--On Thursday, May 06, 2010 16:15 -0700 David Morris
I think the number of meetings 'registered' for is a poor
criteria for familiarity with IETF culture and more important
familiarity with the participation of the potential nominees
being considered for leadership roles in the IETF.
In the pre-day pass days, I paid full fare more than once but
only attended a day or two of each meeting. This year, when I
used a day pass, I might have opted to attend two WG meetings
on two days requiring regular registration but not seen any
more of the IETF culture than I did with a single day pass.
I have some sympathy for your position and, indeed, can figure
out all sorts of ways by which the "3 of 5" criterion could be
fine-tuned. I also know people who attend few meetings in
person whom I'd rather have on the Nomcom (because of knowledge
of the IETF culture, people, etc.) than people who regularly
attend meetings but who have managed to escape having clues on
those subjects. Maybe, as we get better, or aspire to get
better, about remote and "parachute in" participation, we should
completely reopen the question of Nomcom qualifications and make
provisions for alternatives to regular meeting participation.
Alternately, since participation in multiple WGs with different
styles gives a lot more information than just seeing one (no
matter how many days one pays for), perhaps we should tighten
the rules by requiring active participation in more than one WG.
But it seems to me that it should be possible to clarify the
relationship of an experiment to the Nomcom process without
either (i) creating a back door through which to open the Nomcom
selection model or (ii) taking up lots more time than creating
the experiment itself did (although I believe that the IETF Last
Call that the IESG has initiated is appropriate and necessary).
I also find the idea that someone would plan to attend for a day
but make the decision as to whether to get a day pass or pay the
full registration fee based on Nomcom eligibility mildly
appalling. Fortunately, I think it is also unlikely, at least
In addition, if we were going to start tuning, I note that there
are things other than the Nomcom, such as signing recall
petitions, for which we use Nomcom eligibility as a criterion.
If we were to make day passes permanent, it is possible that
we'd want to make different decisions about Nomcom-eligibility
and Recall-eligibility (the IESG should be even less involved in
the latter decision, for obvious reasons).
In those respects, I think there are two things about the
proposed IESG statement that are, in retrospect, not quite
One is that the entire second paragraph ("The IESG observes...")
is irrelevant and distracting. The IESG is welcome to observe
anything it likes, but the IESG doesn't get to second-guess RFC
3777. All 3777 says (See Section 4, paragraph 14) is what the
requirement is. It doesn't say, e.g., "these are the reasons
and, if other arrangements come up that seem to support the same
reasons, they might be used as justification to vary the rules
without formally updating BCP 10". So I would encourage people
to ignore that paragraph and the IESG to drop it.
The second is a little harder to explain. It seems to me that
the real problem here is that, other than in 3933, we don't have
rules about the scope of experiments. While this particular
instance may require quicker action, it seems to me that we need
a clarification that _no_ experiment can be initiated that has
process effects unless there is a document (such as a 3933
proposal) that explicitly describes the intended process
experiment component. That would essentially prohibit creating
"day pass" as an alternate form of registration without sorting
this type of thing out with the community. That idea has rough
edges, but such ideas typically do.
Finally, as Dave Crocker pointed out, complexity in our
operating rules rarely serves us well. Whether the discussion
is about this case or about Nomcom qualifications more
generally, we should not try to do enough hair-splitting to
cover every possible case... if only because we will get it
wrong and then require even more hair-splitting.
Ietf mailing list