Re: [wg-c] Eureka?

[Kent Crispin <kent@songbird.com>]

On Fri, Aug 06, 1999 at 02:57:39PM -0400, Ross Wm. Rader wrote:
[...]
> I went back and re-read Kent's post and no where did I see any mention that
> the registry would have to run as a non-profit.

I make a distinction between the "registry" and the "registry
operator".  A registry must be a non-profit public service oriented
etc, while the registry operator may be any entity.  I specifically 
pointed out that the registry operator may be a for-profit business, 
and in fact, I expect that to be the prevelant case.

This is precisely the arrangement with CORE -- the registry is a
non-profit, it was operated by Emergent, a for-profit company until
the mothballs were applied. 

The question has been raised as to whether you need the interemediate
entity -- why wouldn't the registry operator contract directly with
ICANN, and thus avoid the intermediate not-for-profit entity?  The 
answer, from my perspective, is: that's an implementation detail.  
There is a clause in the ICANN bylaws that explicitly forbids ICANN 
from running a registry, which could mean that ICANN should refrain 
from being involved in the business decisions of hiring for-profit 
contractors.  But there are many ways that could go.

> My understanding is that
> potential registries would bid on the namespace they wanted to run and based
> on the strength of their proposal and business model, one of them would be
> awarded finite rights to administer the TLD in question. The success or
> failure of the namespace would rest with the efforts of the registry, but
> because of the finite nature of the contract, could be rebid within a
> reasonable period of time to allow for continued competition and innovation
> at the registry level. Obviously this can't be a non-profit proposition.

This is not quite my view of it, which is that the registry is still
constrained to run as a shared registry.  Therefore, the success or
failure of the TLD namespace is somewhat secondary to the registry
operator -- presumably in the long term there would be several (many)
distinct registry operators around the world, and if one did a
particularly good job they could reasonably expect to win bids to run
more TLD databases.  That is, registry operators would compete on 
quality of service offered to registrars and ICANN, not through 
promoting one TLD over another.  Note that registry operation is 
largely independendent of whether a TLD has a charter or not -- 
charters are enforced through the efforts of registrars and dispute 
resolution mechanisms.

[...]
[...]
> 
> To sum & clarify...
> 
> 1.) There should be as many new TLDs created as the market can bear. (which
> is presumably lots)
> 2.) These should be administered by the global governing body as a public
> resource
> 3.) The rights to run the registries for these TLDs should be open for
> tender and awarded on the basis of a fixed term, renewable and profit
> centered contract.
> 4.) The business model employed by the registry would be determined by the
> registry, except in the case of "special" TLDs where restricted registration
> is desirable. If the TLD in question is a "special" TLD, both the registry
> and the governing body will be required to negotiate the finer points of
> administration and business model.
> 5.) Registries would be allowed to administer as many TLDs as they want to,
> assuming that they are working within the confines of their pre-existing
> contracts and that they tender successful bids for the additional TLD
> administrations.

Yep -- that's pretty much the way I see it, too.

Kent

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