Re: [wg-c] Short Position Paper

[Kent Crispin <kent@songbird.com>]

On Fri, Oct 08, 1999 at 09:17:14PM -0700, Dave Crocker wrote:
> Kent,
> 
> I like the thrust of your paper and most of its details.  A few concerns:
> 
> At 06:03 PM 10/8/1999 , Kent Crispin wrote:
> >1) Five to nine new TLD names be approved forthwith with the intent
> >that they be run as totally open gTLDs.  No further open gTLD names
> >should be approved until a process for approval of "chartered" or
> >"sponsored" TLDs is in place, and at least as many "chartered" or
> >"sponsored" TLDs are approved.
> 
> Why should chartered be in any way tied to unchartered?  (The concern isn't 
> about whether to have chartered, but rather the coupling.)

Open gTLDs and restricted TLDs both occupy the same name space --
approving .ham as an open gTLD precludes its use as a restricted TLD. 
While in any particular case there might alternatives --
.amateur_radio_operator, perhaps -- I want to give restricted TLDs a
good shot at the name space before the floodgates open. 

It might turn out that there is little interest in restricted TLDs. 
If so then there is absolutely no doubt that this issue would be
revisited.  But I wanted to keep the proposal very short, and not be 
specific in how or when it might be revisited.

> >2) ICANN should publish a Request For Proposal for registry
> >operators.  The goal of this RFP would be the selection of at least
> >five independent registry operators.  At least one registry operator
> >should be selected from each ICANN geographical region.
> 
> You are assuming one name per registry?

No, I wasn't.  In fact, I was thinking in terms of exactly 5
registries being selected, one from each region, and some of them
getting more than one TLD.  But what I wrote doesn't say that, 
obviously. 

> This means no economies of scale for their operation.  Why not do number of 
> assignments on a par with what NSI already has?

Just too many things to juggle...as you can of course appreciate,
this proposal has *many* internal compromises.  Some thoughts on the 
matter: 

Given the geographical restriction I made up, some current ccTLD
registries will be likely bidders, and will already have running
operations.  This would make the economies of scale less important,
because they would already be financially stable.

It is very difficult to predict how the competition will play out. 
One concern about adding a large number of open gTLDs is that they
will compete against each other for a tiny market share, remaining
forever invisible because they are just a crowd, and that
.com/.net/.org will dominate the scene for the next 5-10 years.  I
think a smaller group of open TLDs actually stands a better chance of
building significant market share. 

Plus, the political forces being what they are, I don't think that a
large number of open gTLDs is likely within the next two to three
years, in any case.  I think the 5 to 10 number is really about as
far as we can stretch things.  So therefore, get at least one new 
registry per geographical region, and spread the TLDs that we can 
get among them.

In the best of all possible worlds there would be a fairly large
number of entities that would make proposals for restricted TLDs, and
existing or new registries would bid to run them.  TM interests are
much less concerned about restricted TLDs now, and, I believe, with
some experience with UDRPs will be less worried about new gTLDs a
couple years in the future. 

The idea is not to try to get everything at once, but instead use 
alternatives that build the infrastructure in the meantime.

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