RE: [wg-c] voting on TLDs

["Roeland M. J. Meyer" <rmeyer@mhsc.com>]

Whoa Son! You got an awful lot of speculation out of a single statement,
whose scope was strictly market demand estimates. There were also a LOT of
conditionals, like "appears" and "depending". Get the bit out of your teeth
and get back to reality.

-----Original Message-----
From: owner-wg-c@dnso.org [mailto:owner-wg-c@dnso.org]On Behalf Of Kent
Crispin
Sent: Sunday, March 12, 2000 5:49 PM
To: 'wg-c'
Subject: Re: [wg-c] voting on TLDs


On Sun, Mar 12, 2000 at 09:10:20AM -0800, Roeland M. J. Meyer wrote:
> The only point Dave leaves off here, and rightly, from his perspective, is
> the natural limit of the marketplace. The most optimistic view is that
there
> appears to be a maximum market size of 3000-4000 TLDs, depending on how
they
> are defined.

I don't know where you got your "most optimistic view", but it is
clearly bogus -- from several perspectives.

Most important, it makes the fatal assumption that the only reason to
run a TLD registry is to *sell* registrations.  In fact, of course,
there are a great many different possible reasons to run a TLD registry,
both commercial and non commercial, that do not involve directly selling
registrations.

Moreover, this projection suffers from the common flaw of assuming that
each TLD has its own technical/business infrastructure.  This is a huge
error, and vastly distorts the economics.  In fact, of course, the
incremental cost for adding a new gTLD is close to zero.  Any
technical/business infrastructure that can provide a registry service
for a single gTLD can provide registry service for a hundred gTLDs.  The
only issue is *total* registrations across all the gTLDs, not
registrations in a single TLD.  A registry that is profitably run with 1
TLD with "n" registrations per day will be, for all practical purposes,
just as profitable with 100 TLDs, with "n" total registrations per day.
Moreover, the scaling problem caused by adding a new gTLD is
indistinguishable from the scaling problem of the growth of
registrations in an already existing gTLD.  (*)


Example: .ibm

IBM corp of course has the technical and financial capability to run a
TLD registry for it's own purposes.  What policy does ICANN adopt that
prevents this from happening? If, on the other hand, ICANN allows this,
what is to prevent the hundreds of thousands of other companies that
already have sufficient resources to run a TLD registry from getting
their own? In fact, given the existence of production-grade approved
registries, even small companies could afford their own TLDs.  If one
claims that this can't be allowed, then one *agrees* that ICANN must be
using some evaluation process on TLDs, allowing some, and not allowing
others...


Example: .museum

The .museum TLD was proposed as a sponsored TLD, run by the non-profit NGO
"The International Council Of Museums".  There is no intent at all that
the registrations in that TLD be for the purpose of making money.


Example: TLD to support IP telephony

Here we have a proposed TLD that would be run by a non-profit member
organization of telephony providers that would support the telephone
number - IP address mapping.  In a sense it is a commercial TLD, but it
really only provides a fundamental technical purpose.


(*) In the case of chartered TLDS *where the charter is enforced by the
registry*, this would not necessarily be the case, of course.  However,
in a shared registry context the enforcement cost would be born by the
registrants and registrars; and in a "sponsored" TLD the cost would be
born by the sponsor.

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