[Patents] Re: Patents digest, Vol 1 #162 - 2 msgs

[PILCH Hartmut phm@a2e.de]

Greg Aharonian wrote:

> "1. That the patent laws should recognize that business method and
> software patents are fundamentally different than other kinds of patents."
                                     
>     GREG comment: I disagree.  To the extent one equates business method
>     and software patents, in light of the Doctrine of Equivalents and the
>     latest hardware/software codesign tools, one has to equate software
>     and hardware/electronics patents.

Why?
Here are some possible criteria for distinguishing them:

* whether the patented object is a material good
* whether an industrial process is involved in creating the patented
  object
* whether an industrial process is involved in reproducing the patented
  object (e.g. copying of information and reproduction of life are not
  industrial processes nor are they inventive)
* whether the patented invention is technical in the sense that it 
  - involves the application of natural laws (==> material goods)
  - is automated, involves no human action (==> industrial process,
    no business methods)

see also the FFII proposal for change of the EPC 52

        http://swpat.ffii.org/stidi/epue52en.html

which takes recourse to Karl Poppers 3 worlds theory.

>     Thus from a computational theory
>     and statutorial point of view, it would be very hard to draft really
>     effective laws to distinguish between these conceptually equivalent
>     concepts.  

You need neither computational theory nor jurisprudence to make the above
distinction.  They are rather a philosophical problem.  Too bad that only
lawyers tend to get admitted into the discussion.

>     Sure the flows of prior art are different, but that does
>     not require changes in the patent laws, but rather the workflow
>     practices of the PTO (which the PTO could have done years ago).
 
Maybe that could improve things, but what do you do with one other
fundamental issue: the problem of how to define inventivity.

Novelty is straightforward to define, but exactly how much is an
"inventive step" or "inventive height" (Erfindungshoehe, in German patent
law)?  

One proposal is to at first publish only the "purpose of the invention"
and then conduct a public competition, allowing anybody to submit
solutions.  If more than a certain number of people submit the solution,
it is not considered inventive.

Would that fall under what you mean by "workflow practices" ?

Or is that not rather a problem at the conceptual level, i.e. properly
defining "inventivity"?

Even if it is a "workflow practise" problem, you still have the other
problem, i.e. the problem that software belongs to world 3 (information)
and is therefore not cleanly separable from the idea that is to be
patented.

The basic idea of the patent system is to promote the dissemination of
informational values (world 3) by limiting the freedom of transforming
these into material values (world 1).

This basic idea is violated/perverted once the "industrial application"
itself is also an informational value or, as with business methods, a
human-behavior value (world 2).

Even software patent advocates must recognize this problem.  In fact, imho
the above statement from Mr Bezos is the only one in his letter that I can
really subscribe to and recommend as a progress.  The conclusions he draws
(like limiting the duration of software patents), are of course neither
practicable nor really justified, as you point out correctly.  The only
solution to the patent crisis that I see is:  take software and business
methods out of the patent system and, if necessary, design sui generis
systems for them.

--
phm