Lynn's sTLD Limit (3), DNS Taxonomy Bashed in Amsterdam
Date: Monday December 16 2002, @01:34PM
Topic: New gTLDs

Stuart Lynn's "action plan" for new gTLDs was the real topic of interest in Amsterdam. As Lynn himself remarked on the podium, it was new TLDs that lined 'em up at the microphone for hours, not the "Evolution and Reform" plan, for which two hours of public comment has been scheduled. (And as yours truly riposted to a burst of laughter from the audience, "that's because we've all given up on the Reform, Stuart.")

It was not just the interest in new TLDs that was notable, but the remarkably sane and informative dialogue that took place - one that made it clear that:

a) The Board was listening

b) Almost no one supported the arbitrary limitation to three sTLDs, and many favor more open gTLDs

c) Almost no one, from yours truly to Esther Dyson to Elliott Noss, supported a "taxonomized" top-level name space.

Miracle of miracles, the Board resolution dumped the limit to three, asking ICANN "in as timely a manner as is consistent with ICANN staffing and workload, ...[to solicit] proposals for a limited number of new sponsored gTLDs." It did, however, ask the GNSO to make a recommendation "on whether to structure the evolution of the generic top level namespace and, if so, how to do so."

The retreat from the Magic Number 3 came due to an outpouring of challenges from the floor. The arbitrariness of the limit was recognized quickly, and the Board began to speak of a "range of new sTLDs." Even a spokesperson for one of the rumored favorites for inclusion within the magic three, World Health Organization's .health, said at the mic that she opposed any artificial limitation.

It became generally known at the meeting that former Board member Ken Fockler had been engaged by the .travel folks and ICANN management's attempt to accommodate his lobbying may have been the driving force behind this aspect of the plan. Rumors abounded that the favored three had already been selected, and included both .travel and .health (there was no rumored consensus on the third). One commentor, Lee McKnight, noted that limiting it to three because of insider deals is not a good idea while they are under close review by the Department of Commerce, as it looks "too closed and captured." The Nokia Corporation issued a stirring call for letting the mobile industry develop its own TLD after two years of waiting.

The idea of making the TLD space an exhaustive, mutually exclusive classification scheme took it on the chin even harder. The GA's Thomas Roessler noted that interposing a centralized taxonomy on user identification went against the end-to-end principle of the Internet. Esther Dyson criticized it as anti-innovation and diversity. Tucow's Elliott Noss observed that "one thing we have learned is that we cannot predict whether names are 'useful' or have 'support.'" Frode Griesen, Milton Mueller, and several others also opposed the idea. Only the BC's Philip Shepperd, who drafted the BC position paper that Lynn was trying to put on the agenda, spoke out in favor of it.

In general, ICANN's Board seems to be waking up to the fact that managing the name space is its core business, and that adding TLDs is a critical part of that business. There was some interesting dialogue about why one might need a TLD rather than a SLD(amazing how few staff or Board seemed to understand the economic distinction between owning and renting), how users actually use domain names, and whether ICANN needs to prevent or repair failed TLD registries.

This discussion has been archived. No new comments can be posted.
Lynn's sTLD Limit (3), DNS Taxonomy Bashed in Amsterdam | Log in/Create an Account | Top | 11 comments | Search Discussion
Click this button to post a comment to this story
The options below will change how the comments display
Check box to change your default comment view
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Re: Lynn's sTLD Limit (3), DNS Taxonomy Bashed in
by fnord ( on Tuesday December 17 2002, @09:53AM (#10585)
User #2810 Info
The TLD space was originally designed to be taxonomized, either by content (.com, .edu, .int) or geopolitical entity (.us, .ie, .cc). Veri$ign broke the former, and .tv et al broke the latter, in the name of profit. One cannot put Humpty Dumpty together again, but if we're going to have more new gTLD's, and I support the 50 per year idea, there has to be some sanity employed or the vast majority of end-users (who weren't there to speak or be represented) will shun most all of them due to confusion. I know a number of reasonably tech savvy people who are still surprised to learn that there is now an .info TLD.

If there is no taxonomy one might as well hand out random strings (eg: .0d7z) for all the good it will do. We still do have a taxonomy of sorts, .name or .museum or .biz at least make some sort of semi-intuitive sense, as will .health or .travel. The issue to me is whether this should be left to ICANN or some committee appointed by them, or whether it should be left to the magical forces of the free market. Much as I appreciate the benefits of the latter, what if there is an applicant for .book and an applicant for .books and an applicant for .lit? That decision will still come down to an ICANN call and they're unlikely to allow competitors in a similar namespace due to a fear that both will fail, so there is no true free market anyway. In addition, at what level does one differentiate? Should it be at the level of .food or at the level of .chips or at the level of .fritos? If that is left entirely to the free market to sort out then we will have entries at all levels, leaving a further counter-intuitive mess for years.

I'd personally like to see someone like the ITU handling this as they have a wealth of experience and resources in similar areas. I fail to see what is so hard to understand about using the telephone book model as a reasonable starting point for a taxonomy. I've been told over the years by many who pay attention to such things that the DNS is not intended as a directory system, I always ask why not? and have never received an answer, even a poor one. Paul Mockapetris has said he intended it as a directory system and that's good enough for me. Some wisdom, even some sanity, is sorely needed.

With regards to Ken Fockler's rumored involvement, I really think ICANN needs something in its bylaws saying that former BoD members (to say nothing of sitting ones for that matter, eg: Rob Blokzijl) cannot become ICANN lobbyists for a set period of time as is common with some elected/appointed/employed government positions, as well as in some private enterprises with non-disclosure/non-competition agreements. I would like to see a similar proviso for ICANN staff. Of course I rarely see what I'd like to see from ICANN, but perhaps some successor body could... -g

[ Reply to This | Parent ]
  • 1 reply beneath your current threshold.

  • This article comes from ICANNWatch

    The URL for this story is: