How will ICANN measure the success of its new top-level domains program, and how many defensive registrations is too many?
These questions are now firmly on the ICANN agenda, following the publication of 37 draft recommendations for how to measure the success or otherwise of new gTLDs.
The advice of the Consumer Trust Working Group, published last night and now open for public comment, is a must-read for anyone interested in the emerging new gTLD market.
The recommendations describe myriad ways ICANN could benchmark the performance of new gTLDs three years from now, to fulfill its promise to the US Department of Commerce to study the effect of the program on consumer choice and competition.
While it’s a broad document covering a lot of bases, I’m going to be disappointingly predictable here and immediately zero in on the headline wedge issue that I think will get most tongues wagging over the coming weeks and months:
The working group decided to recommend that, as a measure of success, domain name registrations in new gTLDs should be no more than 15% defensive* three years after launch.
Defensive in this case would mean they were registered during the mandatory Sunrise period.
The idea is that consumer choice can be demonstrated by lots of registrations in new gTLDs, but that defensive registrations should not count toward that goal.
The 15%-Sunrise baseline number was chosen fairly arbitrarily – other suggestions were 20% and 12% – and is designed to spur community discussion. It’s not final.
Still it’s interesting.
It implies that if a registry has 15,000 Sunrise registrations, it needs to sell another 85,000 domains in three years to be seen as having made a successful contribution to consumer choice.
By way of an example, if it were to be retroactively applied to .xxx, the most recent gTLD to launch, ICM Registry would have to get its total registrations up to 533,000 by the end of 2014.
Is 15% too low? Too high? Is it even a useful metric? ICANN wants to know.
Whatever the ultimate number turns out to be, it’s going to be handy for plugging into spreadsheets – something opponents of new gTLDs will find very useful when they try to make the case that ICANN endorses a certain dollar value of trademark extortion.
Because many registries will also accept defensive registrations after Sunrise, two more metrics are proposed.
The group recommends that domains in new gTLDs that redirect to identical domains in legacy TLDs – strongly implying a defensive registration – should be no more than 15% after three years.
It also recommends that ICANN should carry out a survey to see how many registrants own matching second-level domains in legacy TLDs, and that this should also be lower than 15%.
I’ve only outlined three of the working group’s recommendations here. Many of the other 34 are also interesting and will be much-debated as the new gTLD program continues.
This is vitally important stuff for the future of new gTLDs, and applicants would be well advised to have a good read — to see what might be expected of them in future — before finalizing their applications.
* It should be noted that the recommendation as published confusingly reads “Post-Sunrise registrations > 15% of total registrations”, which I think is a typo. The > operator implies that non-defensive registrations only need to be over 15% of total registrations, which I’m certain is not what the working group intended to say.
(UPDATE: this typo has now been corrected).
ICANN thinks the benefits of new top-level domains will outweigh the costs, and it preparing for a scrap when it meets its Governmental Advisory Committee in Brussels next week.
In a number of briefing documents published yesterday, ICANN makes it clear that it does not think the new TLD program will create a huge economic burden on trademark holders.
Brussels is possibly the final major hoop ICANN has to jump through before its board of directors will be able to sign off the Applicant Guidebook and start accepting new TLD applications.
There are a number of areas where the GAC and ICANN disagree. Next week’s meeting is intended to identify those differences, and to try to find ways to resolve them.
What will be the damage caused by cybersquatting? How much money will big business be forced to spend on defensive registrations?
Nobody knows for sure, and none of the independent third-party economic reports commissioned by ICANN ventures anywhere near a comprehensive empirical study.
So the GAC wants another economic study done, to quantify the costs and benefits of new TLDs, and to figure out how voluminous defensive registrations is likely to be.
ICANN disagrees, saying essentially that more studies are pointless, and that demand for defensive registrations in new TLDs are likely to be low.
The Board position is that defensive registrations will increase but not in numbers projected by some trademark holders
Estimates from the Coalition Against Domain Name Abuse, CADNA, last year put the cost to brands of the new TLD program at $746 million. But ICANN isn’t buying that:
The Board also believes that economic studies do not provide evidence that new TLDs will result in demand for defensive registrations. Existing independent studies, that the Board is seeking to validate, indicate that defensive registrations occur only in the very largest, well-established registries only.
the Board is seeking (and will publish) independent corroboration that:
• trademark holders generally do not register their trademarks in all the current generic TLDs.
• Therefore, it is not expected that trademark owners, in general, register their trademarks in new gTLDs, and
• due to the expected costs to run a registry and the expected low number of defensive domain name registrations, there is no economic incentive for an applicant to obtain a TLD for the sole purpose of making money from defensive trademark registrations.
ICANN does not identify these “independent studies”, but the data points cited in the document (pdf) point to a February 2009 article published on CircleID by Paul Stahura, and a comment made on that article by Richard Tindal that cites third-party data.
The Stahura report is arguably the most comprehensive carried out on defensive registrations in existing open gTLDs, concluding that the current cost to trademark holders is very low indeed, and that the bulk of typosquatting and trademark enforcement goes on in .com.
The research suggested that each new TLD would create costs in the tens of thousands of dollars per year, across the whole universe of trademark interests. It used baseline registrar fees in its calculations, unlike the CADNA report, which used sunrise fees about a hundred times greater.
But the Stahura study is “independent” only in the respect that it was not commissioned by ICANN or carried out with its blessing or participation.
At the time it was published, Stahura was president of eNom owner Demand Media, which is expected to be a new TLD applicant. Tindal, apparently also cited in ICANN’s latest report, was senior vice president, registry, for Demand Media.
Independently validating the report’s conclusions will be important, if only to avoid accusations that ICANN is making its decisions based on the views of those who would benefit from new TLDs.
Another of ICANN’s newly published briefing documents (pdf) also address the specific trademark protection mechanisms called for in the Applicant Guidebook.
The GAC has not yet published, or provided ICANN with, its specific recommendations relating to these mechanisms (I understand that will come in the next day or two) but they are expected to call for a tightening of the rules governing the Trademark Clearinghouse and Uniform Rapid Suspension policy.
Unlike several parts of yesterday’s briefing papers, ICANN’s language when discussing these two mechanisms does not suggest to me that it is preparing to substantially compromise.
With trademarks just one of many issues under discussion, Brussels is shaping up to becoming very interesting indeed.