ICANN/GNSO GNSO Email List Archives


<<< Chronological Index >>>    <<< Thread Index >>>

[ga] Alternative Domain Name Distribution Models

  • To: <ga@xxxxxxxxxxxxxx>
  • Subject: [ga] Alternative Domain Name Distribution Models
  • From: "Michael D. Palage" <mike@xxxxxxxxxx>
  • Date: Wed, 5 Sep 2007 07:11:31 -0400


Let me begin by stating our points of agreement. As I stated in my last
email, I believe there should be a presumption of using the current
registrar distribution model for all registries. I believe in reading
your comments, you might find a "stronger" presumption in favor of using
the current registrar distribution. The reason I submitted my comments
with regard to Recommendation #19 was that there was no
latitude/flexibility to discuss varying levels of presumption just a
simple black and white rule. As we saw during our time on the Board in
connection with the 2004 sTLD round, black and white analysis rarely
works. That is why the original sTLD evaluation committees only approved
2 out of the 10 applicants (.POST and .CAT) and why the ICANN Board had
to intervene in connection with (.TRAVEL. .JOBS, .MOBI, .ASIA, and

I fully agree that registries which have strict verification criteria
are generally more suitable to potentially overcome the presumption that
I have discussed. That is because the heightened levels of verification
generally discourage registrars from investing the time and money to
build systems to interface with these registries, and these heightened
levels of verification also generally result in lower numbers of
registrations. In fact according to my quick back of the napkin
calculations, the only sTLD with over 50,000 registrations is .MOBI. In
fact most of the sTLDs are under 10,000, numbers that I believe are
comparable to the .EDU, .GOV, .MIL and .INT TLDs that I referenced.

As I previously stated, my initial thinking was to look toward a black
and white numerical threshold by which registries could seek an
exemption. While this would work for most of the current registries in
today's marketplace (i.e. .MUSEUM) it would not scale to address every
possible scenario. That is why I have been an advocate of the rebuttable
presumption. It sends a clear message to potential applicants that there
is a presumption in favor of the existing registrar/registry dichotomy.
In fact I am even personally open to dialog about wording along the
lines of a strong presumption. I just know that the absolute black and
white rule contained in Recommendation #19 will not scale and will fail
in the next round. That is why I am trying to socialize this idea of
taking a different look at the situation.

I agree that registrars have done a lot within the existing .COM, .ORG
and .NET gTLDs to decrease costs and spur innovation, that is why in my
answer to Elliot I do not see any justification in today's market to
allow a registry such as VeriSign to go direct.

As Eric said on the list, I think there is a lot of common ground shared
by most of the people intelligently discussing this topic and hopefully
this dialog will help encourage potential alternative solutions to
maximize the common ground, and help provide for the best criteria for
the next round of sTLDs.

Roberto, I hope this email provides some additional clarification to my
original posting and finds some additional points of commonality in our

Best regards,


P.S. In the interst of continuing a positive dialog I changed the
subject line to this thread :-)

-----Original Message-----
From: owner-ga@xxxxxxxxxxxxxx [mailto:owner-ga@xxxxxxxxxxxxxx] On Behalf
Of Roberto Gaetano
Sent: Wednesday, September 05, 2007 3:41 AM
To: 'Michael D. Palage'; 'Nevett, Jonathon'; 'Danny Younger';
Subject: RE: [ga] GNSO Council: Taking 21 months to arrive at bad

I am mostly a listener on this topic, and I am very interested in
hearing the different opinions. Incidentally, it is refreshing to read
something on this list that is related to DNS issues.

I have a comment related to Mike's post below:

> Obviously if a registry goes direct there is the need for
> scrutiny to make sure that a registry does not abuse its 
> position as a sole source provider. However, as has been 
> demonstrated in the case of .EDU, .GOV, .MIL and .INT, 
> consumers are not being harmed even though they are I some 
> cases paying higher per domain name registrant costs than 
> equivalent registrants in the .COM space.

There are striking differences to me between .EDU, .GOV, .MIL, .INT, and
.COM. The first one is that everybody qualifies for the latter, and very
few for the former. This is because there are strict rules for the
former, and none for the latter. Connected with this, is the fact that
the business model is quite different. I am not a registrar, but if I
were one I would doubt that I would rely too much on registrations under
.INT to ensure income for the company ;>). This simply means that in
some cases the registry-registrar model would make, IMHO, little sense.
And maybe .MUSEUM is, among the ICANN-introduced TLDs, another example.
However, I think that we have to be extremely careful in dropping the
model in the general case. For general purpose TLDs, or generally
speaking for TLDs that do not have a market niche but are in competition
among themselves (I am talking about cases like .COM, .ORG, .NET, .INFO,
etc.) it would be extremely dangerous, IMHO, to have some of them bound
by current contracts to the use of registrars, while others would have
free hand.

Again, I am eager to listen, mine is not a position carved in stone,
just food for thought.


<<< Chronological Index >>>    <<< Thread Index >>>