[ga-roots] Re: ICANN Policy -- revised version

["Milton Mueller" <Mueller@syr.edu>]

It seems a dialogue is actually beginning.

I think the statement you make below - only the specific part 
that I quote - is quite accurate. Currently, ICANN does not 
specify whether its TLD assignments will conflict with alternate
roots, nor does it make a commitment to avoid conflict. 
That fundamental ambiguity is what I wanted DNSO to explore.
It was my opinion that such an ambiguity represented the
absence of a policy, not a policy per se, and that there was 
a need to fill that vacuum with a DNSO-initiated process.

More comments below prefaced with MM##

>>> "M. Stuart Lynn" <lynn@icann.org> 06/20/01 04:32PM >>>

My paper does not assert that ICANN would 
purposely grant the same TLD name to one that already exists in an 
alternate root. Neither, on the other hand,  did it say that it would 
purposely avoid doing so. What it in effect said is that it would be 
erroneous for any alternate (pseudo or otherwise) root operator to 
*assume* that ICANN would necessarily avoid doing so. 

MM## What you say below may be a reasonable policy, or not,
or it may constitute one half or three fourths of a reasonable
policy. But it is a policy that as far as I can tell lacks any basis
in a DNSO process, the White Paper process, etc.

Lynn:
What the draft 
effectively says is that ICANN's orderly decision process (always, of 
course, subject to improvement) operating within its public trust 
would not be preempted by actions taken by others operating outside 
of the public trust. That is, do not assume that by creating a TLD 
within an alternate root, you receive credit with ICANN's own 
processes and can thereby tie ICANN's hands.

To put it another way. ICANN has no policy asserting that it will 
give credit to any TLD created outside of ICANN's processes. 

MM## Below, I think it's a real stretch. You're trying to go
too far with too little process or documentary support. 

Lynn:
Which means precisely that ICANN can give no credit to any TLD created 
outside of ICANN's processes. Which is what I wrote.

By my reading, however, the White Paper does in fact provide 
guidance on this point. It repeatedly refers to the need for 
decisions about new TLDs to be coordinated in a way that is 
accountable to the community (in so many words). Simply allowing any 
alternate root to preempt names hardly sounds like such 
accountability to me. What would stop such a root from choosing all 
the nouns in Webster's and any other dictionary to preempt future 
competition? Where would community accountability and coordination be 
in such a case?

MM## That's why we need to define a policy. Somewhere between
someone creating a root server and entering all the names of
the dictionary and the current ICANN regime, which requires
everyone to bow and scrape before a central licensing authority
making arbitrary decisions in an environment of artificial scarcity, there
is a reasonable policy to be found. 

I think the New.net concept of letting new TLDs enter the
market on their own initiative and reaching some 
OBJECTIVE threshhold before being entered into the ICANN root
has a lot of merit. But of course it needs to be fleshed out.

As an example, given current evidence it now looks as if the 
(tacit) policy that ICANN actually followed in its assignments
was not as bad as it first seemed. ICANN's Board avoided
conflict with .Web, a long-established alternate root operator.
And while I do not necessarily condone its conflict with .BIZ,
at least it's clear now that most of the registrations there
came AFTER the assignment in the ICANN root had been made 
to Neulevel. So ICANN did not cause as much conflict as it first
seemed. 

But again, I think the economic, technical, and political aspects
of this relationship need to be carefully and rationally explored. 
Demonizing alternate roots, or pretending that its a binary
choice between the current system and utter anarchy
is not useful. 

Lynn:
We now have interesting situations where, quite independently of 
ICANN, there is conflict between most of new.net's names and names in 
other alternate roots. This is the expected result of a regime where 
root-zone decisions are made in an uncoordinated fashion by companies 
based on their individual commercial concerns. Given the incentive 
for a better-financed commercial actor to muscle aside a weaker one, 
the result of this lack of coordination is name conflicts that 
certainly does not make for a stable DNS.

MM## Better-financed commercial actors muscle aside weaker ones
in market competition all the time. That in itself is not problematical.
The real issues are a) whether a business can develop a property
right in a TLD (I obviously do not accept current ICANN orthodoxy
on this, but that's too complex to go into here), and b) whether the 
Internet communty via DNSO/ICANN can define some reasonable 
threshold that must be passed before welcoming independent efforts 
into its root.

Regarding the "discussion draft" and whether it was intended
to pre-empt discussion, OK. I cannot read your mind. I will 
revert to the claim that it DID pre-empt discussion. 

And I reiterate my statement made in Stockholm: if you want
to rebut that assertion in an entirely convincing manner, put
up, on ICANN's home page, alongside the link to your draft,
the briefing paper I developed for the Names Council. 

rgds,
MM

--
This message was passed to you via the ga-roots@dnso.org list.
Send mail to majordomo@dnso.org to unsubscribe
("unsubscribe ga-roots" in the body of the message).
Archives at http://www.dnso.org/archives.html