Subject: Re: Free Software vs. Disruptive Change
From: "Stephen J. Turnbull" <stephen@xemacs.org>
Date: Wed, 10 Jul 2002 14:50:52 +0900

>>>>> "Mikko" == Mikko Valimaki <mikko.valimaki@hiit.fi> writes:

    Mikko> The open source and GPL work merely as a marketing (and
    Mikko> perhaps peer review) tool for them.

Not so "mere."  This is the main point of "The Cathedral and the
Bazaar".  :^)

    Mikko> Hence, MySQL is also one of those projects where the
    Mikko> development process is actually closed: just couple of guys
    Mikko> control all the code put in the official releases and they
    Mikko> take your copyright if you wish to contribute anything to
    Mikko> them.

This is the same as for any GNU Project project.  I wouldn't call that
"closed" (despite the fact that the FSF assignment policy is a _huge_
irritant for me personally).  It's simply a transaction cost imposed
by the use of a very restrictive (in a legal sense) license, and the
"outside" contributor does not participate in profits from commercial
licensing of her contribution.  In the GNU case, the latter doesn't
exist, of course.  In MySQL's case, I think that the barrier to entry[1]
imposed by the GNU GPL more than compensates for the transactions costs.

A device for financial participation that has been discussed on this
forum a number of times (I think Tom Lord has proposed similar things
recently, but it comes up often) is some kind of contributor royalty
scheme.  And of course we've seen the stock options given by Red Hat
and other companies to everybody (more or less) with a ChangeLog entry
in the kernel, etc.  These kinds of participation are hard to value,
however, and pure VCs are unlikely to be happy with making these very
large.  OTOH, if they're _too_ small, they could even be considered
"insulting," like a penny tip to a waiter.

However, I think that, given the culture, it's likely that gestures
like that of Red Hat will occur occasionally on projects that are "big
wins" even before they go commercial.  And for the rest, the "minimal
contribution disincentive" effect I describe in my response to Jean
Camp should keep the patches flowing as much as any sane level of
"outside contributor royalty" could.[2]  This seems optimal to me.


Footnotes: 
[1]  I have both the deprivation of commercial licensing revenue for
use of the well-debugged (free-speech) GPL implementation and the
"competition from a vanilla version" that the (free-beer) GPL
implementation creates in mind here.

[2]  "Should" means IMHO.  Tom Lord may respond "micropayments", and I
just don't (without good evidence) believe they will work.  Doesn't
mean he's not right.  :-)

-- 
Institute of Policy and Planning Sciences     http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
 My nostalgia for Icon makes me forget about any of the bad things.  I don't
have much nostalgia for Perl, so its faults I remember.  Scott Gilbert c.l.py