Subject: Re: patent trolls and X-licensors
From: simo <s@ssimo.org>
Date: Sat, 03 Jun 2006 10:18:53 -0400

On Sat, 2006-06-03 at 17:46 +0900, Stephen J. Turnbull wrote:
> >>>>> Santiago Gala <sgala@hisitech.com> writes:
> 
>     Santiago> El jue, 01-06-2006 a las 18:43 +0900, Stephen
>     Santiago> J. Turnbull escribió:
> 
>     >> Maybe in Europe.  In the U.S. it's quite explicit.  Practicing
>     >> the patent without license, for any purpose, is forbidden.  Nor
>     >> is there a fair use exception, for research or anything else,
>     >> as far as I know.
> 
>     Santiago> I don't understand what you call "practicing the
>     Santiago> patent".
> 
> Implementing the process in a form that can be executed, whether as a
> physical process or a virtual one.
> 
> Given that software patents are allowed in a given jurisdiction, in
> that jurisdiction it would include writing a program.
> 
> According to simo's post, it would appear that such jurisdictions
> include Germany, so I don't know what you mean by

In the European Convention on Patents, software is escluded from
patentability, but the EPO (European Patent Office) has already awarded
30000 or more software patent.
If you have the money usually you can go up to a court and invalidate
such patents, but it is so costly that many just don't have the means to
go to courts (even if it costs less then in US).
The last 5 year I have been working in my free time as an activist
against software patents in Europe, and against an European Directive
(CII) that would have legalized software patents.
The objective was to change the directive to clearly state that software
was not patentable addressing the loopholes that the EPO used to grant
them. The outcome was that the Directive, for the first time in European
history was retired after the second lecture. It wasn't a complete
victory but at least there is still space to fight software patents
there.

>     Santiago> At least in Europe's traditional legislation, only
>     Santiago> industrial processes are subjects to patents,
> 
> Would you like to explain what is meant by "industrial process"?

In Europe we used to define Industrial Property and that comprised
patents, basically our proposal in the amendments for the CII directive
defined patents valid only for physical things under the natural forces
doctrine. A patent could be granted only if it teaches something new in
the manipulation of the physical forces of nature. This would exclude
any abstraction and mathematical processes, and of course software and
business methods.

>   Does
> a paperclip in a novel shape, for example, count as an "industrial
> process"?

No, a shape can at most be awarded a so called design-patent, unless the
shape itself has a property that make the object act in a different way.

>   Is the "Post-It Note" an industrial process?

Do you mean the physical Post-It ?
In this case, what is patented is not the post-it as such, but the
physical process needed to build it.

> As for "physical processes," would a paper-feed mechanism whose timing
> is controlled by cams or other mechanical delays be patentable, but if
> the same timing were controlled by software and electro-mechanical
> actuators it would not be?

The use of software in an invention does not preclude at all the grant
of a patent, but yo have to show something new, if the only difference
is that software controls a well known physical process there is not
much to patent. Instead if, thanks to software, you can show a new way
to control a physical phenomena, then probably the whole invention will
be patentable (not just the software itself, but the whole).

Simo.