[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [ga] new WG on chartered/sponsored TLDs



Hi Kent;

Please sign me up as a supporter of your proposal (to the NC or ICANN,
whatever it takes to get it organized, sooner is better). 

Bill Semich
.NU Domain

At 02:59 PM 2/10/00 -0800, Kent Crispin wrote:
>I am interested in forming a new dnso working group.  I have been in
>touch with Carolyn Chicoine and others on the NC on the topic, and it
>appears that 1) there will be a formal procedure for proposing a new WG
>sometime in the not too distant future (but probably after Cairo); and
>2) on the other hand, if several NC members can be persuaded to support
>it, a WG could instituted through informal procedures.  Either way is
>fine by me, and I will pursue both avenues as appropriate.
>
>In the meantime, I wonder if there might be some interest in the GA in 
>discussing the matter.  Just in case people are interested in such a 
>discussion, here are some random points and issues that occur to me:
>
>A.  Chartered TLDs
>
>0) There are those who disagree with the notion of chartered TLDs on
>basically spiritual grounds.  That is, they believe that *all* decisions
>concerning operation of the TLD, and criteria for use, should determined
>by the registry.  In this view, ICANN/DNSO should have no involvement 
>in such decisions.
>
>I can appreciate that point of view as a philosophical point, but I am
>not interested in discussing it in this proposed WG.  The purpose of the
>WG I propose is not to decide whether ICANN should authorize chartered
>TLDs -- the purpose, instead, is a good-faith exploration of the issues 
>that would be involved if they were authorized
>
>1) Why not do this in WG-C? WG-C has as its charter discussion of new
>gTLDs, with emphasis on the "g", and I think *that* discussion should
>remain in WG-C.  There is some overlap, but there are many unique 
>issues in the area of charters and sponsors, and WG-C already has a 
>rather full plate. (Some might say "too full".  In fact, one of my concerns 
>about DNSO processes in general is that they tend to be very large scale 
>and unfocussed.  More narrowly focussed WGs tend to be more likely to 
>accomplish things, in my limited experience.)
>
>2) One of the frequently mentioned criticisms of "charters" is that they
>would be difficult to enforce.  Such critics point to .edu as an example
>of a failed charter.  Others think that enforcement is not anywhere near
>as difficult as that, and point to .edu as an example of a successful
>chartered TLD.  How rigorous does enforcement have to be?  If .1% of 
>the SLDs in a cTLD would fail to meet the criteria in the charter, what 
>are the potential harms?
>
>3) Enforcement of a charter has costs -- what mechanisms can be used to 
>enforce them?  How would the costs be allocated.
>
>4) The base enforcement widget in a chartered TLD is a contract a
>customer signs with the registrar/registry, which includes their 
>agreement to the terms of the charter.  What are the legal 
>characteristics of a charter and such a contract that would make them
>enforcable in a cost-effective way?
>
>5) Presuming that ICANN decides to implement chartered TLDs, an approval
>process for charters would be required.  Presumably the approval process
>would involve a set of well-defined criteria, as well as possibly some
>form of required public/dnso review.  What are the criteria for good
>charters? What would be the appropriate public/dnso review process? What
>would be the legal liability of a public review process?
>
>
>B. Sponsored TLDs
>
>The term "sponsored TLD" (or sTLD) may be unfamiliar, since it is
>relatively recent.  A "sponsored TLD" is a TLD that has a "sponsor" -- 
>an organization that is given control over policy for the TLD in 
>question.  A sponsored TLD may or may not have an associated
>charter -- the two concepts are somewhat orthogonal.  The European 
>Commission is trying to get .eu approved as a quasi-ccTLD; but if there 
>was a mechanism in place for approving sponsored TLDs, the EC might try 
>to get .eu approved with the EC as its sponsor.  Note that the 
>"sponsor" and the "registry" are distinct -- the EC could contract with 
>any competent registry to provide registry services.
>
>Another potential example of a sponsored TLD is Eric Brunner's .naa TLD 
>-- there the sponsor is a "consortium" of tribal 
>governments/organizations that would assume policy responsibility for 
>the TLD.
>
>Yet another example would be a ".movie" TLD, where the sponsoring
>organization would be a hypothetical international consortium of movie
>production companies that would take policy responsibility for the tld.  
>This TLD might also have a charter, that it would be restricted to 
>commercial sites publicizing movies -- the purpose of the TLD most 
>emphatically does not need to be non-commercial.
>
>And again, there have been discussions of a TLD reserved for technical 
>support of IP Telephony.  The sponsors would be an international 
>consortium of telephone companies.
>
>Two immediate questions come to mind:  
>
>1) what are the characteristics required for an organization to take
>policy control over a TLD? Stability, of course, would be very high on
>the list, because sponsorship would be something that would be very
>difficult to revoke.  I've used the code phrase "international 
>consortium" fairly freely -- what standards of openness etc would such 
>consortia require before they could be considered as a sponsor.  (Note 
>that ICANN is effectively delegating some hypothetical policy 
>authority, and thus it could be argued that the delegatees should share 
>some of the characteristics of ICANN.)
>
>2) what are boundaries of the policy authority of a sponsor?
>
>Any thoughts?
>
>-- 
>Kent Crispin                               "Do good, and you'll be
>kent@songbird.com                           lonesome." -- Mark Twain
>--

Bill Semich
President and Founder
.NU Domain Ltd
http://whats.nu
bill@mail.nic.nu
--
This message was passed to you via the ga@dnso.org list.
Send mail to majordomo@dnso.org to unsubscribe
("unsubscribe ga" in the body of the message).
Archives at http://www.dnso.org/archives.html