Subject: Re: Exploring the limits of free software: Cygnus, and GPL
From: "Stephen J. Turnbull" <turnbull@sk.tsukuba.ac.jp>
Date: Wed, 26 May 1999 14:11:06 +0900 (JST)

>>>>> "craig" == craig  <craig@jcb-sc.com> writes:

    craig> will we begin to take, more seriously, the importance of
    craig> thoughtfully designed, up-front-specified, products and
    craig> platforms?

    craig> It's easy for me to say "yes" based on the assumption that,
    craig> since I believe such things have value, and I believe the
    craig> OSS development model finds value of that sort better and
    craig> faster than the proprietary-development model, the market
    craig> will *make* it happen.  But that's a bit of a cop-out,
    craig> since I don't know, offhand, *how* it'll make it happen,
    craig> which means I don't know if it's possible (to a
    craig> sufficiently useful degree).

One possibility is evolutionary.  When version 0.5-beta exclusively
for Platform X appears on Freshmeat, people using Platform Y will look
at the code, see the incompatibilities with Platform Y and make
requests.

It should be a lot less expensive to change the design at that stage
to account for such than to do so at the equivalent of "public
release" stage for proprietary products.  Ie, the better, more modular 
design wouldn't be done "consciously" but in response to earlier (than
would occur for non-open-source) comments from the peanut gallery.
This could be more effective than conscious design in a proprietary
environment.

Out-and-out standards-writing activity?  Brain-storming is fun; do you
think all the people on this list who have announced they want a
generalized discussion of a certain kind of generalized program
(namely "source exchanges") actually intend to pay no attention to
what others are saying and will not cooperatively evolve their
individual programs and ideas toward some standard---despite that
being their stated intention?  _Of course_ they will generate some
kind of standard.  Some will implement more of it than others do, but
it will be an informal, but explicit, standard.

Whether searching Deja-News is an effective-enough way to discover
what that standard is as a separate Info | App | API chapter would be
is another question; I'll bet it is (and that some dedicated soul will
assemble an API FAQ which will be turned into a genuine API manual by
some other dedicated soul).

-- 
University of Tsukuba                Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences       Tel/fax: +81 (298) 53-5091
__________________________________________________________________________
__________________________________________________________________________
What are those two straight lines for?  "Free software rules."