Hello,
The IESG defines the job and the IESG 'operates' the model. So yes,
formally the IESG controls this issue.
Mutter.
I suppose the seated IESG defines what it does.
Or more precisely, each member of the IESG decides what (sadly) he should do to
best serve in the role.
But don't see the "job requirements" put out by NomCom as the IESG placing
requirements.
The NomCom asks the IESG what skills and tasks are needed in the view of the
current IESG.
The IESG serves up this information and NomCom can do with it as it likes.
More important, the community can and should tell the NomCom what it wants the
IESG to do and not do.
However...
Nomcom can apply pretty much whatever criteria it wants to the selection
process and Nomcom has been know to have some private negotiations with
the IESG. Nothing as profound as the change we are suggesting, but
still...
Exactly.
It would be better for the IESG to take the initiative here and formally
and publicly re-define the job, but it has so far ignored such requests.
Which cuts into what Peter says since from where I sit there are many calls on
the time of an AD and it would be useful to know which tasks the community would
like me to not do (I think that I and my chief sponsor would both be happy if I
spent more time working for them and less working as an AD).
It might be helpful to talk about what could change, such as:
1. Less/no time on document reviews.
Right. Ask:
Do the reviews currently done by ADs make a difference to the quality of
individual documents or the overall canon?
Would we be happy to publish RFCs with lower quality?
Are there other ways to reliably improve the quality (like, for example, the
community doing proper reviews of their own work)?
Could discussions of Discusses go more smoothly and take less effort?
Note:
I spend more time (factor of more than 2) reviewing I-Ds when publication is
requested (from a WG) than I do reviewing at IESG evaluation.
2. Less time managing working groups.
Right. Ask:
Does the management of working groups improve the output?
Would we be happy to allow WGs to "muddle through"?
Could WG chairs somehow (magically?) take more responsibility and get stuff done
(some chairs do)?
Note:
A few of my WGs/chairs take most of my management time.
I don't put a lot of time into managing WGs unless something goes badly wrong.
3. Fewer working groups to manage.
Right. Ask:
Should an AD close a WG with a small number of people doing productive work
slowly?
Should we refuse to open new WGs until we have finished other work?
Could we reduce the number of WGs per AD by increasing the number of ADs?
Note:
Even if management of the WG is not issue, each WG produces documents to be
managed, so reducing the number does have a direct impact.
4. Fewer "extra" tasks (e.g., interfacing with other SDOs).
Not sure that interfacing with SDOs is a big time drag.
But other things do eat into time. These are mainly IETF management tasks.
Arguably this is what the IESG is for. If it isn't you had better tell us!
5. Less/no involvement with BoFs and other early-stage efforts.
Right. Ask:
Should BoF proponents be left to flounder or should an AD help them to get their
work going?
Is a BoF shepherd from the IAB enough?
Could other people from the community fill this role?
6. Less/no involvement with creating the schedule.
Frankly, with the new tooling, this doesn't take more than a few hours per IETF
meeting cycle.
And that does point up a good question to ask the IESG in general - what
additional tools could significantly reduce your work load?
Perhaps some time and motion studies are in order to figure out how ADs
spend their time. In my experience, document reviews required a major
time commitment.
And, since I do track my time at quite a fine granularity...
Email takes up a large lump.
Got to skim the mailing lists for each of my WGs every week.
Got to skim the mailing lists for other WGs in my area every month.
Got to skim IETF-announce and IETF-disgust daily.
Another 15 or so IETF lists that I skim weekly.
A few related lists (such as NANOG) that I skim when I can.
Emails directed to me that need attention "at once".
Please don't assume we ADs want this to be full time work.
Please do tell us what you don't want us to do, and what you would like us to do
less.
Consider whether the "IETF management" could be separated from the "document
quality".
Cheers,
Adrian