ietf
[Top] [All Lists]

RE: isoc's skills

2004-10-11 12:22:40
Dave,

You are correct we need a detailed documentation of the interface before we
deal with any corporate entity. As I see it the differences between your
opinion and others has more to do with your focus on the interface than the
organizational structure others are commenting on. My initial reaction to
your concern about ISOCs track record was that the IETF itself has even less
of a track record, and a poor one at that. Despite the legal difference
between the Administrative office being a separate corporation vs.
incorporating the IETF itself, the backers of both of those choices appear
to assume the IETF will directly deal with the financial issues because
their arguments against outsourcing all say 'they have a different focus'.
As several people have stated, the IETF participants have a technical skill
set and no demonstrable skills at financial administration. Why then are
people so quick to point out that outside organizations have a different
focus when our internal skill sets don't match the need?

Yes before we go off and sign agreements we need to know what the details of
those agreements will be. We do not however need to know all the details
before we set a general direction. This discussion has pointed out several
characteristics about the IETF as a collective:
We need an accountable entity that is responsive to the financial needs of
administering the IETF. We should be simplifying the funding profile for
supporting both the standards development activities and document
publication. We want sufficient autonomy to act as if we independent and all
powerful, yet we are quick to dismiss political & legal issues as
irrelevant. 

Others may add to the list, but taken collectively it should be clear that
scenario C is fundamentally the end of the IETF since it only addresses the
first item. Alternative contractors might cover the last item as well, but
ISOC provides the path that covers all three. Yes we still need details, but
that should be easier to do if we know where we are heading.

Tony 


-----Original Message-----
From: ietf-bounces(_at_)ietf(_dot_)org 
[mailto:ietf-bounces(_at_)ietf(_dot_)org] On Behalf Of
Dave Crocker
Sent: Monday, October 11, 2004 10:00 AM
To: Brian E Carpenter
Cc: ietf(_at_)ietf(_dot_)org
Subject: Re: isoc's skills

Brian,

 ... I believe that policy concerns are best addressed by
 ISOC.  Because ISOC's role in the standards process is at
 one remove, it can work to educate legislatures and
 administrations without appearing to favor one
 participant over the other.
 That sounds wonderful, except that ISOC has no significant
 experience in that work and that work requires skill and
 experience.

 ISOC, like IETF, is largely a volunteer organization as far
 as this sort of work is concerned.  If the community wants
 ISOC to take such a role, the community will also have to
 provide the volunteers.

I do not understand your point.  We are going to hand over all
administrative responsibilities for the IETF to a volunteer
effort?

My guess is that the difference in our views is the difference
between theory and practice.  I am making an assertion about
ISOC's actual skills, based on its history of performance.  You
appear to be making assessment based on the theory of its
framework, or potential, or somesuch.

ISOC "can work" to do all sorts of things.  The question is what
has it demonstrated skills in?  If the IETF is going to increase
its dependence on ISOC, then the IETF needs assurances that ISOC
can perform the tasks that the IETF needs.

When we step away from theory and rhetoric, I believe we find
that ISOC has literally none of the necessary skills.  To the
extent that it has attempted relevant activities, I believe its
track record is poor, at best.

In general what I have noted about the discussion of
organizational structure/home for the IETF is that it pretty
complete lacks clear, precise, stable specification of the job we
want done.  So when I talk with individual about it, the details
of their response float all over the map.

My experience with this sort of variability in responses is that
there is some sort of mystical hope that making some sort of
change will have major benefit.  However no one is able to state
any of this concretely.  And the outcome of such a process is
pretty much certain to be disappointing, at best.

We want to delegate all sorts of responsibilities to ISOC; or
maybe we want ISOC to delegate them to 'experts'.  We want ISOC
to handle the IETF budget, but we do not believe we are handing
ISOC any additional power over the IETF. And so on.

I have tried to list specific problems with the IETF and note
that none of them will be improved by the current structural
work.  Most will not be affected at all.  What I have noted is
the lack of specificity in any responses about this.  It is
significant that this line of enquiry is not pursued further.

As nearly as I can tell, the IETF leadership's current concern is
that CNRI/Foretec have too much power and too little
accountability.  What is being proposed is, frankly, hand over
exactly that same role to ISOC.  CNRI would be replaced by ISOC.

Now the obvious and vigorous responses to this assessment is that
there will be vastly greater accountability, that there will be
an MOU, that ISOC are good people with good intentions, and so
on.

All of that might well be true, but it ignores that
organizational behavior reality that different organizations
always have different goals, at some point.  A relationship needs
to be developed with very precise and appropriate specification
of the details to that relationship.

To that end, I suspect the single most important piece of work is
the MOU.  Rather than discussing high-level structural
abstractions, we should be discussing the precise contents of a
specification for the job we want done.

When we have agreed on those details, we can present them to all
sorts of people and organizations, including ISOC (and, by the
way, CNRI).  What should ensue, then, is a negotiation for
performance of those tasks.

Where is the public discussion and refinement of that work?


 As has been commented to me repeatedly in recent months,
 when someone in government wants to obtain advice about
 the Internet and about Internet policy, they do not
 regularly consult ISOC. ISOC does not regularly testify in
 Congress.
 ISOC is international and is currently active in WSIS, the
 international debate including Internet policy issues. If you
 want ISOC to take part in national policy-setting in your
 country, it's in your hands. That's one of the things ISOC
 chapters can do.

The reference to the US Congress was an exemplar.  And
"participation" in WSIS could mean lots of things.  I have gone
to some ITU meetings, but that does not place me in the role of
providing policy leadership to the ITU.

If someone is going to claim that ISOC is in a leadership
position for Internet policy-setting groups, then it would be
helpful to see description of its activities in the regard that
show actual leadership.  Going to meetings is not enough. Running
a workship is not enough.  Policy-setting is an ongoing political
dialogue.  Where are ISOC's political skills?


 More generally, as folks postulate spiffy functions for
 ISOC, it might be worth asking where ISOC's expertise for
 that function has been demonstrated.
 That includes minor items like operational administration
 of a standards body.

 Well, nobody has demonstrated that skill as far as the IETF
 is concerned, because we've never put *all* the
 administration into one place.

I am afraid that this is another example of how muddied
discussion of this topic is.  (And that's not a criticism of you
in particular.  I think the issue is the entire tone and content
of the community discussion and, apparently, the leadership's
internal discussions.)

I did not say we do not need to rationalize things.  I think the
portion of the Advisory Committee's report that did problem
analysis was excellent.  We need to rationalize accounting,
accountability and authority.

My point is that there seems to be a mystical belief that we can
hand everything over to ISOC and it will solve everything.  In
fact it will solve nothing.

The solution(s) lie in the details that can only be accomplished
by the IETF, itself, unless we simply want to make the IETF a
wholly-owned subsidiary of the organization we hand ourselves
over to.

Obviously, lots of folks do not agree with my assessment.  What
is lacking is anything concrete and detailed to explain why.


 But ISOC has administered
 itself for the last ten years, through good times and bad.

Frankly, Brian, this statement is rather scary.

First of all, the ability for an organization to administer
itself has nothing at all to do with its ability to perform the
job the IETF is considering giving to ISOC.

Second of all, ISOC's extremely problematic history is not likely
to serve as a recommendation for its ability to perform whatever
the heck it is we are currently asking of it.

d/




d/
--
Dave Crocker
Brandenburg InternetWorking
+1.408.246.8253
dcrocker  a t ...
www.brandenburg.com



_______________________________________________
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


<Prev in Thread] Current Thread [Next in Thread>