Re: [Patents] Patents meeting in Cambridge last week - follow up

[Hartmut Pilch <phm@a2e.de>]

> A valuable way of looking at the problem would be to say that the level of
> inventiveness required for a patent involving a software process should be
> one which deserves 20 year protection. RSA would fall into this category,
> whereas trivial software improvements would not.

Indeed a good way of looking at it, but it is not enough to say that RSA
is great.  You have to look at the actual claims.  Possibly the scope that
was claimed does not deserve 20 years.

Also, I see no feasible rule for selecting patents such as RSA and
distinguishing them from "trivial" ones.

        ">=20 twists of the brain?"

If it is 20, will all 20 have to be named in the claim?  What if someone
can then circumvent the claim by combining 19 of these twists with some
other twist?  Would the patent then still be worth applying for?

This is why even brilliant computing ideas tend to result in trivial
patents:  A claim with 20 twists is too narrow.

Besides that, there are other problems with patents such as that of
interoperability, which would have to be given precendence over any
patent, as proposed in one of the ITRE amendments.

Maybe these are not the only ones either.  An RSA patent, even if
reasonably narrow and not preventing interoperability, could slow down
development of some downstream applications significantly by imposing
burdens.  Also it could be a kind of innovation that comes from a sudden
thought rather than a major investment and therefore doesn't need any
monopoly incentive.

> 5. If the EU intends to formalise new rules on software patents they should
> extract concessions from the USA to:
> - a) improve their patent quality;

quality at EPO is not significantly better.

> - b) reduce the cost and delay of their judicial system;
> otherwise the EU will continue to import legal services from US patent
> attorneys.

In many ways the EU patent system is the most expensive one.

> We are encouraging people who were at the meeting to write to their MEPs and
> members of JURI. We will send this note to the JURI members.

A nice note and good thinking.

Still much of this is under the assumption that

        1. We somehow can't get around software patents
        2. There is a lot of good will out there for reforming the patent
           system

I think these two assumptions contradict each other.

If there was good will for all the small reform measures which you are
proposing, then there would also be good will for considering to adopt
a strict approach such as that in the early days of the European Patent
Office, which you find exemplified at

        http://swpat.ffii.org/papers/epo-t850022/

When writing to JURI members, it would be nice if you could put in a few
references, including that to our document about the JURI amendments

        http://swpat.ffii.org/papers/eubsa-swpat0202/juri0304/

and

        Analysis of McCarthy's latest paper
        http://swpat.ffii.org/papers/eubsa-swpat0202/amccarthy0305/

Perhaps instead of many of the proposed improvements it would be more
helpful to point out that McCarthy and Harbour show how their proposals
would elimintate undesired patents, based on the suggestions in the

        Patentability Legislation Benchmarking Testsuite
        http://swpat.ffii.org/analysis/testsuite/

This should be a minimal consensus position which no decent person, no
matter what his position on swpat, can want to refuse.

Also tell them about our meetings

        http://swpat.ffii.org/events/2003/europarl/06/

and come to them!

-- 
Hartmut Pilch, FFII & Eurolinux Alliance              tel. +49-89-18979927
Protecting Innovation against Patent Inflation       http://swpat.ffii.org/
145,000 votes 400 firms against software patents     http://noepatents.org/