Date: 2005-05-18 20:22
From: Dave Crocker <dhc2(_at_)dcrocker(_dot_)net>
Interesting discussion; I'm focusing on the specific suggestions for
change made by Dave.
Working groups are expensive. Very expensive.
Not just in iesg time, of course, but in quite a variety of resources
consumed.
True. But sometimes the alternative (i.e. no WG or premature termination
of a WG) is more expensive or entails other serious problems.
Having lots of working groups seems to mean that a) we give very
poor support to them, and b) we hassle them when they are done.
Or not quite done, as the case may be. [specifically when there is work
needed to advance a Standards Track specification; there are instances
where a WG has no hope of ever getting done without serious reorganization,
possibly including rechartering or invocation of RFC 3934, etc. -- in such
cases "hassling" or more serious measures may be necessary]
If the IESG is really concerned about quality, then it needs to both
faciliate
and apply pressure at the beginning of the working group and continue these
during the life of the working group.
The silliest thing we can do is permit poor charters,
A close second is after an accepted call for rechartering, to allow the WG to
continue to wander aimlessly without actually rechartering (this HAS happened).
floundering working
groups
There exist management tools to guard against this (milestones are supposed
to be one of those, if/when used with status reports and other project
management tools). Many of the tools are designed to bring problems to
light early (i.e. to prevent floundering) by fostering good communications
up the management chain (in this case from WG Chairs to the cognizant ADs).
and then veto their documents after 3-10 years.
That's a symptom of a WG that has been in deep trouble, trouble which has
gone unremedied. When the damage is that severe, sometimes there is no
alternative but amputation, and in some cases, euthanasia.
So, what sort of support?
1. Much, much better charters. For example, we do not even try to enforce
the
requirements specified in the Working Group Guidelines RFC.
A start might be a prototype generic model charter, identifying common items
which should be incorporated and potential trouble spots so that those can
be addressed appropriately during chartering and rechartering. The first
step in generating such a model charter is collection of examples of what
has worked well and what has failed dismally, from present experienced and
past ADs.
2. Much better oversight of new working group chairs, to ensure that the
group gets traction on its effort.
Yes, and oversight (possibly partially delegated to the Chair(s)) for other
key WG positions, appointment to which (e.g. Consultant) may be required
(positions as listed in RFC 2418). E.g. perhaps a gentle suggestion from an
AD for a separate Secretary if the Chair seems to be having trouble keeping
records as well as managing the group.
3. Insistence on review of major decisions along the way.
Yes, but balanced against micromanagement. When a major decision can be
anticipated, a decision point should be entered as a milestone. Once a
major decision has been made, it should probably be recorded in the charter,
so that the same issue isn't constantly re-debated, particularly when there
are many changes in WG Chairs or ADs over the life of the WG.
4. Reorganization of IETF management to handle the current load and
diversity of participation better,
Perhaps Directorates or other assistance to manage large numbers of WGs.
with a particular eye on finding ways
to prevent single-person vetos.
And single-person overrides. And a screaming horde of partisans with no
substance behind the screaming. Also, in these cases and other instances,
careful attention to possible conflict-of-interest issues. I'll repeat
some words of wisdom from a past WG Chair:
One screaming
person does not indicate that there is no rough consensus, but one or
two well-reasoned arguments against a screaming huge crowd does. And
a huge number of "I'd prefer X, but I couldn't care less" votes
versus 2 or 3 well-argued "X will spell doom for the Internet, and Y
will save it" votes *is* rough consensus for Y over X.
That ought to apply to any collective decisions, not only to WG discussions.
And there needs to be a recognition that "a screaming huge crowd" (or not
so huge, as the case may be) might represent a vested interest who believes
that its ox is about to be gored (or perhaps shoved aside to make way for
a freight train).
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf