RE: [gnso-vi-feb10] My thoughts on the proposals

[Jeff Eckhaus <eckhaus@xxxxxxxxxxxxxxx>]

Alan,

This is a great example and actually shows the point I am trying to prove, that 
these concerns are not reliant on cross-ownership but general concerns and can 
occur regardless. 

AAA Registry (no Registrar ownership) can set up a reseller SSS that is selling 
the AAA domain directly. You now have a Registry selling domains directly which 
is a concern and exactly the same as the scenario you have discussed below with 
less steps. 

The point I am trying to show is a Registry can game, scam, do harms and bad 
things, let's work on protecting from those,  co-ownership on its own does not 
lead to harms. But the thinking that because a Registrar is the owner, and that 
is what will lead to harms, is unfair and prejudicial. 



Jeff




-----Original Message-----
From: Alan Greenberg [mailto:alan.greenberg@xxxxxxxxx] 
Sent: Thursday, June 03, 2010 3:13 PM
To: Jeff Eckhaus; VI
Subject: RE: [gnso-vi-feb10] My thoughts on the proposals

Jeff, I guess I don't see it that way. Aside from the obvious requirement to 
audit that in fact a registrar is not authorized to sell it's own domain, the 
issue of resellers confuses the issue several times over. Resellers can be 
wholly owned subsidiaries of registrars, and resellers can have relationships 
with multiple registrars at once.

So in the simplest case, the YYY Registry has a Registrar RRR which does not 
market the YYY domain. But RRR has a reseller subsidiary SSS. SSS is also a 
reseller for an unrelated registrar GGG (which does market YYY). You can now 
have SSS selling YYY (via GGG). This is virtually impossible to audit given 
that ICANN has no information about registrar-reseller relationships (and 
registrars have said that it should stay that way).

Since this is a "simple" case, perhaps you get the idea.

As I said in my note, perhaps in the long term, we can find a way to finesse 
this. But in the short term it adds complexity and uncertainty and is just too 
subject to gaming. And perhaps ICANN compliance not being up to the job in this 
brave new world is a matter for another day. But if what we propose makes it 
even harder for them to function effectively, then I don't think we have the 
luxury of ignoring that.

Alan

At 03/06/2010 04:40 PM, Jeff Eckhaus wrote:
>Alan,
>
>Thanks for the note and I do have one specific point on your note and 
>an issue you brought up. On the subject of cross-ownership rules where 
>the related parties do not market the TLDs you are sympathetic to it 
>but have concerns about ICANN enforcement. I believe the idea of why 
>proposal writers brought this scenario in as a compromise, is that it 
>would not add any additional enforcement needs above and beyond if it 
>was an applicant who did not have cross-ownership.  But would not 
>restrict a class of applicants just because they are an ICANN 
>accredited Registrar.
>
>This is a subject that I really want to be clear on because some 
>believe that if it is a Registrar that applies it will require 
>additional scrutiny even if they do not distribute the TLD. I do not 
>believe that  is true, as it requires the same exact level of scrutiny 
>as any other applicant whether they are co-owned or not.
>
>Now whether or not you believe ICANN compliance is up to the task on 
>new TLDs is a another matter for another day
>
>
>
>Jeff
>
>
>
>-----Original Message-----
>From: owner-gnso-vi-feb10@xxxxxxxxx
>[mailto:owner-gnso-vi-feb10@xxxxxxxxx] On Behalf Of Alan Greenberg
>Sent: Thursday, June 03, 2010 11:25 AM
>To: VI
>Subject: [gnso-vi-feb10] My thoughts on the proposals
>
>
>Since we are getting close to the point where we must decide what we 
>will say in Brussels, I thought I would state my position. To be clear, 
>this is a personal position and not a formal ALAC position.
>Although I have discussed aspects with some ALAC members who support my 
>views, there are clearly others who do not.
>
>First, with regard to timing, although we clearly have to give a report 
>and it is important for the community and the Board to understand where 
>we are heading, I do not see Brussels as a drop-dead decision point. 
>The Board gave us the task of providing an alternative to their "zero 
>cross-ownership" default in sufficient time to allow it to be used in 
>the first round of new gTLDs without delaying that launch. Since we are 
>optimistically looking for a launch near the end of this calendar year, 
>that means that our recommendation needs to be locked in by *about* 
>August to allow for public comment and GNSO/Board approval.
>
>I am approaching this from the point of view of registrants and the 
>overall community. I specifically am not looking at it from the 
>perspective of business opportunities. I favour security and stability 
>of the gTLD ecosystem over innovation at this point.
>Moreover, I am looking at this as a solution for the first round of new 
>gTLDs. The PDP WG can continue to work on a long-term solution which 
>might be quite different. For that reason, I think that it is important 
>to not implement something that may need to be either turned back when 
>the final recommendation is made, or worse, have those who participated 
>in the first round be working under different rules than those who come 
>in later.
>
>Of the proposals on the table, I support the latest one submitted by 
>Brian Cute on behalf of Afilias and others, although I would like to 
>see some modifications to it prior to final acceptance.
>
>I support this proposal because it makes minimal changes to an already 
>complex environment, but it plugs some loop-holes that are already 
>there. I would like to see some exceptions built in to accommodate SR 
>TLDs and small cultural or not-for profit TLDs, but I agree that 
>defining these may be a challenge. Hopefully we can do it before my 
>suggested August date.
>
>At this time, I reject the proposal to relax the cross-ownership rules 
>as long as the related parties do not market their own TLDs. I am 
>sympathetic to it in theory, if it could be credibly enforced.
>But I believe that it will make an already complex environment even 
>more complex, with absolutely no chance of ICANN being able to monitor 
>or take action on infractions. As others have noted, ICANN compliance 
>may be getting more effective, but they still have a long way to go 
>when you factor in how much the TLD space they are responsible for will 
>grow. The issue of resellers has been mentioned, and I think that it is 
>important, particularly since ICANN compliance does not consider them 
>within their scope at the moment.
>
>I cannot consider supporting the CAM model without a well-defined 
>definition of the "market" that I agree with. I strongly believe that 
>there is not just one homogeneous TLD or even gTLD market, a core 
>assumption of many of the economists who would be the arbitrators in 
>this proposal. The really successful new gTLDs (and I do think that 
>there will be some), will be markets unto themselves.
>But that level of success will be impossible to determine ahead of time 
>when the CAM model says the evaluations will first be made.
>Once a decision is made for integration, later success cannot undo it, 
>except by forced divestiture and that is not something we should build 
>into our system. Moreover, I still question whether the appropriate 
>competition authorities would participate in a timely fashion as the 
>proposal anticipates.
>
>Lastly, I reject the arguments raised by some that many ccTLDs are 
>vertically integrated and it works for them. While not denying the 
>truth of this statement, I note that these TLDs operate in a totally 
>different environment. Many or most are not-for profit, often with 
>active or passive government involvement. They often set rules far more 
>stringent than ICANN does (or perhaps can) and then they often 
>diligently enforce those rules. But as noted, I believe that there are 
>certain classes of gTLDs for which this could be both desirable and viable.
>
>Alan