Re: Looking for Area Directors Under Lampposts

Long thread, and yes, some repetition from the past.

What follows in my personal view. This may or may not match with what
other people in the IESG or elsewhere think. Indeed, as noted, even
within the IESG there are differing views about the role of the IESG.
And that’s good.

I actually do think we have a slightly different situation today
than we have had ten or fifteen years ago. Most of that is
cultural, and also dependent on the people who are currently
active in the IETF and the team in the IESG. Maybe the external
world has had some effect as well. I’m sensing that “professional
standardiser” is a less frequent mode of operation in the companies
that we work for; for instance, people are increasingly expected to
combine that with open source development, and be more tightly
connected with business efforts. Which are good things from my
perspective.

Anyway, I think today the IESG has at least some examples of people
working real day jobs alongside of their AD responsibilities and has varying
levels of delegation and different arrangements. All that is good.

But really, the key issue is “how does the IETF work best and best
improve the Internet”? Not whether a particular person is the
one fixing particular  issues from the documents… or even the
workload…  the end result is what matters. And, here’s the
main thing: if you are in a leadership position you are asked to get
the job done, not to do it in a particular way. You could (try) to do it by
fixing everything coming your way, and writing yourself the specs that
didn’t get written by others but should have. Or you could try to make
the organisation itself work better. I personally believe the latter is
a better approach. Our formal rules do not prevent the different
modes of operation.

And Dave is right in that the quality of the work or the success of
IETF efforts does not _directly_ depend on the individual ADs. Yes,
final document review is a part of their job description, but it is
only a part. Their leadership role should not be seen as a direct line
into fixing particular issues in the output of the IETF. They should
indeed be the facilitators, but that involves many things — as Joel
noted, “there is an important organizational component, respecting
fostering new work, managing the area, or coordinating cross-area
work. “ And simply getting the right people in place as working group
chairs and in other roles is important, as John L noted.

I’m interested the direction of thought that started from Brian’s
observations of serious errors and Bob’s suggestion that we might
want to measure that. For those that do not know, Brian is a very
active Gen-ART reviewer, and those reviewers have one particular
vantage point to what the process produces. (But of course, as
with everything else, not everyone will agree to either a draft’s
content or its review; we only have subjective observations.)
An interesting statistic might be how many issues are brought
up in last call or by various directorate reviews. Or the number
of discusses. Or the number of last call/directorate issues vs.
discusses. My gut feeling is that the number of discusses is
smaller, but I’m not sure. Anybody up for writing some scripts
to do measurements of this?

I also wanted to high light some items from the discussion that I very
much agreed with. Thanks all:

John Levine:

> My non-statistical opinion is that the charter is much less important
> than WG chairs who understand the topic and WG members who can (and
> are willing to) write well.
Indeed. And people involved in general, be they editors, chairs, etc.

Tim Wicinski:

> As a fairly new WG Chair, I have always had the impression that submitting to 
> the IESG a document that had not been through strong editorial review for 
> quality and readability was unacceptable.  Perhaps my ADs have done a good 
> job of setting expectations.
Yes.

Joel Jaeggli:

> When I look at  drafts on the IESG agenda I want to know what happened
> in IETF LC GENART/SECDIR/OPSDIR, what the working group thinks of the
> IPR if any and how the document fits with respect to past or forthcoming
> work, whether registry polices are reasonable or complied with. I'm not
> so interested in technical merit or correctness from my own vantage
> point since I am rarely the consumer for whom the document was intended.
> I am certainly not an arbiter of taste.
That is also my approach.

Joel Jaeggli:

> > More broadly, ADs need to stop believing that the fate of the
> > Internet rests on their shoulders.  For that matter, they need to
> > stop believing that the fate of the IETF rests on their shoulders.
> I don't view it that way. I hope others also do not. With respect to
> the long term viability of the IETF as a place to do work, managing a
> subset of the working groups, bofs, and participants in productive
> efforts is far more important imho then late stage document review.
+1

Dave Crocker:

> When performed well, the AD job is one of helping to make sure that
> community ducks are lined up properly.  They facilitate the process
> of organizing and operating an effort.  They don't "direct" the
> actual work.  They don't initiate it and they don't manage the
> internal working group process nor the technical content that is
> developed.
Right.

Randy Bush:

> all serious workers (in the ietf) are overloaded; that's life.  ADs are
> serious workers, we hope.  what i hear them whining about is incoming
> document quality from the wgs.  if we are going to micro-manage the iesg
> as an olympic sport, perhaps we should focus on that.

I don’t want to put blame in any way to WGs, I’m sure the IESG could also
use more oversight and feedback by the community. And I actually
do think most WGs produce very good output, and in particular,
after community review in last calls and directorates we in my
impression suitable quality level for PS documents.

But the WGs are the primary place for IETF work to happen. It
follows that most significant improvements, issues, and successes
are there.

Joel Halpern:

> Dave, as far as I can tell, your description does not match the roles and 
> responsibilities assigned to area directors by the process.  You have in the 
> past written I-Ds and suggested changes in those responsibilities to more 
> closely match your vision.  The community has not adopted those documents.
You are right, although I note that the ADs have a big role in facilitating the 
overall process but they of course also have a role in final document review. 
Although, as noted, the rules also leave a lot of room on how that is achieved.

Harald Alvestrand:

> In my opinion, the main job of the AD is to get others to do *their* job
> - especially to make WG chairs do their job of making the WG produce
> high quality drafts that reflect WG consensus and help make the Internet
> better. (Yes, that's three wishes.)

Right.

Jari

signature.asc
Description: Message signed with OpenPGP using GPGMail