A little over a year ago I wrote an article on CircleID that asked the important question: "Will PDP 3.0 Save the Multi-stakeholder Model? The answer I gave at the time was an emphatic "no". At that time PDP 3.0 was just a series of incremental improvements to symptoms of issues in the multi-stakeholder model, but none of the recommendations addressed the real fundamental problem: Namely, that participants within the ICANN community (including Government representatives) and within the Working Groups have little incentive to compromise, nor do they have the authority or willingness to come to a mutually beneficial resolution on some of the most contentious issues within Internet governance today.
Unfortunately, over a year later, my answer has not changed.
On June 4th, ICANN released its latest document entitled "Enhancing the Effectiveness of ICANN's Multistakeholder Model - Next Steps." It identifies a proposed path forward following community comments on its previous draft work plan, which was included asssss an Appendix to ICANN's "Draft FY21-25 Operating & Financial Plan and Draft FY21 Operating Plan & Budget." As a result of the comment period, according to ICANN, the following six topics were deemed the most important by the "community."
Prioritization of the work and efficient use of resources
Precision in scoping the work
Consensus, representation, and inclusivity
Complexity
Culture, trust, and silos
Roles and responsibilities
Although there is certainly a lot to be said on each of the topics, I am going to focus on the third one. Specifically the community's input that it has difficulty in coming to consensus. The current document describes a number of work items that are currently underway including PDP 3.0 and the "Consensus Playbook." The Consensus Building Institute (CBI), completed the Playbook in April 2020 and includes practical tools and best practices for building consensus. In developing the Playbook, CBI interviewed 14 ICANN community leaders and developed 15 recommended “plays” that can be applied to different phases of a group’s work. For transparency purposes, I was one of the community leaders interviewed (based on my work as a co-chair of the Subsequent Procedures Policy Development Process looking at the rules for the next round of new gTLDs).
First, let me state that I REALLY like the Consensus Playbook. I have tried out a number of the tactics included to get certain issues within the SubPro PDP resolved. I highly recommend that ALL working group chairs read this document thoroughly! More on that document later in another post.
However, neither the Playbook nor PDP 3.0 address the lack of incentives for Working Group members to actually compromise on a solution. Rather, it implicitly assumes that all Working Group members have shared goals to get to a balanced compromise policy position for the benefit of the Internet. It does not address the basic flaw that participants in the policy process all represent different interests, many of which may benefit from no changes to the status quo, and others that would only benefit from changes in the status quo.
Yes, everyone needs to file a Statement of Interest to participate. But, (a) few people read them, (b) little, if any, detail is really provided on conflicts, and (c) no one is required to name their "clients" unless they work for a domain name registry or registrar. Representatives from trade associations, educational institutions, consultancies, etc. do have to name their direct employer (eg., I work for JJN Solutions), but they do not have to disclose specifically who they are representing. Some attorneys argue that they cannot disclose who they are representing for fear of losing the attorney-client privilege. This blog does not argue with any of these assertions (for now).
In addition, you can participate in these groups in your individual capacity, meaning that you may not be accountable to anyone other than yourself (or alternatively put - accountable to no one).
Problem Analysis through Closed Generic Example.
Let's walk through a real example that demonstrates the difficulties with reaching consensus. Should an organization be able to apply for a generic "dictionary word" top-level domain and reserve all "use" of the TLD for its own organization (and its affiliates). This is often called a "Closed Generic" top-level domain. In the 2012 round of new gTLDs, this practice was not prohibited in the Applicant Guidebook (the rule book for applying for new TLDs). Nothing prohibited Google, for example, from applying for .search and proposing to use that TLD for its own purposes without making second level registrations available to any third party.
Making a very long story short, the Governmental Advisory Committee (GAC) gave advice to the ICANN Board that it believed a Closed Generic TLD should "serve a public interest goal." The ICANN Board received a bunch of letters from the community, some of which supported Closed Generics, and others that opposed. The ICANN Board essentially punted the issue by stating that Closed Generics could not proceed to delegation in the 2012 round, but it could have its application for a Closed Generic deferred for consideration in the next subsequent round. As a result of this resolution, all applicants for Closed Generics either agreed to open up their TLDs in some form, or withdrew their applications. No applicant opted to have their application for a Closed Generic deferred to the next round. Issue solved????
Eight years later we are not much closer to solving the issue. Why? There is still a strong rift in the community between those that support Closed Generics and those that oppose them. There are great arguments on both sides as to why they should or should not be permitted. There is also a valid argument stating that we should try to develop a mechanism to evaluate public interest goals (in accordance with GAC Advice).
The truth is that there is no right position and there is no wrong position (despite what advocates on either side argue). At the end of the day, it is a value judgment. And the community needs to find a way to figure out what to do on this topic. Why is it so difficult to come to a consensus on a compromise solution? Why is it not just a matter of coming up with a proposal that encompasses some common elements from each of the sides? The 200+ members of this group are all some of the best and brightest minds on Internet Governance. Surely with their combined brain power, knowledge, expertise and acumen, we should be able to come up with something we all can live with?
We have PDP 3.0 and the Consensus Playbook now, shouldn't we now be able to effectively use the Multi-Stakeholder Model? According to the Board's Work Plan, we should have all the tools needed to solve this problem!
No, we can't yet. PDP 3.0 and the Consensus Playbook are not the complete solution.
(1) The Incentive Problem.
As stated above, Closed Generic TLDs were not allowed to proceed in the 2012 new gTLD round. For most other issues, the SubPro working group has taken the position that where there is no consensus on changing any rules from what happened in 2012, then the default is that subsequent rounds would be launched without making those proposed changes. Also referred to as the "Status Quo."
Only if there was consensus on the changes, would those changes be allowed. You can see where this is going. Those that liked what happened in 2012 are reluctant to agree to any form of compromise because by definition they would (at least in their own minds) end up with something that was worse for them. They view it as a zero-sum game. If we move away from the default which I liked, it would be bad for me or my client. Therefore, if one does not agree with any proposed changes, he or she can refuse to accept them (or even work on improving them), thereby ensuring there is no consensus to make those changes. That person has no incentive to work out a compromise.
With Closed Generics, however, we have the added complexity of defining the status quo default position. In 2012, it is true that the ICANN Board did not allow applications for Closed Generics to proceed in the 2012 round, but they explicitly offered an option (that no one took up) to move that/those applications into the next round subject to policy GNSO policy development on the subject. The ICANN Board also specifically asked our group to come up with some policy in this area. So, some argue that since they did not proceed in 2012, the status quo is not allowing them in subsequent rounds. Others argue that since the ICANN Board did not specifically ban Closed Generics, the status quo may actually be that they are allowed (as they were not disallowed in the Applicant Guidebook).
If there was certainty on the default, this wouldn't help with the incentive problem because those that like whatever the default is would never move off their position and therefore prevent a consensus on any changes from that default.
(2) The Willingness Problem.
Heavily related to the Incentive Problem, if I am going to dislike any solution that moves off of what I believe is the default, then why should I waste my time working on a solution that I dislike or that is worse for me or my clients. Put another way, why should I work my tail off on getting to a compromise for which I feel I will be worse off? If I am representing a competitor to someone that applies (or wants to apply) for a Closed Generic
(3) The Authority Problem.
In order to address this problem, lets discuss the composition of a PDP Working Group. Anyone that wants to join a PDP Working Group can. There is no requirement that you officially represent one stakeholder group, constituency, government, institution, etc. You can just be you.
That said Working Group members are usually comprised of Contracted Party representatives (the domain name registries and registrars that will have to implement the policies), some members that officially represent their businesses, stakeholder groups, constituencies, non-profits, consultant that represent one or some of the above, attorneys that may represent one or some of the above, members of civil society (that may or may not be representing their organizations), government employees participating most likely in their individual capacities, consultants that may be paid to just monitor what is going on, and of course individuals that may also be working for one or some of the above, may be retired, or just someone who is interested in the topic.
To state it another way, some participants are accountable to businesses, organizations, constituency groups, stakeholder groups, advisory committee, governments, clients or perhaps only to themselves. Those that represent others not participating in the working group often are paid to be there to lobby specifically for their clients' interests. Thus, that representative may have received specific instructions to push for those interests and not to move off of that for any reason. This situation is common among lawyers and consultants. Even if they had the incentive and the willingness to find a compromise to the issue, they do not have the authority to do so. This is especially the case where the client behind the scenes is in favor of the status quo. But it can also be the case where consultants are paid to lobby for changes and to not return without those.
So what can/should we do?
In the next blog post, I will put forth some measures that I believe can help with these issues, but also argue that some of these issues are inherent in the multi-stakeholder model which we will need to live with and others which may requiring systemic change.
Sorry for the teaser, but this post was getting too long.
Picture courtesy of ICANN at https://gnso.icann.org/en