Bert, et al.,
Thanks for your note. I too have watched the evolution of the
relationship with CNRI for a long time. I served on the IESG from 1989
to 1994 when Phill gross was the IETF chair, and I served on the IAB for
another two years. I co-chaired the POISED working group which
reorganized the relationship between the IAB and the IETF and instituted
the nomcomm, so I empathize quite a lot with your concerns. That said,
let me suggest three things, the last of which is, in my view, the most
important.
1. In my view, the relationship between the IETF and ISOC is not at all
similar to the relationship between the IETF and CRNI. The IETF already
has a strong management role in ISOC. The IETF has seats on the ISOC
board, it has complete visibility into the budget, and other specific
formal controls. That is, it's not a distant, separated relationship,
but one where the IETF is already deep inside. To a great extent, the
IETF has as much control over ISOC as it would have over a new
organization it might try to establish. And speaking for myself, if the
IETF needs greater visibility and/or representation within ISOC, I'd
support whatever changes in ISOC governance are needed. ISOC was
founded to be a home for the IETF and it has functioned in that role
since its inception. And a large percentage of the ISOC members and
board are people who have grown up in the IETF.
2. I have trouble understanding option C. It's a lot of mechanism that
doesn't have a clear purpose. We can dig into the details as deeply as
desired, but I truly believe it's a quagmire. Some people who advocate
it view it as merely a glitzier form of A or B. Others view it as a
version of D dressed up so it doesn't seem quite so confrontational. I
think it creates a whole series of problems and doesn't provide any real
protection. If a real breach does develop between the IETF and ISOC in
a way that doesn't get addressed sensibly through the very strong
structures and protections that are already built in, then the IETF can
walk away and create another legal home, develop its own source of
funding, and take on the full load of being a separate and independent
organization.
3. As I understand it, the administrative restructuring effort was
stimulated by problems with the support functions. The problems had
reached the point where the IETF's work was not getting done.
Discussions with the supporting organizations, particularly
CNRI/Foretec, didn't resolve the problems, at least in part because the
contractual and management relationship wasn't clear. I believe there
is now clear agreement that much cleaner management and budget authority
is required. One way or another, there will be unified management in
place to oversee the support functions. That much is good. What is not
so good, in my opinion, is the lack of focus on what needs to be fixed
in the support functions. The IESG area directors are perhaps the most
affected and have the most knowledge in this regard, but I'm sure
working group chairs and many others have much to say. Irrespective of
whether one chooses option A, B, C or D as the framework, the real work
will come in restructuring the support functions and working out the
details of who does what. I've been surprised at how little discussion
there's been along this line. If things were broken enough to bring the
relationship with the support functions to a head, then surely it's time
to lay out what needs to be fixed.
What aspects of the IETF's operation hasn't been served properly? Some
possibilities that come to mind are:
O Turn around time in the Secretariat, IANA and/or RFC Editor hasn't
been fast enough.
O Information flow hasn't been timely or accurate.
O Coordination among the support functions and between the support
functions and the IETF hasn't been satisfactory.
O Automation has been weak and stovepiped.
I can't speak authoritatively about these, and perhaps I've missed some
important questions. But it seems to me that this is where we need some
attention with constructive suggestions on how to proceed.
(I left out discussion of the meetings. This is also important, and
there has been some specifics discussed on this list regarding meetings.
This is good.)
Steve
-----Original Message-----
From: ietf-bounces(_at_)ietf(_dot_)org
[mailto:ietf-bounces(_at_)ietf(_dot_)org] On
Behalf Of Wijnen, Bert (Bert)
Sent: Friday, September 03, 2004 6:07 PM
To: sob(_at_)harvard(_dot_)edu'.cnri.reston.va.us; ietf(_at_)ietf(_dot_)org
Subject: RE: Options for IETF administrative restructuring
Scott writes:
... snip ...
I think that option C brings little useful to the table. I fail to
see that incorporating the organizer of the IETF admin functions
solves any existing problem that is not better and more
easily solved
by options A or B. Option C mostly adds the complication
and expense
of creating a corporation whose purpose almost no one
outside of the
IETF, and I expect few inside the IETF, would be able to
untangle from
the purpose of the ISOC.
Scott, my personal view on this is that if we kust look at
TODAY, then
you are correct. And so today I could live with scenario A or
B (I think
I prefer some form of B in that case).
However... in the very distant past, I am sure that the
relationship with CNRI was very good too. And I am sure that
at that time we were happy and that we assumed nothing bad
would happen (if we actually even though
of it). And one would think that if things were to get in a shape or
situation that we (IETF community) no longer find acceptable,
then we would expect that we (or our leadership) takes action
or comes with
proposals for action. Oh well... not so easily. It drags on for long.
When I joined the IESG in 1998, I saw little if any problems
with the secretariat or CNRI. I did not know how meeting
revenues were being used
to support the secretariat and all that. I learned that our
IETF chair at
the time (together with you (Scott) if I recall correctly?)
were negotiating a contract with CNRI. The response that came
back was that it needed some more review on CNRI side and
things were not easy... and a year or so went by. The next
rounds we needed the contract more, because the finacial side
was not transparent and not reported in a timely fashion and not in
enough detail. But again there were all sorts of unclear
issues. And so we went on for 6+ years (yep I have been on
IESG that long). And in the meantime the working relationship
between IESG and Secretariat also suffered. A lot of that had
to do with increased workload, but also with the fact that we
did not have a good contract in place that documented what
the responsibilities and tasks were/are of secretariat and
how we (IESG and rest of community) interface/interact with
secretariat. Let alone that we (IESG/IETF) did not have
control over finances so we could set priorities on specific
pieces of secretarial work that needed improvement.
Now the important lesson I learned here is that I think
initially there were good intentions and good relationships
between CNRI and IETF. But they degraded very slowly year
over year and we still have no contract or MoU or such in
place. And it had to become really bad (my preception) before
we (IETF/IESG/IAB) finally took action to investigate and
document our major concerns in RFC3716.
In other words... things creep on you very slowly. And if you
things go well
for a long time and then start to deteriorate very slowly,
then it is tough
to stand up and cause trouble and REQUIRE/ORGANIZE change in
a commonly agreed manner. Specifically if one side is not
eager to move swiftly.
Now... I assume and hope we will never get into such a
situation with ISOC. But I think that is what we (IETF) also
assumed/hoped many years ago with CNRI.
If we were to go for option C, then in my personal view, it
would have the serious benefit that we are ALWAYS (from day
1) responsible to make sure things work well. And we need to
re-negotiate every so often if we want to keep the
relationships that we have or if we want to change them. So
in my view we would run far less risk to ever get in a
similar situation as where we are today. Yep... initially it
will cost us some more money and effort I suspect. But I
think it is worth the price.
Note also, that I DO want the IETF to keep its curent
relation with ISOC and that I also DO want the new "IETF
secretarial support function" to have very strong relations
with ISOC. We can and should work that out and then document
it so we all know what the relationship is.
But... this is just my personal view.
I hope it helps a constructive discussion.
Bert
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf