Things are getting worse at ICANN
Today, ICANN issued a “Staff Manager's Issue Report on the need for a Predictable Procedure for Changes in the Operation of TLD Registries.” ICANN has been worried about how it justifies Board decisions to amend contracts between ICANN and the registries, and has decided to do what it can to create a process. Here are the problems with the Issue Report:
1. Scope. The report suggests that the process will cover “significant actions by TLD registries that, because of their architecture or operation, could affect the operational stability, reliability, security or global interoperability of the DNS, that registry or the Internet.” This means that anything a registry does — even steps that don't involve Registry Services, such as offering value-added services or changing management structure — could potentially be covered by the GNSO process. The paper notes cryptically that “the unsponsored registry operators have noted the need to account for contractual agreements in the development of the evaluation procedure.” That's right — at the moment, no approval by ICANN is needed or required or appropriate for most actions of a registry that don't amount to the introduction of a new Registry Service. Is ICANN suggesting that it will unilaterally re-do the contracts it has signed with registries? Doesn't it take two sides to amend a contract? How can ICANN establish an “evaluation procedure” for actions that a registry is free to take without evaluation?
A secondary scope issue is that sponsored TLDs have had “delegated” to them the right to introduce new Registry Services without any input whatsoever from ICANN. So even less scope of “evaluation” is appropriate with respect to the sTLDs. Is ICANN proposing to renege on the notion of “sponsorship” it labored so mightily (and over such a painfully long period of time) to create?
2. Getting Bogged Down. It cannot be that ICANN wants to have a process towards a procedure towards a protocol towards a premonition of every sneeze that comes from the general direction of a registry. ICANN will need a staff of hundreds — ICANN will need to become the FCC. Is this the goal? Can this possibly be the underlying motivation for this particular PDP?
3. Legitimacy. ICANN is not a regulator. No government body has delegated to it any particular authority. ICANN is asserting, however, an “APA” model of ICANN's legitimacy — that ICANN's Board is a group of wise individuals who can be relied on to listen carefully to comments and make the best central decisions on behalf of the global Internet community.
The problem with the “Administrative Procedure Act” model is that the ability of the Board to listen does not, by itself, give ICANN the right to make decisions or be deferred to by governments, for a very simple reason: No one gave ICANN the power to make rules in the first place. (Indeed, a delegation of regulatory power from DOC to ICANN (which is a private party) would violate the nondelegation doctrine and raise substantial due process concerns.) While it is perfectly true that the Board of Directors of ICANN has the responsibility to act on behalf of ICANN as a corporation, and it is also true that ICANN has established this elaborate “policy development process” in order to gather comments on noticed rules from interested people, the idea of APA-like notice and comment rulemaking depends on a delegation of authority from a responsible body, including clear standards that limit the scope of agency discretion.
Here, there appear to be absolutely no limits on ICANN's discretion to block any particular action desired by a registry. And no one gave ICANN the power to tell any registry what to do — unless that registry has agreed to do what ICANN asks. Finally, it is very likely that ICANN's actions will be effectively unreviewable. This is therefore an illegitimate assertion of power on ICANN's part, and there is no reason for any registry to accede to it.
The source of ICANN's legitimacy is, and can only be, consensus — not ad hoc, anti-competitive reactions to the requests of special interest groups.
The ICANN consensus contract asks each potential participant whether they will agree contractually to implement and abide by a future rule, sight unseen, provided that most people support it and those parties substantially affected by the policy do not vigorously oppose it (or their objection is unreasonable).
This contract supports ICANN's legitimacy (and helps in understanding ICANN) because it provides a demonstration that each participant in the ICANN regime has affirmatively agreed to ICANN's jurisdiction for the limited purpose of making global rules with which most affected participants agree to go along. It is intentionally designed to produce only those rules that most people agree should be global — and very few rules will fall into this category. Everything that is not the subject of a global consensus agreement will be left to local decision-making. Everything that is not prohibited should be permitted.
More on gaming conference
There is a great thread over at terranova.blogs.com about the conference. And I am very proud indeed of this mention from Richard Bartle:
“Fortunately, the lawyers at the conference did seem to have the “right” opinion (from my point of view) even though most were new to the field and few were actual games-players. I began to take heart during a talk by Yochai Benkler of Yale Law School, who gave a very good overview of the issues (if not any actual solutions). The turning point for me, though, was the talk by Susan Crawford of NYLS. In it, she discussed the notion of identity in virtual worlds, and asked whether people need a “law of identity” to protect these second selves from organisations (eg. VW companies) who can obliterate them at will. She came out against, which I would have stood up and cheered at if it weren't for my British reserve.”