ICANN/DNSO
DNSO Mailling lists archives

[ga]


<<< Chronological Index >>>    <<< Thread Index >>>

RE: [ga] VeriSign May Ditch Domain Deal


At 08:55 AM 5/18/2001 -0700, Mike Roberts wrote:
>At 21:03 +0000 5/17/01, Roberto Gaetano wrote:
>><snip>
>>Outside of the limited timeframe that was intended for this situation, it 
>>will constitute a bias.
>>Seen in the long term, it will give the company that owns the Registry 
>>*and* a Registrar a definitive competitive advantage. And this simply 
>>because it gives the possibility to NSI/Verisign to plan in advance 
>>common strategies between the R-y and the R-ar parts, to make full use of 
>>synergies, to share know-how, and so on.
>>The competitive advantage of NSI-the-Registrar over the other testbed 
>>Registrars was not in the possibility of NSI-the-Registry unfairly 
>>blocking other Registrars or unfairly privilege NSI-the-Registrar (about 
>>which I will comment below), but in the fact that NSI Registrar had 
>>already the knowledge of the environment, operations and protocols that 
>>would have been put in place, knowledge that the others did not have.
>>This is a situation that cannot continue in the future. Technological 
>>change, commercial/technical solutions, and what else, can be put in 
>>place by Verisign-the-Registry after consultation (or at least full 
>>awareness) of Verisign-the-Registrar, and here lies the competitive 
>>advantage. This is why vertical integration has been a no-no since the 
>>early days (as Director Kraaijenbrink put well in MdR).
>
>Roberto -  There are many TLDs with no separation between registry and 
>registrar(s). There would be a great uproar around the world if ICANN even 
>attempted to adopt a consensus policy that TLD registries could not 
>include the registrar function.
>
>So a black and white rule on separation is very unlikely to win 
>approval.  If ICANN were to have a policy on registry-registrar 
>separation, then the policy would have to have some elements of 
>discrimination in it to deal with the circumstances when the community 
>believes that lack of separation is damaging to other registrars and 
>registry operators and to registrants.
>
>As everyone knows from the length and complexity of the original 1999 
>agreements, and of the proposed 2001 agreements, there is nothing simple 
>about dealing with issues related to market dominance.  It's even harder 
>to develop registry agreement language that would be acceptable to the 
>operators and to the community at large.  Beyond that, such language would 
>very likely run into national laws dealing with market dominance, as it 
>has in this case, where both the competition directorate of the EC and the 
>US DOJ have been taking a close look at the proposed new agreements.
>
>So, bottom line, ICANN as a private sector consensus body should stay out 
>of trying to either make or interpret competition law. Participants in the 
>ICANN process can (and have) avail themselves of recourse to national 
>competition bodies where needed.


Mike --

         It's great to see you on ga, but I'm quite confused by 
this.  Roberto wasn't suggested that ICANN should "make or interpret 
competition law."  Nor was he suggesting that ICANN promulgate a 
one-size-fits-all policy on registry-registrar separation, to apply from 
Andorra to Zanzibar.  Rather, he was urging that ICANN, in drafting and 
entering into agreements, should be mindful of the consequences of its 
actions for free and fair competition -- and that the ICANN/Verisign "plan 
B" is bad for competition.  Now, it's certainly possible to argue that 
ICANN should not pay attention to such concerns, and should attend only to 
purely technical matters, but in fact ICANN, in dealing with the gTLDs, has 
always concerned itself quite strongly with competition policy.  Here's a 
quote, for example, from ICANN's first status report to the Department of 
Commerce:

>[T]he most critical immediate challenge facing ICANN and the DOC remains 
>the creation of a fully competitive environment for the registration of 
>names in the global Top Level Domains -- in particular, .com., .net, and 
>.org. The transition from monopoly to competition in these domains is 
>necessary for the long-term success of the privatization approach endorsed 
>by the White Paper . . . . Obviously, full and fair competition requires 
>that all have the same opportunities, and to the extent that there are 
>consumer protection or other requirements, that all meet them equally. 
>Thus, it is critical to accomplishing the White Paper objective of 
>maximizing competition that (1) NSI's registry and registrar functions be 
>fully separated, so that NSI as a registrar does not have any structural 
>advantage over its registrar competitors; (2) NSI accept community 
>consensus policies relating to registrars, as reflected in ICANN's 
>accreditation standards; and (3) the relationship between NSI as registry 
>and all registrars does not allow NSI to impair or adversely affect the 
>development of competition because of its continuing monopoly position as 
>registry operator.

         (You're familiar with this document; I assume you wrote it.)  It's 
one thing to argue that the new Verisign contract is good because it's 
consistent with these concerns -- but your message above seems to suggest 
that ICANN shouldn't be troubling itself with these concerns in the first 
place.  Is that right?  Or is your point just that ICANN should not address 
competition issues in the *ccTLD* context?  (If so, I agree with you there.)

Jon


Jonathan Weinberg
weinberg@msen.com

--
This message was passed to you via the ga@dnso.org list.
Send mail to majordomo@dnso.org to unsubscribe
("unsubscribe ga" in the body of the message).
Archives at http://www.dnso.org/archives.html



<<< Chronological Index >>>    <<< Thread Index >>>