Re: [soac-newgtldapsup-wg] GAC Communique on JAS

[Eric Brunner-Williams <ebw@xxxxxxxxxxxxxxxxxxxx>]

To follow up on the CATALAN vs CAT issue, some example communities:

Absentee Shawnee Tribe of Indians of Oklahoma
Agua Caliente Band of Cahuilla Indians
Bad River Band of Lake Superior Tribe of Chippewa Indians
Barona Group of Capitan Grande Band of Mission Indians
Bear River Band of Rohnerville Rancheria
Big Pine Band of Owens Valley Paiute Shoshone Indians
Cachil DeHe Band of Wintun Indians of the Colusa Indian Community
Central Council Tlingit & Haida Indian Tribes of Alaska
...

The registration rules the US established when it created the NSN.US 
and more importantly, the -NSN.GOV namespaces, allow shorter forms of 
the name, e.g., the Aroostook Band of Micmacs is allowed to use 
micmac, with the -nsn.gov suffix.
Now obviously, no Tribal Government located in North America is going 
to meet the Board's example, which we've adopted, that of being an 
applicant from a least or lessor developed economy, but point is that 
the policy of using names that are shorter than the formal identifier 
of a community has been examined by the United States and found to be 
workable.
Therefore there is an "errata fix" to recommend to DAGv5 which is 
addressed in part by the comments of Amadeu (copied), Avri (wg 
contributor) and others. One which would allow a Catalan application 
to select "cat" rather than the longer "catalan" or the accented 
catalán form. Of course, this ignores the possibly relevant fact that 
"cat" is reserved in iso639 for Catalan, just as "nai" is for "North 
American Indian".
Eric

On 12/15/10 2:02 PM, Richard Tindal wrote:
> I was specifically asking if you think CAT would get 3 or 4 points under the 
> Registration Restriction component of scoring.   My sense is that CAT would
>
> In terms of the 14 points,   CAT possibly wouldn't get those 14 points if they 
> applied under the new process for .CAT,  but if they applied for .CATALAN they 
> would get 14 or more.
>
> I think thats why the scoring is structured the way it is.   If a group 
> representing the Catalan community applied for .CATALAN they would beat all 
> other possible applicants,  but if that same group applied for .CAT (a more 
> generic term) they might not beat all other applicants.
>
> A problem you're noting is that if they apply for .CATALAN as community they 
> might be saddled with registration restriction procedures that are too 
> expensive for them to manage.  Thats what I'm trying to get a handle on.     
> How costly are these procedures currently,   to say MUSEUM or CAT?
>
>
> On Dec 15, 2010, at 1:16 PM, Eric Brunner-Williams wrote:
>
> > Hi Richard,
> >
> > It is actually an open question if the .cat application would get 14/16 under 
> > the current guidelines.
> >
> > You are correct that where the applicant thinks that their string will not be 
> > in a contention set, and does not adopt a restrictive registration policy, 
> > assuming their belief is correct and the application is not eliminated through 
> > auction, it may not have adopted binding restrictions on its registration 
> > policy.
> >
> > However, if the applicant is willing to risk the application fee with no 
> > benefit from the application being self-identified as community-based, and with 
> > only future cost (in the form of revenues not received due to a restrictive 
> > registration policy) from identifying the application as community-based, a 
> > standard application from the same applicant making the same cost-benefit 
> > calculations is equally likely.
> >
> > So the net is that we may see no needs-qualified community-based applications, 
> > except those which have reduced viability due to the adoption of restrictive 
> > registration policies contained in the application, and difficult to modify 
> > subsequently.
> >
> > Looking at the first of your three points, as Amadeu has pointed out, the set 
> > of values present in scoring criterion 3 overlap, with some degree of 
> > incoherence. This should be fixed before we engage the 13 or 14 question. As to 
> > the second, the possibility of two community based applicants and the 
> > consequences, which range from none, if only one application meets the current 
> > criteria, to blocking all, if two or more applications meets the current 
> > criteria, is a hypothetical that should not have unintended adverse 
> > consequences where only one application meets the current criteria. A high 
> > score under a poorly designed scoring system protects all communities by 
> > eliminating their applications, which is not the usual form of protection 
> > sought. The same observation applies to low scores and the vulnerability of 
> > community applications. A multi-application hypothetical should not have 
> > adverse consequences where there is only one application.
> >
> >
> > Eric
On 12/15/10 12:44 PM, Richard Tindal wrote:
> Hi Eric
>
> To clarify for all (and I know you're aware of this) getting a 14
> point passing score only matters when there is another applicant for
> the string. If there
> is only one applicant then no scoring occurs (however as you note the
> applicant is tied to their promised Registration Restrictions)
>
> Most community strings will in my opinion, only have one application
> therefore, the 14 points wont generally matter. (i understand there
> are possible exceptions like .GAY)
>
> I understand the point you're making that even if there is only one
> application the Registration Restrictions commitments made by the
> applicant can affect their success.
>
> I think there are very strong reasons to keep community scoring high
> (I've attached some below). It seems to me that managing Registration
> Restictions could be done in a way that is not overly damaging to a
> TLD. You're familiar with the .CAT methods. I think they would get 3
> or 4 points for their Registration Restrictions if they were assessed
> under the current rules. Could future Community operators not use
> something like what CAT uses?
>
> Thx
>
> RT
> --------------------------------
>
> 1.Lowering the Score Can Harm Registrants
>
> An important component of the scoring method is “Criterion #3:
> Registration Policies (0-4 points)”.Having defined a precise community
> the applicant must then show that through restrictive registration
> policies only members of that community will be able to register
> second level names.By definition then, a community TLD is not
> available to everyone at the second level.You cannot have a
> ‘community’ that anyone can later join.If the scoring threshold is
> lowered it will be easier for applicants to obtain community status on
> strings that should be available at the second level to a very wide
> variety of registrants.A low score means that widely used terms could
> be captured by communities of opportunity who will then be prohibited
> from making second level names available to the general public.This
> consequence of low community scoring is not in the public interest.
>
> 2.A High Score is the Best Way to Protect Real Communities.
>
> Community scoring _only happens when there are two or more applicants
> for the string_(i.e. if there is only one applicant for the string
> there is no need for a community applicant to pass the 14 point
> threshold).In a situation where there are two or more applicants for a
> community claimed string it is in the interest of that community to
> have the scoring threshold at a high level.The higher the scoring
> threshold the more likely the string will be awarded to the applicant
> who most closely represents the community in question.If the scoring
> threshold is low it will be easier for all applicants to achieve a
> pass score, and this will negate the advantage of the applicant who
> best represents the community.
>
> 3.A Low Score Will Allow Successful Objections to Legitimate Communities
>
> The standards for successful Objection to a community application are
> based on the standards required to achieve the 14 point score.If the
> scoring threshold is lowered it will be easier for groups, who may not
> be closely associated with the community, to successful object and
> block the applicant.It is in the interest of real communities to have
> a high score.
>
> *__*
>
>
>
>
>
> On Dec 15, 2010, at 12:11 PM, Eric Brunner-Williams wrote:
>
> > Colleagues,
> >
> > Applications which are needs-qualified and "community-based" are,
> > due to the current string contention eligibility requirements,
> > either certain to fail if found to be in a contention set, if
> > meeting only 13 or fewer of the 16 evaluation points, and lacking
> > access to auction capital, or are certain to fail if meeting the
> > 14/16 evaluation point goal due to highly restrictive registration
> > policies.
> >
> > While the issue of community evaluation scoring and string
> > contention policy are outside the apparent remit of the JAS, if the
> > implementation in the current DAG is the final text, then we are
> > highly unlikely to be able to identify contention set surviving, or
> > economically viable community-based applicants, except where the
> > applicant identified the application as community-based, but made no
> > restriction intended to meet the 14 of 16 threshold.
> >
> > I suggest that this is an issue we can usefully bring to the
> > attention of the Board, and others, as the elimination of
> > community-based applicants from the beneficiaries of the JAS effort
> > is unlikely to have been the Board's intended goal, nor that of the
> > GAC and others contributing to the issue of developing a
> > community-based application type which is not inherently non-viable.
> >
> >
> > Please see the comments submitted, particularly those of Amadeu,
> > Avri, and Elisa Cooper of Mark Monitor on the subject of scoring
> > community-based applications.
> >
> > http://forum.icann.org/lists/5gtld-contention/msg00002.html (Amadeu)
> > http://forum.icann.org/lists/5gtld-contention/msg00001.html (Avri)
> > http://forum.icann.org/lists/5gtld-contention/msg00000.html (Elisa
> > Cooper of Mark Monitor)
> >
> > Eric