Straw process outline for dealing with structural and practices problems

[John C Klensin]

Avri, Elwyn, Brian, and WG participants,

I want to repeat and expand on what I said Thursday evening,
in the context of this proposal and a few other things.

Before I go on, I want to stress that I think the "problem
statement" effort has been very useful.  And I would like to
personally thank the people who have spent large amounts of
effort drawing it together.  The question in my mind is how,
and when, we translate study, design, and proposals into
action.

Sooner or later, as the saying goes, push has to come to shove.
One of the things I took away from Thursday evening was a lack
of community enthusiasm for more WG efforts, special design
teams, etc.  My personal take on that at this point is that:

	(i) Elwyn's proposal is an interesting mix of design team
	and directorate ideas.  But the thing those ideas have in
	common is that they are subject to IESG (or WG Chair)
	management.  Elwyn's proposal, as I read it, is for a group
	that is really independent of either.  That makes it just
	the right thing if we are really in a meltdown state.  But
	if the percentages Thursday night are representative of the
	community, then we really aren't: the support for the "we
	are in serious and urgent need of fundamental structural
	change" (my paraphrase) questions was definitely
	underwhelming.

	(ii) Support for "more WGs", "more design teams", and
	"appoint someone to decide" was also fairly low.  I don't
	think that makes a strong case for variations on those
	themes, although I could be wrong.

	(iii) While the question wasn't asked explicitly, I am
	convinced from what was, and was not, said on Thursday
	that, had Harald asked whether the community wanted the
	entire IESG to resign, effective as soon as replacements be
	found, that proposition would also have received little
	support.  Add to that the lukewarm (at best) support for
	the proposition that we need major structural change to
	survive in the fairly short term, and I don't think 
   we are in the middle of a
	meltdown.  We can cast this situation in terms of "trust",
	or other language, but the bottom line seems, to me, to be
	that we are still willing --perhaps anxious-- for the IESG
	to manage the IETF process and to exercise some leadership
	about it.

So, assuming we consider all of this important, sooner or later
a proposal has to be turned over to the IESG, and the IESG
needs to take action on it (or not).  And these efforts are
either are important or they aren't.  If they are not important
enough to accept some delays in the IETF's technical output,
then no proposal, including Elwyn's, is going to work (or be
worth the cost... if it involves a fairly large number of
people (and even a dozen is fairly large) who could otherwise
be doing technical work, reviewing documents, etc., then the
decision to do it is a decision to slow our technical output
down.  If it involves predominantly people who would rather be
working on procedures than doing technical work, or even are
more qualified to work on procedures than on engineering, it is
a recipe for disaster.  No free lunches here.

And that brings me back to my Thursday evening comment.  I
think it is time to _do_ something, not design structures for
talking about what to do.    I think we should go through the
issues in the problem statement document, one at a time, and
quickly generate and document some possible approaches for
each.  I don't care if we reach a clear consensus about any
given approach, indeed, I think that trying to do that might be
harmful.  We should then pass the package -- issue and possible
approaches together-- off to the IESG with

	* A statement that the community, or at least some
	significant fraction of this WG, considers the issue
	important.

	* A request that they pick whichever one of the approaches,
	or one of their own, makes sense given their perspective on
	this issues.

	* The clear expectation that they will consider the above
	and expeditiously make that decision and implement it, even
	if it delays other work, _or_ that they explain why it
	really isn't an issue or that they couldn't find an
	acceptable solution.

	* We make it clear that IESG members who are uninterested
	or unwilling to engage on those issues are expected to
	resign, that the Nomcom is expected to replace, as a
	priority, any who don't resign, and that any who are
	problematic who are not up for review this year (or who are
	especially problematic) will be recalled.

The bottom line is that we are either serious about this or we
aren't.  And "serious", at this stage, implies that we are
willing to have other things delayed in order to move forward
and that we are willing to fire IESG members who don't want to
play.  The good news, from my standpoint, is that I know the
majority of the IESG fairly well, and they really want to do
the right thing, including working to the community's
priorities
if those are clear and realistic (e.g., "add these additional
things to your workload while speeding up the rate at which you
do other things" is not plausible).  I strongly suspect all of
them do.

All a more complicated process will accomplish, IMO, is delay
and wasted cycles.  Regardless of how a proposal originates
and how specific it is, the IESG is still obligated to figure
out
what makes sense and what doesn't and what can be practically
implemented and what isn't, regardless of the level of
consensus from some group that is, inevitably, not
representative of the community.  So why spend more time
developing procedures to develop procedures, taking people away
from technical work or driving everyone who isn't focused on
process details as-an-end away, producing a less and less
representative "consensus".

       john

p.s. To save generating another note, there will come a time at
which the procedures and models we have discussed about WG
management should be applied to this WG.  Have we reached the
point of diminishing returns and how do we figure that out?
For
obvious reasons, if more energy isn't going to produce enough
more in the way of results to justify the investment, it would
be better if the WG "voted" itself out of existence than to
force Harald to unilateral action.  But, while I can't speak
for others, I expect him to do his job as AD.  And that
implies that, if he concludes that the WG has produced about
all it is likely to produce relative to the cost of having it,
he should shut it down.

regards,
     john


--On Friday, 18 July, 2003 11:05 +0200 avri
<avri at apocalypse.org> wrote:

> I think it would need a consensus process for this group to
> pass this on to the IESG, though of course Elwyn could so,
> or the IESG which I believe is well represented on this
> list, could just pick it up and combine it with whatever
> other processes ideas they were looking at.
> 
> In the meantime I think it a good idea for this group to
> keep talking about this and any other process suggestions
> and see if we can come up with something we can reach
> consensus on.  If the IESG reaches a decision point before
> ...