Re: [ietf-nomcom] NomCom 2023 IETF Chair/GEN AD Term Reset Proposal

[John C Klensin <[e-mail]>]

Martin,

With the understanding that I'm often in the rough about these
things these days but I hope not in this case...

I think a a two-year term is entirely reasonable and that
neither the one-year option nor juggling things to allow a
nearly three year term makes sense for reasons your note did not
mention (I assume deliberately).  For the one-year option, we
have frequently had ADs tell us that it takes several months --
sometimes six or more months -- to really get up to speed and
that is usually with the assistance of a co-AD who has been on
the job for a while.  We've been lucky with many IETF Chairs,
including Lars, who have come up to speed much more quickly, but
intentionally getting ourselves into a situation where an IETF
Chair is just coming up to speed and then has to be distracted
by end of term issues just feels like an opportunity for a
self-inflected wound on the community.  Conversely, forcing a
three-year term on someone who might not work out or who might
be able to line up the resources for a two-year term but not
three would risk problems for the individual and the community
and might reduce the pool of possible candidates.

While I prefer the scenario Lars announced of his stepping down
in March, there is another possibility neither of you mentioned,
which would be for him to resign effective as soon as a
replacement can be found, having the NomCom handle finding a
replacement on an accelerated basis, and then, as an exception
due to the circumstances, appoint Lars as Chair Emeritus and
allow him to assist with the reading-in and transition
processes.  That would also allow relieving Lars of some of the
Chair responsibilities much sooner.  Based on what I know, not
my preferred solution, but I don't know all the facts and it
would be a logical possibility.

More generally, thanks for approaching the issue this way.  This
is obviously an exceptional situation.  Exceptional situations
happen and, by definition, the next one will probably be
different.  I think we are far better off dealing with them as
exceptions, proposing something sensible for the details of the
particular case, and getting community input and signoff, rather
than changing procedures reactively in ways that create more
specific rules and risk tying our hands the next time something
unusual happens.  

FWIW, I see most of draft-eggert-ietf-chair-may-delegate in the
same light.  Viewed as a way to deal with an exceptional
situation of which the community is just now becoming aware (and
a quibble or two aside), it makes perfect sense and we should
not need a BCP, updates to multiple documents, and the long time
that takes to allow it.  Viewed as a more permanent change, it
raises (but does not appear to address) serious questions of
accountability, job descriptions for other ADs, etc., possibly
including the optics of the IETF Chair or IESG making a proposal
about how they get their work done and then being the body that
decides whether to approve that proposal (after community input,
of course).  Nor does it seem appropriate for the Chair (or
IESG) to propose a permanent change but then behave as if it had
been approved without public discussion and a public Last Call
first.  At the same time, temporarily rearranging things with an
IESG with which the community is familiar seems far more
reasonable (and safe) than requiring the NomCom to select future
ADs, not just for the skills associated with their Areas, but
for readiness to take on selected IETF Chair responsibilities.
So, IMO, we should deal with it as part of the same exceptional
situation and move on rather that reactively making permanent
changes to important parts of our organizational model.

thanks,
   john


--On Tuesday, September 19, 2023 00:05 -0700 NomCom Chair 2023
<[e-mail]> wrote:

> All,
> 
> In discussions of the process for replacing Lars as IETF
> chair, it was observed that the IETF chair role is
> exceptional. With the vacancy coming at the March IETF, it is
> possible that the term could be reset.
> 
> According to Section 3.5 of RFC 8713, when filling a vacancy
> of precisely one year, the new term is a single year. Though
> this does not apply in this case, a vacancy that occurs after
> the meeting results in a term that is almost three years.
> 
> For other positions, it is necessary to ensure that the usual
> cadence of replacements remains synchronised with other
> appointments to the same position. The IETF chair has no such
> need.
> 
> In requesting that the NomCom fill this vacancy, the IESG have
> expressed a preference for a two year term. However, this
> would deviate from the documented process.
> 
> The question therefore:
>     Should the cycle of appointments for IETF chair (GEN AD)
> be reset? 
> 
> If the term is reset in 2024, someone selected for IETF chair
> would have a two year term. Subsequent IETF chairs would then
> be selected for 2026 and subsequent even-numbered years
> thereafter, or until a similar reset occurs.
> 
> The NomCom would like to solicit community feedback on this
> suggestion.  Input on this process is welcome over the next
> month at [e-mail], ending on 2023-10-20.  As the
> responsible body for RFC 8713, the IESG will be responsible
> for judging consensus and whether this process variation is
> acceptable to the community.
> 
> If the community does not reach consensus, the rules in RFC
> 8713 will be followed strictly. The next IETF chair will be
> seated for one year.  The possibility of term variation will
> be made known to prospective nominees as part of the upcoming
> call for nominations.
> 
> Martin Thomson
> [e-mail]
> 
> _______________________________________________
> IETF-Announce mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/ietf-announce