PUBLIC PARTICIPATION IN ICANN - THE AT LARGE DEBATE
Joe Sims*
* What follows is a personal evaluation of these documents. It does not necessarily represent the views of any member of the ICANN Board or staff, and is obviously a single perspective on a complex set of issues. I hope it will contribute to the ongoing dialog.
INTRODUCTION
One of the important remaining issues relating to ICANN's organization is how to achieve an appropriate level of public participation in ICANN without impairing its ability to effectively carry out its principal mission - to preserve the operational stability of the Domain Name System. While there is little disagreement on the concept of public participation, the type and approach have been vigorously debated since before ICANN's creation. In 2000, ICANN failed to find a consensus position after considerable effort, and fell back to a temporary compromise - the direct election of an ICANN Board member from each of five geographic regions, and the establishment of a blue ribbon At Large Study Committee (ALSC), headed by the former Prime Minister of Sweden Carl Bildt, to seek a final resolution that could achieve consensus support throughout the ICANN community.
The ALSC has now released what it describes as a draft final report. It will take further input during ICANN's meeting in Montevideo (now underway) and will produce a final report to the ICANN Board prior to the 2001 Annual Meeting scheduled for Los Angeles in November of this year.
As part of this debate, the Markle Foundation, an organization interested in global communications policy, funded an effort by a group of public interest groups and academics to examine the 2000 elections, and to make recommendations to the ALSC and the ICANN Board on the subject. This group has now produced the NGO and Academic ICANN Study (NAIS), a lengthy report containing both evaluations by region of last year's elections and the authors' views on the proper approach for ICANN moving forward.
Concurrently, the Markle Foundation commissioned a survey of American public attitudes on "governing the Internet." The results of that survey, described by Markle's President Zoe Baird as intended to examine "who should be accountable for governing the Internet," were also recently released. Given Markle's role in funding both the 2000 election and the NAIS study, it seems useful to note generally here, as background and context, the results of this survey as provided by Markle in a report entitled "Toward a Framework for Internet Accountability."
SUMMARY
The Markle survey asked those interviewed or surveyed how the Internet should be governed With apologies for necessarily over-simplifying, the report concludes (after 84 pages) that the answer is not clear: The Internet requires some new paradigm of governance, the report says, but it does not try to identify it. The report does conclude (not surprisingly) that, whatever that paradigm turns out to be, the public wants to be involved in some meaningful way.
The NAIS report turns out to be two essentially separate documents: an evaluation by region of the 2000 elections and a set of recommendations as to the future approach to public participation that seem to be completely disconnected from those evaluations. The research findings could be fairly summarized as follows: Most potential At Large members have never heard of ICANN, and have no interest at all in it; this was even true of most of those who actually registered as At Large members for the 2000 elections, since many if not most of them did so for some reason other than genuine interest in ICANN. The 2000 election had many flaws, including particularly the postal authentication system, language barriers, and unfamiliarity in some regions with the voting process. Many people would prefer different, smaller geographic regions to recognize cultural and other differences that are subsumed in ICANN's current five regions. Only in North America were there no major complaints, but the universe of people in North America interested in ICANN is trivial. The biggest driver of registrations was national pride, and this is likely to be an even bigger force in future elections, now that people see what can be done. It is not at all clear that even a massive outreach effort would produce a meaningful increase in the level of real interest in ICANN and its activities.
The NAIS recommendations seem to completely ignore these research results. They call for regional and global elections of half the ICANN Board, with the electorate made up of all "interested" humans, and suggest using the same approaches that the evaluations concluded were seriously flawed. In addition, the NAIS authors recommend a massive outreach and education program apparently designed to convince people that they should be interested in ICANN. While conceding that their recommendations do not enjoy consensus support, the NAIS study nevertheless asserts that they represent the only way for ICANN to gain what NAIS calls "legitimacy," although exactly what NAIS intends that to mean is not clear. The NAIS study also recommends the imposition of a mandatory tax on all domain name registrations to pay for what will be an extremely expensive effort to interest people that are not now interested in something that they probably should not be interested in, and then to have that artificially created electorate select half the directors of an organization whose primary responsibility is maintaining the stability of the DNS. And finally, the NAIS authors recommend that ICANN take a variety of other steps they see necessary for "accountability" and "legitimacy", including adoption of the United Nations Charter on Human Rights and the creation of an independent global judiciary. This is a wish list of what are clearly not mainstream positions; it is not a credible set of recommendations deserving of serious consideration.
The ALSC draft report shares many of the underlying premises of the NAIS report - the importance of appropriate public participation, the value of an organized and informed At Large constituency (they suggest the creation of an At Large Supporting Organization, while NAIS recommends the creation of an At Large Membership), and that the direct election of a portion of the ICANN Board is the preferred vehicle for the expression of public participation in ICANN. It also agrees with the NAIS study that there is very little interest among even those knowledgable about ICANN in the structure and performance of an At Large participation, and that there is certainly no consensus on a solution. Where ALSC parts company with NAIS is that it has attempted to craft a solution that has "the best chance of consensus support," - in other words, instead of demanding that the world accept a particular choice, the ALSC has tried to find a solution that most of the ICANN world might find acceptable. It recommends that the At Large electorate be made up of those individual domain name holders who choose to pay a small additional fee in order to participate in an At Large Supporting Organization, which would elect one-third of the ICANN Board. The ALSC believes that this approach will (1) be self-funding, thus eliminating the need for taxing those domain name holders who choose not to participate; (2) will maximize the chances that the At Large electorate will be truly interested in ICANN (as demonstrated by a willingness to pay a fee) and knowledgable enough to have obtained a domain name; and (3) has at least the potential for generating consensus support from the ICANN stakeholders. It recognizes that its proposals would leave some people out, and that there are still considerable logistical challenges to be faced even with the approach it recommends, but it concludes that this approach is "defensible, executable, effective and most likely to gain consensus." While there are a number of serious potential flaws in the ALSC approach, it is a serious and credible proposal by an independent body that appears to have taken seriously its instructions to do a "clean sheet" study and to try to produce an approach that could gain consensus.
These and other approaches will continue to be debated until the ICANN Board meeting in November, at which time the Board is scheduled to receive and act on the final ALSC report.
THE MARKLE SURVEY
The Markle survey report contains a large amount of interesting information, but only part of it is directly related to this issue. It was obviously intended to justify and explain a need for some new governance paradigm for the Internet, but equally obviously not intended to suggest the answer to the question. According to the Markle survey, the public wants to be involved in some meaningful way in Internet governance. Unfortunately, the survey results also demonstrate that a majority of those surveyed believe that the general public does not know enough to make a meaningful contribution to that governance. (page 45) This was the majority view not only of the general public surveyed, but also of the "Internet experts" surveyed. With respect to the latter, the report noted an interesting consequence of the desire for public involvement and the perceived lack of ability of the public in this area: "One result of this [view that the public does not know enough to be involved in setting rules], for these experts, is that they often say that they tend to discount the public's opinions as they work on developing rules for the Internet, and instead rely on their own judgment about what is good for the public." (p.45-6)
The most interesting part of the Markle survey for these purposes was the fact that not a single person in any of the diverse focus groups surveyed had ever heard of ICANN, and only about half of the "Internet experts" could identify ICANN. This broad and statistical affirmation of what to many seems intuitively obvious -- that the apparent "public" interest in ICANN is actually only the intense (and loudly expressed) interest of several dozen academics and policy activists -- is also consistent with the NAIS study results, as discussed below. It also sets the stage for what is a key question in the debate over public participation in ICANN: how much of the debate today truly reflects the public interest, and how much reflects the personal and institutional agendas of the proponents, who like the "expert" quoted in the Markle report "know" what the public would want if they only knew enough about the subject.
THE NAIS REPORT
The NAIS report, also primarily funded by Markle, seems to this reader to be a striking illustration of the Markle survey finding that, as a "director of a non-profit that focuses on Internet policy says, 'I don't respect necessarily what the public wants. So I would make the decisions more on what I think the public needs.'" (p.46) This seems the only reasonable interpretation of the fact that the NAIS report's conclusions are clearly not drawn from, and indeed bear little resemblance to, the factual evaluations of the 2000 election reported in the body of the study. It is apparent from both the discussion in the report and its conclusions that the NAIS recommendations are based more on pre-existing policy preferences than on factual analysis; indeed, the "research" portion of the report seems largely a throw-in, included to give the recommendations some additional weight. This certainly does not disqualify the recommendations from consideration; they are entitled to whatever weight is justified from their inherent persuasiveness. But it does suggest that the policy recommendations contained in the NAIS report are and should be viewed as simply those of its authors, not the result of some scientific study or survey.
Again with apologies for over-simplifying (and with the promise of providing examples below), the NAIS report can be summarized as more a statement of desirable objectives (at least from the authors' perspectives) than a proposed solution to a problem. The report recognizes that "there is far from a broad consensus about ICANN's proper role, and there is even greater variation of opinion about the best way that public representation can keep ICANN on the right course (or whether there is any role for public representation at all)." (p.90) But the NAIS authors then assert that the only possible solution to the problem is one that has already been demonstrated to lack consensus support. In a consensus development organization, where it will take a 2/3 vote of the existing Board to adopt any solution, this seems an odd way to advance your objectives.
The NAIS report documents, in its regional analyses of the previous At Large election, significant executional problems. It then also ignores those problems in asserting that a global election among all who are "interested" in ICANN (but presumably only after significant outreach, because the NAIS research work documents, consistent with both the Markle and ALSC conclusions, how few people actually are "interested" in ICANN,) is the only way for ICANN) to gain "legitimacy," and thus presumably to do its job (more on this later).
The report recognizes that global elections would be very expensive (although it probably seriously underestimates the cost). It then avoids dealing with that problem by asserting that (unlike all the other subgroups of ICANN, which are self-funding) it is the responsibility of those individuals and businesses who pay for domain names to shoulder this expense, whether they want to or not. It is notable that the report rejects the notion that domain name holders should have the ability to elect any of the At Large Directors, and seems not to appreciate the fact that an additional fee imposed on all domain name holders would likely be considered just the kind of "Internet tax" that so many successfully advocated against earlier in ICANN's existence.
The NAIS report clearly began with an assumed premise -- all interested individuals should have a voice in ICANN. It then proceeds to the conclusion that the only way this could appropriately happen is to hold direct elections of all interested humans on the globe for half the seats on the ICANN Board. How it gets from "a voice" to "elect half the Board, no matter how much it costs or how hard it is to do" is never clearly explained. This analysis attempts to pull together the logical and rhetorical threads, and figure out how NAIS manages to weave them into this particularly unattractive piece of fabric.
1. The Purpose of Public Participation in ICANN. The NAIS report contains several somewhat inconsistent explanations for why public participation in ICANN is important [my comments in brackets]:
-- "The mission, character and history of ICANN requires global public participation and representation for its long-term legitimacy and stability." (p.3) [This seems uncontroversial].
-- "[ICANN's] potential authority is largely unbounded. . . . ICANN is likely to face increasing pressure from those seeking ways to control a range of behavior online. The broader this pressure, the more sweeping the potential impact, and the greater the need for public participation and representation." (p.4) [This seems circular; if ICANN "requires" public participation today with its limited responsibilities, it certainly would if it had broader responsibilities. But the NAIS report authors appear not to perceive that the reverse is probably also true -- if ICANN actually becomes a global democracy, as the NAIS report finds essential, there will be little excuse for it not taking on additional policy responsibilities. Today, ICANN can truly say it is not the government of the Internet; if the NAIS recommendations were adopted, this would be a much more difficult argument.]
-- "Public participation is . . . a safeguard against domination by governmental or commercial interests on a national level, and as a source of input as [sic] for global interests as ICANN's functions transition away from US control." (p.4) [This is the first of several times in the NAIS report where the authors display a significant anti-government bias -- not just with respect to ICANN governance, but also the notion that global public participation in ICANN is somehow a protection against actions by the governments elected by those same public participants. This concept reappears regularly throughout the NAIS report -- that governments are not truly representative of their citizens, and that direct global elections are a vehicle around that problem. Whatever the merit of that notion, it hardly seems an appropriate justification for any particular level or kind of public participation in ICANN. And the notion that public participation in ICANN provides some real protection from perceived governmental abuses seems naive. In addition, the NAIS report is harshly critical of ICANN's Government Advisory Committee, saying it has "poor mechanisms for transparency, and for public input and participation because it operates largely in a closed and inaccessible fashion." (p. 32) Later, it asserts that "The GAC speaks on behalf of governments as institutions, not on behalf of the public users of the Internet as individuals," and that it acts in "an unusually secretive fashion," which makes it "an inappropriate channel for the public voice within ICANN." (p.105) This odd view of the world, as if governments were disconnected from the populations they represent, seems to underlie much of the approach of the NAIS authors, who apparently give no weight at all to the notion of representativeness if it is not embodied in direct public participation. The logic behind this is, at a minimum, not obvious.]
-- "ICANN will gain requisite legitimacy from the creation of a membership that allows for substantial participation and representation by interested members of the public . . . . Since ICANN activities have the potential to impact all Internet users and in fact the public in general . . . . membership should be open to all who express interest . . . " (p.8) [Here the rationale for broad "membership" seems to be the potential of ICANN, not its current reality. Indeed, the NAIS authors apparently think that "ICANN will become more, not less, relevant and visible to the public" in the future. Since this is the diametric opposite of the stated goals of virtually every other ICANN stakeholder (and the management of ICANN), who have consistently resisted adding the kind of functions and authority that would produce this result, it seems the NAIS authors have a very different ICANN in mind than the rest of those involved in the ICANN process. In addition, this notion of "legitimacy" appears frequently in the NAIS report, but it is never clear exactly what it means. At one point, the report states the obvious -- that if businesses and governments choose not to continue to look to the ICANN root as the authoritative root, ICANN will lose its reason for existence. (p.21) But the NAIS report never seems to connect this practical point with its calls for broad global elections. Perhaps the NAIS authors assume that, without such elections, governments and businesses would be more likely to look elsewhere than ICANN for coordination of names and numbers, but they never so state, and the notion seems unlikely. There appears to be little enthusiasm for broad global elections by all "interested" humans among the governments most active in ICANN, and absolutely no support for the notion from businesses. "Legitimacy" in the NAIS context, as best can be understood from its report, appears to mean in the eyes of those very small number of humans who have ever heard of ICANN, and especially those who have been generally critical of t
he amount and methods of public participation in ICANN to date (which would include at least some of the NAIS authors). According to NAIS, this "legitimacy" can only be garnered by reducing the level of ignorance about ICANN out in the world, and encouraging large numbers of people who currently don't even know that ICANN exists to participate and vote for directors on the off chance that, at some point in the future, ICANN might do something that they might want to have a say in.]
-- "The central part of the [criticism of ICANN] is that ICANN's organizational structures and activities do not comport with the ethos of good and democratic governance." (p.15) [This cramped perspective of the source of all ICANN's criticism illustrates how the NAIS authors manage to come to some of their odd recommendations. In fact, "democratic governance" has never been a condition precedent to a viable and successful ICANN, which as a practical matter depends on voluntary acceptance by those with whom it contracts and interacts. That acceptance is itself premised on ICANN's success in providing an effective mechanism for consensus policy development that contributes to the maintenance of the stable operation of the DNS. It seems highly unlikely that the root server operators, the address registries, the name registries, the ISPs or the registrars of the world perceive "democratic governance" of ICANN as the critical characteristic upon which its success rests.]
-- "The need for user participation and representation to legitimize Internet governance -- and consequently ICANN itself --is the proper basis of all debates with regard to the At Large Membership." (p.18) [This is one of the places where the Markle survey and the Markle-funded NAIS study show their common roots. The Markle survey report obviously sees ICANN as an entity engaged in Internet governance (see p.45). As this quote reveal, so does the NAIS report, and this perception appears to drive much of its policy focus. The NAIS report concludes that ICANN is important, can affect people directly or indirectly, and thus "public participation in ICANN is . . . a prerequisite to ensure that the 'public interest' is taken into account when implementing its mission." (p.20) NAIS does not explain why direct public election of the managers of this particular important resource is so critical, when the managers of perhaps even more important resources - national military forces, electricity and water systems, the telephone system, etc -- are almost never directly selected by the public they serve.]
-- "Without a broad, participatory and representative At Large Membership, ICANN will not gain or hold legitimacy, and its long-term survival will be threatened." (p.22) [Of course, these words have imbedded in them a number of policy judgments that might not be universally shared. Since the most important value that ICANN serves is stability, and since there appears to be significant concern from many parts of the ICANN community about the potential instability effects flowing from direct elections by a global electorate almost totally ignorant of what ICANN does, this assertion seems more a proposition than an axiom.]
There is little debate about the value of some form of public participation in ICANN. What ICANN does will have effects on many Internet users, albeit mostly small and highly indirect effects. More generally, the Internet is (or should be considered) a global public resource; as such, all Internet stakeholders, including the users, should have some way to participate in and influence ICANN. But it is not intuitively obvious that "the public" in this context must be defined as "anyone that can be persuaded that participating in ICANN would be interesting," which is essentially the NAIS recommendation.
It is not clear how the NAIS authors can so confidently assert that "[i]nterested ICANN stakeholders include many Internet users who are not necessarily domain name holders" (p.8) when NAIS's own research shows that there are very few people in the world that have any interest in ICANN at all, whether they hold a domain name or not. Given that the vast majority of people in the world don't even know ICANN exists, and that to change that fact would require the expenditure of many millions of dollars that, under NAIS's plan, would be extracted from current and future domain name holders without regard to whether those individuals (or big businesses, or individual proprietorships, or small businesses) wanted to fund this cost, wanted to participate in ICANN or even thought that an At Large membership was a good idea, the case for inventing an "interested" electorate has, at a minimum, not been made.
"Public participation" in ICANN, if that is truly the goal, can certainly be accomplished in many less expensive and complex ways, and with much less risk to ICANN's basic mission of protecting the continued stability of the DNS, than the approach suggested by NAIS.
2. The Relationship Between Public Participation and Direct Elections. The portion of the NAIS report devoted to an analysis by regions of the previous ICANN election catalogues a large number of problems, differing somewhat by region. The research presented in the report concludes that:
In Africa, there are a relatively small number of people with access to the Internet; most of those had never heard of and did not care about ICANN; many could not understand the voting process; and the frequent failure of the postal system made it difficult for many to authenticate their registration. The researchers recommended significant outreach to inform potential members about ICANN, a need to allow much more time for paper mail to reach its intended destination if that process was used again, a change in Web-based registration and voting, since few have Internet connectivity, and consideration of an email voting system notwithstanding its potential for fraud.
In Asia-Pacific, there was also little knowledge or interest in ICANN until some in Japan began a serious effort to protect what they saw as the "Japanese" seat on the Board; this caused reactive efforts in China and elsewhere; and nationalistic competition became the driving force for registrations, not real interest in ICANN or its work. The researchers recommended a new regional structure to account for heterogeneous linguistic and cultural backgrounds, significant outreach efforts to overcome language barriers and ignorance about ICANN, and the creation where they do not exist of forums for grass roots activism.
In Europe, registrations were driven heavily by where the popular press became active, but the low voting turnout was attributed to unfamiliarity with PIN numbers, or confusion, or summer vacations. [The researchers apparently did not consider apathy, which seems a possible explanation since registrations were not primarily the result of serious interest in the work of ICANN but more the product of considerable confusion about what the role of ICANN was in Internet governance.] The researchers called for more outreach, some way of taking the different cultures of Europe into account and not requiring a single voice to represent the whole of Europe, and greater support for more languages.
In Latin America, again the researchers reported that the number of people "concerned with ICANN-related issues is quite small." In addition, familiarity with elections is relatively low in some Latin American countries. The researchers also noted that most people who registered "did so more from the recommendation of others, than from any particular personal motivation or commitment to ICANN." The researchers noted problems with the postal mail service: "In Latin America, few would have faith in any election system where the right to vote depends on postal services or the capabilities of failure-prone servers." And to the researchers this suggested a broader point: "can we find meaningfully representative systems when we are working with electoral universesof undefined character and scope?" Only about half the people interviewed supported contined direct elections; the remainder proposed some kind of indirect mechanism for ensuring public participation. Like in Europe,the researchers recommended rethinking the geographic regions, since Latin America contains several different cultural and geographic communities.
In North America, the most connected region on Earth, 3449 people voted in the At Large election, out of a potential electorate of at least 100 million. Here too, only a very small fraction of the population has ever heard of ICANN, and among those who have, "there is no consensus about the true and proper nature of ICANN's mission. . . . [E]ven among those who share similar views of ICANN's mission, there are differences of opinion about the role of public representation within that framework." Many of those interviewed "expressed skepticism about whether [the low turnout] could be substantially increased without a far more concerted and widespread public education campaign." This despite the fact that there were multiple online resources available, a widely publicized public forum for all candidates, and "a proliferation of voter education and advocacy resources online, highly accessible to interested members of the North American electorate." The researchers concluded that the "technical (and to a certain extent, arcane) nature of ICANN's mission, combined with a relatively low level of sophistication among North American Internet users regarding ICANN's policy issues, seems to make substantially larger electorates unlikely without dramatically greater public education, and perhaps an altered vision of ICANN itself." [One wonders what this "altered vision" would look like?] The researchers also concluded that there was a "lack of consensus about the purpose of the At-Large Membership -- an important observation about this election as a whole, and a common refrain heard from participants on every side of this issue."
These findings could be fairly summarized as follows: Most potential At Large members have never heard of ICANN, and have no interest at all in it; this was even true of most of those who actually registered as At Large members for the 2000 elections, since many if not most of them did so for some reason other than genuine interest in ICANN. The election had many flaws, including particularly the postal authentication system, language barriers, and unfamiliarity in some regions with the voting process. Many people would like some different, smaller geographic regions to recognize cultural and other differences. Only in North America were there no major complaints, but the universe of interested people in North America is trivial. The biggest driver of registrations was national pride, and this is likely to be an even bigger force in future elections, now that people see what can be done. It is not at all clear that even a massive outreach effort would produce a meaningful increase in the level of real interest in ICANN and its activities.
One might have expected from these findings a very healthly skepticism about repeating the past, and have hoped for some creative alternative suggestions. But instead the NAIS authors offered the following recommendations:
-- Because almost no one has heard of ICANN, and even those who have are not interested in participating in it, those who purchase domain names should be taxed for a massive outreach and voter education campaign to explain to the public why they should care, and why they should want to participate in ICANN;
-- Notwithstanding that the past election experience would easily convince an objective observer that a repeat was likely to be a horribly expensive failure, ICANN should undertake an even larger effort with essentially the same tools that, by NAIS's own evaluations, failed last time;
-- Despite complaints in several geographic regions that they were too big for appropriate representation, we should use the same regions because they are convenient and easy [these attributes do not appear to be positive with respect to other options on other issues]; and
-- Half of the Board of ICANN must be elected through this flawed process, because that is the only way "to provide the public voice with meaningful representation," even though the portion of the "public" that has any interest in ICANN is minuscule.
The only logical explanation for the dichotomy between NAIS's research and its recommendations is that the recommendations were preordained and not affected by the research. Given the funding source, perhaps this is understandable; it is clear that Markle has a particular vision of public participation on the Internet, and is prepared to spend considerable of its funds to advance that vision. In any event, this disconnect between the research and the recommendations means that the recommendations have to stand on their own merit, and thus we turn to the specifics of those recommendations now.
3. The Basic Recommendations of the NAIS Report. The NAIS report concludes with some very specific conceptual recommendations about how to produce public participation in ICANN. It makes no pretense that these recommendations reflect any kind of consensus among ICANN stakeholders; in fact, the report explicitly concedes that there is no consensus, but then goes on to assert that the only "appropriate" way to generate public participation is through (1) global elections (2) by all "interested" persons (3) for half the Board seats. It then offers specific suggestions about how to accomplish those goals. My comments on these conceptual recommendations follow.
Global Elections. The case for global elections in the NAIS report is very weak. The regional studies demonstrate empirically what was already common knowledge -- most people have no knowledge of, and no interest in, ICANN. The NAIS report justifies ignoring this apparently decisive fact on the grounds that ICANN is the first step toward the creation of global democratic mechanisms that can avoid national governments and allow people to directly control important public resources. Putting aside for the moment whether this is a broadly shared view or goal, NAIS nowhere explains why ICANN is an appropriate vehicle for this experimentation, nor how the other stakeholders in ICANN can have any assurance that chasing this goal will not interfere with ICANN's basic mission -- preserving the continued stability of the DNS. To NAIS, the problem is public ignorance, which can be corrected with massive outreach and voter education -- to alert them to the existence of ICANN, to explain that it is important to them and their future, and to convince them that they should vote for people who are going to protect their interests (which, since the public does not have any current interests, will presumably be the interests of those conducting the outreach and education).
It is hard to imagine a better illustration of the "I know what the public needs" view; even when their own research shows that the public doesn't care, the NAIS recommendations express what NAIS thinks "the public needs." Since this recommendation ignores the serious practical problems that would be associated with a global election of potentially millions of people, assumes away the funding problem by requiring all those who purchase rights to use domain names to pay the (probably) tens of millions of dollars that would be required to manage such an outreach program and election, and assumes that ICANN staff (already seriously overworked) can simply buckle down and spend the time necessary to do something right that has never been done before, it is not persuasive. This may be a wish or desire, but it is not a credible recommendation entitled to serious consideration.
All Interested Persons. The NAIS report documents that there are very few people truly interested in ICANN. But since this fact would not support its call for global elections, and since the authors clearly believe that more people "should" be interested in ICANN, they would remedy this "problem" by a massive outreach effort. Nowhere are the pros and cons of this recommendation compared to the basic purpose of this exercise: to find a practical and appropriate way to ensure public participation in ICANN. Obviously, there are many possible ways to accomplish this objective, and NAIS lists some: use domain name holders as the electorate, have At Large directors appointed by governments, have At Large selected by intermediary organizations or by the Board. But the NAIS authors then reject all these options because "Direct elections, while imperfect, are more likely to provide ICANN with global legitimacy than other proposed options." (p.9)
This concept of "legitimacy" is worth more attention, since it is the core principle upon which the NAIS recommendations rest. Legitimacy is clearly not a self-defining term; it means different things to different people and groups. Some of those differences are highlighted in the research portion of the NAIS report; one example is the reference to practices in the Asia-Pacific region during the last election that some in the Western World would find illegitimate (corporate encouragement to register and vote). It is highly unlikely that ICANN's "legitimacy" in the eyes of many (if not most) of its stakeholders rests on the presence of globally elected directors selected by a vote of all interested persons in the world. To NAIS, however, this is the only definition of legitimacy. That view is not widely shared among other ICANN stakeholders, or this issue would not still be unresolved.
Even the NAIS authors appear somewhat ambivalent here. At one point, the NAIS report asserts "the point of ICANN elections is to choose directors that are representative of the public's perspective . . . . " (p.9) But later, the NAIS authors assert that "ICANN is a membership organization, not a government. The elected At-Large directors are not meant to "represent" the entire worldwide public at large, nor even all Internet users. The elected directors are chosen by the members of the organization and, in the service as Directors, are expected to bring the perspectives of those members to the Board." (p.109) One might ask why it is necessary or desirable to engage in extensive outreach and voter education, if ICANN is a membership organization and not a government? Why would it not be sufficient, for example, to randomly select several persons from an At Large membership (or even an ALSO as proposed by the ALSC) who by joining (and even paying a fee) had demonstrated that they were "interested" in ICANN, and have them bring the necessary public perspective to the Board? The answer may be that the real motivation of the authors is not simply appropriate public participation in ICANN but some broader strategy aimed at developing global democratic institutions dealing with Internet governance.
The NAIS authors flatly assert that "Without a broad, participatory and representative At-Large Membership, ICANN will not gain or hold legitimacy, and its long-term survival will be threatened." (p.22) This seems more a hope than a prediction; ICANN's "long-term survival" would seem to depend more, for example, on the continued cooperation of root server operators to recognize ICANN as the DNS policy development body, on ISPs' continued willingness to treat the ICANN root as the authoritative root, and in the willingness of TLD registries to contract with ICANN, than on the presence of an At Large membership, however organized. The NAIS authors seem to assume that, because in their eyes "democratic governance" of ICANN is so important, so will all other ICANN stakeholders, but they provide no reason in their report to believe that is actually the case. Indeed, it seems much more likely, as the ALSC finds, that continued operational stability of the DNS, and ICANN's contribution to that, is really the critical ingredient in ICANN's survival.
Later on that same page, the NAIS report goes on to say: "Membership in ICANN should be open to any individual with an e-mail address and the interest to join. The low barrier to entry ensures that anyone with the time and interest to participate can do so. That very openness provides the opportunity for participation to those who want it, and in so doing validates the public legitimacy of the process. It is precisely the lack of barriers to public participation as a member that legitimizes the election as a selection process, not its "representativeness."" (p. 109) This makes it clear that the goal of the NAIS authors is global democracy, not just public participation in ICANN. For NAIS, ICANN is apparently merely a vehicle for advancing a broader agenda, with little (or no) regard for the possible consequences on the ability of ICANN to carry out its limited but important mission. One does not need to be a corporate apologist to recognize that global elections by all interested persons is probably the most difficult, the most expensive, and the least likely of all the possible ways to produce useful, relevant public participation in ICANN.
Half the Board. This recommendation is, if anything, even less well supported than that of just discussed. It is clearly an article of faith, not the result of analysis. Indeed, it is only mentioned in 5 places in the report, probably occupying a total of just over a page of text out of the more than 150 page total [my comments in brackets]:
-- "Balance and of [sic] appropriate representation require that the Membership directly elect at least the same number of seats on the ICANN Board as the various Supporting Organizations in total (nine currently)." (p. 7)
-- "In order to provide the public voice with meaningful representation, At-Large Directors should balance the policy authority of the Directors chosen by Supporting Organizations." (p.9) [This part of the report goes on to note that this would mean that by-law changes (which require a 2/3 vote) would have to have "at least some support from both the At Large and SO Directors. In our view, this would provide an appropriate level of 'checks and balances' on the board, and would provide the At-Large with enough authority to block changes to the bylaws that might negatively impact the public voice." Or, to translate into practical terms, the At Large alone of all the ICANN stakeholders should have veto rights over all ICANN actions. It is hard to take this proposition seriously. The report goes on to say "We believe that reducing the At-Large Directors to less than one-third of the Board would seriously undermine the legitimacy of ICANN." The authors give no explanation for this ipse dixit, and leave open the question of whether having one-third of ICANN's directors come from the public (as the ALSC recommends) would have the same effect.]
-- "The importance of public participation should be recognized by ensuring the Membership a number of Board seats at least equal in number to the representation accorded other stakeholders in aggregate (currently the Supporting Organizations)." (p. 108) [Here, the NAIS authors explictly state what is implicit in the entire report -- that their view of the world is that the public ("all interested individuals") is on one side, and everyone else -- governments, scientists, providers, business organizations (large or small) -- are on the other. It is "us versus them," and thus to be "fair," each side has to have an equal number of players. This view of the facts is idiosyncratic at best. You can see part of the rationale for this bizarre perception in the NAIS report's perspective on the existing ICANN Supporting Organizations: "It is not clear by what measure of legitimacy or representation a board director selected by the DNSO, for instance, can be said to validly represent the interests of all the various members of the various constituency groups within the DNSO. Few in ICANN criticize the DNSO because only a handful of businesses, out of the millions of businesses worldwide, actively participate in the business constituency group and in the selection of the Names Council representative. The same is largely true of the ASO and PSO, and of the legitimacy of the directors they select." (p. 109) In fact, the ASO is populated by every single one of the address registries in the world, and those registries are themselves large membership organizations composed of numerous entities involved in the allocation of IP addresses and address policy. The ASO directors are by definition highly representative of that collection of ICANN stakeholders. The PSO is composed of every single body that generates global Internet protocols, and each of those organizations is itself a body composed of many individuals or entities; thus, the PSO Directors are also highly representative of that collection of ICANN stakeholders. The DNSO is a more complex body, and may indeed benefit from restructuring to make it more representative, but the business constituency itself is made up of a number of membership organizations (e.g., the US Chamber of Commerce) and other entities that collectively represent many millions of businesses. It is hard to see why this is not a representative body for this collection of ICANN stakeholders. The notion that "all interested individuals" have legitimate interests equal to the sum of all other ICANN stakeholders is certainly not obvious, and the NAIS authors never try to justify this odd proposition.
-- "The only board selection mechanism that can assure ICANN's legitimacy is a direct election of At-Large seats on the board of directors. The number of these seats must be at least equal in number to the aggregate number of seats granted to the Supporting Organizations." (p. 121) [By this stage of their report, the authors must assume they have made their case, since they offer no justification for this assertion here.]
-- "We believe that an elected one-half of the board represents an appropriate level of representation to assure the accountability of ICANN's policymaking endeavors." (p. 122) [Here, the NAIS report presents its most thorough justification for this point, citing balancing "stakeholder interests represented in the Supporting Organizations with the diverse public interests represented through the At-Large Membership;" providing a "meaningful counterweight against the actions that the Board's vested interests might otherwise impose by majority on the public and so satisfies needs for accountability and legitimacy in ICANN;" enabling a "diversity of viewpoints on the Board;" and being "consistent with concepts of historical balance expressed in ICANN's founding documents." Again, the arguments reflect the "us versus them" perspective noted earlier, and are otherwise on their face persuasive on the merits.
4. The At Large Membership. To be fair to the NAIS report, it says that its recommendations on elections are stated to be secondary to their main point -- that an ICANN At Large Membership needs to be created that is globally representative, inclusive, and active. Electing members of the ICANN Board is only one of the tasks of the ALM as NAIS's authors see it; as or perhaps even more important is energizing a large body of people around the world to "balance" the corporate and governmental interests that NAIS sees as dominating ICANN today (and perhaps creating a base for global public participation on other issues).
One must ask: Is what ICANN does so important to humanity that it justifies the most comprehensive global collection and organization of individuals in the history of the world? That is what the NAIS report recommends: the "functions" it envisions for the ALM include "the creation of local, regional and global associations that will allow members to receive and discuss information about ICANN in the language most convenient for themselves;" "deliberation and efforts to affect ICANN policy, such as (virtual) discussion platforms, working committees, petition tools, and outreach meetings;" "enabling the development of communities or networks around specific issues and concerns to facilitate informed debate;" and "ensuring transparency of the policy-making process, including consultation mechanisms that help inform and justify Board decisions." This utopian view of what it takes to gain appropriate public participation in ICANN seems highly disproportionate to the problem. Many would undoubtedly feel that the enormous energy required to carry out these recommendations might be more productively focused on some other more appropriate objective, such as feeding the hungry or treating disease.
The NAIS report recommends "a concept of membership in which an individual may join simply by registering his interest in participating." This membership would require a Membership Council, which would also have to be directly elected; a Secretary-General (undoubtedly with helpers); ALM Working Groups or Committees, which "could be initiated upon request of individual AL Members, registered ALM associations or the Membership Council;" and ALM Local/Regional Associations, which would be created by "submitting a description of the association's purpose and support within the Membership to the ALM Secretariat." Any such association so registered would "achieve access to the membership roles of the ALM."
Considering the almost total lack of interest by the general public in ICANN, and its very limited mission, the rationale for this extraordinary construction is hard to understand -- unless it is simply a vehicle for the creation of a global public policy infrastructure that could then be redirected to other issues. This explanation would at least explain, if not justify, the cost and complexity of creating a structure that on its face is more complex than any existing international body -- including those other organizations with Secretary-Generals. In fact, this recommendation seems to have very little to do with providing public participation in ICANN, and thus is not a credible suggestion for solving that particular problem.
5. Other Recommendations. The other NAIS recommendations range over many subjects. A few deserve mention:
-- Notwithstanding the problems with postal authentication documented by its own research, the NAIS report concludes that it is the best of several flawed options. [It does not consider at least one other possibility -- that the unavailability of unflawed options might lead one to conclude that global elections are not a viable method of gaining public participation in ICANN.] It goes on to suggest sending the PINs by postcard because it is cheaper than a letter -- of course, since the PIN is an antifraud device, sending it open to the world seems inconsistent with its purpose.
-- The NAIS Report asserts that "a conceptual pillar of ICANN's funding structure has been the notion that those groups deriving financial or other benefits from ICANN's operation should should the responsibility for supporting ICANN's activities." (p.129) This is flatly and obviously incorrect, and why the NAIS authors would put forth such a demonstrably false proposition is hard to understand. The accurate statement of ICANN funding is that those who provide name and address services to users are the conduits for ICANN funding, but all of those entities undoubtedly pass those costs on to their customers -- name registrants or resellers, and address recipients. ICANN funding has nothing to do with the commercial success or failure of any of the entities with which ICANN contracts; its funding needs are determined by its costs, and do not vary with the commercial results of any of those who serve as funding conduits. Because the NAIS authors put forth an inaccurate and confusing picture of how ICANN is funded, they obscure the real effect of their funding recommendations -- that name registrants and those who obtain goods or services from the recipients of address allocations will pay for the costs of NAIS's public participation recommendations.
In addition, NAIS's cost estimates are hopelessly understated; they take no account of the massive outreach and voter education initiatives they say are necessary to create a group of interested public electors, and they ignore completely the costs of the additional staff that ICANN would inevitably have to hire to establish and service the massive infrastructure and global election mechanism they recommend. The NAIS authors reject membership fees as "contrary to basic notions of democracy and fairness," but seem to have no trouble at all with the fairness of taxing domain name holders for the costs of creating an artificially "interested" electorate.
-- The NAIS report recommends that ICANN should "guarantee individuals and organizations around the world the protection of their individual liberties (of the sort contemplated by the UN Universal Declaration of Human Rights), their property, their expectation to be treated fairly and with due process." This includes the "right" to participate in the DNS "without distinction of any kind, such as race, color, sex, language, religion, political or other opinion, national or social origin, property, birth or other status;" to "freedom of opinion, which includes freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of national frontiers;" and "the right of all individuals and organizations worldwide to privacy." The NAIS authors apparently did not see any need to explain why these were relevant to ICANN's limited mission, or how ICANN could every hope to accomplish these lofty goals. .
-- The NAIS report recommends the creation of a global "judiciary" that "serves as a real oversight body." Apparently having a membership of all humans "interested" in ICANN is not enough protection against what the NAIS authors refer to as "majoritarian impulses," and there is a need for an analogue to the International Court of Justice to protect against such abuses. Interestingly, these advocates for public involvement reject consensus decisionmaking, the base on which ICANN was established, as probably not "meaningfully achievable for contentious policy issues." Instead, the NAIS authors would rather depend on a complex and expensive global bureaucracy -- which begs the question why they are so hostile to using an existing multinational bureaucracy instead of creating a new one.
-- The NAIS report still does not all find these solutions sufficient, and therefore recommends the creation of Board and Staff Codes of Conduct; fair administrative procedures, and an active independent review panel. The relevance of these to the issue of public participation is not obvious.
-- The NAIS report has a short section recommending reform of the DNSO, but this is clearly an afterthought, and its recommendations and rationale raise more questions than they answer. In addition, they are markedly inconsistent with other aspects of the NAIS report. While earlier criticizing the ASO and PSO as unrepresentative, here the NAIS authors concede that they have "effective internal processes" that produce "basically effective" dialogue and "substantive work products." They seem to attribute this to the fact that these bodies have evolved from pre-existing organizations. (p.139) [It is odd that they do not draw from this recognition the clue that perhaps using pre-existing organizations would be an effective way of dealing with the public participation issue generally.] The DNSO, on the other hand, is seen as dysfunctional, in part because it is in "the awkward position of attempting to gain consensus from competing 'supply' and 'demand' perspectives." (p. 139). The NAIS authors, consistent with their doubts about the consensus policy development process noted earlier, appear to forget that ICANN itself was designed from its creation to be an organization with the "awkward" goal of generating consensus from "competing" perspectives; this is a feature, not a bug.
The NAIS report suggests that it would be an improvement to separate providers from users, and to create a "Name Providers" SO and a "Consumers" SO. This is remarkably similar to the conceptual framework that led the ALSC to its recommendations (see below) that the Board should basically consist of representatives selected from each of three general categories -- developers (ASO and PSO), providers and users. To the NAIS authors, with their view that "interested individuals" must have half the Board seats, this insight leads instead to the suggestion that the DNSO's current three seats be subdivided into some for providers (in which they would include both gTLD and ccTLD registries and registrars) and users (where they are not specific in their recommendations, but seem to contemplate that NGO name holders, academic name holders and individual name holders would have essentially equal power to the business, ISP and IP constituencies). The net effect of these recommendations would be to even further reduce representation on the Board from the business-related stakeholders of ICANN, and increase that of groups such as those represented by the NAIS authors. The NAIS authors seem to be oblivious to the fact that this particular recommendation is even less likely to gain consensus support than their other recommendations.
The one point on which the NAIS report is very clear with respect to SO restructuring is that it should not be connected to what it sees as the critical need to create the proper level of public participation in ICANN. Since it would seem to make logical sense to deal with restructuring issues as a whole, both to ensure that in the end the right balance is struck and to avoid having ICANN go through a long period of contemplating its internal naval to the detriment of its main mission, this conclusion seems purely tactical -- an effort to avoid having the public participation issues dealt with in the context of the broader issue of how ICANN should best be structured to produce appropriate involvement by all its stakeholders.
6. Conclusion. The NAIS report is a policy prescription for the creation of a globally democratic organization with extensive non-governmental review and appeal mechanisms. It offers a point of view, but makes no pretense to reflect a consensus position, or a position that could ever gain consensus support. Instead, it is presented essentially as a demand -- the minimum process and procedures for ICANN to obtain "legitimacy" in the eyes of the NAIS authors. The NAIS authors never take the trouble to explain why their views should prevail over all other ICANN stakeholders, nor why a public that neither knows or cares about ICANN should have veto power over the aggregate of all other stakeholders. As such, it does not provide any useful assistance in solving ICANN's present problem, which is to find a method for appropriate public participation in ICANN that at least potentially could obtain broad support from ICANN's entire stakeholder community.
THE ALSC DRAFT REPORT
The ALSC draft report starts from a very different premise than NAIS. Instead of attempting to justify a preordained conclusion, the ALSC was instructed to do a "clean sheet" study, and to consider the full range of options. These included, for example, whether there was any need for public participation in ICANN, and if so, under what conditions; whether public elections were necessary or appropriate for some or all Board seats; and what other role(s) (if any) an At Large membership should play in ICANN. As best can be seen from its draft report, the ALSC took these instructions seriously, and considered the full range of possibilities in coming to its draft conclusions.
There are two important points of agreement between the NAIS and ALSC reports: both conclude that there is today very little interest in ICANN or the structure and performance of an At Large membership, and both agree that there is no consensus on how to structure appropriate public participation in ICANN. As the ALSC put it, "[d]espite extensive efforts, the ALSC has found it difficult to generate a high degree of interest in these issues. We have not found profound and widespread opinions favoring any one solution to the At-Large issues." (p.1) From this common ground, however, the two reports diverge dramatically. The NAIS authors assert that there is only one acceptable solution to the At Large problem, whether others find it agreeable or not. The ALSC took a different approach: "We have concluded that the main interest of the wider Internet community is in the stability and reliability of the Internet itself, and that a structure for participation and representation that is seen as creating the best possibilities for this would meet with its tacit approval." (p.1) The ALSC then concluded that "the approach with the best chance for consensus support" involved "organizing ICANN along stakeholder interest or functional lines of developers, providers and users, by defining At-Large members as 'individual domain name holders' (since they have a strong and tangible vested interest in ICANN activities, not just the Internet in general), and by providing these At-Large members with the opportunity to fully participate in ICANN and select one-third of its Board." (p. 6)
The ALSC and NAIS reports also agreed that public participation in ICANN was important, and that some form of elections of a certain number of directors by a broadbased electorate would best serve that end. But where the NAIS authors insisted that directors must be elected by a global electorate of "interested individuals," and must have half the Board seats (and thus an effective veto over all important Board decisions), the ALSC concluded that there should be a balance among directors selected by developers, providers and users on the Board, and that individual domain name holders were a satisfactory source of public input into director selection. The ALSC describes its recommendations as "an approach to Internet user participation and representation in ICANN that the ALSC believes is defensible, executable, effective and most likely to gain consensus." (p.8) This analysis now turns to the specific suggestions contained in the ALSC draft report.
1. An Electorate of Individual Domain Name Holders. The ALSC concluded that "structured involvement of individual Internet users in ICANN policy formulation and decision-making is needed . . . " (p. 11) Contrary to the NAIS authors, however, the ALSC did not "accept the notion that users' interests are somehow exclusively or even best protected by the direct election by e-mail address holders of half of ICANN's Board. Clearly this approach does not have consensus support." (p.11) The ALSC concluded that "any attempt to reach individuals on the basis of e-mail addresses [would] be administratively and financially unworkable on a global scale and fraught with potential dangers ranging form capture to outright fraud." (p.11) Its alternative -- what the ALSC termed a "viable compromise" -- was to look to individual domain name holders as the electorate for At Large directors.
Clearly this suggestion has pros and cons. Individual domain name holders are an identifiable group that has already expressed some interest in the subject area managed by ICANN. They can at least theoretically be reached in a decentralized way through registrars, assuming the latter's cooperation, and they could provide a way for the At Large to be self-funding, thus avoiding the tax recommended by the NAIS authors. On the other hand, domain name holders are largely entities, not individuals; by some estimates there are only about 6-8 million individual domain name holders. The ALSC report is not absolutely clear on this point, but it seems to assume that some individual would be designated by each entity holding domain names to vote one time on behalf of that entity. How this could be administratively managed is not clear. Using registrars to sign up those interested and collect a fee (recommended by ALSC both to ensure interest and to provide a way to self-fund the At Large) will work only if the registrars agree to take on this task; since it will cost time and money for them to do so, it is not clear they would be wiling to do so. There would remain the potential for capture and fraud, although certainly less so than in a broader, less defined electorate.
On balance, the ALSC approach seems more workable than the NAIS recommendations. While both would require outreach, this is obviously easier with those who already have a connection with ICANN than with the entire human population. If the ALSC recommendation was adopted, it might be simpler to limit voting to just the individuals (and not the entities) holding domain names; this would not be "fair" in some sense, since it would exclude, for example, single proprietorships and other small businesses that may well have no other practical way to participate in ICANN. Still, even the NAIS report agrees that the objective here is not a representative election in the governmental sense, but merely a way to obtain "appropriate" public participation in ICANN; there does not appear to be any reason to insist that any person with interest must be able to participate to accomplish that goal. Another possibility would be to simply invite all who are willing to pay a nominal fee -- just enough to ensure some level of actual interest -- to become ICANN "members" and then to select (perhaps by vote) the appropriate number of ICANN directors. Perhaps the biggest practical hurdle to the ALSC approach is obtaining the cooperation of the registrars, who might insist on being able to pass on the cost of this particular administrative cost to their customers.
2. Creating an At Large Membership. As does the NAIS report, the ALSC report recommends an outreach effort to create an informed and engaged At Large membership. The ALSC's suggested approach is to create an ALSO, with dedicated staff funded by ICANN and open to participation by all interested individuals who are wiling to pay a membership fee. The ALSC recommends 6 geographic regions, each electing a Director and with the five candidates receiving the next highest vote totals becoming the At Large Council for that region. [It should be noted that the ALSC suggests no measures to ensure geographic balance within the regional councils, which is potentially a serious flaw with this specific suggestion.] Each region would select two of its council members to serve on the ALSO council, which would be responsible for managing the ALSO.
There are two significant differences between these recommendations and those of the NAIS report: six rather than five regions, and a membership fee. The additional region would encompass India, parts of Western Asia and the Middle East; the desirability of this addition to the status quo is certainly open to debate, but given the ALSC recommendation that the ALSO elect 6 directors, it would simplify the logistics of that task. Another option would be to stay with the current five regions, and select (perhaps by blind draw among the persons who got the second highest vote totals in each region) a sixth director. The fee is seen by the ALSC as necessary for self-funding, and as a way to limit membership to those that have "demonstrated a commitment to supporting At-Large activities." (p.15) The ALSC apparently does not agree with NIAS that this is discriminatory or unfair, since it concludes that what is necessary (and workable) is informed public participation, not participation by all "interested" persons. The ALSC sees a need, as did the NAIS authors, for "local and regionally based (and perhaps eventually issue based), informed participation by individuals in ICANN," (p.16), but would leave the details of how to reach these goals to the ALSO once formed.
3. One-Third of Directors. The ALSC takes a diametrically opposite approach to that taken by NAIS. The NAIS authors conclude that the At Large must elect half the Board to have an equal say with what they view as the "vested interests" in ICANN. The ALSC rejects the notion of "dividing the Board between At-Large and all other interests . . ." (p.17) as "having an unsound logical basis . . . ." (p.18) The ALSC concludes that a functional approach "makes more sense considering ICANN's evolution and responsibilities, and it has the greatest likelihood of achieving a consensus. (p.18) Thus, it suggests that developers, providers and users should each have equal representation on the Board, in part because this "prevents any one of the constituency groups from exercising undue influence within ICANN." (p.18) This contrasts sharply with the NAIS recommendations, which insist on the At Large stakeholders having the practical ability to veto any decisions by the Board.
The ALSC draft recommendations are certainly not free from debate. Six At Large directors is three fewer than the number contemplated by ICANN's original bylaws, but it bears noting that that structure was in fact commanded by the person with decisional responsibility in the US government at the time of ICANN's original recognition as a price of that recognition, and never had consensus support in the Internet community, as has been repeatedly demonstrated by the inability to find consensus support for any implementation of that concept since ICANN's creation. Six At Large directors, selected as suggested by the ALSC, would require the creation of a new ICANN region, and that effort would undoubtedly be the subject of much debate before it was completed. Limiting the electorate for those directors to individual domain name holders does not eliminate the practical and logistical problems of such elections, but certainly reduces them to at least potentially manageable proportions. On the other hand, it seems hard to argue that six directors would be insufficient to permit public participation in ICANN, and it seems at least possible that this suggestion could garner consensus support, whereas it has clearly been demonstrated that the NAIS position (half the Board elected by an At Large membership) cannot.
4. Conclusion. The ALSC makes it clear that it has suggested a solution to the public participation issue that it believes is both workable and at least potentially capable of garnering consensus support. It also makes it clear (although its draft report preceded the NAIS report) that it finds what turned out to be the NAIS recommendations "to be unworkable and likely to engender fraud or capture." (p.19) Its recommendations have obvious (and potentially serious) flaws: the reliance on the cooperation of registrars that may or may not be forthcoming, the potential disenfranchisement from participation in ICANN of small businesses and individual proprietorships, and the potentially divisive necessity of creating new geographical region boundaries. Nevertheless, the ALSC report is clearly superior to NAIS in at least two respects: it is at least an attempt to identify a consensus solution, and it focuses on what is practical and workable, not simply what is conceptually attractive to its authors.