[Fwd: Re: [council] Demonstrating Pre-Existing Polices]

[Louis Touton <touton@icann.org>]

Erica,

As you accept in the fourth paragraph of your message, it is quite
difficult (probably impossible) abstractly to define in advance what
items are helpful to establish prior policies and what items are not. 
Most documents that have been discussed, for example, contain several
different statements within them.  Different statements can be pertinent
to different aspects of pre-existing policy.  Artificial restrictions 
on what categories of items may be considered can turn out to operate 
in unexpected ways, ultimately frustrating the search for the true 
situation.  Indeed, in some circumstances reference to materials other 
than documents may be appropriate.

As you note, the Green Paper, even though a discussion document, does
show an important part of the path by which ICANN was formed, and as
such may be relevant.  This illustrates the flaws in deciding what is
and is not pertinent without focusing on the exact issue to be
determined.

On your question about RFC 1591, I appreciate that you are employing it
only an example, in view of the non-ICANN financial stakes you may have
or represent regarding the matter.  But let me respond to the example
simply for illustration purposes only.

First, the mere manner in which something is described, without
reference to the context or the consequences of that description, does
not strike me as a matter of ICANN "policy" (not meaning here to include
editorial policies such as whether to capitalize "Internet").  While you
are correct that policies often categorize things, it is not necessarily
true that all categorizations have policy consequences.  Even where a
categorization has consequences for one policy, it may have different
consequences (and the categorization may not be appropriate) for another 
policy.  A careful review of the purpose of the categorization is 
important.

Turning to your example, let me suggest that more care may be needed in
your reading RFC 1591 to devine whether there are only two (as you
posit) or more categories of TLDs.  A full reading of RFC 1591 shows
more than just two "categories" being mentioned:  It mentions "generic
TLDs" and "country TLDs".  (It does not use the abbreviations "gTLD" and
"ccTLD".)  It further categorizes (see page 2) the then-existing generic
TLDs as either "international in nature" or "restricted to use by
entities in the United States."  Thus, it does not seem that RFC 1591
would support an assertion that "it is contrary to ICANN policy to
describe a gTLD as a global TLD"; RFC 1591 itself describes some generic
TLDs as international, though it describes others as limited to a
particular country.

One must also keep in mind, of course, that RFC 1591 is by no means the
last word.  It was published in March 1994 and represented a statement
of information about the domain-name system as of that time.  Many
things happened between then and the time ICANN was formed in late
1998.  Moreover, in the realm of TLDs there has been significant
development of policies within the ICANN process.  In April 2000, the
Names Council recommended without dissent that ICANN "establish a policy
for the introduction of new gTLDs in a measured and responsible manner 
. . . ."  The Names Council's recommendations contemplated a regime
under which many different categories of TLDs would emerge:  "The Names
Council takes note of the fact that the WG C report indicates that
several types of domains should be considered in the initial
introduction, these being: fully open top-level domains, restricted and
chartered top-level domains with limited scope, non-commercial domains
and personal domains."  See
<http://www.icann.org/yokohama/new-tld-topic.htm>.  The recommendation
of the Names Council was adopted by the ICANN Board in July 2000
(resolution 00.46).  Today we have a large array of different kinds of
TLDs in the process of introduction, and they may be categorized in many
different ways:  uTLDs, sTLDs, restricted TLDs, unrestricted TLDs, etc. 
These events seem to disspell the notion that there are only two
catgories of TLDs.

Best regards,

Louis Touton


-------- Original Message --------
Subject: Re: [council] Demonstrating Pre-Existing Polices
Date: Wed, 11 Jul 2001 22:20:23 +0800
From: "erica" <erica.roberts@bigpond.com>
To: "Louis Touton" <touton@icann.org>, <council@dnso.org>
References: <3B4BDD64.9C56200D@icann.org>

thanks Louis - this is very helpful.

I agree that any claim that 'x is a pre-existing policy' needs to be
supported by clear documentary evidence.  It is therefore important to
cite chapter and verse of relevant documents when any claim of this type
is made.

This will help us distinguish between what policies pre-dated the
creation of ICANN and what some advocate hoped policy would become (or
thought it was).  However, my question relates to the composition of the
relevant pool of documents.  I'm a bit surprised that you include the
Green paper in this.  I guess it could be used to throw some light on
the meaning of some sections of the White paper, but since the Green
paper was a discussion paper, I don't see how it could be deemed an
authoritative policy doc.

Having said that, I accept that it is not always easy todefine what can
and cannot be deemed  an authoritative policy doc and what precisely,
can be derived from them.

Let me try an example that has arisen recently:

There has been some discussion (protest) about a recent ICANN doc that
suggests there are currently four global commercial TLD. This
classification seems to conflict with RFP 1591 which distinguishes
between two classifications  - generic TLDs and country code TLDs. 
Given that all TLDs resolve globally, it would not seem to be consistant
with RFC 1591) to imply in any way that gTLDs can be distinguished as
global TLDs  (with the implication that ccTLDs are not global).  Any new
classification system strikes me as a a very significant part of any
policy framework.

However, I guess my real question is whether, on the basis of RFP 1591,
we can assume that it is pre-existing policy to classify TLD into two
basic types - generic TLDs, and country code TLDs?  Obviously, if this
is true, then the corollary,  that 'it is contrary to ICANN policy to
describe a gTLD as a global TLD' would also be true.

Please be assured that I'm not trying to score any points here.  Rather,
what I am trying to do is to highlight the difficulties in determining
just what can and cannot be deemed to be pre-existing policy.  The
question relates not only to the documentary evidence but also to what
is, and is not, a policy issue.   I appreciate the difficulties here,
but the issue is an important one which goes to the role of the NC. 
Without clarification (from staff, the Bod and/or NC??), this could
prove something of a time bomb for us all.  I see benefit in diffusing
it early  rather than waiting for it to explode at some later date.

regards,
erica




----- Original Message -----
From: "Louis Touton" <touton@icann.org>
To: <council@dnso.org>
Sent: Wednesday, July 11, 2001 1:00 PM
Subject: [council] Demonstrating Pre-Existing Polices


> Erica,
>
> To state the proposition most directly, to demonstrate what policies
> were in place at ICANN's inception one should look to the evidence that
> illuminates events occurring at or before ICANN's formation.
> Ordinarily, documentary evidence will be the strongest, since it is not
> subject to shifting recollections and the like.  This includes not only
> ICANN's formation documents (Green and White Papers, Articles, bylaws,
> etc.) but also such documents as RFCs, IANA news memos, and other
> documents that record policies that were followed.  One should be
> careful to distinguish between what policy existed and what some
> advocate hoped the policy would become.
>
> I hope this is helpful to you.
>
> Louis
>
>
> -------- Original Message --------
> Subject: Re: [council] Finalization of "A Unique, Authoritative Root
> for  theDNS"]
> Date: Wed, 11 Jul 2001 13:19:08 +0800
> From: "erica" <erica.roberts@bigpond.com>
> To: "Louis Touton" <touton@icann.org>, <council@dnso.org>
> References: <3B4B40EC.33F58001@icann.org>
>
> I am increasingly confused about ICANN policy and would appreciate
> clarification of how to determine what policies were in place previous
> to
> ICANN's creation.
> I assume that any claim that something is a pre-exisitng policy must be
> evidenced by reference to the Articles, By-laws or RFPs.
> However, this is clearly a foundational issue that requires
> clarification.
> Louis: Can you assist in this please.
>
> erica