[Gnso-newgtld-wg-wt2] Work Track 2: Single Base Agreement

[Rubens Kuhl]

Hi Kurt. 

Responses inline. 

> On Oct 3, 2016, at 8:53 PM, Kurt Pritz <kurt at kjpritz.com> wrote:
> 
> All:
> 
> Without fully fleshing out my thoughts, which I hope to do shortly, I want to express my support for Ray's idea that there be one agreement. It was certainly courageous given the clamor for additional TLD categories. Here are a number of points to support my thinking: 
> Crafting agreements for brands and other categories of registries opens the door to a myriad of categories that will be impossible to adequately develop. In a policy-level discussion, we will never be able to pre-determine the number of "categories" that might arise, nor the accommodations those category members might seek. In fact, it is presumptuous for us to do so.
> Our inability to score 100% on the test above leaves us in a position where, from time to time, we are asked to create new versions of the agreement for a new interest group or changes for an existing TLD-type that was inadequately considered at the outset. The door, once opened, will not be able to be closed. Having created a "modest" number of additional registry agreements, who are we (or who is ICANN) to reject bone fide requests for more? By omitting certain "categories," will we stifle a form of innovation or choice that will benefit us all? Better to leave the door open with a general approach.

Although I agree with the principle, I have to note that the current agreement is strongly biased towards traditional revenue-generating business models. Some aspects of that:
- The fee is based on domains under management, which imply that domains are part of a revenue proposition, a cost factor to ICANN, or both. Neither is true for instance if a brand TLD registers millions of dictionary words to itself in a quest for search relevance.
- Data escrow is required even for exclusive-use TLDs, where the single registrant of all domains is the TLD owner
- Wildcard prohibition (but keeping dotless prohibition) could be revisited for exclusive-use TLDs
- Requiring EPP for Brand and Exclusive-use TLD, while even having registrars that could be adequately solved by batch processes, different APIs
- Requirement of registrars for all >100 domains (also mentioned by you below)
- WHOIS service for Exclusive-use TLDs (for which the IANA WHOIS is enough)

I know some people also group EBERO, CZDS and DNS SLA in requirements that could not apply for such TLDs, but I beg to differ and think of those as requirements for all gTLDs. 

> The one take away from the sTLD round in 2004 (and proven in the current round), is that structuring the TLD landscape according to categories does not work. In the most recent round, 90% of the pain in evaluations was a result of two types of categorization created by the Guidebook: community and geographic names.  Community consideration of TLD applications did not work in the objection or CPE processes. Government approval of geo TLDs remains problematic. More categories = more objections, IRPs, litigation because ICANN is drawing artificial lines where there are a continuum of business and operating models. Quite a bit could be written about this but I cannot overemphasize the pain or underestimate the benefit caused by well-intended categorization of TLDs.

Can we address those issues without a TARDIS ? 
> 
> To me, the more elegant, professional and policy-consistent approach is to develop a single agreement and processes for relief & criteria for certain provisions. I don't think there are many contractual provisions that are in play so this need not be complicated. The process / criteria generally are: 
> Here is the provision
> Here is the policy reason for the provision
> Here is the reason why a specific relief request does not upset the policy. 
> 
> This, I believe, is how the brands positioned the argument to ICANN that there should be exception for brand-registered TLDs. But brands and other "categories" are not one-size fits all. E.g., why cannot a brand self-register some names for infra-structure purposes and allow registrar-made registrations for others? I agree that the current situation is a patch and the cure is a more flexible approach not a different, but still hard-wired categorization.  
> 
> Brands have two stakeholder groups advocating on their behalf, which is how they were able to undertake an individual negotiation with ICANN achieve accommodations in the registry agreement. Even with that, I have spoken an owner of brands that is dissatisfied with the current structure. I don't think any ICANN working group is situated to negotiate (because that is what this discussion is) on behalf of possible TLD category advocates for the accommodations that that group might deem suitable. Instead, exception to contractual provisions should be able to be routinely requested and not unreasonably withheld.

The problem here is also principle x practice; ICANN has a strong track record in unreasonably withholding relief from provisions, including in the 2012 round Alternative Launch Programs (where only a couple TLDs endured the years to get those approved and even so strongly capped) and since the beginning of times the WHOIS Conflicts with national law issues. 

There is also within staff a bias towards "if something is not explicitly allowed then it's denied", which prevents such reasonable, sensible ideas from being used. 

> 
> I think the "firm demand for separate agreements"that we talk about is really a demand for relief from certain contractual provisions.

For most clauses, yes. The ones involving fees, likely no. 
> 
> I think the single-agreement approach has the potential to be long-lasting and adaptable and the multiple-ageement approach will be an inadequate attempt to address the current landscape. 

That would assume a source of reasonableness within ICANN as whole (not only staff) that comes very close to MLK's "I Have a Dream". I'm not trying to kill the dreamer, but asking to count on the dream requires a leap of faith. 



Rubens




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mm.icann.org/pipermail/gnso-newgtld-wg-wt2/attachments/20161003/78158ef4/attachment.html>