On Wed, Jul 31, 2013 at 08:38:26AM +1200, Brian E Carpenter wrote:
>
> It's been pointed out before that in a group with very diverse languages,
> written words are usually better understood than speech. It's a fact of life
> that you can't have a full-speed cut-and-thrust discussion in a group
> of 100 people, half of whom are speaking a foreign language. Sitting in
> a circle does not fix this.
While that is true, I think it misses the point of the objections to
the sit-and-watch-PowerPointTV.
First, I observe that we already _have_ a great deal of written words:
the drafts. I continue to believe that altogether too much time in WG
meetings is spent "introducing", "presenting", or otherwise showing
off ideas in an existing draft to participants in the WG. I
acknowledge that (particularly in early stages of WG life, in topics
with a lot of different work, and in cross-WG presentations) these
"intro" presentations are a fact of life. But I think we are
extremely bad at holding the reigns on them.
In a WG meeting, I think such "intro" presentations about drafts
really can be kept to three pieces of information: the name of the
draft, a slogan describing the problem it is supposed to solve, and a
pointer to the beginning(s) of discussion thread(s) on the draft. If
the person promoting the draft can't give the elevator pitch, they
don't know their own draft well enough to summarize it and shouldn't
be presenting it. Any additional discussion in the presentation ought
to be exploring, as much as possible, one or more of the following
topics:
- a particular issue
- is $issue a real problem
- alternatives for solving $issue
- motivation for $issue solution choices
Each such slide, it seems to me, ought to encourage at most a couple
minutes of exposition and then some discussion. The _reason_ to get
together in a big room with other people is to use the high-bandwidth
opportunity to hash out the extent of a problem. The back and forth
of "you forgot this", "no that won't work because it explodes foo",
and so on, is the value here.
Notice that none of that includes complicated flow-chart diagrams that
explain in detail a proposal. There _is_ a place for those, however:
an actual presentation that gets made after significant discussion on
the list has made it clear that nobody understands the proposal. At
that point, those 10-15 minute presentations of some proposed
mechanism are important, if only to inspire commenters to go back to
the list and say, "Ok, _now_ I get what you were trying to say, and
your text needs to be improved along the following lines." But these
full explanation presentations happen too often when there has not
been such confusion.
Of course, all of the above depends on us going back to the list and
working out the details there, and it depends on people having read
the drafts and having a list of questions themselves that have been
deferred from the list for the face to face discussion.
I believe presentations in meetings are also sometimed useful if they
are exploring a problem space. In that case, I believe what one needs
is _short_ presentations of the sort, "Here's what I think the
problems are," and then a lot of well-moderated discussion.
Unfortunately, actually running meetings this way is a lot of work,
requires fairly careful planning, and requires an indifference to
nasty remarks on the part of presenters who would much rather listen
to themselves for 20 minutes than to others. But I think it'd make
for better meetings. (Yes, along with room layouts that were more
suited to getting people to the mic.)
> The old days are gone.
Yes, and we need to figure out how to use meeting time effectively
here in the new days. That effective use does not, I think, involve
expanding to fill all the time in the year with 20 minute low-content
presentations summarizing the draft that you can read in the span of
the time it takes to get through the presentation. (Perhaps I'm
wrong. Perhaps people find that the only time they have now to read
the drafts is during the presentation of the draft. I sure hope not.)
Best,
A
--
Andrew Sullivan
ajs@xxxxxxxxxxxxxxxxxx