Community Evaluation Criteria

[Amadeu Abril i Abril <Amadeu@xxxxxxxxx>]

With regard to the Community Priority Evaluation, we are worried that the 
current version of the scoring  leads to two undesirable results: on one hand, 
it promotes adopting unreasonable registration policies, while on the other 
hand, it would prevent not just the most reasonable, but even the most 
restricitive existing community/sponsored TLDs to pass the test.

We think that Eligibility (who can register), Name Selection (which names can 
be registered by each registtrant) and Content & Use Policies (how the domains 
can/cannot be used) are not separate, isolated, cumulative steps as presented 
in the Guidebook. They all are used for the same purpose: making sure the TLD 
is used as it should. But if there are strict eligibility and name selection 
policies, we won't see much about content & use. 

In fact, strict eligibility and most specially, strict name selection can only 
be carried out in "membership-card carrying" communities: prfessional groups, 
associations, organized industry branches, etc. These points, plus the one on 
"strict delineation" leaves most real-world community-based would be 
applications below the required score. And in the cases strict eligibility and 
name selection is present, not much enforcement would be necessary. 

Once again, we urge the staff to re-check what would happen if .aero, .cat, 
.coop, .museum or .travel would be required to past the test (to name some 
among the most restrictive gTLDs, with different approaches but all of them 
quite successful in both guaranteeing that the TLD is really used by the 
intended community and in preventing negative externalities, ie, damage to 
third parties for IP infrinements or non-compliant use of the namespace). There 
is no requirement to tailor the process to fit any given past or future 
application. But if too many real-world cases among those universally 
considered as real community-based TLDs fail or risk failling the test, then 
perhaps we should conclude that it is not refined enoguh to identify genuine 
community-based TLDs.

The net effect of these shortcomings is that the applicatns would have strong 
incentives to propose absurdly and impractical restrictive registration 
policies while still risking to fail the test and, hence, risking the auction 
(and none of this is the goal of the Community Priority Evaluation, quite the 
conrary: identify the genuine and reasonably policed community-based TLDs and 
offer them immunity from auctions against non-community applicants, in exchange 
of guranteeing the community nature of the TLD through reasonable registration 
and enforcement policies).

Our suggestions are:

*Ideally, Eligibility, Name Selection and Content & Use should be considered 
together, and rate the result from 0 to 3. At very least, Name Selection and 
Content & Use should be considered alternative, not cumulative solutions, or

* Decfrease the "pass" scoring to 13.

Amadeu Abril i Abril
CORE Internet Council of Registrars