MIXER example (mailing list size/activity)

[James Kempf]

Keith,

FWIW, I support pruning of WGs that are not making progress, or otherwise
reorganizing (like giving the parts of the work that need to be done to other
groups if that makes sense). I think the WG process needs to be managed, and one
aspect of management is pruning when stuff doesn't get done. People need to be
clear about why the work is being terminated, and they need to be given enough
warnings, perhaps a formal review, all of which takes up the AD's time. Perhaps
that's the key issue.

            jak


----- Original Message -----
From: "Keith Moore" <moore at cs.utk.edu>
To: "Margaret Wasserman" <mrw at windriver.com>
Cc: "Harald Tveit Alvestrand" <harald at alvestrand.no>; "Keith Moore"
<moore at cs.utk.edu>; <problem-statement at alvestrand.no>
Sent: Thursday, March 13, 2003 4:56 PM
Subject: Re: MIXER example (mailing list size/activity)


>
> On Thursday, March 13, 2003, at 08:59  AM, Margaret Wasserman wrote:
> >  I guess I'm trying to figure out whether
> > the lack of timeliness of IETF work (which I think we all agree
> > is an issue) is caused by a process problem or a resource
> > problem.
>
> does it have to be an either-or?  it's fairly clear to me that we have
> both kinds of problem (at least many working groups have trouble
> working effectively, which I'd claim is a problem with our process),
> and both kinds of problem affect timeliness.  actually, they compound
> each other, because the failure of working groups to produce good
> quality output in a timely fashion increases the resource drain on IESG.
>
>
> >> I think the number of new issues was actually very, very low; mostly
> >> it had to
> >> do with the way to treat delivery notifications - the rest was just
> >> integrating
> >> and fine-tuning previous work.
> >
> > This might have been a process problem.  Would it have made
> > sense to organize the work differently to allow the parts of
> > the protocol that did have consensus to move ahead?
>
> Not really.  IIRC, the problem simply wasn't an inability to reach
> consensus, there were very few divisive issues.  It was that the two
> people who were believed to be absolutely critical to the document's
> success had trouble finding time to work on document revisions.  I
> didn't help that X.400 is so baroque, and that the nature of gateways
> is that there are lots of weird details to specify.  Bringing on
> another document author would not have helped, because the 'absolutely
> critical' people would still have had to review and comment on the
> other author's contribution.
>
> > Would better WG chair management have fixed this problem?  For
> > example, should the WG chair have fired the authors/editors and
> > found someone else with more time/energy to complete the work?
>
> Probably not.  Again, the two primary authors were viewed as essential.
>   Very few people had the expertise to write that document.
>
> Another question worth asking is whether the WG should have been shut
> down or declared dormant when it failed to make timely progress.  IESG
> is very reluctant to shut down working groups - it makes people angry
> and it looks like failure.  But if there is a shortage of people to do
> the work, maybe the work isn't as important as it seemed at first.  Or
> if the work is important, maybe it's better done by means other than an
> IETF working group.
>
> our working groups are like Maslow's hammer - we tend to act as if they
> were the solution to every problem, even though we should know better.
>
> Keith
>
>