From jgb@gsyc.escet.urjc.es  Wed Nov 22 00:19:48 2000
Date: Wed, 22 Nov 2000 01:19:48 +0100 (CET)
From: Jesus M. Gonzalez-Barahona jgb@gsyc.escet.urjc.es
Subject: [Patents] natural forces

Gregory Aharonian writes:
 > Finally, if you can induce infringement by showing someone source
 > code that infringes a patent, you can induce infringement by showing
 > someone a circuit diagram that infringes a patent.  The expression
 > of the device (circuit diagram, source code) has no bearing, 
 > especially if use software (VHDL, Verilog) to transfer information
 > about a digital circuit to someone else.  I argue that this is yet
 > another aspect of the hardware/software codesign equivalence -
 > with VHDL I can represent a new device/invention in a generic way,
 > and then let the computer automatically generate a circuit diagram
 > or source code to implement the new device/invention.  Showing the
 > VHDL to someone else represents both the hardware and software worlds.

Imagine a whole operating system with thousands of applications which
is distributed as source code. Imagine a system for automatically
download it, compile it, and install it. Imagine that the source code
hits hundreds of patents. Imagine that any user around the world can
do this donwload-compile-install (and use) process.

When I describe this system to lawyers, they inmediately talk to me
about contributory infringement. I'm not a lawyer myself, adn it seems
that this figure is not well defined, to say the least. But it seems
that if a company distributes programs this way, they are looking for
trouble.

Well, this system is not really virtual. It exists and is called
Debian GNU/Linux or FreeBSD, for instance. Do you mean that both
Debian and FreeBSD can be distributed in source code and that they are
never going to have legal problems related to patents? I'd like to
hear informed opinions aobut this, since I'm mostly interested on
libre software, where you can always design systems this way.

On the other hand, I still have to see a VHDL *design* doing something
useful. Of course, the harware you build using the design can do lots
of useful things. In the case of software, there is no diference
between design and "the thing that does useful things". Source code
(and even binary code) is a detailed design. And source code (or more
usually binary code) is "the thing that makes useful things". Here
lies (one of) the difference...

 > 
 > Look, I have no problem with people's criticisms and attacks ono
 > software patents here - some have legitimate grounds.  I just
 > don't believe that software is really that different for those
 > attacks to be applied to other areas of technology.
 > 
 > You don't like a patent holder getting control of a new invention
 > for 20 years?  Fine, but those in other industries (and consumers)
 > have the same complaints as well.  You don't like patents
 > restricting the flow and communication of ideas?  Fine, but those
 > in other industries have the same complaint as well.  Software is
 > not that special to be treated differently under the patent laws.

I understand that many others may have compalints on patents in other
industries. And maybe they are right. But we have something different
here, in the same sense that a mathematical theorem is different,
and (for now) not covered by patent law. Of course somebody can come
and say that mathematics are not that diferent, and that they should
be covered by patent law. And to some extent tehy are doing it with
software patents. But that doesn't mean that I should agree that
mathematics is not different to other "indistries" with patents ;-)

Saludos,

        Jesus.

PS: I see you very interested in comparing with other industries. And
maybe that is an interesting point of view. But I prefer it the
other way around. Let's discuss whether software patents provide
enough benefit for the society, when compared to the harm they
cause. Maybe from this discussion people knowing other industries can
extract conclussions valid in other industries. But at least we should
be able of extracting conclussions valid for the software industry.

-- 
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




