On estimates of applications arising from any "EOI" and zone file access

[Eric Brunner-Williams <ebw@xxxxxxxxxxxxxxxxxxxx>]

Kurt,

In the November 24th briefing paper on zone file access [1], there is 
a turn of phrase that is also surprising.
The Issue section, paragraph 2, begins thus:

"The ICANN process to expand the namespace to tens or perhaps hundreds 
of new gTLD could result in scaling challenges for interested parties 
to continue to collect and compile gTLD zone files for legitimate 
purposes."
I draw your attention to "tens or perhaps hundreds."

The briefing paper speaks to zone files, a consequence of delegations, 
and their access.
It anticipates a numeric range consistent with my earlier comments 
"the rate of change of the IANA root is a limited resource, and the 
limit per year or recurring application period is about half of the 
number of entries of the IANA root at present" at [2].
The purpose of my comment, not accurately captured in the issue 
summary at [3], was to point out that, independent of all other 
issues, if the rate of change is the limiting resource, that other 
"data", which the proposed EOI model is intended to acquire, is 
decorative, not informative.
Of course, this was not the majority view, and in a technically 
informed, bottoms-up, multi-stakeholder, consensus-based private 
organization, it is reasonable for broadly held policy goals to only 
be critically re-examined when they cannot be realized with available 
mechanisms.
The Draft Expressions of Interest/Pre-Registrations Model [4] 
continues to anticipate a volume of applications several multiples of, 
if not an order of magnitude greater than, the range of anticipated 
delegations in the briefing paper on Zone File Access.
How are these two anticipations of volumes to be reconciled?

Should we assume that as many as nine out of ten applications by city 
governments, regional governments, and cultural and/or linguistic 
institutions, using the best application of record, CORE's 2004 sTLD 
application on behalf of PuntCAT, as modified to reflect the 
requirements in the DAGvX, will fail to result in delegations?
Should we assume that as many as nine out of ten applications for 
"generic" strings will fail due to string contention, and possibly 
defects in their applications, or other reasons?
Should we assume that applications will not fail at a nine out of ten 
rate, but that the number of non-failed applications will be 
significantly greater than the number of delegations?
I'm sure there are other possible means to reconcile the Zone File 
Access briefing paper and the Draft Expressions of 
Interest/Pre-Registrations Model paper. Even typographical error is 
possible.
As in my two prior comments, I'm not trying to re-argue that the EOI 
can't be useful, I've already made that case, and I'm pleased that 
Staff's summary of comments caught the more obvious points I made. I'm 
also not trying to argue for any particular form internal to the EOI 
model. I'll turn to that later, or my colleges at CORE will.
My sole interest is in the Staff's statements of anticipated volumes, 
of applications, and of zone file access agreements.
I am of course, employed by CORE, which has an interest in the 
outcome, though as usual, this is offered in my individual capacity.
Cheers,
Eric

[1] Zone File Access and HSTLD Briefs 091124 v2.doc, attachment to 
Request for volunteers for Expert Advisory Groups, email to various 
GNSO Stakeholders, received 27 November.
[2] http://forum.icann.org/lists/eoi-new-gtlds/msg00067.html

[3] http://www.icann.org/en/topics/new-gtlds/eoi-model-18dec09-en.pdf
At page 5, reads

> In addition, some comments indicate that
> an analysis of available market capital indicates that demand will not exceed 
> 500 applications.
> Therefore, according to the comment, the EOI process will not provide data to 
> resolve demand and root
> scaling issues beyond that which is already known. Therefore the EOI cannot 
> serve any useful
> quantitative purpose, i.e., the EOI is not necessary for ICANN to address 
> root‐scaling concerns.
There was no mention of "market capital" in my comment, only of the 
marketing (public awareness) campaign ICANN has committed to since the 
new gTLD process began. The summary appears to have conflated two or 
more comments, one by a "market capital" informed commenter, and my 
technically informed comment.
[4] supra.