[CWG-Stewardship] Update on the Integrated model.

[Andrew Sullivan]

Hi,

I'm not going to speak in this note about the RIR case, because while
I've paid attention to that I have way less familiarity than I do for
the IETF case.  I should note, for those who may not know, that I'm a
sitting member of the IAB (I've just been reappointed by the nomcom
for another 2 years) and that I'm currently the program lead and
program chair of the IAB's IANA evolution program.  I was also pretty
active in the IANAPLAN WG at the IETF.  I want to be very clear that I
am in this message (as in any where I don't state that I'm speaking on
anyone's behalf) speaking for myself only; I haven't sent this note
around to anyone to comment on, and there have been no consensus calls
of any sort about what I write below.

On Mon, Feb 23, 2015 at 12:03:39AM +0000, Milton L Mueller wrote:
ut
> > this is a pretty low-overhead, outsourced-operator approach. 
> 
> I really don't see how the proposed configuration differs much from that. 

The model, particularly in the proposed configuration, requires the
IETF to appoint something like 3-5 people in a shared service
arrangement with two other communities.  At least one of those
communities has proved to have somewhat different priorities than the
IETF for the evolution of the Internet.  At least one of those
communities also has dramatically different views from the IETF about
what the correct mechanisms are for ensuring able administration and
sufficient accountability of those placed in positions of trust by the
community in question.

Involving the IETF in a joint venture under these circumstances
strikes me as fraught with risk.  We have a critical but fairly dull
function that we desire to be administered in the same professional
and consistent way it has been administered approximately forever.
With respect to that administration, we have a straightforward way of
addressing matters in the event of irreconcilable difference.  The
three configurations of the model presented in the document Avri
distributed upthread all require the IETF to do way more work than
that, and to involve itself more intimately with other communities
that have rather different methods of working, assumptions about the
right ways to ensure accountability, and so on.  When presented with
the options of doing that, or else finding another way to run the
relevant registries ourselves, I am not convinced that the IETF's
trade-off calculation would come out either quickly or as many seem to
believe it would.  

>from this, but I do think IETF has a very substantial long term
>interest in a more rational, well-structured approach to IANA.

What is that interest?  The IETF definitely has an interest in
ensuring the NTIA transition happens.  I'm sure that the IETF should
prefer a clean and tidy functioning of the IANA functions with respect
to the root zone.  But the IETF has carefully avoided expressing
opinions about the output of the ICANN policy sphere except where it
has very grave technical consequences, and I think that is
appropriate.  As for how IANA is function from the IETF point of view,
the community has been totally clear: we're happy, and expect to
remain so.  We don't need additional accountability mechanisms.

>Also, in earlier NTIA proceedings the IETF has
>expressed a preference for keeping the IANA functions coordinated and
>under one roof

But this has always been a pragmatic goal.  For my part, I prefer that
because I think it works better.  But if the facts change, I'll change
my mind.

> interest. An example of that would be the IETF's participation in
> the Montevideo Statement, which helped to kick this process of IANA
> globalization into motion.

That is actually an interesting example, because in fact the IETF did
_not_ participate in that statement.  The IETF chair and the IAB chair
did.  At least in the case of the IAB chair, I can say with some
knowledge that the participation was with the knowledge and support of
the rest of the IAB.  But that was not the IETF participating.  There
is exactly one way for the IETF to participate in anything, and that
is for the IETF to come to rough consensus on the matter.  Formally,
outside of that, there is no IETF opinion at all.  

This goes right to the heart of the IETF's principles of organization.
To the extent that the I* group (which produced the Montevideo
Statement) has come to be any sort of formal organization, the IETF
(and to a lesser degree, the IAB) fits rather badly in it, because the
IETF is not organized in a way to "participate" like that.  This is
the same reason why the proposed joint board makes me uncomfortable.
We have ways to accommodate such appointments; but the entire notion
of representativity, which seems quite important to the names
community, is not really a thing for us.  Also, we're all volunteers,
and have day jobs.
 
> So if you understand those benefits there is hope that a meeting of the minds can be arrived at
> 

I should have hoped it obvious that what I hope for here is a
mechanism, pretty much of any sort, that will promote the stable,
predictable, and continuous operation of IANA in the way it has been
working for the Internet community.  Whatever complaints one might
have about the various policy, financial, or commercial decisions of
ICANN since its forming, I think it is hard to say that the IANA
portion of the organization has been at the bottom of those.  I just
want to make that continue in the world where we no longer have the
NTIA contract.

Best regards,

A

-- 
Andrew Sullivan
ajs at anvilwalrusden.com