The proposed changes to the ICANN bylaws are ill-advised in a variety of respects. These brief comments speak first to the central concerns raised by the proposal for an "At-Large Council" and indirect voting for the At-Large members of the ICANN Board. Next, they address other issues implicated by that proposal. Finally, they address a provision in the proposed bylaws intended to affect the relationship between the DNSO Names Council and General Assembly.
1. The proposed indirect election scheme for At-Large Board members is insufficient.
The notion that ICANN's membership should not elect Board members, but instead should elect members of an "At-Large Council" whose sole function is to select Board members in a manner insulated from the public, was floated by ICANN staff two weeks before the Santiago meeting. ICANN staff released two documents: The first, titled "Staff Report: ICANN At Large Membership," appeared to assume that the membership would elect Board members directly, and posited an At-Large Council that would serve other functions. The second, titled "Analysis: Statutory Members versus Nonstatutory Members for the ICANN At Large Membership," suggested towards the end of the document that the ICANN should "consider" the alternative of indirect elections as a means of avoiding the requirements of California nonprofit corporations law. This suggestion did not excite significant public attention. When, in Santiago, therefore, the Board adopted a resolution that "the At-Large Directors should be selected by an At-Large Council" rather than by the membership, there had been no meaningful opportunity for public comment. The first real opportunity for comment is now, in connection with the draft bylaws changes that ICANN staff have proposed.
The At-Large voting scheme proposed in the bylaws is distressing. The essence of the scheme is to distance the actual choice of directors from the voting membership, leaving users without meaningful voice. It tosses away the possibility that ICANN could gain legitimacy and buy-in by giving individual users a stake in its structure and processes. Few users will consider it meaningful representation to have the right to participate in the election of people who will then themselves participate in an election of Board members; few users will believe that this structure gives them a voice in the process. I would be surprised if many thought this sole privilege worth the price of ICANN membership fees. Indeed, it is hardly clear why anybody would consider it worthwhile to *serve* on the At-Large Council.
Some suggest that ICANN needs such a mechanism because the membership in a direct election would elect kooks and loons. This, I think, is entirely unwarranted (and if it were true, then instituting an indirect-election system probably wouldn't help much). The closest thing we have had to direct voting so far has been the balloting process leading to the nomination, by the DNSO General Assembly, or DNSO candidates for the ICANN Board. Although the number of persons who participated was small, the candidates who got the largest number of votes in that process were highly serious and capable individuals (and, indeed, were more technically proficient than the candidates the Names Council ultimately selected).
ICANN staff have suggested that direct voting would carry with it an impermissible risk of derivative lawsuits. Professor Froomkin has explained, in his comments, the degree to which this concern is overblown. In any event, I don't think any sensible business would take as its primary consideration in structuring its organization and operations the desire to avoid frivolous suits -- that's ultimately a much less important concern than the other representational issues that the decision implicates.
Finally, ICANN staff have raised the argument that users would be better served by an At-Large Council with the power to act as an effective voice for users before the ICANN Board than they would be served by having a mere direct vote for Board members. It's true that the indirect election proposal would seem more attractive and coherent if it came coupled with other meaningful powers for the membership council — if it *in fact* created an entity within ICANN that could act as an advocate, before the Board, on behalf of the general membership. Such a council might have two powers. The first would be to initiate policy proposals. This would be valuable because it would allow the membership to put on the agenda proposals that had broad support, but were nonetheless bottled up in the relevant SO Council because of process failures in that body. The second would be to review policy proposals initiated by the SOs. This would be valuable because it would give the membership a focused voice against ill-considered proposals reflecting the narrow or skewed views of a particular SO.
The fact is, though, that the bylaws changes now before the Board don't incorporate any such powers for the At-Large Council. The mere existence of the council will do nothing to create an advocacy mechanism for At Large interests, since the council will have no functions other than meeting periodically to select Board members. No member of the Internet public will accept the creation of the At-Large Council on the theory that it will be an effective advocacy voice, unless the Board enacts provisions that allow the Council to fulfill that role. Nor is it sufficient for the Board members to promise to consider such changes at some point in the future. Rather, ICANN should define the powers of the At-Large Council at the same time as it defines its structure. It's no way to run a railroad to enact the current bylaws proposal (at best incomplete, at worst incoherent) with the promise of revisiting these concerns as an afterthought, sometime later on.
In short, if the creation of the At-Large Council is to be justified on the ground that users would be better served by having the Council as a voice for their interests within ICANN, then ICANN should act *now* to give the Council the necessary powers. That means pulling the current bylaws proposals off the table, and rewriting them so that the At-Large Council has the same powers of policy initiation and review as do the SOs.
2. The proposed membership provisions are problematic in other ways.
The bylaws have several problematic features unrelated to the indirect voting scheme. Art. II, sec. 3 restricts members' communications with each other using the registration list to "such ways and such circumstances as the Board determines are appropriate and desirable." To the extent that ICANN staff (entirely reasonably) wishes to avoid spam using the registration list, it should craft a more narrowly tailored prohibition. This blanket assertion of control over the members' speech activities, in the Board's unfettered discretion, is far too broad and carries with it the potential for abuse. Similarly, the provision in Art. II, sec. 4(g) that the Board or Council can terminate a member for "conduct prejudicial to the Corporation's purposes" seems the sort of tool beloved of despots, and wholly unnecessary for ICANN. If it responds to a specific concern, it should be drafted narrowly to target that concern. Art. II, sec, 2, in allowing the Board to establish a membership fee, should provide explicitly that the amount of the fee cannot exceed the sum necessary for fair recovery of costs associated with the membership structure.
Art. II, sec. 6, appears to incorporate a simple (though serious) drafting error. It carefully lays out the consequences if two-thirds of the Board members determine that the At-Large selection process was fair and effective, and the consequences if two-thirds of the Board members determine that the At-Large selection process was *not* fair and effective. But suppose there is no two-thirds supermajority in either direction?
3. The proposed change respecting the power of the DNSO General Assembly should be reconsidered.
The proposed bylaws contain one change apparently intended to affect the relationship, within the DNSO, of the Names Council and General Assembly. That relationship, currently, is somewhat one-sided: All policy authority is in the hands of the Names Council (and this, the constituencies that have gained official recognition); almost none is in the hands of the Genreral Assembly.
Recently, in connection with the election of ICANN Board members from the DNSO, participants' attention became focused on Art. VI-B sec. 4(d), which provides that "[t]he GA shall nominate" candidates for the ICANN Board. Some General Assembly members urged that the GA *as a body* should play a role in the nomination of candidates. As Roberto Gaetano put it
"Nominations . . . should also not be a purely formal (and trivial) process. In other words, in order for the GA to have a *real* role in the nominations, the output of the process should be a shortlist of candidates that *the GA* nominates and supports, not a raw list of candidates that "10 out of I-don't-know-how-many-people" nominate and support. . . . This is the added value that makes the list of candidates a "qualified list" - this is the added value that makes the nominated people "nominated by the GA". . . . This procedure will "give a role to the GA". The current procedure does not." <http://www.dnso.org/clubpublic/ga/Archives/msg01077.html>.
ICANN staff have now proposed to amend Art. IV-B sec. 4(d), so that — rather than providing that "[t]he GA shall nominate" candidates, it provides instead that "[p]articipants in the GA shall nominate" candidates. The apparent purpose of the change is to remove any basis in the bylaws for suggestions like Gaetano's, and to cement in the bylaws the proposition that all institutional authority in the DNSO rests with the Names Council.
It is by no means clear, though, that this change would be a good one. In my view, the DNSO would benefit from a process in which the GA had an affirmative role to play in screening candidates for the ICANN Board, subject to later election by the Names Council. There surely has been no adequate public ventilation of ICANN staff's reasons, if any, for believing that to be a bad result as a policy matter. This change, buried deep in a raft of proposed bylaws changes and largely unremarked, should not go through absent that public discussion.