Architecting, Steering, Chairing [was Re: Longer or more meetings ?]

[Natale, Robert C (Bob)]

Hi,

Referring to Dave Crocker's posting of Mon, 9 Dec 2002 19:04:35 -0800:

dc> Perhaps the real challenge is to find way to do more work online

Doing more work online efficiently and effectively would be both very
helpful in many ways and very impressive in terms of demonstrating
the potential value of collaborative applications based on IP.

dc> and/or more work in design teams that are legitimately responsive
dc> to working group rough consensus?

Yes, that might be necessary too, mainly as a derivative of what I
personally see as the root cause of many of the problems (including
many of the related symptoms as well) --  Lack of active leadership:

   - The IAB does not do enough proactive architecting, whether of
     the normative (long-term), prescriptive (near-term), or remedial
     (post facto) varieties.

   - The IESG does not do enough steering, especially in terms of
     proactive longer-term guidance of Area direction and cross-
     Area interactions.  Here I am talking about coarser granularity
     up-front stuff rather than charter reviews, milestone checking,
     and Last Call document reviews (at which most ADs do a good
     job most of the time, IMHO, considering the workload and the pay).

   - WG leadership is, on the whole, far more inclined to over-
     compensate on the side of "democracy" than on the side of
     sticking to the mission [which, since there is often more
     than one legitimate way to accomplish the mission, often
     means making some executive decisions and pressing forward].

Now, I am in no way suggesting that we should change the basic
nature of the IETF process...just that, IMHO, we may have let the
process drift too far towards its (laudable) ideals and, as a
result, away from the course of external reality.  Again, I am
not arguing that we institute a dictatorial culture...just that
we apply a process correction that results in a greater measure
of leadership from our leaders.

If necessary (and I don't think it is) such an adjustment could
be accompanied by a correspondingly moderate increase in checks
and balances, accountability, and "recalls" for cause.

In the event that my suggestion does not find favor with a
significant number of other participants, I'll fall back to
supporting something like or on the order of Dave's suggestions.
I'm not in favor of a big "problem statement" effort on this
stuff...as opposed to trying out a small number of sensible
process improvement steps and gauging their impact and
repeating as necessary.  The process problems -- whether
via root cause analysis, visible symptoms, or perceived
effects -- are fairly clear and we can look to other
standards bodies for comparisons and contrasts for additional
data points.  Enhancing our output by enhancing our process
in a timely fashion is a crucial goal now, I think.

Cheers,

BobN


Cheers,

BobN