[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: What do we want to achieve from reorganizing our functions?




--On 28. juli 2003 21:34 -0400 Margaret Wasserman <mrw@windriver.com> wrote:

At 10:08 AM 7/28/2003 -0700, Harald Tveit Alvestrand wrote:
--On 26. juli 2003 13:22 -0400 Margaret Wasserman <mrw@windriver.com>
wrote:
The IETF runs extensively on a concept called "two-in-a-box" --
where two people are jointly responsible for most functions.
This has some benefits, but it actually takes considerably more
combined work than having a function run by one person (because
of the need for the two to communicate), and things happen
slower.
two reasons for doing this is because single-person functions have led
to  single-person "empires" (Marshall Rose and the NM area was the most
blatant example; John Moy has been accused of the same thing wrt OSPF;
he's now got 2 co-chairs, but still gets accused of it), and because
we've  had numerous instances of chairs dropping out on relatively short
notice  (or with no notice).
Makes sense.  I just think that we should understand that there
is a cost to this, and that the cost doesn't scale linearly as you
add more people to the group of people who are "in charge" of
something.
yup. I regard 3-chair WGs as an emergency measure most of the time - there may be many reasons, for instance when one is really "group secretary who wants a blue dot" or "chair emeritus who is kept on to make sure he still feels part of it", or "part of a transition strategy". It's the same reason why I've resisted repeated suggestions over the years that we should have 3 ADs in the APPS area.

This gets even worse when there are 3 or 4 people responsible
for day-to-day management of a function.  And the idea of
having more than 4 people responsible for managing anything is
almost completely unworkable.
if you have more than 2 people, is it required to designate a "lead"
among  them? This frequently happens in practice, even when they are
formally  equals.... I am not sure of the value of formalizing it.
I have been in groups of three or four that have functioned
without an assigned or defacto lead, but it only works (I
think) when the three or four people are very closely aligned.
In most cases, it works better for their to be a lead, although
it doesn't have to be assigned from the outside.

The point is that, once you get to a large enough group, you've
got "management by committee", and I'm not sure that's ever a
good idea...
No - but neither is the "rule of one" a paragon of good ideas - it's too easy to slip over into the authoritarian/dictatorship style.

Balance in all things, including the striving for balance....