MSJ and others asked about carrots.
I was thinking about that at the time, and at the time, I came up blank.
Thinking today, I remembered something from a Eliyahu Goldratt book
_Critical Chain_. Stick with me a moment please.
{I strongly recommend The Goal, and learning about
https://en.wikipedia.org/wiki/Theory_of_constraints.
His book are novels, with characters, who learn}
Critical Chain deals with project management, with people in the fictional
class coming from (house/building) construction, and also software (modem!)
construction. An example of slipping schedules was a housing example where
the drywall people would be scheduled for Tuesday, would show up, and find
that the stuff-in-the-walls were not done/inspected. They would then go to
their next job.
The walls would be finished at 2pm, but the drywallers would not return for another week.
The same situation would occur in the modem vendor, where the DSP experts
(the critical constraint) couldn't start on time because something else
wasn't done.
The answer from all of this was the micro-optimizations of having the
drywallers or DSP people never be idle was the problem: it costs a bit more
to have them wait around half a day, but it improves scheduling **overall**
Herein might be the carrot.
If you get things done according to the schedule,
then
-- the cross-area reviewers are already ready
-- the AD already has their review time in their calendar
-- the IESG telecon is already scheduled
Would this really work? Would any WG being able to satisfy this?
Which of these is the critical constraint that we need to keep from being idle?
My feeling is that it's the AD review.
{Thank you for those that read to the end!}
--
Michael Richardson <mcr+IETF@xxxxxxxxxxxx> . o O ( IPv6 IøT consulting )
Sandelman Software Works Inc, Ottawa and Worldwide
Attachment:
signature.asc
Description: PGP signature
