RE: PAB [kent@songbird.com: Onward]

["Antony Van Couvering" <avc@interport.net>]

> -----Original Message-----
> From: owner-participants@dnso.org [mailto:owner-participants@dnso.org]On
> Behalf Of Kent Crispin
> Sent: Thursday, January 28, 1999 2:46 PM
> To: DNSO Mail List
> Cc: participants@dnso.org
> Subject: Re: PAB [kent@songbird.com: Onward]
>
>
> On Thu, Jan 28, 1999 at 12:55:04PM -0500, Antony Van Couvering wrote:
> [...]
> > >
> > >     The DNSO recognizes that current registries have already existing
> > >     relationships with their registrants and, in many cases, with
> > >     sovereign governments.  The DNSO recognizes that there is a
> > >     strong presumption of continued stable operation of these
> > >     relationships.
> >
> > Kent, you talk about these relationships, but I just don't know of very
> > many.  Which of the many ccTLDs that have relations with sovereign
> > governments are you talking about?  If you mean that
> governments are aware
> > of their existence, then I suppose that's a relationship, and you are
> > correct.  If you mean that there is any formal or contractual
> relationship,
> > I believe you are gravely mistaken.  Would you care to give
> some examples?
> > Of domains with more than 15,000 names in them (the top 20), I
> can think of
> > only two.

And John Charles Broomfield has added a couple more -- Italy and Spain.  I
think he's wrong about Switzerland, they just had a bit of a dust up there
on the subject, but you'd have to ask SWITCH about it.

>
> No, I don't mean contractural relationships.  I am only interested de
> facto relationships, of which there are many, whether supported by
> contract or not.

In that sense, I have a relationship with you, as do your dear friends
Michael Sondow and Eberhard Lisse.  But ordinarily we wouldn't call that a
"relationship."  Look, in real life a government is very powerful, always
have been, always will be, and if they really want to, and really work to
take over a ccTLD domain, no doubt they will succeed.  That doesn't mean
that we have to say to governments, "Forget RFC 1591, which has governed the
administrations of TLDs to date.  Instead, here's this "sovereignty"
language, which is a green light to you that not only doesn't discourage
your control, it practically invites it."  The fact that a government knows
about, and communicates with, a given ccTLD is *not* the same thing as
recognizing its control.

>
> > Also, rather than devolve the responsibilities and rights of
> registries into
> > a mere power relationship, especially with nice-sounding
> language that means
> > very little when rubbed, polished, and examined, why not simply ask for
> > continuation of the current guidelines which TLDs should
> follow.  How about:
>
> You misunderstand my intent.  I am very well aware of your and IATLD's
> position concerning rfc1591.  But the evidence is incontrovertible that
> it is a controversial position -- the response on this list makes it
> clear, but it has been obviously controversial for some time.

I have to demur.  Everyone's been going along very nicely with RFC1591 for
quite some time, thank you.  The controversy, such as it is, has to do with
the presumption that the ground rules can be changed (vide "sovereignty")
without so much as a nod to the people affected.  I don't think, by the way,
that the evidence is incontrovertible.  If you want to use participation on
these lists as a measure, then you would have to say also that it is
incontrovertible that there is practically no support for a trademark
constituency either.  I have met no-one who heads up a TLD who thinks that
RFC 1591 is a bad idea -- it would hardly be credible if they did --, they
are just worried about tweaking their governments, or feels that "resistance
is futile" (a la John Charles Broomfield).  And that, in my mind, is exactly
why it is necessary to have some agreed-upon standard that *at the very
least* establishes some kind of due process.

Look, the fact is that all TLDs have been guided (if not governed) by RFC
1591 for some time -- as well as other relevant RFCs.  Now, in its place --
nothing, except some new language, invented out of whole cloth, concerning
sovereignty.  Nothing about the responsibilities of TLDs, nothing about lame
delegations, nothing about non-discrimination.

> That
> is, there is no widespread community consensus behind it.  And, from
> what I can see, there is no consensus even in the ccTLD community --
> only about a third of the ccTLDs belong to IATLD, and almost none of
> the big ones.

And CENTR has even fewer.  What's the point?  That the big ones should
dictate terms, or that there has to be consensus among registries before a
position is accepted?  Perhaps we should just go with what NSI says, since
there is "consensus" among gTLDs -- .COM, .NET, and .ORG all agree...

To use an analogy, there is certainly no widespread consensus about the
United States Electoral College.  In fact, just about everyone considers it
an annoying anachronism, it has nothing to do with the actual results of
elections (and shouldn't).  But because there hasn't been a consensus to
*get rid of it*, it stays.  RFC 1591 is not, in the main, anachronistic,
which doesn't mean that it can't be improved.  But since it has been there,
working, referred to, and used, and there hasn't been any consensus to get
rid of it, why on earth should it be trashed?


>
> Given that fact, I am trying to propose language that we can all
> agree to, as a starting point -- language that we can reach something
> approaching consensus on.  And I think there is widespread agreement
> to the above language -- it seems to me a simple recognization of
> fact -- and it does give you a bit of what you want.
>

I have to say that "widespread agreement", absent any method for determining
agreement, is in the mind of the beholder.  How can you say there's
widespread agreement when you've only just introduced the language?  If you
want to include something, but don't want to refer to RFC 1591, why not
include it wholesale?  What is in it that you object to?

> Obviously, you would like more.  Obviously, insertion of language
> concerning rfc1591 is going to be a rough road.

And getting rougher by the minute.

> Let's try a
> different tack:
>
> All the TLDs, regardless of the sovereignty issue, have expressed a
> desire for some extra control over their destiny -- supermajority,
> pre-recommendation review, selective veto.  But every other
> constituency would like the same thing, and the DNSO has a
> requirement for *fair* processes.  How about the following language
> (modified Section 8 in the merged draft):
>
>   Within the Names Council decisions shall be arrived at through a
>   rough consensus basis, to the extent possible.  Such decisions
>   shall be recorded as "consensus decisions".  Consensus will be
>   assessed by the Chair of the Names Council.  If two or more Names
>   Council members formally object to the measure in question, the
>   Chair shall determine that a consensus has not been achieved, and
>   shall call for a formal vote.
>
>   In such cases where consensus cannot be achieved and a formal vote
>   is deemed necessary, the matter shall be publicized before the DNSO
>   for 15 days, and then a formal, public, recorded online vote of the
>   Names Council shall be taken.  The measure shall pass if two thirds
>   of the votes cast are in the affirmative.  However, in any policy
>   recommendation the count shall be reported in full to ICANN, so
>   that ICANN may independently judge the extent of support for the
>   proposition.
>
>   For any policy recommendation, regardless of the vote, the Names
>   Council shall include any dissenting opinion developed through fair
>   and open consensus processes from any Constituency.
>
> This is a generalization of the CENTR proposal.  It treats all
> Constituencies equally; it gives "producers" (registrars and
> registries) a *near* veto on any measure that affects them; and,
> under any circumstances, requires that any strong dissenting opinion
> *must* be conveyed to ICANN.  This is also similar, but not
> identical, to NSI's "pre-review" mechanism, but it still allows a
> strong majority opinion to be carried to ICANN over the objection of
> any single Constituency.
>
> This doesn't address the sovereignty issue directly.  But if, as you
> seem to believe, the ccTLDs all feel as you do, then they (and their
> registrars) will have a collectively a strong say and a near veto in
> any policy that touches that issue.

Why don't we bypass complicated parliamentary procedures and introduce this
language:

"Registries shall adhere to relevant RFCs until such time as the ICANN, upon
recommendation by the DNSO, shall mandate changes."

Does anyone object to that?  (I know Kent will.  How about anyone else?)

>
> I am interested in any comments anyone may have on the above
> proposed language, incidentally.
>
> --
> Kent Crispin, PAB Chair				"Do good,
> and you'll be
> kent@songbird.com				lonesome." -- Mark Twain
>