ICANN ICANN Email List Archives

[gnso-vi-feb10]


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

Re: [gnso-vi-feb10] Competition authorities

  • To: Gnso-vi-feb10@xxxxxxxxx
  • Subject: Re: [gnso-vi-feb10] Competition authorities
  • From: Richard Tindal <richardtindal@xxxxxx>
  • Date: Mon, 26 Apr 2010 17:36:05 -0400

In today's email I was talking about the actual registry, not the back-end 
provider.

As with the JN2 proposal,  I think the only time a back-end provider should be 
treated like the registry itself is when the back-end provider has control over 
policies and/or pricing.

RT


On Apr 26, 2010, at 5:28 PM, Eric Brunner-Williams wrote:

> On 4/26/10 4:23 PM, Richard Tindal wrote:
>> 
>> I think this is a scenario JN2 are trying to address in their proposal.
>> 
>> Under the CORE, Afilias and PIR proposals,  a large domain reseller,  let's 
>> say Yahoo,  could become the registry for .WEB and still offer  .WEB names 
>> to consumers.   Yahoo would simply become  a reseller for WEB,  buying names 
>> from an unaffiliated registrar at a fraction above the registry price.   
>> This would give Yahoo the effective market presence of a registrar, even 
>> though they were only a reseller.     
> 
> I confess I missed this in our last off-list Q&A.
> 
> RT: if WEB LLC (not owned by eNom) is the registry for .WEB can eNOM
> be the back-end registry provider?
> 
> EB: Yes.
> 
> (here Yahoo is the hypothetical back-end registry provider, rather
> than eNom)
> 
> EB: Try .cat (.web), PuntCat (WEB LLC) and CORE-Registrar (eNom).
> 
> RT: If so,  can eNom be accredited and sell WEB names?
> 
> EB: No. See .cat (.web), PuntCat (WEB LLC) and CORE-Registrar (eNom).
> 
> EB: CORE could sell .cat, but we don't because we can see where the
> conflicts could arise.
> 
> (here, in Yahoo having a $6 true cost advantage over all other parties
> engaging ultimately in registrations of .WEB)
> 
> So I suggest there is a slip of the pen, at least for the CORE model
> being one which allows the nickle price of self-dealing. I'll let
> Brian and Kathy examine their proposals to see if they think they've
> caught the nickle self-deal. Alternatively, show me how to construct
> the nickle exploit under CORE's proposal, as that will be a surprise.
> 
>> For example,  if the registry price was $6.00 Yahoo could probably buy names 
>> from an unaffiliated  registrar for $6.05.    Even though Yahoo the reseller 
>> paid $6.05 per name,  $6.00 of this flowed back to Yahoo the registry,  and 
>> so Yahoo would have the presence of a registrar for an incremental cost of 
>> only $0.05 per name.
> 
> The example has fundamental value, not just showing how to set up the
> nickle exploit. The true cost of domains in bulk is pennies over the
> wholesale price. Of course, this is only of interest to parties that
> traffic in domains in bulk.
> 
> Eric
> 
>> The JN2 position is that Yahoo could create the same potential harms as a 
>> .WEB reseller they could create as a .WEB registrar,  hence JN2 seek to 
>> treat these affiliated resellers like affiliated registrars for the first 18 
>> months of TLD operation.
>> 
>> RT
>> 
>> 
>> On Apr 26, 2010, at 10:28 AM, Antony Van Couvering wrote:
>> 
>>> 
>>> Yes, in general I think this is the out -- become a reseller of a 
>>> registrar, which is not a registrar, and go from there.  This may not be 
>>> ideal for some, however, and is probably not a long-term solution for 
>>> many...
>>> 
>>> Thanks Tim,
>>> 
>>> Antony
>>> 
>>> On Apr 25, 2010, at 8:03 PM, Tim Ruiz wrote:
>>> 
>>>> Antony, I am sure we could help them get something going through our
>>>> reseller program, either turnkey or API. Then they can put it where ever
>>>> they want on their own drop down. The only catch is they may need to do
>>>> some of their own translation for the site.
>>>> 
>>>> Tim 
>>>> 
>>>> 
>>>> -------- Original Message --------
>>>> Subject: Re: [gnso-vi-feb10] Competition authorities
>>>> From: Antony Van Couvering <avc@xxxxxxxxxxxxxxxxxxxx>
>>>> Date: Wed, April 21, 2010 7:13 pm
>>>> To: Avri Doria <avri@xxxxxxx>
>>>> Cc: Gnso-vi-feb10@xxxxxxxxx
>>>> 
>>>> 
>>>> Because there are likely to be -- if this doesn't take so long that
>>>> everyone's completely exhausted, morally and financially, before the new
>>>> gTLD round starts -- small registries that are simply not going to be
>>>> interesting to registrars (because of their size), or for which existing
>>>> registrars will not be appropriate (because they don't support the
>>>> registry's language, for instance). In these cases, it makes perfect
>>>> sense to have a registry and registrar integrated. 
>>>> 
>>>> This is the case for many small ccTLDs, for instance, and they are a
>>>> good case in point. Even if (to pick on them) GoDaddy does decide to
>>>> carry .bt (Bhutan), it will be pretty hard to get to (low on a drop-down
>>>> list), and it certainly won't be in the Bhutanese language or alphabet.
>>>> That same dynamic will apply for .zulu or .kurd or .berber.
>>>> 
>>>> 
>>>> On Apr 21, 2010, at 5:11 AM, Avri Doria wrote:
>>>> 
>>>>> Why do people think that there will lots of application that include 
>>>>> cross-ownership? for example in AVC message I felt like this was going to 
>>>>> be a road block for every poor little new registry and I did not 
>>>>> understand that.
>>>> 
>>> 
>>> 
>> 
>> 
>> 
> 




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

Privacy Policy | Terms of Service | Cookies Policy