ietf
[Top] [All Lists]

Re: draft-klensin-nomcom-term-00.txt

2005-07-27 06:09:08
I have to disagree somewhat with this line suggesting stricter limits on serving duration. I agree that a lack of bench strength is a real problem that should be addressed.
I suspect that we may have more bench strength than we think.
I strongly suspect that with some of the other changes being discussed (I like the separate review idea, although I think it needs some work) there will be more capability to do more sane jobs.

However, defining the process so taht if we turn out to have insufficient bench strength we produce a disaster seems extremely bad design. (When folks bring me protocol designs like that, I ask them ~are you really, 100% sure that problem can not occur so you don't need to design for it?~) The statements as written in the draft have the nice property that there is some wiggle room if things look wrong.

My biggest worry is the one piece of structure that has no wiggle room. As defined, if the Nomcom in phase 1 decides not to reappoint the incumbent, there is no way to recover if that turns out not to work. I like the idea of considering incumbents on their own. But I can not find a way to make the two-phase system work without severe risk of backing ourselves into a corner.

Yours,
Joel M. Halpern

At 08:42 AM 7/27/2005, Spencer Dawkins wrote:
I too like this draft and agree that having most IESG members serve for two terms is ideal and making it more the exception that people serve for three or four terms. I also like the flexibility it gives the NOMCOM without creating strict term limits.

When someone is "needed" for more than two terms, what does that say about the state of their area?

The IETF is based on the commitment of community participation, rather than the brilliance of individual leadership.

If we do not have multiple, acceptable choices for an AD slot, then we have a deeper problem with the Area (and/or with the job of being AD, of course.)

What would happen if the term limit were firm, with no exceptions?

John Klensin keeps telling me that we do better when we don't have absolute prohibitions in our processes.

In this case, I think he's trying to have enough flexibility that (for instance) if you have just replaced one AD in an area and some event happens that makes it impossible for the newest AD to serve, you don't have to replace the other AD AND the first AD's replacement in the same NOMCOM, just because the BCP says "two terms and out".

I'm OK with this level of flexibility.

Do I think we could survive replacing both ADs in one NOMCOM? It hasn't happened often, but it has happened (isn't Routing the most recent example? and we still have a Routing Area).

I am sympathetic to Dave's concern about areas that don't have "AD bench strength". The draft still allows ADs to serve in a fourth term. If you can't replace an AD after three terms (one for training, one for effective leadership, and maybe one for transition), I think that's really, really bad.

I would hope most NOMCOMs would be comfortable with replacing both ADs in the same cycle, but if the longer-serving AD has a bunch of working groups that are winding down, the NOMCOM might reasonably say "we'll replace this AD next year" - and if the BCP allows this, it might be the right thing to do.

Spencer


_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf


_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf