[Editors Note: This is the third installment in our series looking at the ongoing ICANN reforms. If you haven't already, be sure to read Part 1, Part 2, Part 3, Part 4

[View/print PDF]

The LSE Report’s harsh diagnosis of the problems of the GNSO were widely praised, even by those working within GNSO. In the next step, ICANN’s Board Governance Committee formed a Working Group (BGC-WG) to consider the report, and make specific proposals for reform. The Working Group was formed on 30 March 2007. It was chaired by Roberto Gaetano and included six other Board members or liaisons: Rita Rodin; Vanda Scartezini; Raimundo Beca; Tricia Drakes; Susan Crawford; and Vittorio Bertola. The BGC-WG was dominated by what one might call “ICANN idealists,” people who favor a bottom up approach to policy making modeled on the institutions of the Internet technical community.

The BGC-WG operated for about a year. Three months after its formation (June 2007), it released its Preliminary Report and then, after several rounds of comments and modifications, it released its final report in February 2008. The GNSO Improvements report of the BGC-WG proposed three main changes in the way the GNSO would work. One was to adopt an IETF-style Working Group model as the basis for policy making. The second was to regroup GNSO constituencies into four broader “Stakeholder Groups.” The third was to restructure the GNSO Council and its voting along lines close to what the LSE had proposed.

Working Groups
The core of the BGC WG recommendations was its reliance on IETF-style working groups to make policies. Open, consensus-based Working Groups would become the foundation of policy development work in the GNSO. As noted above, ICANN idealists believe that ICANN’s global policies should be based on consensus arrived at through open, deliberative processes. This view is innocent of many of the sobering insights of political science and political economy. Social scientists studying policy and policy making processes tend to look at public policy differences as the product of disputes over the distribution of costs and benefits. By their very nature, policy processes are conflictual rather than consensual. The resolution of these conflicts typically produces winners and losers, as the policies adopted skew benefits toward some stakeholders and impose costs or harms on others, based on how much power they have.

In contrast to a political science or political economy view, the idealists equate ICANN policy making with what they think they know about deliberation over technical standards. They see policy making in ICANN not as self-interested bargaining among organized interest groups, but as a process of deliberation and discussion among unaffiliated individuals seeking the best solution. They believe, therefore, that ICANN should model its policy processes on the Internet Engineering Task force (IETF) and the World Wide Web Consortium (W3C). These Internet technical institutions rely on open working groups and use “rough consensus” among the participants as the basis for forging agreement on technical standards documents.

Stakeholder Groups
The BGC seemed to like the LSE Report’s recommendation that the GNSO should be composed of broader groupings of stakeholders. It eventually proposed that four broad groups be formed that would serve as the foundation for most GNSO functions: Registrars, Registries, Noncommercial Users and Commercial Users. These would be called “Stakeholder Groups.” But unlike the LSE Report, which proposed to just replace existing GNSO Constituencies with the new, broader groupings, the BGC proposed to add another layer of organization. Constituencies would be retained; Stakeholder Groups would be composed of multiple constituencies under the umbrella of a single SG.

In the end, the BGC-WG recommended a 19-person Council consisting of 16 elected members, four from each of four stakeholder groups. Contracting parties (Registries and Registrars) would each receive four seats on the Council. The other two stakeholder groups would represent users; i.e. those “affected by the contracts.” Commercial registrants and Non-commercial registrants would each receive four seats on the Council. The remaining three seats would be appointed by the Nominating Committee.

The GNSO Council
In line with its idealist view, the BGC-WG wanted to move the Council away from being “a legislative body focused on voting” towards a more strategic body that manages and oversees the policy development process. Voting, the BGC-WG claimed in its report, can encourage division rather than cooperation. Working Groups managed by a Council, they believed, would produce a more collaborative, inclusive approach. Under this new regime, the BGC opined, the Council’s most important functions would be to guide the establishment of working groups and to monitor their progress. “The Council should ensure that each WG has an appropriate charter and timeline, an experienced and neutral Chair, perform adequate outreach and have adequate technical expertise. Once a WG completes its work, the Council would review its conclusions and could prepare a statement informing the Board on how well the WG performed its function.” In the initial proposal of the BGC, in other words, the Council would not even vote to approve the WG’s policies; it would simply prepare a statement that would accompany the policy proposal on the way to a Board decision. While this proposal would indeed have eliminated the GNSO Council’s “legislative” functions, it would basically move the decision making power to the Board, centralizing even more authority over GNSO policy.

Critique of the BGC report
The BGC-WG made a good-faith effort to reform the GNSO based on the well-researched recommendations of an independent expert report. The Board Working Group reflected the views of well-meaning “liberals” on the ICANN Board and was motivated by values of inclusiveness, consensus, fairness and efficiency. Still, there were obvious problems with the report, and many stemmed from what seemed to be a willful refusal to recognize the political and regulatory aspects of ICANN’s policy making processes.

By creating broader Stakeholder categories, the BGC-WG wanted to make it relatively easy for new Constituencies to form under them. But – despite warnings from the NCUC and others – it completely ignored the heavy burdens that would be imposed on civil society actors by adding yet another layer of organization onto ICANN’s already complex system of participation and representation. Volunteer participants were now confronted with three distinct layers of self-organizing structures: Working Groups, Constituencies and Stakeholder Groups, each with their own officers, rules and processes.

One good thing about the BGC’s proposed reforms was that the working groups would be open; i.e., participation in them would not be restricted to delegates of GNSO constituencies or Council members. This would allow the group to draw on willing volunteers with expertise from any quarter. The outcome of such WGs might also be more flexible: “In a more open, inclusive working group setting,” the report claimed, “participants would be able to analyze and debate problems and potential solutions without feeling that they have to develop or assert a particular, or fixed, ‘constituency’ position.”

At the same time, some of its recommendations reflect a certain naïveté about political processes. The regulatory and economic policies ICANN produces for the domain name industry are not the same as technical standards, because in the standardization space people have much stronger incentives to come to an agreement. The openness of the proposed WGs raised questions about their manageability and about commitment and representation. Would consensus among a WG, which might not include all the relevant players, mean anything? Who would determine whether “rough consensus” among this amorphous group existed or not? What made consensus binding? For example, what if people associated with major stakeholders agreed to something in a WG and then others in the stakeholder group did not go along with the consensus? What if they agreed at first but withdrew their consent later?

In its comments on the report, civil society stakeholders warned the BGC-WG that consensus-based processes are biased toward the status quo; in effect, they give anyone who is happy with the way things are the power to veto any change. They also questioned the attack on voting. “While it is true that the GNSO has been seriously deadlocked in the past, it is mistaken to interpret this failure as a product of ‘voting.’ And it is utterly naïve to believe that the conflicts of interest that created those deadlocks will go away if there is no voting.”

Next: Field Guide to the ICANN Reforms, Part 4