Subject: Re: Releasing under OS, what License?
From: "Stephen J. Turnbull" <turnbull@sk.tsukuba.ac.jp>
Date: Mon, 25 Jun 2001 16:51:25 +0900

>>>>> "DJ" == DJ Delorie <dj@delorie.com> writes:

    >> Yes, they do.  The group of copyright holders has the advantage
    >> that they can release improvements under proprietary terms.

    DJ> See my other reply.

To Greg Broiles?  There's an important difference, which is the word
"improvements."  Ie, not yet released.  Now, you're "religious" about
it, but consider the several proprietary webservers built on Apache,
or proprietary X servers.  Some businesses may see a strategic
advantage in opening the source to some parts of their product line,
and not to others.

    DJ> Plus, if they plan on doing this, they cannot take advantage
    DJ> of what others build upon their GPL'd work.

They _can_.  It just becomes more expensive.  One could, for example,
distribute separate packages, targeted at different audiences.  In
your Cygwin example, the extra expense is prohibitive.  But Aladdin
doesn't seem to suffer much expense from having a "GNU" version of
Ghostscript that can be distributed with libreadline support.
Different cases, different results.

    >> The LGPL makes this more nearly symmetric.

    DJ> Why?

Because (depending on the application) non-copyright holders can
provide proprietary improvements, and distributed combined binaries,
just as the copyright holders can.

    >> Note that mere size confers an advantage under GPL.

    DJ> Yup, but, IMHO, that's "fair".

[*sigh*  When will I learn to cut the asides?]  I'm not really
interested in "fair" (except as public perception of being fair may be
a marketing plus).  Rather, I'm interested in the business advantages
a developer may be able to leverage.

I don't think you've really managed to poke any holes in my argument
about the business advantage.  I mean an advantage out of proportion to
the resources they currently and in the past put in.  My argument is
that with equal resources (what else are VCs and "behemoths" for?) the
incumbent will win, because the newcomers cannot provide _better_
service for long enough to induce customers to switch.  _Even if the
newcomer is a firm made of developers who secede from the incumbent._

As you wrote earlier:

    DJ> Competition is based on what each side can do for the user,
    DJ> not what they can do for the software.

Corollary:  If I can always do for the user what you can do, we may end
up in a race to do the least for the software, as you (or more likely,
your VCs---programmers are eternally optimists) despair of ever
beating me in the race to do more for the user.  AKA "market failure."


-- 
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 straight lines for?  "XEmacs rules."