[NA-Discuss] A comment on the Andruff Bundling Letter

Eric Brunner-Williams ebw at abenaki.wabanaki.net
Fri May 20 16:38:52 UTC 2011


The Joint Application Support Working Group (JAS-WG) has generalized 
from "script" to "language" to better accommodate the real needs of 
communities which "code switch" between two or more languages, and 
possibly the same or a smaller number of scripts, and includes this as 
a support selection criteria, along with "need" as a criteria, which 
includes an inability to pay one or more application fees and 
subsequent application expenses.

The JAS-WG has identified applications for underserved languages, and 
includes this as a criteria, along with "need" as a criteria, which 
includes an inability to pay a single application fee and subsequent 
application expenses.

Thus, the proposed benefit of a "bundling" of two or more applications 
by a single applicant for two or more strings would be to those 
applicants who are not capable of meeting the several criteria for 
support proposed by the JAS-WG.

A non-exhaustive list of applicants not-qualified for support under 
the JAS-WG proposal at present are:

	o applications made from highly developed economies, e.g., by 
Verisign, located in Reston, VA, for a dozen or more representations 
of the string "com" under one or more linguistic transformation rules,

	o applications made for trademarks, e.g., by a trademark holder for a 
set of two or more trademarked strings,

	o applications made for brands, e.g., by a brand manager for a set of 
strings used as brands,

	o applications made by governments, e.g., a government for two or 
more strings,

	o etc.

There are reasonable grounds to ask that ICANN modify its one-string, 
one-application model.

It fails to conform to the reality that a very substantial population 
using domain names in Han Script (Chinese) view the distinct 
characters of the "Simplified Chinese" reform as interchangeable with 
the associated "Traditional Chinese" characters.

It fails to conform to the reality that some applications will share 
substantive properties with other applications, and regardless of how 
the competition policy question of whether Verisign should be allowed 
additional registries, in its own right or as a "technical registry 
backend services provider" to captive tenants, the utility of 
expending fee-based resources two, three, or thirty times to determine 
if Verisign is technically capable of operating a registry is 
extremely limited.

It fails to conform to the reality that registry continuity, like 
registry escrow, is a service which is "cheaper by the dozen", and any 
reasonably diverse "pool" of registry operators can provide 
"continuity service", whether due to regional infrastructure failure 
such as earthquake or hurricane, war, or ordinary commercial failure, 
at negligible cost to the unaffected registry operators.

There are, quite simply, several sound reasons _for_ bundling, and 
from a process perspective, the assumption that each application shall 
be evaluated independently of all others is, as a problem of method, 
the single least efficient, highest cost, most wasteful, method to 
adopt. At best all it can tell the evaluator (ICANN) is that two 
identical applications can both pass AND fail, which is not a very 
useful result to the applicant in evaluation, or to ICANN as the 
evaluation process owner.

I have on several occasions asked CEO Paul Twomey and Chairman Peter 
Dengate Thrush, now CEO Rod Beckstrom and Chairman Peter Dengate 
Thrush, to redefine an application to be for the resources necessary 
to avoid harm through the promotion of language loss in plural 
language markets and communities, rather than to be simply a single 
string. I wish I could relate that I had a reason to suppose that my 
requests were carefully considered, but I cannot.

However, I will not join Ron Andruff, whom I've worked with on several 
occasions, most recently on the GNSO's Operations Steering Committee 
(OSC), GNSO Council Operations Procedures Work Team (GCOT), on GNSO 
Council operations, and respect highly, or the many others, many of 
whom I've also worked with in the past and respect highly, as the 
mechanism proposed is both redundant where linguistic necessity 
exists, and unrestricted where competition policy concerns remain 

I see no reason to suppose that Verisign has a right to strings that 
have some association with "com" in scripts other than Latin, nor that 
it has a right to a lower cost to apply for such a string than any 
other applicant for the same, or a "confusingly similar" string (in 
the SWORD algorithmic sense of forming a contention set), which may be 
making only one application, intending to independently serve the 
needs of users of that script, and not the shareholders of the Latin 
script registries, still not transitioned from monopolies to 
competitive markets.

I participate in the Joint Application Support Working Group (JAS-WG). 
All errors or omissions in the representation of the current draft 
work product of the JAS-WG, above, are mine.


More information about the NA-Discuss mailing list