From phm@a2e.de  Sat Oct 28 19:01:06 2000
Date: Sat, 28 Oct 2000 21:01:06 +0200 (CEST)
From: PILCH Hartmut phm@a2e.de
Subject: [Patents] Re: patents at three design levels of software

> > You write in
> > http://petition.eurolinux.org/consultation/
> > 
> >   the intellectual property created by the software creator can be divided into 
> >   three levels of design, which have been called level1, level2 and level3 in 
> >   this document. Level1 is shown to be equal to a business model and should be 
> >   patentable as such.
> > 
> > Have you not forgotten a "not" here?

> I did not intent that to be there, no. Is my understanding correct that 
> business models can not be patented (there is no technological innovation in the form
> of an implementation). 

Again:  should a "not" be inserted into your above sentence.
Or do you intend to argue that business models "should be patentable" ? 

> As such the above sentence means that level1 design is not patentable,
> just like a business model is not patentable.

In theory, they aren't.  However the EPO has found ways to make them
patentable.
 
> > I find this three-level division quite interesting and would tend to agree
> > that patents do most harm on level 2 (communication infrastructure) and mean
> > nothing on level 3 (implementation).  Still I think they also do serious harm
> > on level 1 (workflow model).  Just as level 2 patents work against
> > interoperability, level 1 patents work against free market competition and are
> > even questionable under some free trade treaties.
 
> Correct, the reason I left that level largely untouched is because I
> believed that business models can not be patented. So all we would
> have to proof is that level 1 patents are the same as business models.
 
> Am I wrong on the patentability of business models?

see above.
 
> > And the investment needed
> > for innovation at Level 1 is no bigger than at Level 2.  Level 3 is the only
 
> <snip>
 
> As I don't really understand german, I am wondering if the concept of
> different levels of design has been used elsewere. Is it a viable defence 
> agains the ignorance in the courts about software in general and software
> patents in particular?

Not really.
The real defense lines lie elsewhere.
The argument about different design levels was carried out in the 70s and
led to the court decisions saying that none of these levels is patentable. 
Especially Dispositionsprogramm 1976.
 
> Creating the legal devision in patents, and only allowing patents at level 3
> will be a huge step in the right direction (allthough no patents at all would
> be better)

The goal of the consultations is to find a general rule for delimiting
what can be patented and what not, without direct reference to software.

We should stick to asking for the reassertion of the underlying theory of
the current law, see my text "The ignored default option".  Compromise
proposals are not needed:  the politicians will themselves know how to
find a foul compromise.  That is their job, not ours.

What we really need are direct comments on the EU proposals with reference
to the personal or, even better, business experience of the author.

The consultation page needs to be rewritten a bit so as to contain more
hints to ongoing discussions and to how proposals should best be written.
Philippe Aigrain gave some useful hints, and they should be incorporated
into the consultation webpage.  Also, a link to the FFII's proposal for
reassertion of the underlying theory of Art 52

	http://swpat.ffii.org/stidi/epue52/

could help.  This is one of the options that this consultation is about,
and I would wish that many writers refer to this option, either supporting
or criticising it.

-phm





