Latest news of the domain name industry

Recent Posts

Here’s how Donuts wants to resolve its 158 new gTLD contention fights

Kevin Murphy, October 23, 2012, 18:30:25 (UTC), Domain Registries

Donuts is backing a private auction model designed and managed by Cramton Associates as its preferred solution for resolving its 158 new gTLD contention sets.
The proposals, spelled out by auction design expert Peter Cramton during private sessions with new gTLD applicants, caused a bit of a buzz — not all of it positive — at the ICANN meeting in Toronto last week.
But Donuts co-founder Jon Nevett told DI today that Cramton has addressed rivals’ concerns and that Donuts wants to handle as many of its contention sets as possible via private auction.
The idea is that private auctions will be faster and cheaper for applicants than the process set out by ICANN as the “last resort” method for resolving contention sets.
In the ICANN model, all of the proceeds of the auction would go to ICANN, to be distributed to worthy causes at a later date. But with a private auction, the winning bidder pays the losers.
This makes it more attractive to applicants, according to Donuts.
“The cost of losing an ICANN auction is greater than the cost of losing a private auction,” Nevett said. “If you lose an ICANN auction you get nothing, zero, you lose your asset.”
But with private auctions, “it doesn’t hurt as much to lose, so the theory is the second-place guys won’t stretch as much,” he said.
Cramton is a professor of economics at the University of Maryland. A long-time auction specialist, he’s been involved in designing processes for selling off wireless spectrum around the world.
For new gTLDs, Cramton proposes an “ascending clock” auction. At each stage, the price is increased by the auctioneer and the bidders/sellers can either commit to pay that amount or drop out.
The last man standing wins the gTLD, paying the amount that the second-highest bidder was willing to pay.
The money would be divided equally between all the losing applicants. According to Cramton, the advantage over proportional distribution is that it does not encourage applicants to over-bid, keeping costs down.
Cramton’s original plan, which left some applicants scratching their heads last week, was to run the auctions in the first quarter of 2013, before ICANN announces the results of Initial Evaluation.
That would mean that losing bidders would get a 70% refund of their ICANN application fee, which may be an attractive percentage in the case of low-value strings.
But it also means that an applicant could win an auction and later discover its application has been rejected. The other applicants would have withdrawn, so the gTLD would just disappear into the ether.
Judging by a series of videos shot last week and published on Cramton’s YouTube account, many applicants are in favor of running the auctions after IE results have been announced.
Another complaint expressed by Donuts’ competitors last week is Cramton’s “all or nothing” approach, in which Donuts’ rivals would have to commit to use the auctions for their entire portfolio of applications.
According to Nevett, that idea is no longer on the table.
“In the beginning he was discussing that it would have to be all your TLDs or none, and I think a lot of applicants told him that was unacceptable, so he changed his view,” he said.
The idea now is that the auctions would proceed on a TLD-by-TLD basis.
Given that winning bidders are giving money to their competitors, another concern is the ordering of the auctions. You don’t necessarily want to give your rivals a big wedge of cash they can use to out-bid you on the next lot.
The preferred solution here appears to be a simultaneous auction, with all the participating contention sets being resolved at the same time.
There was also a deal of suspicion in Toronto about whether Cramton would be biased towards Donuts, given that Donuts is responsible for finding Peter Cramton and introducing him to the gTLD program.
But Nevett said that Donuts has not contracted with Cramton. Peter Cramton showed up in Toronto on his own dime and has not required an up-front payment from Donuts, Nevett said.
“Every applicant has a veto on whether to participate, and it won’t happen unless every applicant wants to do it,” Nevett said. “Our incentive is to have an auction provider who is attractive to every applicant.”
“Our goal is to get as many applicants to participate in a private auction, so we need the auction to be designed in a way that is simple, fair and inclusive,” he said.
But there’s no denying that Donuts has a greater incentive than most to have a consolidated auction. By its own admission, it’s an eight-person operation without the manpower to negotiate 158 contention sets.
Cramton’s materials from last week’s Toronto sessions can be found at or here.

Tagged: , , , , ,

Comments (11)

  1. Rubens Kuhl says:

    I see some challenges to private auctions:
    1 – jurisdictional laws, as sometimes the binding terms required for such auctions might go against them.
    2 – Not getting all the members of some contention sets to go along
    3 – Contention sets with declared as community applications (will likely elect CPEP first)
    4 – Timing of other possible private auctions and/or ICANN auctions bringing back the possibility of money received from the Cramton private auction to be used in the other auctions
    I wonder if all applicants in private auctions would agree to only get their money back in (for instance) June 2014 with letters of credit, trust funds or other banking instruments that preclude withdraws but recognize ownership.

  2. Thanks but no thanks. Why would I want to enrich Donuts in their carefully constructed and pre-planned attempt to engage in profiteering at the expense of the new gTLD program and non-portfolio applicants?
    Perhaps Donuts should be worrying about passing their background screening rather than lobbying for private quasi-ponzi type of auctions focusing on exploiting the gTLD program to deepen their pockets and bringing an ROI for themselves and their investors whether they lose or win.
    Smart move but “rigging” the new gTLD program to win even if you lose is not something that should be taken lightly even if it benefits some non-portfolio applicants. Smart move by Donuts but I would never agree giving any monies to Donuts, a company that piggybacks others’ work and marks for their personal financial gain, especially if the background screening allegations are true.
    In 2009 Paul Stahura testified before the U.S House Judiciary Committee’s Subcommittee on Courts and Competition in favor of the new gTLD program to spur competition and now his company applied for 307 TLDs, a truly anti-competitive action. So much for competition and leveling the playing field. How is that in the public interest?
    Of course the Donuts approach is using the gTLD guidebook policies to their advantage to circumvent issues that may arise that are negative to them. But the guidebook is not the holy bible nor is it perfect. Donuts is not above ICANN nor bigger than the Internet. It is amusing that they do believe they have everyone fooled.
    I have to admit it was a smart and carefully planned gameplan by Stahura, Nevett and Schindler but I expect obstacles since their approach does not serve the public interest, just their financial interests over the new gTLD program. Win when you win and win when you lose. Who wouldn’t take those odds at Las Vegas?

    • Rubens Kuhl says:

      It might not be on public interest, which would be ICANN getting the money and doing something good with it, but it is on all applicants best interest to settle with a private auction.
      But as you pointed out, the contention sets need objections and backgrounds checks finished so all its members can really know who are the challengers. May be Donuts and/or Demand Media won’t make it.
      (some would add initial or extended evaluation, but ICANN is signaling that all applications will eventually get approved)
      Not going into private auction just not to give money to Donuts is somewhat idealistic and is an option for those who feel that way, but in a simple “Adam Smith” economics point of view, pursuing private auction is the best path for each player so it ends up being good for society.

  3. Sedari says:

    So there are some reasonable principles here for resolving contention.
    The first is that the contending applicants share the pool of money from the winner. That at least means that applicants can recoup some of their investment.
    The second is that it’s in the control of the applicants to manage the timing for the resolution of that contention. There is, it seems, nothing stopping any known contention set from conducting an auction today, notifying ICANN and for the losers to withdraw from the Initial Evaluation and get some money back.
    The third is that it is open to any contention set to choose their auction provider. There is nothing stopping any applicant from choosing one; suggesting the model to the others and seeing if a draw from a hat (for free!) would choose the auction provider.
    Next, the money for resolving contention does not go into the unspecified ICANN pot. Whilst I appreciate the sentiment for doing good works, to quote Bruce Tonkin from last week, we need to be really clear what ICANN doesn’t do. Distributing funds to worthy causes is one of them.
    Finally, whether one likes the auction system or not, it seems to be the way the resolution of contention is going to go. I have not yet seen any moves for contending applicants to merge, co-operate, withdraw or change tack. So, in the end, there needs to be a forcing mechanism.

  4. To anyone who was there, I am interested by any feed-back regarding .wine.
    Thank you.

  5. The evidence against Demand Media is conclusive. It is clear they will be disqualified. The question arises. Will Paul be disqualified? Or will the 107 applications that were done in partnership with Demand Media be disqualified? Thankfully ICANN will not be doing the evaluation but PWC. There rep is worth more than ICANNs desire to make program work for Donuts.
    The only reason they were pushing the auction before IE is that they believe they might be disqualified.

  6. zack says:

    Who was named in the UDRP cases? Was it Demand Media or was it Stahura or any of the other Directors or investors in the Donuts applications? If it was only Demand Media and they are not part of Donut’s application (apart from their back-end provider) then I doubt any of Donut’s applications will fail the background check.
    As I understand it the background check is only on the applying organization and key officers, directors, investors.
    That aside, at the end of the day there will only be one winner for a string…so you can follow ICANN’s process and give the auction proceeds to ICANN or you can come up with another process that allows the funds to go back to the applicants. If a private auction doesn’t work then please someone come up with another process that does instead of just complaining about it

  7. Rubens Kuhl says:

    I’ve read the Cromton Associates documents, and I agree with all but one of their ideas: making the losing profits available for bidding other TLDs if the clock for one TLD stops and a winner is declared.
    That idea shifts power to applicants that on a $/app basis would be weaker, as they will earn more $ to win other apps. I’m not saying that CA was trying to favor Donuts with this; it seems they thought of that as an added stimulus for applicants to not overbid, to not be afraid of losing.
    The effect though is as such because Donuts is the most fragmented contender of all for contested applications. Donuts might not have noticed this advantage, but it’s there. (and now they know it)
    The Cramton docs also address what happens if the process fail to get much adoption, but not what would happen many private auction providers appears and fragment the process.
    I’m not sure the idea can survive fragmentation or not. Will think more about it.

  8. Tom G says:

    It seems unlikely to me that the Donuts guys would be able to raise that kind of investment if they could not pass the threshold of the background check, or that they would even try.
    I’ll be surprised if they don’t pass the bg check.

Add Your Comment