Subject: Re: open source policy across the globe
From: "Stephen J. Turnbull" <stephen@xemacs.org>
Date: Wed, 10 Mar 2004 17:10:51 +0900

>>>>> "L" == L Jean Camp <jean_camp@harvard.edu> writes:

    L> On Monday, March 8, 2004, at 04:06 AM, Stephen J. Turnbull
    L> wrote:

    >> [Restricting trade] is bad public policy.  I believe in
    >> democracy, and if a democratic society chooses to do things
    >> that way, I will acquiesce, but not shut up.

    L> Markets are good public policy ONLY when they function as they
    L> should, they are not good per se.

The first clause is mistargeted, although the consequent is correct.
Markets are good public policy precisely when they accomplish public
goals better than alternative institutions do.

    L> Markets are bad public policy when the exclusion and rivalry on
    L> which they depend are not present, and the cost of exclusion is
    L> greater than the resulting increase in value.

The first clause is demonstrably false, as markets for emissions
permits show.  Artificial creation of scarcity can be a very useful
tool if it allows decentralization of decision-making about allocation
of related resources.

In the case of software copyright, it's a rough approximation?
Definitely.  Good or bad?  That's an empirical question.  If you have
hard numbers, I'd love to see them; I don't know where to get them.
(They're certainly not on offer from the SPA.)

    L> It appears that the means (markets) have been dreadfully
    L> confused with the ends (improving the lot of humanity).

Jefferson may have got the cost-benefit analysis wrong, but he did not
confuse means with ends.  As for latter-day economists, I'll content
myself with the observation that even Milton Friedman signed the
economists's amicus curiae brief to the Eldridge case, opposing
copyright extension.

There _are_ people who are confused (Sonny Bono's widow and the US
Congress), and clearly Bill Gates and Disney, Inc, et al would love to
take advantage of that (and may even be confused themselves).  But I
don't think any of them hang out on FSB, at least not for very long.

    L> The software market "works" under an exclusion model even when
    L> there is proof, in running code and viable businesses, that the
    L> exclusion model is not necessary.

May I point out that Cyclic Software disbanded having taken in a bit
more than $100,000 in its best year (according to Jim Kingdon's home
page), while BitMover annually contributes services of that order of
magnitude of value back to the community?  As a matched pair (same
market segment) you could take that as "proof" that there are cases
where the exclusion model is necessary, even socially beneficial!
Statistics may lie, but examples are downright shameless!

But let's take your unspecified examples seriously.  They're not
working under the exclusion model, then.  Is there a problem here?  It
would seem that those businesses that are successful without exclusion
aren't using it, while those businesses that need it, do use it?
Sure, there's a grey area in the middle where I think the FSB
consensus is that there's too much proprietary software, and that's
where the open source movement comes in, posting "Eric Raymond's Top
Ten Reasons Why Open Source Improves Quality and Lowers Cost".[1]

So it's not obvious to me that there is a need for a _social_ policy
other than education here.  The ordinary business case for specifying
open source licensing[2] for software acquisitions seems extremely
strong to me; balancing that against the automatic cost-sharing and
the production incentives (== rapacious profit-taking) provided by
non-open source-licensing seems like precisely the kind of thing that
the market is reasonably good at.  Again, a "lucky" approximation
because semi-open licensing entails a lot of lock-in risk[3], so smart
customers will insist on fully open licenses.  But close enough that I
think "educate, then wait and see" is the best policy.  YMMV.

If that business case applies to government's operations too, as I'm
sure you agree it does, it makes a lot of sense for government
purchasing guidelines to specify open source.  But that's not what I
mean by "public policy", that's just good stewardship.

    L> I don't think we should prohibit closed software, don't get me
    L> wrong, but that a strict property model makes more money for
    L> some subset of the people involved in software does not make it
    L> superior in policy terms.

Who advocates "making more money for some subset" on FSB?  The policy
argument here is about which produces more benefits in terms of the
somewhat conflicting goals of greater variety of software distributed
to more people.

Note that this _is_ on-topic, as one important question to many would-
be FSB entrepreneurs is "how do I support my hacking jones?  Can I get
paid for it, or do I have to hack for compensation that will not
directly feed my kids and put my dog through college?"  The answer at
present seems to be that it's pretty hard to get paid for basic
development of apps through free software business models.  On the
other hand, there's plenty of money to be made in customization and
configuration.

Some hackers are perfectly happy with the latter kind of work, and for
them life is good indeed.  But others would like to be writing basic
infrastructure and getting recognized in the coin of the realm, which
doesn't seem to be nearly as easy.  Yet their work is presumably more
valuable socially (it's potentially useful to more people).

    L> Luckily for the fsb, bad economics is often good business.

Not luck at all!  Economic theory tells you that in the long run,
you're going to make zero profit.  So good business is all about
getting there before the economics really get a hold on you.


Footnotes: 
[1]  ESR reminds me more of Dave Letterman than Bruce Perens does.

[2]  Consider how Microsoft's recent manoeuvers only make it clear
that "open source with some rights retained" leaves the customer
hardly better off than with out-and-out closed source.

[3]  Or maybe not so "lucky"---why would a smart vendor use non-open
licensing _except_ to generate lock-in?

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