Let me clarify.
Keith's fear of IESG being besieged with requests for IETF adoption
of any work done outside the IETF without a WG Review is bogus as
long as all work to be adopted must go through THE IETF WG process
before it gets to the IESG.
the conern isn't just about IESG resources, it's about IETF as a whole
becoming even more bogged down than it already is by trying to fix
mistakes made by people who were so lacking in clue that they couldn't
get a WG created in the first place - so they decided to try and bypass
the IETF process and get IETF to retroactively adopt their work later.
it's far easier to fix a mistake just after you make it than it is if
you invest a few more man-years in that same direction.
But, to suggest that it must never have been worked on elsewhere, if
you want the IETF to bless it,
no, that's not what I said. I said that it's a bad idea to encourage
people to take their work elsewhere if they want IETF to bless it.
in fact IETF has at least occasionally taken on work started elsewhere,
sometimes with good results. it's not that it can't be done, but it's
not a blanket strategy that should be encouraged, as you seemed to imply.
All you are really arguing about is whether the OPES BOF GROUP can
have meeting space at the IETF meeting,
you're not even in the same solar system.
first, this discussion isn't just about OPES.
second, WGs take a LOT more resources than meeting space. (though meeting
space is indeed precious) they take the time of ADs who monitor the groups,
and chairs and participants who try to make sense of what is going on.
they divert energy from these individuals who might otherwise be working
on other (perhaps more useful) things. they take resources from the
secretariat. and they take energy from folks within and without IETF who
assume that the WG is somehow indicative of IETF direction in that subject area,
and who waste their energies trying to understand that WG's work and reconcile
it with sanity.
third, "problem" WGs - that can't figure out how to work and play well with
the rest of IETF - take FAR more than their share of resources. get rid
of a few problem WGs and the organization as a whole becomes far more
effective.
not whether or not they can
work on it elsewhere and bring the work forward when it is more fully
baked.
of course people can do work elsewhere and bring it forward when it
is more fully baked. but the best time to bring the work forward is
not when it is fully baked, but when it's still easy to change the
recipe. and people who can't describe what they're trying to cook
in the first place probably aren't going to be able to convince IETF
to eat what they've cooked even after they claim that it's done.
Of course they can work on it anywhere they want, and time that they want.
But, the attitude displayed here clearly says that one should
"do it all inside the IETF!"
Nope. The attitude is: if you can't make a good problem definition and
a good case for working on it, you need to go off somewhere and do more
work on *that* - NOT on the protocol design.
I'm not going to comment on the rest of Stef's message, except to say
that if you see any smoke here - it's coming from what he's smoking.
Keith