From jp@smets.com Tue, 04 Apr 2000 22:07:40 +0200
Date: Tue, 04 Apr 2000 22:07:40 +0200
From: JPS jp@smets.com
Subject: [Patents] Re: [Freesw] Analogy (was: Do patents forbid writing books about them?)

There is a very nice article explaining how the EPO has slightly
misinterpreted the Munich convention.
It it published in the very classic "Lamy Informatique" on IP and
software (one of the best text books in France on IP and software)

It says for example in conclusion that :

"national and conventional rules are clear : software can not be
patented by principle. The game which people are playing nowadays is to
bypass them from one way or another, by considering for example that the
combination of hardware and software is equivalent to a virtual machine
which could (one day) be patented. From this point of view, one can
mention the existence of software patents. But such patents, whatever
the way they could be granted, have only the value one wants to grant
them - and one should not refuse the possibility that a consensus could
lead everyone to grant such patents a real value. The efficiency of this
way to bypass legal rules largely depends on the fact that there is or
there is not a consensus to accepts this game - against the positive
rules written in the law -. Therefore, the question whether software can
be patented or not is not stricto sensu a juridical question anylonger.

From this point of view, it is likely that one should consider the
possibility to change written rules (in terms of removing the
exceptions".

Basicaly, it says that
	- Software can not be patented in Europe according to the law
	- The EPO has biases the law with new rules which contradict the law
	- software patents granted by the EPO could be considered as "void" by
a judge
	- The EPO rules are stronger than written law if there is a consensus
in favour of EPO rules

If one then considers that only the member states of the EPC could sue
the EPO in front of the international court of justice (La Haye) for
wrong interpretation of the Munich Convention, then, the fact that no
state ever did it, and no state is likely to do it from now (10.000+
patents would be canceled by such a legal action), leads to the
conclusion  that member states of the EPC have de facto already
legalised software patents in Europe and that it is very unlikely that
they will change their opinion.

Considering that the "one click" patent has been granted by the EPO, it
is now certain for me that Europe will just legalise patents in a way
which will be very similar to the USPTO.

Now, it does not mean that this legal evolution process is fair,
democratic, or based on sound economic rationale.

JPS.



Bernard Lang wrote:
> 
> Our issue here is not what is reasonable, or even what is done,
> legally or not, but what the law says, or will say and how it may
> impact technological and economic development.
> 
> Actually, as far as I remember, infringing for personnal use is
> illegal in some countries, and possibly legal in others (I am not
> sure). Same when infringing for research purposes.
>   But infringing in using a tool for work on hire is definitely not
> permitted.
>   <RANT>
>   And do not believe a second that the legislator uses common sense,
> or even public interest as a guide. Main factors are lobby pressure
> and ideology. Main reason for that is that they do not
> understand. Main proof of what I say : they will not bother studying
> the possible impact of decisions before taking them.
>   </RANT>
> 
>   To come back to the original question Jesus asked I forgot to
> develop the initial point:
> 
>   A major question is whether only the use of a program is protected
> by a patent, or whether the program itself is (either directly, or as
> contributory infringement, which essentially has the same effect).
>   The original Munich conventions states that a program "as such"
> cannot be patented. No one really knows the meaning of "as such". This
> can be interpreted as meaning that a program running on a general
> purpose computer is not patentable, unless is is part of an invention
> that includes an inventive process that spans more than that program.
>   I.e. "as such" means "just a program running on a general purpose
> computer".
> 
>    Then allowing patents on a "program as such", which is now intended
> by changes in regulations, could just mean that the execution of a
> program can be covered by a patent.
> 
>   But some people want it to mean that the "program as such" is the
> program on some storage media, independantly of any execution, or
> equivalently to apply contributory infringement to programs on a
> storage media. That means that a program, i.e. the representation of a
> process, can no longer be freely disseminated.
>   And that is going much further than the patenting tradition in any
> other technical field. Indeed, it is going against the very purpose of
> patenting, which is to allows free dissemination of descriptions of
> the patented process or product.
>   It really is an abuse, even when considering existing rules in other
> fields.
> 
>   Then the discussion I presented in my previous message is about this
> situation. Where is the line ?  What descriptions are legal, and what
> aren't.  It there any logical way to draw the line, or shall we have
> to resort to (incompetent) courts to draw arbitrary (absurd) lines.
> 
>   The fact is that there is an attempt to pervert the very nature of
> the patenting process... and no one knows where that leads.
> 
>   Having such infringement threats on reproduction of programs would
> actually kill the development of open-source (for example by bullying
> ISPs to censor the publication of software by their customers, by
> means of infringement allegations - true or not), and also the
> dissemination of shareware. The latest trick against free speech is
> holding ISPs responsible for their customers, even when they do not
> know whether the customers are doing something illegal.
>   see for example:
>     http://news.bbc.co.uk/hi/english/sci/tech/newsid_695000/695596.stm
>     http://news.bbc.co.uk/hi/english/uk/newsid_696000/696289.stm
> 
> Does that clarify things a bit
> 
> Amicalement
> 
> Bernard
> 
> On Mon, Apr 03, 2000 at 08:52:55PM +0000, Tim Cook wrote:
> > Bernard Lang wrote:
> >
> > > >       And with a broader scope, I had the idea that patents only
> > > > cover actual "artifacts", which in the case of software would be
> > > > binary code.
> > >
> > >   which is no more an artifact than source is. Translating from source
> > > to object is mechanical. The source code need not be very detailed
> > > ... it all depends on the sophistication of the compiler technology.
> > > And source may become very high-level with time (like math proof, that
> > > seldom give many details, but are nevertheless sufficient).
> > >
> >
> > How does this patent analogy apply?
> >
> > There is a patented mechanics tool for a specific purpose. The
> > drawings and a technical description are available from the
> > patent office. If I need one of these tools and I fashion one for
> > myself have I infringed on the patent?
> >
> > There is a patented software tool for a specific purpose. The
> > source code (or part of it) and a technical description is
> > available from the patent office. If I build one for my own use
> > have I infringed on the patent?
> >
> > In the first case most reasonable people would say no. Many
> > handyman types and homeowners do this all the time and even share
> > it with friends. Have you ever heard of one being prosecuted for
> > it?
> >
> > The lawyers/legislators (without even realizing it) will let the
> > 'manufacturing costs' influence their thinking process. Because
> > it costs so much less to build a software package than to 'build'
> > a wrench. While those who do 'build' software know this isn't
> > necessarily true most people don't realize it.
> >
> > Isn't the purpose of a patent to allow information sharing while
> > protecting the original developer in the 'commercial market'?
> >
> > -- Tim --
> > Open Source Health Care Alliance
> > Join us to help build solutions!
> > http://www.openhealth.com/en/healthcare.html
> >
> >
> > _______________________________________________
> > Freesw mailing list
> > Freesw@conecta.it
> > http://mail.conecta.it/mailman/listinfo/freesw
> 
> --
> Bernard.Lang@inria.fr             ,_  /\o    \o/    Tel  +33 1 3963 5644
> http://pauillac.inria.fr/~lang/  ^^^^^^^^^^^^^^^^^  Fax  +33 1 3963 5469
>             INRIA / B.P. 105 / 78153 Le Chesnay CEDEX / France
>          Je n'exprime que mon opinion - I express only my opinion
>                  CAGED BEHIND WINDOWS or FREE WITH LINUX




