Re: [gnso-vi-feb10] Single Registrant TLDs

["Neuman, Jeff" <Jeff.Neuman@xxxxxxxxxx>]

I think there is a little bit more to the debate.  One of the issues is with 
the single registrant whether there is any need for the registry or registrar 
functionality at all.  In other words, is there a need to have a separation of 
the registry/registrar functionality?  Is there a need for an SRS (shared 
registration system) at the registry level? Is there nothing more than pure 
DNS? Why go through the charade of having to set criteria for registrars that 
no one other than the registry itself could meet?

So if there is no true concept of a registrar, then is ownership an actual 
issue?

Let me explain....

An organization (commercial or noncommercial or nongovernmental) could 
establish a SR TLD whereby it owns all of the names and allows others to use 
those names, but there is no concept of having others" register" names.  For 
example, what if the International olympic committee wanted .olympic and they 
only had 200 registrations (1 for each country's olympic committee...i.e., 
USOC.olympic (for the US Olympic Committee). In that case, you could have one 
person at the International Olympic Committee enter all of the relevant 
information into a database (notice I didn't use the word registry database).  
There is no SRS or WHOIS functionality per se normally thought of as a 
registry, and no real registrar functionality of going through a EPP 
registration process.  There is no value add of having to go through an entity 
we today call a registrar.

This is a made up example, but I think is very real and in some ways similar to 
what I have heard .post is today.  I don't have any inside information for 
.post, but as I understand the application from years ago, .post may want to 
give the second level registrations to the national post offices to use as they 
see fit.  The charade of having to select registrars because of legacy 
requirements just makes no sense.  There is no real concept of a registrar in 
that sense nor any value add of requiring the ultimate recipient of the .post ( 
the national post offices) to have to select a registrar.  There is no need for 
the complexity of setting up a shared registration system at the registry level 
either.  

I see traditional brand tlds the same way and very real. Although no Brand TLDs 
may have filed public comments, let me tell you we have talked to some and this 
is a very real example (though I made up the olympic example).


Jeffrey J. Neuman, Esq.
Vice President, Law & Policy
NeuStar, Inc.
Jeff.Neuman@xxxxxxxxxxx



----- Original Message -----
From: owner-gnso-vi-feb10@xxxxxxxxx <owner-gnso-vi-feb10@xxxxxxxxx>
To: Gnso-vi-feb10@xxxxxxxxx <Gnso-vi-feb10@xxxxxxxxx>
Sent: Wed Apr 07 22:50:56 2010
Subject: [gnso-vi-feb10] Single Registrant TLDs


All,

We may be getting a little wrapped around the axle on this topic (this is 
no-one's fault as it's complex) so I'd like to take a shot at summarizing where 
it stands.   

We're not debating whether or not Single Registrant (SR) TLDs should be 
allowed.   They are allowed  -- and have been allowed from the first version of 
the DAG.      Any registry can register names just to itself and no registry is 
required to provide open access to registrars.

Also, no rule we devise will prevent SR TLDs.   We're making rules about who 
can own registries and registrars, not about who can own domains.  An SR TLD 
can exist if we recommend zero cross ownership and it can exist if we recommend 
100% cross ownership. 

What we're debating is whether or not,  in order to register its names,   an SR 
TLD registry must be accredited as a registrar  (and, importantly,  pay the 
fees that accompany that registrar accreditation).  This is the area of 
contention.   

If anyone feels I've mischaracterized the issue please jump in.


Also, I agree with the argument Volker made yesterday.    I think we should 
first see if we can find a rule-set that suits all types of registries.  This 
may be possible.   If we find the overall rule-set doesn't suit a particular 
registry-type then we can drill down on exception cases.

RT