Subject: Re: The term "intellectual property" considered useful
From: "Stephen J. Turnbull" <turnbull@sk.tsukuba.ac.jp>
Date: Tue, 30 May 2006 17:16:05 +0900

>>>>> "Norbert" == Norbert Bollow <nb@bollow.ch> writes:

    Norbert> In view of this goal I need to have a mental model of the
    Norbert> economic system, in which the economic goods are not only
    Norbert> tangible goods, but where digitalizable information and
    Norbert> "what is in someone's head" is also considered in view of
    Norbert> the possible commercial use of information goods,
    Norbert> knowledge, etc.

This is very hard.  Economic models of "information itself" are mostly
black boxes; they are just value that attaches to certain activities
such as copying it or embodying it in a marketable product or
listening to it.  The first two are "derived values", the third is
"utility", but in practical models in the technical literature they
are almost always primitive assumptions.  Then we add definitions of
constraints on exchange or production to make them "look like"
information in some sense.

Cf. Tom Lord's recent posts, which refer to this kind of thinking.  I
think Tom has sufficient credentials to not be written off as a
brainwashed economist.  Consider why he might accept such models for
purposes of discussion.

    Norbert> I still think that that kind of model is alright as long
    Norbert> as it is used for purposes of economic analysis only, but
    Norbert> in drawing my moral conclusion from this model I made the
    Norbert> mistake of implicitly assuming that moral distinctions
    Norbert> which don't exist in the model also don't exist in
    Norbert> reality.

I think that is correct reasoning.  However, keep the "no moral
distinctions" model in mind if it seems good to you; it may be
possible to work forward from there to a legal system the implements
it, and then to the economic consequences.

    Norbert> This objective of drawing conclusions (from a given set
    Norbert> of moral convictions) about how the legal and economic
    Norbert> system should be changed seems to be more difficult that
    Norbert> I had at first expected it to be.

    Norbert> Is there helpful relevant literature?

Milton Friedman's "Capitalism and Freedom" and Friedrich von Hayek's
"The Road to Serfdom" are excellent examples of how to blend moral
philosophy and economics.  You probably won't like the conclusions
very much, especially not Friedman's, but they do demonstrate how to
do it.  Marx's "Economic and Philosophical Manuscripts" are useful
examples, though they don't bear directly on the IP problem, they do
consider property carefully.  ("Das Kapital" unfortunately is
garbage-in, garbage-out; it's not really Marx's fault that he was 15
years too early to benefit from the insights of the "marginal
revolution" in economic theory.)

The GNU Manifesto is a good example.  I disagree with a lot of it, and
from the point of the view of an economist it makes a lot of dubious
leaps of faith, but I have always been impressed with the care that
Richard took to give the "accepted truths" of technical economics due
respect.

The writings of Ronald Coase, especially "The Problem of Social Cost",
and of other members of the "Chicago School" of law and economics,
especially Richard Posner (very influential judge in the 2d Circuit,
IIRC).  I doubt you'll agree with their conclusions, but these are the
guys who provide the theoretical underpinnings for the idea of
creating markets by creating property rights.  Coase is generally
fairly intelligible, even if you don't know the economic theory he
refers to---lots of simple examples.  Posner is a lawyer; sometimes
that's harder than economics, sometimes easier.

Again not very close to IP as far as I know, there's the work of
Amartya Sen who has been trying for decades to bring moral philosophy
back into technical economics (at least as a motivator).  I think
you'll like Sen.

Finally, the works of E. F. Schumacher ("Small Is Beautiful") and
Hernando de Soto (the Peruvian reformer) are very useful as antidotes
to the "Chicago School" tendency to turn everything into utility
calculus.  It's possible that the "Small Is Beautiful" ideas could be
incorporated into formal theory via the economics of complexity and
"bounded rationality".  But the models I've seen aren't very
convincing.

    Norbert> When interpreted as allowing software patents, patent law
    Norbert> infringes at least the last point in this list of
    Norbert> freedoms: Even if in providing cemmercial services I used
    Norbert> just my own computer and software that I've written
    Norbert> myself, I'd still be in trouble for patent infringement
    Norbert> if my code uses a patented idea for which I don't have a
    Norbert> patent license, and the patent holder sues me.

Sure.  That's precisely what patents are intended to do.  When we live
in a community, we accept the abridgement of some of our freedoms as a
matter of course.

Eg, the "linking is deriving" interpretation maintained by the FSF is
no different from a patent in this respect.  The FSF has claimed that
code which contains not a single copy of any text from any program
licensed under the GPL must nonetheless be distributed under the GPL
if it refers to the API of a GPLed program (specifically the FSF
forced Ghostscript to remove a stanza from its Makefile which enabled
linking to libreadline via a GPLed stub, neither of which was
distributed with the AFPL-licensed version of Ghostscript).

Even the GPL makes it quite plain that freedoms can be traded against
one another.  Further, there are plenty of advocates of permissive
licensing who disagree with the GPL's claim to be maximizing freedom
in some sense.

So "it infringes my freedom" is not in itself an argument against IP.

    Norbert> but we should be able to make sure that whatever legal
    Norbert> systems we create have the effect of decreasing the
    Norbert> disadvantages of those who who are disadvantaged through
    Norbert> no fault of their own, rather than increasing the effects
    Norbert> of those disadvantages or the number of people who are
    Norbert> severely disadvantaged through no fault of their own.

Whatever that means.  It's not obvious.  This is one of the things
that Amartya Sen in particular is noted for trying to clarify.  I
don't know if he's written non-technical books, but I would imagine he
has, or his students have.  Another reference in this area is Rawls,
"A Theory of Justice".  That book is definitely accessible to any
intelligent layman, though it's not easy.  There's also a followup by
Robert Nozick, taking an extreme libertarian point of view, I forget
the title offhand though.

In an earlier draft of this message I wrote

    If that's the goal, shouldn't you advocate banning the writing of
    new software, as that clearly increases the disadvantage of the
    90% of mankind who still don't have real access to computers,
    through no fault of their own?  Or are those people too poor (!) 
    to be counted as "severely disadvantaged" for the purpose of
    discussing software patents?

but deleted it because I didn't see how to express my real point.  Now
I see how to make it: Rawlsian theory is very useful in addressing
that kind of specious criticism, because he speaks of "disadvantage"
in terms of "primary goods" such as "freedom of speech" and
"employment opportunity".

It's not yet clear how to bring those ideas into economics, but that's
not your problem, it's ours!

-- 
Graduate School of Systems and Information Engineering   University of Tsukuba
http://turnbull.sk.tsukuba.ac.jp/        Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
        Economics of Information Communication and Computation Systems
          Experimental Economics, Microeconomic Theory, Game Theory