Subject: Re: Donation Models
From: "Benjamin J. Tilly " <ben_tilly@operamail.com>
Date: Thu, 11 Jul 2002 23:44:54 +0800

Russell Nelson writes:
> Simon Cozens writes:
> > The first is that open source projects are made from code, and the best
> > donation I can give them is not money but code.
>
> That's just you, though.  Many people can't program but want to help.

Two other dynamics deserve mention.  

The first is that there can be a considerable overhead in "coming up to 
speed" on an unfamiliar code base.  This can make it more efficient to 
encourage someone who is familiar with it to do more than it is to
actually do the work yourself.

The second is more relevant to the case under discussion.  Software
involves a number of different kinds of work.  Many of them distribute
nicely.  Some do not.  In particular the Perl 6 grants sponsored people
who were engaged in basic design work - an activity which is notorious
for not scaling.  In fact when people do try to distribute it, what you
get is lots of useless discussion followed by people creating pieces
that don't fit together.  (Said useless discussion is often in the form
of long email threads.)

This is why I decided that, given the limits on what I was realistically
willing to commit to Perl 6, it was best for me to wait and familiarize
myself with final design documents, but avoid muddying the waters while
it was in process.  When the opportunity to donate showed up, I saw that
as something moderately useful that I could do to sponsor needed design
work which I was not in a position to contribute more directly to.

For the record, you can see how early I made this decision from
http://archive.develooper.com/bootstrap@perl.org/msg00714.html.
Ironically that email also includes the only significant influence that
I may have had on the design of Perl 6.  The pluggable front end idea
has become central, and my "perfectly grotesque idea" has become a key
design decision.  (But to be fair, I believe that others - particularly
Damian Conway - were already thinking along similar lines.)

In short, large open source projects sometimes need focussed attention
from a small number of closely cooperating people.  In those situations
it can be very helpful to have that team be supported in that work.  In
many cases companies will do it.  (For instance IBM did this with LVM.)
If no convenient company is available, a community donation model
deserves consideration.

Cheers,
Ben
-- 
_______________________________________________
Download the free Opera browser at http://www.opera.com/


Free OperaMail at http://www.operamail.com/

Powered by Outblaze