Re: [Patents] Halloween X

[Jonas Maebe <jonas.maebe@elis.ugent.be>]

On 10 mrt 2004, at 17:33, Gregory Aharonian wrote:

> I assume you mean that "information processes when expressed in 
> software
> should not be patentable, but information processes when expressed in
> hardware should be patentable"?

Information processes should never be patentable. Saying that only 
information processes carried out by physical means should be 
patentable is non-sensical, because it's always done that way: with a 
computer, with pencil and paper, with your brain (chemical processes, 
electrical signals!), ... The reason is that humans are physical, and 
can't interact with non-physical things (just like a computer cannot 
physically interact with a computer program, or a cook cannot 
physically interact with a recipe).

That does not, however, mean that all hardware carrying out information 
processes is not patentable (the "as such" part of art 52 EPC). It's 
quite possible that some hardware construction involves a new technical 
teaching (and is no-obvious and industrially applicable), and therefore 
is patentable. Whether this hardware is used to perform information 
processing or pumping raw oil to the surface is irrelevant.

As far as chip layouts are concerned (described in VHDL or in another 
way), that's exactly what we have a sui generis protection for, I 
thought.


Jonas

PS: since this is my first post to this list, I'll introduce myself for 
those that don't know me yet from other lists:

I'm a PhD student in computer science at Ghent University, Belgium. I'm 
researching dynamic instrumentation techniques (modifying/observing 
programs while they run) and applications of it. I'm actively involved 
in the software patents debate since June last year.