[rfc-i] RFC Editor structure

John C Klensin john at jck.com
Fri Sep 26 04:53:32 PDT 2008

Olaf, while I think you hoped for a very short answer to these
questions, like "yes", I can't provide one. See below.

--On Thursday, 25 September, 2008 18:17 +0200 Olaf Kolkman
<olaf at NLnetLabs.nl> wrote:

> John, and others.
> I tend to write shorter and less eloquently than yourself but
> let me try to summarize your two perspectives.
> I read the following high order bits in your feedback:
> 1. Separation of the RFC Editor function from the Independent
> stream Editor is a 'no go'. These functions are much to
> intertwined. And, one needs a forceful role to push back to
> the powers that be (the IESG mainly).

I think that conclusion from my remarks is correct.  But I
believe that Craig Partridge's comment about groups with little
or nothing to do is as, or more, important than "too
intertwined" in the long run.     If the issue were only "too
intertwined", one could specify two groups, appoint the same
people to both, and then let them diverge gradually.

> 2. The IAOC/IAB are not in the position to select  the
> Independent Stream Editor, the RFC Editor function, or the
> combined function; they lack the experience and the
> perspective. In the light of that we should be having an
> subject matter expert group to select those functions.

i might choose different terminology than "subject matter expert
group", since historical perspective and continuity are also
important, but yes.

> The reason why I want to see this high order bits stick out is
> because although I think I understand you argument I would
> like to know if it is supported by a consensus. As far as I
> can tell there is no strong consent with these two positions.

I believe that is correct.  I also believe (i) that few people
in the community have thought deeply enough about these issues
to get beyond that which seems obvious and intuitive and the
community has come to accept the assumption implicit in the
Nomcom model, which is that almost anyone in the community with
strong technical skills and the ability to work with others can
effectively perform any of the roles that require specialized
management skills or skills outside the IETF mainstream.
Given that, I do not believe one can get informed and meaningful
"strong consent" about any of these  alternatives.  I believe
that implies that the IAB is going to need to take
responsibility for thinking through and discussing the various
options and making an informed and rational decision rather than
relying on a popularity contest.

> I think that there is at least consensus that both roles are
> supported by what is initially the current RFC Editorial
> board. I think that if we leave the wording sufficiently vague
> the RFC Editorial board can take up both roles or evolve.

See above.

> I
> would think that everybody agrees to not formalize the role of
> the RFC Editorial board to much. But in those discussions I
> have not read that the Independent Stream Editor and the RFC
> Editor function should be merged. What is clear is that they
> will need to communicate to each other structurally and the
> RFC Editorial board might create the necessary glue for that.

The current Editorial Board consists of volunteers, some even
busier than I am.  I can't speak for the others, but I fear that
the sort of choices of either Independent Stream Reviewer or RFC
Editor that would be made by the processes that select the
Secretariat function or Nomcom or IAB appointees could lead to
your having a smaller and less functional Editorial Board to
work with.  See your point 2 above. 

> On point 2, the only point of view that I've seen is that the
> RFC Editorial board should take up the responsibility for
> selection. If that plan would have demonstrable support then
> it needs significant work. But I am not sure that we should
> create yet another formal body within the umbrella of IETF
> organizations.

In what I would consider a more perfect world, the IAB would
have the interest, commitment, experience, and skills needed to
perform most or all of the Ed Board tasks itself.  I don't see
that happening.  Even if today's IAB were willing and able, I
don't see any way to guarantee that these tasks could be
well-supported into the future.  I am also pessimistic about RFC
Editor and Independent Submission functions (separate or
combined) being chosen on either a low-bid basis or by people
who don't have a deep understanding of the roles.  I could be
wrong, but I don't believe that the members of the Ed Board are
willing to sign up to train people in those roles, especially
people who may not be interested in being trained.   That
combination of concerns and conclusions leads me to believe that
the Ed Board is the one potential selection resource we have
that would work.  It is not a conclusion with which I'm
completely happy, but, I believe that getting this done well is
more important and whether or not we formalize another body that
already exists (I also disagree with "within the umbrella of
IETF organizations", but maybe that is a quibble).

> At some point I, supported by the IAB will need to do a
> consensus call. My plan is to fix parts of the model in such a
> way that the IAOC can proceed with its work, working out the
> details on how the RFC Editorial Board works with and
> interfaces with the Independent Stream Editor and RFC Editor
> function can be ironed out. I would prefer to do that next
> Wednesday. Depending on how this thread develops.

Thanks for listening and good luck.


More information about the rfc-interest mailing list