Re: [ietf-nomcom] Changing the candidate selection model

[John C Klensin <john-ietf at jck.com>]

--On Friday, June 19, 2009 19:29 +0000 Christian Huitema
<huitema at windows.microsoft.com> wrote:

> I understand the desire to not have term limits, but such
> limits are actually the best way to avoid the failure mode I
> fear most, the "old boyz club". If there is not some kind of
> limit, or at least encouragement to the nomcom to "get new
> blood", incumbents can be there for life, as long as they are
> not grossly incompetent, or bored.

Christian,

I agree with you about the motivation for term limits although I
find them pretty distasteful.  You might recall that the
now-ancient predecessor to this draft, while not actually
including term limits, encouraged the Nomcom to think in terms
of "default return" after one term and "default retire" after
two.

The difficulty I see with term limits is that they cut several
ways.  If they are set short enough (but at more than one term),
they become an entitlement: retiring someone earlier becomes a
big deal.  If they are set too long, while they may still be
effective at preventing lifetime appointments, they are not
effective in removing those who have burned out.  And there are
always edge cases in which term limits dictate outcomes that are
not desirable in particular situations.

My sense is that, if we look at results, the Nomcoms have been
reasonably good at getting rid of the nearly grossly incompetent
and that they, in combination with incumbent good sense, have
been reasonably good at getting rid of the bored and exhausted.
Where they have been less successful is in detecting the
incumbents who performed well early in their time in office but
who are well along the path to burnout, who have developed
abusive behaviors, have developed an exaggerated view of their
own (or the IESG's) knowledge or importance and hence are in
need of retirement before things get worse.  Lack of success in
that area --possibly combined with the "possibly better than the
alternatives" attitude that I discussed in my response to Joel--
has resulted, IMO, in several people being reappointed to serve
one term too long.

Our hope is that giving Nomcoms a clear and separate focus on
incumbent performance without the distraction of thinking about
possible replacements will permit better input and analysis of
trends in incumbent performance and hence retirement of those
who are showing signs of burnout or cumulative poor behavior
before those symptoms develop to the point of causing damage.

There is a risk in the approach which I'm surprised that no one
has noticed: If a Nomcom got lazy, returning the incumbents
without evaluation could save them a lot of work.  However, I
don't think our experience indicates that Nomcoms have any
propensity to get lazy in that way, so I'm not concerned about
it in practice.

    john