The biggest buzz from the Paris ICANN meeting was that the board accepted last fall’s proposal for a streamlined process to add new TLDs. A variety of articles in the mainstream press, many featuring inflammatory but poorly informed quotes (from people who probably got a phone call saying “We go to press in five minutes, what do you think about ICANN’s plan to add a million new domains?”) didn’t help. When can we expect the flood of TLDs? Don’t hold your breath.

Last fall the GNSO, the part of ICANN that concerns it self with generic (non-country) top-level domains finished a lengthy sequence of meetings and consultations leading to a report proposing a revised process for new TLDs, replacing the former process which everyone agreed didn’t work for the 2000 and 2004 rounds of new TLDs. The new process has been called simpler, but as you can see in the following flowchart (original PDF source), it’s still byzantine, and there is no chance of large numbers of new TLDs emerging from it.

The process will involve a process outlined in a set of 20 recommendations and a lot of money. First, there’s the application fee, which hasn’t been set but is unlikely to be cheap; certainly five figures, maybe six. The old fee was $50K, but you also needed full time lobbyists to move your proposal through the ICANN unprocess, something that I doubt will change. The recommendations outline a series of hurdles, both reasonable ones to ensure that applicants are able to run a TLD and the name isn’t too similar to an existing one (ruling out .CMO, presumably), along with some hopelessly fuzzy ones.

Rec 3 says:

The name “must not infringe the existing legal rights of others”, and lists some sample treaties. Presumably this is to keep people other than IBM from registering .IBM, but maybe it’s also to limit .INUIT to natives of the North American Arctic.

Rec 6 says:

The name “must not be contrary to generally accepted legal norms relating to morality and public order”, again listing some treaties which along with the usual human rights treaties include a bunch of trademark treaties, since ICANN has never been able to resist the IP lawyers.

Rec 20 says:

An application will be rejected if an expert panel determines that there is substantial opposition to it from a significant portion of the community to which the string may be explicitly or implicitly targeted.

These last two were the most controversial part in Paris. ICANN’s utter failure to deal with the .XXX proposal gives me no confidence that they’ll be able to deal with any other contentious domains, and their confident assertion that they’ve solved the problem by outsourcing the job to outside experts just confirms the problem, since outside experts can only make reasonable decisions if they have reasonable criteria to start with, which they don’t. The usual examples of domains that Rec 6 would forbid include .JIHAD and .NAZI, but nobody’s been able to tell me what might happen if a group of Holocaust museums wanted .NAZI to publish historical and educational material, sort of like the way the NAACP owns nigger.com.

Then, if there’s multiple proposals for the same name, ICANN can auction it to the highest bidder, involving yet more money. We don’t know what will happen to the auction proceeds, but ICANN has always been a believer in its need to hire more staff.

Also, although ICANN simultaneously said that they’re on track to allow non-ASCII domain names, there’s still no clear rules about who has prior rights to what domains. They say that the existing domains don’t have rights to translated versions of the names, but it’s hard to imagine them giving the Chinese name for China to anyone other than the Chinese government. Given ICANN’s history of caving in the face of lawsuits, I expect Verisign to sue if they try to assign, e.g. .RED (Spanish for network) to anyone other than them since it’s just like .NET. Given how vague rules 3, 6, and 20 are, expect endless threats and lawsuits about them, too. Despite the historic fact that most new TLDs have been egregious failures, there seems to be an endless supply of speculators who are sure that if they can just get .WEB or something, they’ll get rich, so even though there’s an appeal process (the orange boxes on the left), expect everyone who’s turned down to sue, too.

So anyway, in view of all this, I can imagine that perhaps a handful of minor names might arrive. For example, there’s .BERLIN, which a couple of persistent Germans supported by the local tourist and phone book industry have been trying to get ICANN to accept, and which seems pretty harmless, along the lines of .CAT for the area around Barcelona. But other than that, I’ll believe it when I see it.