--On Tuesday, 26 April, 2005 14:23 -0400 Jeffrey Hutzelman
<jhutz(_at_)cmu(_dot_)edu> wrote:
So, I chose to cite a common rationale used by the nomcom
("there were no other choices") and suggest that the regular
occurrence of that rationale is sufficient indication of a
deep, serious problem that needs addressing.
This is indeed a problem. The time commitment required for
these positions is so large that it is essentially a full-time
job. That means the nomcom has to try to find someone who is
qualified, is willing to set aside whatever they're working on
for 2 years, and has a source of funding for that period of
time. Few employers are willing to provide that kind of
support, and few individuals (or at least, few individuals
qualified to serve in the IETF leadership) can afford to do so
on their own.
This is a problem that has been mentioned many times, with
many proposed solutions of varying practicality and
effectiveness. I'm not going to comment here on any of those
proposals, but I do think the problem needs to be addressed.
This is not easy, though, and will require some study and
significant involvement from present and past holders of those
jobs.
Jeff,
There are some ways to fix, or at least improve, parts of this,
but there doesn't seem to be any will in the community to do so.
I'm not going to say much that wasn't said during the Problem
Statement exercise, but perhaps a review is in order.
First, the IESG job has become so burdensome in part because of
decisions by the IESG about how much work they need to do and to
what level of detail they need to address documents. To take a
handy, but deliberately non-specific, example, if the IESG
considers it appropriate to hold a document up over the wording
of a title or abstract, and to debate that wording, the cycles
that go into the discussion are, however indirectly, a
contribution to the IESG's becoming a full time job. I would
rather see the IESG fix that than having the community apply
whatever blunt instruments it has --and there are some signs
that the current IESG might make some steps in the right
direction. But, to the extent to which the Nomcom selects
people who are happy having the IESG be a full time job, rather
than dedicated to getting that workload down, the Nomcom and the
community are endorsing what we then complain about.
Second, there are well-known phenomena in many political systems
that lead people to say "well, that body consists mostly of
scum, but my representative is a good guy". The US Congress has
turned into a particularly striking example, but it is by no
means unique. The attitude creates a very strong bias toward
reselection of incumbents. I suggest that the community, and
the Nomcoms, end up suffering from the same phenomenon and that,
if we don't like the symptoms that seem to come with people who
have been on the IESG for too long, we need to start thinking
about ways to adjust the system to prevent the "too long"
situation.
Finally (at least for this note), the rate at which incumbents
are returned if they are willing to put up with more punishment
suggests that Nomcoms are asking not whether someone new could
do the job about as well but whether they could do a better job
than the incumbent. Unless an incumbent is perceived as having
really screwed up, the answer to that question will almost
always be "no". If nothing else, IESG roles are complex and
anyone new will lose some number of weeks or months reading into
the role. Also someone new is always an uncertainty and it is
rational and normal for, e.g., a WG Chair to advise the nomcom
that a particular AD should be returned based on logic of "I
have figured out how to work with this one, even though it is
sometimes difficult; I wouldn't look forward to breaking in a
new one, risking a worse relationship, etc." In addition to
the "too long in position" issues raised above, this interacts
with the workload and volunteers problem: Indicating
availability for one of these positions, and then staying
available, is itself a big commitment. That is a commitment not
just for an individual but for a company. It is tolerable if
there is a reasonable expectation of getting the position and
having an opportunity to serve. But, if the odds are high that
the incumbent will be returned no matter what, the incentives to
get the needed approvals and resources lined up are really very
poor. We are hence seeing problems convincing people to put
their names in the pot against the name of an incumbent who is
doing even a passable job... and I predict that will get worse.
There are ways to fix, or at least significantly improve, things
along each of those dimensions. But I don't think the community
is really interested enough to consider them seriously, much
less push them through. And that, I suggest, gives us exactly
the leadership structures and behaviors we deserve and
presumably want.
john
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf