Dirk Krischenowski heads up dotBERLIN, which seeks to get a full top-level domain assigned to the German capital, arguing that Berlin has more Web sites than 150 of the two-hundred-odd country-code TLDs, and that other cities should follow suit as a branding measure.
Three US cities have proposals on the table: New York (.nyc), Las Vegas (.vegas), and San Francisco (.sfo). (Krischenowski maintains a list here.) I’m not quite sure how well this would work once you get into the smaller cities, and Krischenowski cautions that it might be too large an undertaking for them:
The application process for a city top-level domain is not a matter of course. Applicants have to meet a variety of technical, operational, economic and political criteria and must demonstrate that their suggestion has the support of the community concerned. Support or at least a statement of non-objection of the respective city government, as well as that of the relevant national government, is also mandatory for an application. In many cases the high cost of an application (around US$ 500,000 to US$ 1 million) and the annual operational costs for the top-level domain (around the same) means that applications for communities with a population of less than half a million people are scarcely economically feasible.
It seems clear that “destination cities” will find the most value in this technique, and that smaller venues and offerings within those cities will likely benefit the most, since they’ll automatically be afforded the same level of association that accrues to the bigger attractions.
There is, I assume, a limit beyond which the current TLD structure cannot be stretched, though I also assume that technological advances will push this limit farther outward; we now have, for instance, domains named in Arabic script rather than Latinate, and other character sets are on the way. Which could make matters interesting for, say, dot-Bangkok.
(Via the Urbanophile.)