Is this really where we want to go? (was: Re: Selecting leadership, take 2)

Jari Arkko jari.arkko@piuha.net
Sat, 07 Dec 2002 10:56:36 +0200


Randy Bush wrote:

> maybe shortness of meetings actually is a problem.

I think it is. The more time (face-to-face, phone-conference,
mailing list, own work) you put to the system, the more output
you get from the system. Our system has been rigged for extremely
short amount of face-to-face and discussion time. Even big, radically
new, and important base technology components get a max of 3 hours
per YEAR. Say, a WG concentrated around some issue has a base
protocol that gets discussed one hour per meeting slot and various
extensions and other issues consume the rest of the time.

In some other standards orgs there may be a montly meeting that
lasts days on the *single subject*. This is tens or even a hundred
times more discussion time.

These other orgs sometimes do badly in terms of e-mail discussion.
We do it much better at IETF, and I'd like to keep it that way.
However, e-mail goes only so far and I think the current mailing list -
meeting time ratio is too extreme. This is one of the reasons why
folks have interim meetings, conference calls, design teams. Plus,
the human nature is that often nothing happens until the meeting
deadline approaches. At IETF, we don't have too many of such
deadlines (only three per year).

> most other sdos have much longer wg meetings.  not that i am
> enamoured.  but if it really would speed convergence and increase
> quality, i would have to seriously consider it.
> 
> and, as someone here or maybe wgchairs list said, few wgs get a
> serious piece of work out without an interim.  
> 
> is a consequence that it becomes *very* hard to be interested in
> many wgs without being a [hisssss] 'standards professional'?  and
> would it lose us what i see as a real benefit, folk with wide
> perspective?

A wide perspective is of course desired. The question is how to get
that wide perspective. Does it require you to be present on all discussions
of another WG? When they debate whether the field should be 8 or 16 bits?
Or is it enough to participate the big discussions?

Perhaps we need a set of 'main meetings' (like the ones we have now) plus
a series of 'working meetings'.

Jari