I think the crucial point to pick up on is something that chromatic has
pointed out very well in any number of use.perl journal postings over the
past year. That is, Perl 6's creation is dependent on how much time people
put into it, and how many people put in time. The volunteer effort to date
has been exemplary and inspirational. When you think about the universe of
possible things intelligent and energetic people could be doing with their
time, that so many have put so much into Perl 6 is a tribute both to the
worthiness of the Perl 6 project and to the fundamental goodness of the
volunteers.
Funding is the piece of the puzzle that allows us to buttress and enhance
the contributions of volunteers. Someone who can contribute 5 hours a week
to p6 development could possibly contribute 30 hours a week if they took on
a reduced workload at their day job. But that doesn't mean their
responsibilities just disappear: a mortgage / rent to pay, insurance
policies need maintaining, kids have to be clothed and educated, and
everyone has a powerful need to eat. Funding makes it possible to bridge
this gap.
To Richard's point, a systematic development plan is a tool that can be
helpful in acquiring funding. The plan is meant to acquire funding, and the
funding is meant to be applied against the plan to make it come to pass.
Done correctly, it's a virtuous circle that Gets Things Done. I completely
agree with chromatic that a plan without resources put against it is
neutered. I don't want a plan that has calendar dates on it. I want a plan
that has major pieces of work and their dependencies on each other reflected
(i.e. a GANTT chart) and a sense of the man-months of required effort for
each work-piece. At that point, the implementation volunteers have done
their job. It then becomes the responsibility of the funding-acquisition
volunteers to take the plan and with it seek out funding to make the
man-months happen.
Cheers,
- Richard
PS I often think of it like this:
Distance = velocity x Time (D = v x T)
When people ask for a release date for Perl 6, what they're implicitly
saying is,
T = D / v, solve for T
chromatic has been the #1 expositor that v is unknown, and therefore we
can't solve for T. In this he is quite correct. (And when we think hard
about it, D can be sometimes hazy as well. If Perl 6 had been implemented
100 times before we'd know D pretty well. But we're still figuring out what
D is.)
The idea behind a plan is to firm up D, at least to a certain minimum
acceptible level, and to allow for what if scenario planning to be played
with potential funding sources. (i.e. if you can give us this much v, we'll
have a decent shot of T happening in the 8-16 month timeframe afterwards,
etc.)
On Tue, Mar 25, 2008 at 2:36 PM, chromatic [EMAIL PROTECTED] wrote:
On Tuesday 25 March 2008 10:50:15 Richard Hainsworth wrote:
What the perl6 language needs now is a systematic development plan, with
broad aims and clear goals that will lead to good quality software and
to the tools to enable ordinary programmers to use perl6 for a variety
of tasks.
Richard Dice mentioned that I should elaborate, lest it sound like I'm
trying
to lecture Richard Hainsworth (not my intent, and I apologize for doing
so).
It's important to keep in mind the degree to which one or two volunteers
going
on vacation can slow the progress of Rakudo (for a recent example) or to
which one or volunteers putting in a few extra hours of visible work can
improve the progress of Parrot (for a slightly less recent example).
A plan that includes some degree of funding will help Perl 6 arrive much
sooner. Previous plans glossed over this part, which is one reason they
didn't work out in the long term.
I just want to make sure that any discussion of a plan acknowledges that
there's a fixed amount of work to go and an unknown amount of available
resources to implement the plan.
-- c
