On Wed, 11 Oct 2000, Nathan Torkington wrote:
> Then again, remember the hassles we had with the perl6-* lists?
> Nobody knew how to deal with topics that overlapped lists. You had
> to know all the groups to decide which it was appropriate for. Are
> these big enough hassles to suggest that perhaps the perl5-porters
> All In One list wasn't such a bad idea after all?
To a point. Instead of top-down or bottom-up, you shoot for somewhere
in the middle, although probably closer to the bottom. Start with the
lowest-level stuff that still has commonality with other group. Hash
out the similarities, differences, and directions. While the groups
and community build up from there, a group of people stick around to
fill in the pieces of their portion. You certainly don't want to
isolate similar development, but there's no need to flood discussions
with specific patter, either.
I picture it much like the mega-conglomeration merger talks. Two big
honchos sitting across the table from each other, hashing out the rough
direction of whatever end agreement they want to take, all the while
taking in advice from advisors, shareholders, etc.
Once they've roughed out an outline of what the agreement *should be*,
they dismiss the little people to flush out the details - cross the
i's, dot the t's, that sort of thing.
This would probably work better for more open development, too, as it
would give more junior folks a chance to contribute what Damian calls
"trivial implementations." You're going to find far fewer people who
are able (or willing) to code some perl functionality from end to end
than if you were to have the few able bodies able to hash out a
thorough framework, and have the nugs contribute bits and pieces within
the framework....
Take, for instance, the whole async i/o issue - I've not Clue One (I
think) about how to go about writing my own i/o lib, async or
otherwise, even though it's an area I'd like to contribute to. Given
the current end-to-end scheme, I'd never get to do anything, and Uri
and Company are going to be coding in the dregs.... There's no reason
that they couldn't get the base down, work more on flushing out its
functionality, and turn over some of the more trivial stuff to some of
us trying to be a little more active.
Hmmm, I seem to have digressed from the original point....
-- Bryan C. Warnock
([EMAIL PROTECTED])
