Comparative Evaluation: Good criteria; not so good details
We favour the procedure established for Comparative Evaluation, with the caveats contained in this and other comments to this forum. But we think that some of the criteria are not polished enough, or simply are mistaken:
* First of all, it makes no sense that an application for a string that matches an “established institution” within the community gets a 3 and one that choses a string that matches or represents the community itself gets a 2. This is to say, taking some historical examples: .sita = 3; .aero = 2. Or .Gettytrust = 3; .museum = 2. .IEC (Institut d’Estudis Catlans, the official langauge academy for Catalan) = 3; .cat = 2.
This is EXACTLY the contrary of what ICANN should be promoting. We udnerstand that it is “simpler”, but in community-based TLDs we should favor identification with the community, NOT promoting the (corporate) ego of a given institution as if it was the whole community.
* Secondly, the identification of community-based with strict (airtight) registration requirements is a bit of an exaggeration. Most notably if this restrictions are intended as pre-registration manual checks. Experience show that this is not the only way to prevent unauthorized registrations. But reading the current language anything short of such individual prescreening seems to fail into the “predominantly intended for”, and therefore getting only a 2.
* We believe that those two examples show the need of a redrafting of such principles, or at least, allowing a much wider range of scoring (up to 5 instead of 3). As it is now, a perfecly sensible application that does the right thing (choosing a string that refers to the whole community and not a more or less random instituion therein; that really offers the TLD for that community, but has no pre-existing complete, perfect pre-identification of any possible registrant) would completely fail.
We urge you to give more thought to this part, as it promotes on one side unnatural choices (stupid strings just to score the vital extra point) and discourages realistic and responsible registration policies. Most notably when the current outcome of failing such evaluation is... the famous “effiicent allocation method” that is out of reach, scope and even dignity for any real, serious, community- based applicant.
Amadeu Abril i Abril CORE Internet Council of Registrars