* The economic impact study promised in 2006 [needs to be] released and evaluated.
This was the principal reason advanced by the NTIA, but it rests on a faulty premise. The NTIA assumes that the primary purpose of introducing new gTLDs is to compete with .COM and the other incumbents. They believe -- as do I -- that new gTLDs will not alter the market power exercised by the incumbents in the TLDs they operate. Based on this assumption, they then make the leap that ICANN may not need to introduce new gTLDs at all.
I read it as a little more sophisticated than that; they have the not unreasonable concern that new TLDs may exist mostly to shake down existing registrants who'd want defensive registrations in new domains.
As I have always seen it, new gTLDs will serve new communities, and in some cases serve those communities with their own languages, not served by the current suite of gTLDs and ccTLDs.
I know that those are the standard examples, but the more I think about it, the less sense it makes. The idea that there are tiny language groups hanging around saying "oh, if only we had a TLD then we would do all sorts of Internet stuff" is rather implausible. They need software that works in their language, perhaps they need people to put their literature online, or to make libraries of existing material, or to host community mail systems and web sites, but what they do not need is a $100,000 vanity TLD to suck up their time and attention. If we look at the new TLDs to date, they fall into three general categories: generic clones like BIZ and INFO, communities of interest like MUSEUM, TRAVEL, and COOP, and technical hacks like TEL, the rejected MAIL, and maybe NAME and POST. If we look at the communities of interest, they've consistently failed to attract membership from their communities. The linguistic CAT is sort of successful but not much of a model since Catalonia is unusually rich, wired, on good terms with their national government, and had an enthusiastic advocate on the ICANN board. The community MOBI is somewhat successful, due to a combination of heavy financial backing and a useful one-off technical hack (web sites that work on your phone.) Other than that, they're all failures. Even if we add in the ccTLDs that have tried to turn themselves into community domains, they've all failed, too. How many people from Los Angeles have a .LA domain? How many doctors have .MD? Given the price and the hassle involved, I'd expect mostly to see corporate vanity domains like .IBM and various attempts to make money from communities. All of the latter will fail, and in desperation they'll do all sorts of sleazy things, the way that .TRAVEL is going all squat all the time. This tells me that the economic analysis and compliance are both important to minimize the damage.
We can make contractual and compliance issues a focus of our work without delaying the introduction of new gTLDs.
Experience shows that ICANN is phenomenally distractable. If they set up new gTLDs without compliance in place, it'll never happen. It's taken the better part of a decade to sort of do compliance for the rules they have now. Why would they be any faster in the future? R's, John