[Gnso-newgtld-wg] Notes and Action Items - New gTLD Subsequent Procedures PDP WG - 03 June 2019

[Rubens Kuhl]

Follow inline some comments on today's discussions.

> Discussion:
> -- The third-parties don’t have a contract with ICANN so don’t see how you could have a publicly available list.  Not sure there is any value.

There are two questions here; one is the feasibility of making the list that ICANN Org already has of RSPs public or not. Every TLD currently in GDD portal has an RSP identifier, but there is no public list of all current RSPs in operation.

The other is if there is value; I don't see much value either, but this is far from impossible. In fact it would be very simple, provided it serves a purpose.

> 
> -- Maybe instead call them “pre-evaluated” RSPs.

This option was floated before, and "pre-approval" seemed combining more acceptance and less rejection from the WG. But "pre-evaluated" wouldn't be wrong at all.


> -- QUESTION: Does RSP pre-approval apply regardless of the services to be provided in connection with a particular application or is the ability to meet the needs related to proposed new services part of the evaluation even if the RSP is Pre-approved?  QUESTION


That's likely to be impossible to generalize. I suggest leaving to the evaluators of the proposed "non-vanilla" services if those services require specific additional evaluation or not. That could take the form of clarifying questions to the proposed services.

> -- If there are new registry services proposed then there would need to be a new evaluation.

Yes, for some definition of "new". This point will come in later discussions in the sections about evaluation.


> -- What might be at issue here is what's the term of the pre-approval? It holds until the completion of the application and evaluation process. From then on the registry operator is responsible for meeting the technical requirements.

> -- The pre-approval holds until the completion of the application/evaluation process.  After that the RO is responsible for meeting the technical requirements.  Not sure how you would remove pre-approval.

> -- Whether an RSP is on a pre-approved list or not: Should be that at a point in time this RSP has qualified against a standard set of questions and that this stands for a period of time.  There still would need to be monitoring by the RO.

> -- Need to decide how long the pre-approval lasts and when another evaluation is required.
> -- the assumption was that the RSPs would retain their quality for that period of time, subject to pre-delegation testing.

It could hold until the evaluation criteria stays substantially similar to the one in which RSP got pre-approval on. I believe that content matters more than time here.

On PDT, we should be careful that ICANN blames the policy and AGB for using the same PDT to do first time delegations and to do RSP transition. While some tests are indeed the same, we should at least free the test at policy-level from being an exact replica of one another, and leave it to fulfilling the stability mission for how these two tests are similar or not.


> -- Agreement on the need for more information from ICANN on the EBERO-triggering incidents.
> -- This was supposed to provide efficiencies in the process.  This was never meant to be an accreditation.

At least not in how loaded the term accreditation carries meaning within the ICANN framework. The word in its natural habitat, the English language, doesn't seem to be so strong as its usage in ICANN matters.
But it what it is now, so better to avoid the accreditation terminology.


> Next meeting: Start with Factoring in the Number of TLDs the RSP Intends to Support.

While load capacity is something that appeals to intuition, it's not how registry systems work. An RSP could handle a million brand TLDs and still fail to support the land rush of a single open TLD that is either too cheap or too popular at pre-registrations. Landrush and steady state load patterns are very connected to commercial definitions,  and one constant theme in the 2012 program is how TLDs moved from one model to another.



Rubens





-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mm.icann.org/pipermail/gnso-newgtld-wg/attachments/20190603/875bb4b9/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 529 bytes
Desc: Message signed with OpenPGP
URL: <http://mm.icann.org/pipermail/gnso-newgtld-wg/attachments/20190603/875bb4b9/signature.asc>