose of this paper to describe how to leverage new technologies for dispute resolution on the web and not to address these issues in particular, only those issues that were considered while designing NRC's pilot will be discussed.

As a starting point it is useful to identify the conflicts. People want privacy and they also want freedom to speak. In the global electronic village these two desires are often in direct conflict. Simply put, does one's free speech right entitle one to send an Email to someone who doesn't want to receive it? Another conflict unique on the web is storage. One's electronic message finds its way onto and into many systems on its way to the recipient. Should the government encrypt messages to prevent noncitizens from reading them because they might end up in Australia on their way to California? What if a citizen of Japan comments on the U.S. rule raising a previously unconsidered point? Does the agency have to respond? Fortunately, so far as can be determined, these issues were largely academic.

NRC did provide for anonymous registration of a participant's identification so that their personal Email address did not have to be revealed. No one took the opportunity. As noted earlier, a method was devised to prevent pornographic and defamatory material from being posted in the RuleNet forums. These modest precautions appear adequate for the time being but further development of legal doctrine in this area appears both required and inevitable.

go to table of contents

CONCLUSION

Information technology provides important new opportunities to improve dispute resolution through reinventing the traditional processes of written and oral communication. To realize these opportunities, interdisciplinary approaches need to be pursued with vigor. No one group of professionals possesses unique qualifications and insights in how to adapt these powerful communication tools to more effective processes. To design effective tools requires a good understanding of the processes of communication, dispute resolution, interest analysis, available alternative computer technologies, and a healthy skepticism about the ways in which human/machine interfaces are first received. RuleNet is a baby step in the direction of building better, more democratic processes. One should not lose sight of the possibility that one might be building a process that eliminates the need for an agency drafter between the legislator and the citizen. At the same time, the facilitation skills the agency brings to the table may be in even greater demand.

go to table of contents Footnotes

Footnotes

1 I am mindful of Professor Henry H. Perritt, Jr.'s convening document, "Electronic Dispute Resolution, an NCAIR Conference," dated May 22, 1996 at page 16 where he describes the difference in time he invested to that of his students when participating in an electronic study group. In RuleNet, a number of improvements were made in facilitator tools to address such concerns. In addition, participants were expected to interact without intervention by the facilitator most of the time. Nevertheless, one should expect that the average participant in an electronic forum will spend less time than the convenor/facilitator/moderator.

2

RuleNet had three types of forums. The "Commons" where any citizen could post comments and participate in threaded dialogue. The "Caucus" forum was discussed earlier. The agency could accommodate up to seven simultaneous caucuses. The "Forum" was a threaded dialogue run under majordomo and hyperchat where anyone could read the dialogue but only registered participants could post comments and files. Registration required tendering a valid Email address which was subsequently polled. When the Email address checked out as valid, the participant was given a password to access the write level of the forum.

The views expressed by authors are their own and do not necessarily reflect the views of Resourceful Internet Solutions, Inc., Mediate.com or of reviewing editors.