Like a clinging, overbearing parent, the U.S. Commerce Department just can’t seem to let go of ICANN. Yesterday Lawrence Strickling, the Assistant Secretary in charge of the NTIA, sent a stern letter to ICANN’s CEO in an attempt to dictate the course of domain name policy.

Strickling called for yet another delay in the implementation of new top level domains. Why? Incredibly, the Commerce Department is claiming that the issue hasn’t been studied enough! Strickling thus ignores ten years of research, deliberation and debate both inside and outside ICANN – some of it commissioned by the Commerce Department itself. It ignores the fact that the current policy process began in 2006 and made a decision to add new TLDs long ago. It asks ICANN to perform a “comprehensive economic analysis” comparing the costs and benefits of new top level domains – as if we were in the earliest stages of policymaking. The NTIA’s claim is so patently absurd and so late in the game that it can almost certainly be written off as a political gesture by the Commerce Department to show the trademark and law enforcement interests who are hostile to new TLDs that it is supporting them.

There is a huge disconnect at work here. From ICANN’s perspective, we are in the end game of a (sort of) bottom up policy process that has been going on since 2006, a process of multiple rounds of comment, studies, revisions and consultations. The process has consumed enormous amounts of time and effort. Certainly there are still legitimate grounds for debate about a few remaining implementation details (such as the censorship issue), but to ask the basic question “do we want new TLDs at all?” at this stage is pretty ridiculous. We’ve been there and done that.

OK, OK, I admit that, so far, no systematic study addresses whether new gTLDs cause cancer.

Let’s pretend that this is a sincere and good faith request from Mr. Strickling. For his benefit, we recommend some homework, a little bedtime reading to bring him and the rest of the NTIA up to speed on the ongoing debate over the economic implications of new top level domains. The following list includes commissioned consultancy reports, reports of ICANN working groups that dealt with economic issues, and reports commissioned by the US government itself. The list below was assembled hurriedly in about an hour in my office; much more could be found.

The list

In 1998 the Commerce Department asked staff economists at the Federal Trade Commission to provide advice about introducing competition into domain name registration; the FTC responded in March 1998. http://www.ftc.gov/be/v980005.shtm

An expert panel that included distinguished economists, trademark lawyers, policy experts, information scientists and technical experts spent four years putting together the study “Signposts in Cyberspace”(authorized by Congress in 1998, funded and commissioned by the Commerce Department and National Science Foundation in 2001, executed by the National Research Council, completed in March 2005) (I served on this panel)

Survey of the usage of the .biz top level domain, Harvard Berkman Center

The August 2004 Summit Strategies report analyzing legal, policy and economic results of the new TLDs created in 2000

Revisiting Vertical separation of Registrars and registries, Charles Rivers Associates International, October 2008.

Preliminary Report of Dennis Carlton Regarding the Impact of New gTLDs on Consumer Welfare (March 2009)

Preliminary Analysis of Dennis Carlton Regarding Price Caps for New gTLD Internet Registries
(March 2009)

Michael Kende’s critique of Carleton report, commissioned by AT&T.

(Final) Report of Dennis Carlton Regarding ICANN’s Proposed Mechanism for Introducing new gTLDS (June 2009)

A list of ICANN workshops held on vertical integration of registries and registrars in the context of new TLDs

The Salop and Wright study Registry-Registrar Separation: Vertical Integration Options (January 28, 2010)

WIPO reports and discussions regarding new gTLDs


An Economic Framework for the Analysis of the Expansion of Generic Top-Level Domain Names
, by Katz, Rosston and Sullivan (June 2010)

To anyone who actually makes their way through this material, it will be obvious that the issue has been debated exhaustively. That is not to say that these studies are masterpieces, but simply that no new consultancy study commissioned by ICANN is likely to have a major impact – either on our understanding of the issue or on the level of political support enjoyed by any option.

6 thoughts on “Why ICANN should ignore the NTIA’s letter

  1. I’ve waded through a significant amount of the material including the reports from Denis Carlton which are quite frankly embarrassingly poor. I’m surprised you even cite them to substantiate any position. Even the latest reports look at very narrow “selected” subset of very obvious problems with the proposed GNSO new gTLD implementation, problems I might add that have been there since first draft applicant guide book.
    For example:
    What happens if Microsoft secures .search?
    How does Google or any startup search provider for that matter feel about it?
    Microsoft may be happy to allow Google to register google.search if they can point video.search and news.search to Bing.
    If there is a wholesale migration to the right of the dot then yes this matters because once users come to perceive entities to the right of the dot as superior ICANN has managed to create a series of private monopolies in perpetuity in every vertical in the world.
    I really fail to see how granting the most economically advantaged private corporations such implicit DNS branding advantage can ever be in the pubic interest? I can see the [would be] contracted parties interest and I can see ICANNs interest; but where is the public interest required under the AOC?
    The sad thing is there are so more equitable ways ICANN could have approached the introduction of new gTLDs and what really surprises me Milton is why you are now not condemning ICANN for what appears to be their current contempt of the community.
    You shouldn’t let your prior dealings cloud your judgment on the NTIA because this letter looks like a much needed attempt to curb some very bad decisions, which as they stand, are almost certainly not in the public interest.

  2. FYI, GPM group, which posted this article, is a domain reselling business with hundreds (thousands?) of second-level domains that it is trying to sell. The value of their “stock” of domains would probably go down if we continue to maintain artificial scarcity by preventing new TLDs.

  3. New gTLDs are unlikely to have significant impact on existing domain name prices in the secondary market. Why? – Primarily because domain names are not substitutable.
    Further, there are only small percentages of risk takers; most people prefer to follow what others, whom they feel are successful, have done. The more new gTLDs ICANN releases the more people will play safe and go with longer names in existing dominant TLDs.
    Each new gTLD introduced makes it harder and harder to gain the critical mass needed for new gTLDs to survive into the long term, because the task of successfully branding new gTLDs is likely to prove gargantuan. .com isn’t sold or advertised by VeriSign. It is sold and branded implicitly by virtually every major corporation using it day in day out across all their communications. It’s that simple.
    The real competition to supply the back end registry services will almost certainly remain with a handful of existing registry services companies. The distinction between registries and registry brands in important because .net .name and .com do not introduce competition to registry services as they are all operated by VeriSign. Additional new gTLDs are unlikely to be a good way of providing competition at registry level.
    The big problem though is the damage the current GNSO plan for new gTLDs will do to one of the most successful systems in the history of mankind.
    Many of the problems stem from ICANN’s insistence for a one size fits all approach when the issues concerning different types of new gTLDs are clearly miles apart.
    This was one of the reasons why the VI group was unable to reach consensus. If ICANN had allowed categories then my guess is the VI group would have much more easily been able to reach consensus with differing models for different categories.
    The one size fits all approach is ridiculously constraining for some categories and will almost certainly lead to weak positions where it is impossible to enforce contract compliance for others. Quite frankly its madness and ICANN should have been brought to task on this a long time ago.
    As I indicated in my last post ICANN has behaved very badly with the Economic studies and only studied a subset of the issues.
    There is a huge difference between awarding generic gTLDs with companies like .info to Afilias & .com to VeriSign whose primary function is registry services and to awarding the generic gTLDs to a single entity whose business is looking to branding advantage within the term.
    Allowing a [would be] contracted party to leverage DNS through its relationship with ICANN to compete against all their competitors [who are forced to use the second level] in the same vertical is indefensible. Trademark Law doesn’t allow it nor should ICANN.

  4. Milton,
    You forgot… the OECD study on gTLDs from circa 2003 or 2004…and oh yeah the best ever gTLD study IMHO ; ) – Milton Mueller and Lee W. McKnight, “The Post-.COM Internet,” Telecommunications Policy, August/September 2004, Vol. 28, Issues 7-8, pp. 487-502.
    Lee McKnight

  5. ICANN nor Doc/Ntia have listened to the user internet community vis new TLD's. As a result there seems to be more confusion accordingly. Too many new TLD's if such is possible might destabilize the marketplace for existing TLD's or may also make IT security much more difficult to accomplish or enforce adaquately.

  6. > may also make IT security much more difficult to accomplish or enforce adaquately.
    I note the careful use of “may” – care to offer any empirical evidence that would support these claims?

Comments are closed.