How many new gTLDs will drop out?

Stéphane Van Gelder, Registry Relations and Strategy Director for NetNames reports on the latest developments in Internet governance

On June 13, 2012, ICANN revealed it had received 1930 new gTLD requests.

Since then, that number has been going down. ICANN has logged 13 withdrawal requests, 6 of which have already been approved: AND, ARE, EST, CHATR, KSB and CIALIS. The remaining 7 are still being processed.

So it's only 1,917 applications now. And the potential for further attrition is there: some applications are bound to be pushed aside during the evaluation process. Also, governments will be making their determinations on which TLDs they feel should not be approved. Anyone, including ICANN's own independent objector, may also raise alarms at some proposals. Even applicants themselves may opt out as they either run out of the money, patience or strength needed to see the process through.

So how many TLDs can we really expect to see when the program crosses its own finish line some time next year?

Getting your money back

Examining why some have already been withdrawn might provide answers. All 6 listed above were uncontested, i.e. only one entity applied for them. AND, ARE and EST were all made by Google's alter ego Charleston Road Registry LLC. Their withdrawal was clearly dictated by a close proximity to geographic names (Andorra or Estonia for example) which would have precluded them from being approved.

No official reasons have been given for CHATR (requested by Rogers Communications Partnership), KSB (KSB Aktiengesellschaft) and CIALIS (Cialis Trademark owner Eli Lilly). But the way the new gTLD program's refund scenario is structured, applicants with second thoughts are much better off giving them substance sooner rather than later.

The scheme puts a strong premium on early withdrawals. Applicants dropping out within 21 calendar days of any GAC (governmental) Early Warnings (official communication from the GAC that it has issues with a proposed TLD) recoup USD 148,000 of their USD 185,000 application fees.

Those dropping out after GAC warnings but before ICANN publishes the results of its initial evaluation (currently expected circa June 2013) get USD 130,000 back. Waiting until after ICANN has posted evaluation results to withdraw means only getting USD 65,000 back.

The last opportunity to bow out and still get something back is if an applicant has to go through extended evaluation, dispute resolution or string contention (i.e. someone else has requested the same TLD). Should those procedures leave the applicant with the impression that getting his TLD will prove very difficult, cutting his losses means getting back 20% of the application fee (USD 37,000).

Predicting the dropouts

This refund timetable can be read as a set of potential withdrawal trigger events. DotBerlin promoter Dirk Krischenowski has obviously been looking at this closely and at last week's Domain Forum in Sofia, Bulgaria, he gave his analysis on how many gTLD applications might not make it.

Going through the initial 1930, Krischenowski worked out that 521 would drop out because they are in direct contention sets and a further 80 would be in indirect contention with another string (strings that are similar enough to risk causing Internet user confusion, but not identical).

He also suggests further drop out numbers:

  • 40 due to successful objections;

  • 20 geographic applications missing the required governmental support;

  • 10 due to clashes with the country code ISO 3166 list (e.g. Google's AND or EST applications);

  • 15 failing extended evaluation;

  • 24 applicants going bankrupt during the evaluation process and,

  • 20 being blocked by GAC advice, i.e. governments saying they would request the ICANN Board not approve them.

Krischenowski's analysis may just be guesswork. But he has been heavily involved in the new gTLD program from day one and has spent years building an understanding of what motivates applicants and what constitutes likely new gTLD success or failure.

If he's right, more than 700 applicants may end up not seeing the program through. This would leave less than 1,200 actual new TLDs to be added to the Internet. Still a true paradigm shift considering there are only 22 active gTLDs at present, but a lot less than the June 13, 2012 "reveal day" numbers might have suggested.


Written by Stéphane Van Gelder, Registry Relations and Strategy Director for NetNames

21 November 2012