> Anybody interested in domain name policy issues should read
> it at:
>
> http://www.ntia.doc.gov/ntiahome/domainname/blileyrsp.htm
>
What a cautionary tale! The Policy-wonks and the Commercial-
wallahs *combined take on the Techno-nerds and fall on their face!
Throughout the doc (and consistent with the cooperative agreement
with NSI, through *14 amendments), the DoC reveals its blithe
ignorance of any distinction between registry ('authoritative root
server') and registrar (emphasis added):
"For a number of reasons, primarily having to do with the global
nature of the Internet, .com, .net, and .org today enjoy a dominant
position in the most commercially valuable Internet registrations.
(1) ... In more conventional economic terms, registering a name in
these domains is so commercially attractive that an exclusive
provider of *registry or *registrar services for these domains would
be able to exercise market power in dictating the terms for the
provision of those services, because its terms are not likely to be
subject to competition from alternative name *registration options.
Whether or not competition will develop in the future is not possible
to predict, but today - and for the reasonably foreseeable future -
there simply is no competitive alternative.
"Consumers and businesses around the world began to complain
about the absence of competition in this lucrative domain name
*registration market. Governments around the world complained
that it was inappropriate that these services were available
exclusively through a monopoly created and controlled by the
United States government."
"... the White Paper identified as a high priority the rapid
introduction of competition in the provision of domain name
*registration services. The interim board of the new corporation was
to develop policies for the addition of new TLDs and to establish
qualifications for domain name registries and registrars. The
Department of Commerce committed to enter into an agreement
with NSI by which NSI would agree to take specific actions,
including commitments as to pricing and equal access, designed
to permit the development of competition in domain name
*registration and to approximate the conditions that would be
expected in the presence of marketplace competition....4
"Under Amendment 11, NewCo's Responsibilities specifically
include the establishment and implementation of DNS policy and
the terms, including licensing terms, applicable to new and existing
gTLDs and registries under which *registries, *registrars and gTLDs
are permitted to operate." (Assistance to NewCo clause and
"Transition" Section of White Paper.)
Network Solutions has indicated that it is not obligated to enter into
a contract with ICANN because the Department of Commerce has
not "recognized" ICANN by transferring authority over the
authoritative root system to it....
"Amendment 11 unambiguously contemplates a contract between
NSI and ICANN under which NSI will recognize that ICANN has the
authority to carry out its responsibilities under the White Paper and
ICANN will accredit NSI as a *registrar and *registry. These parties
have not yet reached agreement on the terms of that contract. As
we have discussed, we believe it is clear that the terms must be
consistent with the policies set forth in the White Paper and, in
particular, with the policy mandating robust competition in the
provision of registrar services."
The confusion is clear from the outset; note 1 states, "In addition to
the gTLDs there are over 200 country-code top level domains
(ccTLDs) that might have been expected to provide some
competition. But relevant data indicate that the ccTLDs have not
yet presented a serious challenge to the commercial dominance of
the gTLDs...." -- that is, it has never been an issue of registrars,
but of the single *registry, which NSI 'developed' as IP in order to
fulfil its contracted function as a *registrar.
But saying that, it is surely obvious that the strighforward way out
of the USG/ ICANN/ NSI contractual logjam is for ccTLDs to
collaborate with the other alternate gTLDs (indeed, perhaps with
ICANN as well) to produce a second 'central' *registry.
a) There is no reason to suppose that any *existing DN under
.com, .net, .org or .edu need be accepted for registration. If using
whois is too slow, they could be promptly checked against a copy
of NSI's registry files (which I understand can be obtained for a few
thousand dollars); thus *when the time comes* for collating
('sharing') the root zone files, there whould be a minimum of
conflict.
b) The cost of registration can be set independently of NSI's
charges for maintaining the *registry.
c) The 'attractiveness' of the present 'market' in gTLDs is an artifact
of the naivet� with which millions of registrants entered the field.
Five years later, this single admission could do more to return
competition to the management of inter-networks than any number
of governmental and quasi-governmental protests as to what NSI
should or should not be doing.
So what, in other words, if NSI does keep control of something
called the 'A' root server in perpetuity? Whether one visualizes a
single 'authoritative' server or multiple ones, 'A' doesnt have to be
part of it. Indeed, as far as singleness goes, its problematic
whether the "separate contract [which] would be required to
obligate ICANN to operate the authoritative root [and which] would
mirror the obligations under which NSI currently operates the A root
server" -- that is, yet another government-guaranteed monopoly --
would resolve any conflict for more than a short period of time. The
clear solution is to take advantange of the opportunity NSI has held
out on a plate: open *registry competition. If the ICANN model
cannot compete, it can join the ranks of the thousands of other
start-up operations in the world, and we can look to see what other
possibilities there are.
kerry
