From phm@a2e.de Thu, 30 Dec 1999 19:01:45 +0100 (CET)
Date: Thu, 30 Dec 1999 19:01:45 +0100 (CET)
From: PILCH Hartmut phm@a2e.de
Subject: [Patents] Re: [Patents] Re: cute campaign?

> > The borderline is: industrial vs non-industrial.
> > Information is not industrial, because it can be replicated by copying.
> > Life is not industrial, because it replicates itself.
> > Aspirin can be patented, because in order to replicate it you need to go
> > through the industrial process that is patented.
> 
> ...or find a different process that produces the same chemical, which
> would no doubt get you sued.  I don't thing this analogy is perfect,
> since I'm in favor of eliminating all software patents (at the expense
> of the tiny minority that are "good"), but I'm not sure I want to
> allow companies to get around drug patents by (wastefully) developing
> different synthesis processes.

My suggestion doesn't necessarily mean that no more "product by process"
patents can be granted.  

I don't like the idea of "product by process", since it makes discoveries
patentable.  And I don't find the development of workarounds wasteful at
all.  Patentability advocates usually claim that they want to have patents
in order to stimulate the development of workarounds.

Nonetheless, allowing "product by process" patents is well in line with my
above borderline definition.  You only have to demand that whatever
workaround processes are comprised in the scope of the patent, they must
be new, inventive and industrial, i.e. not simple copying or
self-replication.  If they don't fulfil these criteria, they should have
been properly described in the first place, and no "product by process"
patent would have been necessary.

> It's a question of judgment on the part of the patent offices.  If
> software patents were granted at 0.01% the rate they are now, there
> would be no objection to software patents.

I find it very hard to imagine such a case.  Software patents are the
result of patent inflation in the first place.  To reduce the number of
patents by 1/100, a radically different patent law would have to be in
place.  I am just designing possible criteria for such a law.

> I wonder if this whole issue will be resolved by simply waiting until
> all the most obvious software patents have expired and patent offices
> get better at identifying inventive and useful software processes.

You mean waiting until the US has tackled the "patent crisis" ?

> Until then, we would be better off banning software patents.

You mean keep the EPC 52 as it is, I guess.

That would be a possibility.  However we would then have to press for a
change in the practise of applying this EPC.  Probably the concept of
"technicity", which the lawyers are currently using as the hinge for all
of their argumentation in favor of deviating from EPC 52, and which they
have written into the TRIPS treaty, will have to be given an
interpretation which includes "industriality".

Traditionally, it comes very close already.  "Technicity" implies the "use
of natural laws without human intervention".  This can be further
specified to mean the production of non-self-replicating material goods.

--
phm





