On Wed, 16 Mar 2005, David Golden wrote:
> That's why their "but but but we're only codifying current patent
> office practice" is such an incorrect and dangerous argument: the
> current practice of the EPO runs counter to what e.g. I as a
> european citizen would regard reasonable reading of the current
> european law. The EPO is out of control, explicitly considers
> itself _above_ european law, and needs to be reined in, basically.
Well, it _is_ based on current office /practice/. I believe the
commision when they say that.
Further, AIUI, this practice didn't arise simply cause the EPO (and
other offices) don't give a stuff for the wording of the EPC - it
came about because of rulings in various member states (Germany most
particularly AIUI).
> Beware quoting the commision's opinion on the matter: they are far
> from an unbiased body.
To be honest, I think the commision's stated opinions (least those
i've read - mccreevy mostly) and objectives are quite fair. It's just
their priorities are quite different, namely harmonisation. There is
currently a distinct possibility of member states taking
contradicting positions towards patents, between those who have
already ruled that (essentially) the technical effects arising out of
computer programme innovation can be patented, the EPO's position
(which arises from this 'precedent' set by those member states) and
other member states whose status either do not allow software patents
or else whose statutes have not yet been definitively interpretated
on the subject.
That is, I don't really think the commision (or vast majority of
politicians really) care about the gorey details of whether software
should be excepted from patenting. I think they're approaching this
from the pragmatic angle of just harmonising whatever is the
generally acceptable practice - which is not a /bad/ goal per se.
Further, there are other concerns, eg TRIPP. Any EU position would
need to be compatible with it.
(And I'm sure that just as programmers sometimes can be obsessive
about achieving unambigious and clean specifications or
implementation of things, that lawyers too prefer clean unambigious
law. A principle/spec/law that requires a list of exceptions isnt a
very 'clean' principle/spec/law.)
> But you're "right": counter to what a naive programmer might expect
> from the pretty plainly worded current law (article 53 of the
> European Patent Convention of 1973 - see [ref. 1] below ), the EPO
> and the patent offices of certain member states have effectively
> been acting as if software patents were allowed.
Well, the problem is that when you have:
"This broad principle shall apply, except for the following specific
excepted things a, b, ..."
Eg: "Ideas can be patented, except for software programmes running on
computers, mathematical papers, ..."
It is *very* easy to find a way to interpret some parts of $something
as falling outside the exceptions, and hence being covered by the
broad principle. Eg, software patents - you can always 'extract' the
idea from the code and define it in terms of interaction with
'objects' outside of the software (this is, IMLU, how "software
patents" have been deemed valid by certain member states and the EPO,
despite the wording of the EPC) - is the idea when expressed in this
way still to do with software/computer programmes? It's very easy to
argue not obviously.
This is obviously quite a pessimistic view on the subject of whether
trying to get software excepted from patents is an achievable goal
(short or long term). I'm actually starting to come to the opinion
that is not practically achievable, that no matter how you word the
exception patents will still have a bearing on software.
> But this has not been tested in court for the most part AFAIK, it's
> only the patent bureaucracy's documented and deliberate
> reinterpretation of the law to maximize patent lawyer profit -
> history at [2].
My understanding is that it *has* been tested in the courts, refer to
the commisions proposal for this directive - it's why we already have
"software patents".
> And the corporations are afraid to test that in court, because
> right now the law is pretty clear that software CAN'T be patented,
> despite the patent bureaucracy just sort of assuming it has the
> authority to decide it can be
I believe you're mistaken here - unless the commision outright
blatantly lied in the proposal. Further, see the recent article on
Groklaw[1] which discusses some of the history (as told by a patent
lawyer). (btw, the ILUG/kde.ie/FSIE letter got on groklaw!).
> With the new directive, they'd suddenly be able to quote a new law*
> supporting their position and their 30K currently basically
> unenforceable software patents - instead of their current deeply
> suspect argument, which is along the lines of:
No, it wont be a new law. It'll be a new /directive/ which member
states will have to conform to. For at least a few member states this
will not require *any* new law - because they are already conformant
(AIUI), for some other member states it will require only minor
adjustments (because their position isnt clear), for some others it
will require actual substantive change in their position.
> """
> We the patent office, having failed to delete exceptions in Art 52
> silently and without public scrutiny back in the 90s, hereby interpret
> "as such" in Art. 52(3) with reference to Art. 52(1)c to mean
> software on its own, when not running on a computer.
> """
I dont think this is fair. Also see above about the "Broad principle
shall apply, except for specific thing" problem.
> For some reason, the corpies are pretty afraid to test in court the
> EPO's argument. That's also why they try to shift debate to "what
> is a technical contribution" rather than "should software patents
> exist at all".
See above. This is a general problem with having specific exceptions
to a broad principle.
> * and quoting law is more powerful in civil law jurisdictions of
> continental europe than common law jurisdictions like Ireland and
> England, where precedent is more important.
I have no idea about how the continental napoleonic-derived (amongst
others) judicial systems work.
> Personally, I disagree with all patent law - I don't see why anyone
> should have the right to stop me building stuff. whether it's
> similar to what someone else is doing or otherwise, not do I seek
> to restrict others building stuff. Patent law should only apply to
> patent holders.
Having it apply only to patent holders is impossible ;) - who'd file
a patent? :)
I think current patent law and practice has a lot of problems. There
are two problems, one major (simply a general truth) and one minor
(of our own doing).
1. "Innovation" seems to increase exponentially
The more a society innovates, the more innovations there exist to:
a) Make a society more productive/efficient (including at
innovating)
b) Build further innovation upon
Eg, the internet firstly makes communication and collaberation far
far more efficient and secondly provides a basis for even more
innovative inventions to build upon.
2. Given that the broad principle of patents revolves around "ideas",
we have broadened the scope of patent law ever and ever more.
Initially patents were scoped primarily to cover technical ideas.
That scope has increased and increased to cover ideas in more and
more fields, where now it literally does seem to cover "ideas" in
general (with a few narrow exceptions). (this is i think a
consequence of the "broad principle, except ..." problem above).
The obvious consequence is that a system which relies upon patent
offices properly examining patents for validity simply becomes
swamped with applications and ceases to be able to examine claims
properly leading to logjam of applications and/or granting frivolous
patents.
I think it is possible to achieve wide consensus on tackling this
problem. It wont even need lobbying by relatively small groups of
computer 'nerds'. The annoyance of software patents potentially could
be resolved within the context of such reform.
As to whether, within the scope of the principle of being able to
patent ideas, sofware should be excepted, I dont know. It's
fundamentally an inconsistent position, I don't think broad consensus
for this is possible. I'm not even sure what my position, as a matter
of principle, is on this - even though I'm sure that software patents
would be a bad idea given the current state of patent systems.
> or doing business, and programs for computers;
Yes, but what is a progamme for computers exactly? :) (NB: a
rhetorical question... the fact this would require a long long, and
likely inconclusive unless you define it literally, discussion should
make that obvious ;) ).
1. http://www.groklaw.net/article.php?story=20050314114813919
regards,
--
Paul Jakma paul at clubi.iepaul at jakma.org Key ID: 64A2FF6A
Fortune:
Praise the sea; on shore remain.
-- John Florio
Maintained by the ILUG website team. The aim of Linux.ie is to
support and help commercial and private users of Linux in Ireland. You can
display ILUG news in your own webpages, read backend
information to find out how. Networking services kindly provided by HEAnet, server kindly donated by
Dell. Linux is a trademark of Linus Torvalds,
used with permission. No penguins were harmed in the production or maintenance
of this highly praised website. Looking for the
Indian Linux Users' Group? Try here. If you've read all this and aren't a lawyer: you should be!
