[Patents] natural forces

[Jesus M. Gonzalez-Barahona jgb@gsyc.escet.urjc.es]

Gregory Aharonian writes:
 > >What is clear however is that any patenting of information works
 > >infringes basic freedoms such as the freedom of expression and
 > >therefore needs to be excluded ....
 > 
 > Which is an argument against ALL patents, not just software patents.
 > I "express" my new digital hardware circuit with a printed circuit
 > diagram just as I "express" my new software algorithm with a source
 > code listing, just as I "express" my new mechanical design with a
 > blueprint, all three of which circuit/code/blueprint, are
 > computer representable, computer designable, computer optimizable,
 > etc.  All of these expressions should have equal freedom of
 > expression considerations applied to them.
 > [...]

There is (at least) one basic difference when you express your digital
hardware cirucit with a circuit diagram and when you express your
source algorithm with the source code listing. 

In the case of the circuit diagram, it is information describing the
circuirt. But it is not the circuit itself. In order for the circuit
to work, you have to build it. And the patent gives you a monopoly on
building the circuit, not on the diagram. In fact, the diagram is
usually part of the patent application, which is public. The diagram
can be included in books about electronics, and can be discused with
coleagues, and even be left in a web site for public information. And
nothing of this is covered by the patent (in fact, the very idea of
having patents is to encourage the spread of inforamtion about the
covered technology, while giving exclusive rights over its use).

In the case of the software algorithm, the source code *is* the
algorithm. When two computer professionals want to talk about the
algorithm, the best thing they can do is write it as a computer
program source code. To teach it, you use the code listing. When you
want to disseminate that technology, the best way of doing it is by
publishing a code listing. But it happens that that code listing *is*
the software. Therefore, there is no difference between the code listing
(which is the mean for expressing the algorithm, like the circuit
diagram is in the case of the electronic circuit) and the software
algorithm: they are the same thing. In fact, I was told several times
that distributing a source code covered by a patented software
technology is considered contributory infrigment of the patent, since
anybody who downloads the code can quickly compile it, and use
it. When I cannot distribute the source code, my freedom of expression
is at stake: I cannot distribute information about the software
techniq ue in the best way it can be done.

Of course, this is by no means strange, since computer programs are
fully equivalent to proofs to mathematical theorems, which are in
themselves information, and just information...

BTW, it is interesting how software patents are accepted without the
slightest trace of source code. I wonder how a software algorithm can
be expressed in detail without including source code... (oh, maybe I
forgot that software patents don't describe software techniques, just
ideas that can be implemented somehow in software...)

Saludos,

      Jesus.



-- 
Jesus M. Gonzalez Barahona                | Grupo de Sistemas y Comunicaciones
jgb@gsyc.escet.urjc.es / jgb@computer.org | ESCET, Universidad Rey Juan 
Carlos 
tel: +34 91 664 74 67                     | c/ Tulipan s/n
fax: +34 91 664 74 90                     | 28933 Mostoles, Spain