Re: [Patents] [Fwd: "Computer Implemented inventions" and TRIPS]

[Hartmut Pilch <phm@a2e.de>]

> Therefore not only would it be untrue to say that the WTO (TRIPS) treaty
> requires the EU to treat software as a patentable invention as opposed
> to a literary work . Quite the opposite, under this treaty, any
> signatory who does not protect programs "as literary works" could be
> said to be in breach under WTO rules. Patenting of computer programs of
> any description could therefore be said to be in violation of this
> treaty, because it would mean no longer protecting computer programs as
> "literary works" but as "inventions" and they cannot be both.

The EPO crowd may have noticed this and has therefore changed their
wording.  In 1997 they still talked about patenting computer programs when
they are technical.  In 2000 they started talking about
"computer-implemented inventions" and saying that these are not computer
programs, even when the claimed object is, since 1998, a "computer
program, characterised by that .... " and even though the definition of
"c.-i. i." as promulgated by EPO and CEC covers nothing but computer 
programs in a patent claim context.

Since the inception of the law-change project in 1997, the EPO has created
caselaw for a sui generis software patent, in parallel to the sui generis
software copyright that we already have.  All TBA doctrines in this
direction were quickly adopted as official line by the EPO administration
and by Commission and Council.  The body of doctrines from IBM 1+2 to
Pension Benefits System and CEC+Consilium proposals constitutes a set of 
rules which are not accepted anywhere else in patent law and which deviate
from Art 27 TRIPs.

Art 27 TRIPs gives a finite list of admissible criteria for patentability.
However, in the case of computer programs, an extra criterion of 
"technical contribution in the inventive step" is introduced by the EPO.
Moreover a principle of "assessing the technical character of the 
invention by considering the claim as a whole" is introduced which is 
unheard of elsewhere.  Even worse, the technical character is affirmed 
even in the case of program claims, where the claim as whole refers to
pure information and thus contains no technical features (but at best
symbolic entities that can be interpreted as referring to technical 
features).  In this case one might as well affirm the technical character 
of an instruction manual or a novel that describes some new technology.

The departure from the normal scope of patent law (industrial products and
processes in fields of applied natural science) into the category of
informational entities is itself in no way mandated by Art 27 TRIPs and is
at odds with the requirement of treating information entities as
"copyrighted literary works" of Art 10.  One breach leads to another
breach: software patents are integrated into the TRIPs system only by
means of sui generis software patent law, which again is not permissible.

The cleanest interpreation of TRIPs is imho the one which the European
Parliament adopted under our influence:  computer programs are not
industrial products and data processing is not a field of applied natural
science but rather a layer of abstraction that is applied to all fields of
applied natural and social science.

The TRIPs treaty is clearly on our side, even without Art 7-8 which demand
that the patent system should be framed in a way so as to balance
interests and serve innovation.

        http://swpat.ffii.org/analysis/trips/

It is strange to see the Council pressing ahead against the Parliament and 
against TRIPs in secret negotiations now.



-- 
Hartmut Pilch, FFII & Eurolinux Alliance              tel. +49-89-18979927
Protecting Innovation against Patent Inflation       http://swpat.ffii.org/
300,000 votes 2000 firms against software patents    http://noepatents.org/