Hi,
I agree with Joel. Though, the metric is simple (and in that regard
possibly good to track) I believe it measures the wrong thing. I think
that if the WG does work and there is no need for drawing conclusions on
the WG mailing list, or the WG chair has chosen not to contribute
technically - possibly to stay impartial in a technical argument where
he later has to do conclusions.
However, I think measuring the timeliness of a WG based on the WG
mile-stones and how they are kept measure the right thing.
In these kind of problems, often measuring something is enough to
correct the problem or at least address it: If you measure how many
mails the WG chair sends, he will send more mails. On the other hand, if
he is measured by the WG milestones, he has the incentive kick his
editors or WG to do the work in time.
Cheers,
Jonne.
On Tue, 2005-06-21 at 18:57, ext Joel M. Halpern wrote:
Actually, based on experience in effective and ineffective working groups,
I don't think the 1 week (or even two weeks) suggested below is a
reasonable measure of activity.
When I was a WG chair, there were often multi-week periods when I did not
post anything to the list. Sometimes this was because nothing was
happening. Sometimes it was because the design teams were making good
progress. Note that this could mean that the design team was working to
get something coherent to post to the list, and therefore nothing appeared
from anyone on the list.
In other working groups I have been in, there have often been multi-week
periods where the work is going well and there are plenty of posts on the
list, just none from the chairs because there is no need for them to post.
So, I think measuring the chairs posting rate is not a good measure of
anything.
Measuring activity on the working group mailing list is probably
sensible. But a one week time horizon is usually wrong.
Yours,
Joel
At 10:34 AM 6/21/2005, Harald Tveit Alvestrand wrote:
--On 20. juni 2005 07:39 -0500 Spencer Dawkins
<spencer(_at_)mcsr-labs(_dot_)org> wrote:
Two related problems here, as you pointed out in another posting - when
the WG is only active for six weeks per year, and when the WG chair is
only active for nine weeks per year. I don't see how we can focus on this
with our current milestone tracking ("no, really, we'll finish that draft
by the NEXT meeting, this time for sure"), so your comments in the
"front-end delays" thread apply here as well.
Let me offer a simplistic metric.....
if a WG chair has posted nothing to the WG mailing list for a week, and
that WG chair has not told the WG he's on holiday, that WG chair is
probably not doing his/her job.
If NOBODY's posted to the WG mailing list for a week, it's time to close
the WG.
Harald
_______________________________________________
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
--
Jonne Soininen
Nokia
Tel: +358 40 527 46 34
E-mail: jonne(_dot_)soininen(_at_)nokia(_dot_)com
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf