Re: [wg-c] proprietary TLDs

[Kent Crispin <kent@songbird.com>]

On Thu, Nov 25, 1999 at 11:56:07PM -0500, Milton Mueller wrote:
>Don't forget Brunner. Just as Ambler claims a property right in .web, Brunner
>claims a sovereignty right to .naa.

Ambler is indeed claiming a property right.  He has filed suit over
that claimed right. 

Brunner, on the other hand, is not claiming any right to the name
".naa" whatsoever.  He is making a proposal for a TLD to serve a
particular purpose, oriented towards a particular community, and
proposing a body to represent that community in policy terms.  Part
of the purpose in the policy is to create a portion of the namespace
where a particular view of "rights" concerning *some* names exists. 
That is not at all the same as claiming a sovereignty right. 

 [...]

>The Brunner case is really a lot more interesting than people seem to notice,
>having as they do the commercial generic TLD uppermost in their minds.
>Position Paper E is not really a position paper at all, but an application to
>run a registry.

Neither Eric nor his company is applying to run a registry.  Eric is
proposing that a registry be created that will be run by a community. 

>One could criticize Brunner on this count. All of the other
>people positioning themselves for TLD rights, though decried as greedy
>capitalists by some, would never be so purely self-interested as to dress up
>their claim for a specific TLD as a generalized policy position and insert it
>into the discussion at this juncture. Brunner gets away with it, I guess,
>because of the sympathy factor. 

Oh, do tell.  Could you elaborate on this just a little more?  I 
would love to hear your explanation of the "sympathy factor".

>But let's leave that aside for the moment and
>take it as an example of a real-world application. As such it raises some
>valuable issues to consider regarding how ICANN might entertain such
>applications.
> 
>There is, for example, no term limitation in the application. And indeed, how
>could there be? Would we want to force this particular group of native
>americans to "rebid" the registry every four years?

Could be.  The proposal doesn't go into that detail, but indeed, the 
registry could be rebid every 4 years.  In fact the proposal makes a 
distinction between operational and jurisdictional scope, and the 
operational part could well be rebid frequently.  The policy 
authority is clearly longer term, but that is effectively delegated 
to a government body.

>If so, ICANN will be put
>in the position of judging who "best represents" the native americans if there
>were competing applications?

This sounds very threatening in theory.  In practice it is not.  Of 
course, it is in principle no different than ICANN deciding who runs 
the registry for Moldavia.

>What is Roeland Meyer prepared an offer to run
>the .NAA registry that was cheaper and demonstrably more technically capable
>than the one run by NCAI/AFN?

How would Roeland establish that he best represented native
americans?

Anyway, the .naa proposal is for a non-profit, public benefit
registry. 

>Is ICANN in any position to decide which group
>of native americans best represents "native americans?"

No more than it is in a position to decide which group of Palestinians
best represents Palistinian interests.

>Should it be able to
>change the registry operator regardless of the wishes of the community served?
>If not, how does it ascertain those wishes? Even if it was capable, do we want
>it to have that power?
> 
> Of course, if the delegation is not rebid, then Brunner has the perpetual
> property right that some people on this list seem to fear and loathe.

There is a significant distinction between delegating a domain to
well-established representatives of a very large community, and
delegating a domain to a private individual or a private corporation
for the purpose of their private exploitation. 

>PPE also claims that it puts forward a shared-registry model. But does this
>mean that any registrar in the world can register names within it? No. The
>proposal calls for a "geographically distributed set of initial registrars
>operating on the basis of mutual trust and common policy within a specific
>jurisdictional scope."

Of course you jerked that language out of context, from the section 
on cost and risk reduction.  The full quote gives a rather different 
flavor: 

  There is no risk anticipated to the registry not adequately
  addressed by the adoption of a geographically distributed set of
  initial registrars operating on the basis of mutual trust and
  common policy within a specific jurisdictional scope. 
  
>In other words, in order to register names one must be
>trusted by, and conform to policies set by the owner of the delegation, which
>in this case is the National Congress of American Indians/Assembly of First
>Nations. In effect, this model means that NCAI/AFN dictate the terms under
>which registrars are permitted access to the database.

This is true for any sponsored or chartered TLD -- any registrar that
registers names in a chartered TLD must of course adhere to the
policies defined in the charter; any registrar that registers names
in a sponsored TLD must adhere to the policies adopted by the
sponsors.  There is absolutely nothing wrong with that.  The proposal
says:

  We wish to draw attention to the role of policy in the definition
  of a registry.  In the Co-Chair's example, ".family", the
  underlying issue is a shared policy model for registrars accessing
  a registry, in contrast to the registrar discretionary model.  A
  similar construction will arise for shared jurisdictional scopes
  for registrars accessing a shared registry, in contrast to
  the implicitly non-scoped model. 

>I have no problem with that, because I believe that such relationships should
>in fact be determined by voluntary contract.

Of course you are still compelled to bring it up as if it were
something sinister... 

>As Weinberg's position paper made clear, it is likely that many similar
>applications will be received and that those who wish to make registrations
>within a TLD conform to a specfic policy may want more direct control over
>which registrars are permitted to enter names in their database.

That is not what is proposed for chartered or sponsored TLDs -- they 
are explicitly open to any registrar who agrees to follow the 
specified policies.  Any delegation from ICANN will also of course 
be through a contract that will specify fair access and fairly 
enforced policies, with defined appeal processes, etc.  All that is 
part of the challenge of setting up a chartered or sponsored TLD -- 
the problem is not unique to .naa.

>This does not
>pose a problem for those of us who believe in a heterogeneous model. It poses
>big problems for those who think that the relationship between registry and
>registrar ought to be dictated and standardized by ICANN.

Eh?

-- 
Kent Crispin                               "Do good, and you'll be
kent@songbird.com                           lonesome." -- Mark Twain