Subject: Re: BSD, GPL and macroeconomics (This is a private question to JSS)
From: "Stephen J. Turnbull" <stephen@xemacs.org>
Date: 19 Feb 2002 15:13:40 +0900

>>>>> "shap" == Jonathan S Shapiro <shap@eros-os.org> writes:

    shap> I said the cost of maintainance, not the price of support.

Ah, thanks.  I should have known known what you meant, but
"amortization" is an accounting term, not an economic one, and
triggered the wrong associations.

It is not clear to me that conventional[1] open source organization
actually doesn't _raise_ total cost of maintenance, though.  Steve
McConnell's arguments specific to maintenance costs
(http://turnbull.sk.tsukuba.ac.jp/Research/OSS/McConnell-PrimeTime.pdf)
seem pretty plausible to me, despite my co-authorship of the rebuttal.

I should note that it also occurs to me that even if total cost rises,
it may also be reallocated in such a way as to justify the additional
cost.  Ie, it although open source makes the total cost of correcting
specific faults higher, it also somehow manages to distribute those
solutions more effectively, and therefore increases net customer
value.  It's a very fuzzy idea, though, and I don't really know how to
explain further.

    shap> The proprietary world does not, as a rule, accept such
    shap> donations easily. Ask any SAS customer.  Further, the equity
    shap> imbalance in the license structure reinforces the customer
    shap> in deciding not to donate something of value: there is no
    shap> quid pro quo.

I'm not really thinking of customers.  I'm thinking of small
consulting shops based off a larger app.  They would establish a
relationship with the larger company.  True, without open source they
would always face the danger of having their business cannibalized.
However, one suspects that some boilerplate license could be written
to handle the most common cases, in the same way that GPL or MIT
handle almost all the interesting cases in traditional open source.

It seems to work for GAMS (a simulation package I use occasionally,
which provides a front end to numerical analysis engines).  The
community is small enough that the relationships are easy to manage,
and the occasional third party contribution is always to integrate a
new solver.  The wording of the licenses (use of some solvers is
covered by licenses from both GAMS and from the third party) and the
pricing suggests there are royalties changing hands.

    shap> So if you are asking: "does this model work?", my answer is
    shap> "In practice, probably not."

Expert opinion noted.  The mindset objection by itself is quite
powerful.



Footnotes: 
[1]  You've come a long way, baby!  I hardly think "conventional open
source" is an oxymoron any more.  :-)

-- 
Institute of Policy and Planning Sciences     http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
              Don't ask how you can "do" free software business;
              ask what your business can "do for" free software.