Subject: Re: FW: Why would I pay for Ximian software?
From: "Stephen J. Turnbull" <stephen@xemacs.org>
Date: 25 Dec 2001 19:39:17 +0900

>>>>> "Bernard" == Bernard Lang <Bernard.Lang@inria.fr> writes:

    Bernard> but ...  this wisdom evolved for the economy of the 19th
    Bernard> century, with high investments and significant marginal
    Bernard> costs in the production of goods, and with generally
    Bernard> tractable network effects.

    Bernard> In the economy of intangibles, none of these premisses
    Bernard> are true, and

tonstanco redux.  Please, no.

The shape of the marginal cost curve is interesting for its slope,
which determines economies of scale _if_ fixed cost ("investment") is
relatively low, and for the level of unit cost at which marginal cost
equals average cost, which is the most plausible definition for "the
natural unit of price" in this context.  That is, if marginal cost is
decreased dramatically, then the level at which "investment" becomes
"significantly high" also drops dramatically.  To make your case,
you'd really need fixed cost dropping dramatically faster than
marginal cost, but I believe the opposite is the case.

I believe that the importance of network effects is drastically
overestimated by most free software advocates, but I know of no
empirical studies on their strength.  However, my argument to
Mr. Stanco about the amount of social contact you can have with 6
billion human beings in one human lifetime is indicative.  Metcalfe's
Law is hugely optimistic.

True, we have the phenomenon of Linux, which is argued to be network
externalities on the production side.  This has the advantage that the
"network" in question need only be hundreds or thousands of
individuals, so externalities are still feasible.  I'll comment on
that in a response to another post.

You are absolutely correct, all this needs to be examined carefully,
and may even require development of new analytical tools and data
sources.  However, at this point I tend to side with Qoheleth.  "There
is nothing new under the sun."  I believe that network externalities
are just a new variation on the old theme of increasing returns.

Entrepreneurs are welcome to place their bets optimisitically, of
course.  I hope I'm wrong, and I hope it's the first hat in the ring
that proves me wrong, if I'm going to be proved wrong.  :-)

    Bernard> there is good reason to believe that other ways of
    Bernard> allocating goods, of creating incentives might be more
    Bernard> efficient,

Yes.  One of them is the DMCA, some of whose provisions (and exactly
the provisions which possess the most efficiency-enhancing potential)
are inhuman, unacceptable invasions of privacy and infringements of
basic freedoms.

So watch your step.  Remember that Marx also aimed at freeing the
human spirit from the alienation of the property relationship.  True,
the material conditions seem to be qualitatively changed.  But it is
the human spirit that defeated Marxism, and I see little change in the
relevant aspects of that.

    Bernard> if only to avoid monopolies that are the unavoidable
    Bernard> result of current economic structures in the realm of
    Bernard> intangibles.

Er, relatively large fixed cost and marginal cost low and flat is
called "natural monopoly" for good reason, only partially mitigated by
the fact that if IP is abolished, intellectual assets only need to be
developed once.  And significant network externalities, if present,
should only enhance the tendency.

Avoiding these monopolies is a very very hard problem.  Schumpeterian
analysis strongly suggests that the best bet is regulatory judo: award
strong monopoly franchises with sharp teeth for enforcement over
limited domains (sounds like copyright in software to me), and then
let the process of "creative destruction" do its thing.  Ie, creating
rapid technological progress via "serial monopoly".

    Bernard> The whole set-up is artificial anyway, based on how
    Bernard> exchanges are regulated.

The set-up is _not_ artificial.  It closely corresponds to a
mathematical structure that is very natural in its place.  IP could
very well be _the wrong place_ for using it (I believe it is
plausible, obviously).  But calling it artificial is incorrect.

    Bernard>    For example I am still trying to find a single
    Bernard> economic study that asserts that the cuurent extension of
    Bernard> patenting

Ehem.  Who mentioned patents?  Not me; I recanted my (lukewarm, devil's
advocate) support of software patents almost two years ago.  AFAIK all
responsible economists (eg Hal Varian, Franklin Fisher) are at worst
not in favor of them in principle, and all are opposed to the current
regime which makes a mockery of even the PTO's weak definitions of
"novel" and "prior art".

    Bernard> If not ... are the statements below fully justified  ?

Yes.  They are warnings to people who are out to change the world to
be the way they would like it to be, without really considering that
only 2 of 100 human beings own PCs, and only a small fraction of those
are developers, and many of them develop as a "day job" and are
primarily interested in compensation packages.  This is a situation
ripe for "counterrevolutionary activity" (eg, Sklyarov's plea bargain).

There is no question that the structures currently in place serve the
rich and powerful.  But this is not an unnatural consequence of
artificial structures snuck into place by a few wiseguys who figured
out how to cheat the rest of us, and once on top keep us down by
pointing the guns and tear gas of the police at us.  It's a fairly
natural consequence of the fact that it takes fairly large changes to
the system to substantially change the shape of the distribution of
wealth, and even more to change that of the distribution of political
power.  These changes scare the bejabbers out of the large mass of
less powerful because they involve substantial harmful dislocations.

Unfortunately, small tweaks of the system can produce huge differences
in which small group gets to be "the rich and powerful".  This is why
revolutions are so easy to subvert.

Now, you might ask, "why not do it slowly and non-coercively?"  I have
no problem with that at all.  However, my expection is that the
proponents will be sorely disappointed (cf sourceXchange and
CoSource.com).  Open source enterprises can not pay as well as
proprietary enterprises, and "micro payments" according to a
Reputation System will likely sum to a small fraction of the
compensation package that can be offered by an entrepreneur, or
entrepreneurial manager in a large corporation.


-- 
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.