From Luuk.van.Dijk@xs4all.nl  Thu Feb 15 21:44:34 2001
Date: Thu, 15 Feb 2001 22:44:34 +0100
From: Luuk van Dijk Luuk.van.Dijk@xs4all.nl
Subject: [Patents] Dutch MP's demand triviality criteria in the law before
 allowing SW pats

Ram'on Garc'ia Fern'andez wrote:

> 
> Although you probably know, let me remind you the discussion between me
> and Steve Probert where he said that "it is difficult to persuade
> programmers that 'easy' inventions are not patentable".
> 
> http://swpat.ffii.org/stidi/frili/indexen.html

I (and the MP's) realize that currently the laws put no lower bounds on 
these. The news is that the MP's which to see effective, clear and sharp
criteria, in the law, otherwise swpats should remain explicitly banned.

That is: not reviving the existing system, but to extend it so that it functions as intended.  Otherwise: no party.  They do not intend to
have some sloppy criteria (and the VOSN will keep its eyes open!) and
have a fake solution to this, as they even took the trouble to point out
to FENIT that their pro enthousiasm was not in the interest of their 
members.

Incidently: even if this is all arranged as proposed, the MP's are in
favour of explicitly listing business methods as non-patentable in a new
52.2, just in case someone uses the 'its with a computer therefore it
is technical' argument.

All 'n all I think the Dutch MP's have shown to posess a brain: either
way its good: or obviousness is effectively banned, or the ban on 
swpats remains.

> 
> The report by Joaquin Seoane and me, available at
> http://www.dit.upm.es/~joaquin/report.en.pdf has a chapter on
> the problem of trivial patents. It explains the history of the
> problem. I believe that it can be useful to you.

It already was, thanks. (btw its report_en.pdf, your link was wrong)

> 
> I think that the right thing to do is to put a criterion of
> difficulty. The invention should require a good amount of work.
> Ideas that are easy for any programmer should not be patentable. A
> design that take two hours of paper and pencil should not be
> patentable.

I think it will be difficult to make such things objectively measurable.

I was thinking more along the lines of arranging it
as a social game: providing the right incentives
for all parties involved.  I remember a link
on this list to a system where the applicant publishes
the problem along with a 'bounty' fee, the rest of the 
world gets a certain time to present a solution (new or 
already existing) and if succeeds gets the bounty, or otherwise
the monopoly is granted.

I'd be interested to hear other proposals.

Luuk van Dijk

___________________________________________________
Mind over Matter         Mind.over.Matter@xs4all.nl
The Netherlands                tel +31 6 224 97 227
___________________________________________________




