Software Patents in Europe - May conference + update
I posted this to the wrong place a couple of weeks ago so it has been
overtaken by events a little. There has since (16th June) been a vote by
the JURI committee which essentially allows complete patentability of
software - there have been no significant provisions to help Free Software
or reduce the effect on small businesses.
The final vote is on June 30th - that's less than 10 days - it has suddenly
been brought forward from September. If that goes the way of the last one
(and it currently looks like it will) then the EU wil have US-style
software patents and all the associated badness. It's very important that
MEPs all over europe hear about how the current legislation will be very
bad for SMEs as well as Free Software (unfortunately the opinions of SMEs
counts for more than that of Free Software people with most MEPs at the
moment). If you are going to say anything about this then you need to do so
right now - next week will be too late. This is your last chance to try and
prevent software patent disaster in Europe.
The Greens/EFA and FFII (Foundation for a Free Information
Infrastructure) organised a 2-day conference in Brussel
(Bruxelles/Brussels) on 5th/6th May 2003. The issue of software
patents is something that should be of great interest to all
programmers. There were 3 DDs present that I was aware of (Wookey (me), Jama
Poulsen and Rene Engelhard (on the second day)). There may have
The conference consisted of one day of panel discussions and talks, half a
day of hearing at the EU offices, then a demonstration, and finally some
more panels and presentations. Below I describe the events of the
conference and some of the most accessible arguments put forward by
attendees on the subject. It should serve as a useful introduction to the
In case you haven't got time to read all this I'll start with the summary:
the current draft of the EU 'Directive on the Patentability of
Computer-implemented inventions' will allow unlimited patentability of
software, and quite probably business methods too. This will be bad for
Debian, Free Software, and most European software companies. It will be good
for a few large companies, especially American ones that already have huge
patent portfolios, and lawyers.
There appears to be plenty of enthusiasm for the current draft in the
European Commission and parliament, although many of those supporting it
claim that they do not want business method or software patents, but don't
understand the details well enough to realise that the current wording still
The next vote on this is in the JURI committee, which deals with legal
affairs and the internal market, on June 16th, and there will be a final
vote in the parliament around the end of June. If you want to avoid a
future where you will almost certainly infringe patents every day you do
some coding, it is time to explain to your MEPs why the current directive
is not good enough and what they should support instead. Time is short and
the subject is complex - we all have many urgent things to do, but whatever
deadlines you currently have, this is almost certainly more important in
the long term.
To give you a bit of an idea of my background - I didn't know anything about
software patents until a few months ago, but since reading some of the
papers on the FFII site, hearing Stallman speak on the subject and attending
this conference I have become a great deal better informed, and very worried
about the way things are heading. As well as seeing that swpats are clearly
a bad idea for software freedom I also have a practical interest as Chief
Nerd of a small European company that writes software. We could easily be
sued out of business by a company with a relevant patent, despite us having
no idea that such a patent exists - how can that be right, fair or
Over the 2 days there was a huge amount
of information given so I won't attempt to summarise it here. I'll
just give some of the best and simplest arguments I heard which give
some idea of the overall opinions. There were some significant
speakers: Lawrence Lessig, Richard Stallman, Mozelle W Thompson
(Commissioner US Federal Trade Commission), Brian Kahin (Prof. for
information policy studies, Uni. of Michigan), Reinier Bakels
(Amsterdam Uni, author of EU-sponsored swpat study), Jean-Paul Smets,
Peter Holmes (UK economist - one of 3 authors of EU report).
The most obvious thing was the continuous stream of SMEs who unanimously
stated that they didn't want or need swpats. There were representatives of
very small companies (such as myself), and individuals up to quite famous
and successful companies of both Open Source and Closed Source flavours
(MySQL and Opera respectively). There were people from Norway, Hungary,
France, Germany, the Netherlands, the UK, Ireland, Poland, Portugal and
America present (probably many other countries too). Not one expressed the
opinion that the ability to gain patents on their work would help their
businesses to innovate, or indeed help their businesses in any way.
The biggest complaint was that swpats cause uncertainty
impossible to run your business in such a way that you are safe from
unexpected attack by lawyers. Patents are written in a special
language which even experts in a field find difficult to
understand. In order to decide if you infringe a patent you have to
ask a patent lawyer, and they generally charge several thousand
dollars per patent for an opinion. There are already an estimated
20,000 patents granted in Europe and it is impossibly expensive for
small companies to check these against their work for infringement.
This means that they can really only ignore the problem but of course this
means that one day, should you become a little bit successful, a letter from
a patent lawyer will arrive on your desk accusing you of infringement. The
typical cost of defending such a case is about $500,000, even where prior
art is shown to exist or the patent is not actually relevant. How many small
companies can afford even one instance of this, never mind several?
A number of people made the point that the only protection they needed in
the market was that of copyright and being innovative. Having a good idea
and implementing it is plenty of protection against the competition in
software - they will always be behind you if they re-implement your ideas -
and copyright prevents them simply copying your software and thus catching up.
Patents would just slow you down and add costs to the system. There is
nothing wrong with the simple competitive market in software. Adding
state-granted monopolies to this is equivalent to having a law which says
that new software companies can not be created. No-one would pass a law like
that so why pass this directive?
The existing software market is not broken - software and software
companies developed just fine without patents (Microsoft never had any
patents before it got be huge, for example), so the premise that they
are necessary to encourage investment and innovation is clearly
wrong. Remind your MEP that to change the software legal environment
in this fundamental way needs some very good reasons and
evidence. They must ask for such evidence before making this
change. Reasons like 'we need to compete with America' are simply
ignoring the question - where is the evidence that doing this _will_
help us compete. Allowing software patents would actually remove an
advantage we currently have over America and thus be a massive own
goal. EU companies can already get US software patents if they need
them for use in the US market. It seems there is no good evidence
that software patents help the economy, but most people take it for
granted that they are 'a good thing'. This assumption must be
The only evidence that does suggest SMEs benefit from patents in the
US includes in its definition of SME companies which do no research or
development of their own at all. They just buy up patents from defunct
companies and then go round suing other companies which are making
real products. Is this the sort of development we wish to encourage?
Ask why politicians normally in favour of the free market think that 20-year
monopolies are necessary for software? The 20-year aspect is clearly highly
inappropriate for software, which may well have a life of 18 months or so. It
seems that this is written into all patent legislation, including world
agreements such as TRIPS so it is not possible to have a compromise of a
shorter patent for software (e.g 3 years). This all-or-nothing aspect is a
good argument for nothing in this case.
Your MEP will say 'other inventions get patent protection - why should
software be different'. The reason is that programming is the unobvious
combination of obvious things, which is just the same thing a chef does. She
takes a selection of ingredients which are commonly available but then
combines them in an unusual way to make an interesting meal. Should recipies
be patentable so no-one else can make that dish? If not, then neither
should software. In fact, many of the arguments against swpats also apply to
other patents and it may well be the case that the economy and freedoms
would benefit from the restriction of other patents too, but that is a much
larger argument than we want to have now and arguing for the complete
dismantlement of the patent system is a good way to get yourself dismissed
as a loony, so it is important to restrict ourselves to arguing against
patents in the field of software, where copyright protection already
Patenting works reasonably well where one patent corresponds to one product,
as in chemistry and pharmaceuticals. But every program of any significance
has thousands of ideas in it, and thus can be affected by thousands of
patents. This is not a practical proposition.
Those who promote patents do it because they have a financial interest.
Large corporations do it because they can afford the costs and use patents
to supress competition. Patent lawyers do it because they make a living from
it, and it never occurs to them that patents might not actually be a good
idea in a wider macro-economic sense. MEPs must be made aware of this bias
and consider what is good for _everyone_, not just large corporations and lawyers.
Wording of legislation
When considering wording you need to understand that patent lawyers will
twist language to an extraordinary extent. The current EU law (the European
Patent Convention of 1973) currently already says that 'Computer programs as such'
are not patentable, and something has to have a physical effect to be
patentable, but lawyers have successfully argued that 'Computer
programs' are not 'computer programs as such', and that because computers
are physical things that generate heat etc there is a 'physical effect', so
thousands of software patents have already been granted despite perfectly
clear words saying they can't be. I was astonished to hear of this sophistry.
Any new legislation needs to be written bearing this sort of thing in
mind. And it is similar arguments about the current proposed
legislation which allow MEPs to say they don't want software patents,
but still to allow them in the legislation which uses the biased term
'computer implemented inventions' throughout, and requires a
'techncial contribution' but doesn't define technical, which means
that it doesn't specify or limit anything.
The drive behind this is that patent lawyers want to increase their
sphere of operations, and have been doing so for several years. Clear
legislation saying 'software, ideas and algorithms cannot be patented' is
needed with a line drawn between genuine physical effects ('controllable
forces of nature') such as a chemical process which happens to be controlled
by a computer and software, (which could be patented) and software (which
can't, no matter what you use it for).
The other area that also really suffers from patenting is standards
bodies. JPEG, for example was created specifically as a patent-free
standard, even forgoing some speed in order to avoid known
patents. But now, 15 years on, one of these lawyer-troll companies
having acquired a patent on an aspect of JPEG stills, has started
enforcing it. They have got millions of dollars already from Japanese
companies, and sued the head of the JPEG group _personally_ for libel
when he said that their patent was invalid due to prior art.
Various things need to be done to make standard-setting useful. Time limits
on saying you have a relevant patent, use in software implementing the
standard being exempt from infringement etc (!!). Even this is not
really good enough for free software where some code might be re-used for
another purpose. This should not suddenly become liable.
A few MEPs (Neil MacCormick, Marco Cappato, Mercedes Echerer) came to the
conference, and a few more were present at the Hearing. Nearly all those who
heard the arguments seemed strongly persuaded of our position, but there are
several hundred other MEPs who do not know how SMEs feel on this issue- they
need to if we are to have a chance of either correcting or defeating this
On the second day we went into the parliament building for the hearing.
There were over 100 people who turned up to support the FFII position -
there wasn't enough room for everyone to get a seat.
Many people spoke convincingly, and some emotionally, about swpats, almost
universally condemning them, apart from a couple of people reporting studies who
merely said that there was no evidence they did any good. These included
Stallman (again), David Axmark (Director of MySQL), Opera Man !!, Laura
Creighton, (a venture capitalist), Lui ??, Brian Kahin, etc etc
Laura Creighton explained that she could invest less if swpats are
introduced as she will have to keep money back for the inevitable lawsuits.
She has already been involved with two in the US. In both cases the suits
were frivolous and the patents obviously stupid, but it was still cheaper to
just pay up than go to court to prove a point. We currently have an
advantage and should not throw it away.
Again a stream of people argued that it would reduce employment, destroy
some companies, hand others over to the US, and increase costs for all. It
would confer no advantage at all except allowing lazy venture capitalists
and accountants to value startups.
Opera Man described how they already get lots of hassle from US lawyers and
waste a lot of time that would be better spent writing software.
Towards the end of this session Ellie van Plooj, the draughter of the report
for the Commission on !!, turned up. She told us that there had to be a
directive, and it had to be like this. She then listened to only a couple of
responses before leaving. This was extraordinarily rude - she clearly didn't
want to listen to any arguments contradicting her position, and effectively
said that whilst we didn't like swpats, she had even thicker piles of paper
on her desk from people who did like them. The argument that whoever sends
in the thickest documentation in support of their arguments gets the
legislation they want is laughable.
This was the most depressing part of the proceedings. An enormous amount of
very good argument had been made by many people. This was clearly being
ignored by the most influential person on the committee. This is not how
democracy is supposed to work - we should at least get our arguments
After the hearing we went to the nearby Place du Luxembourg where some food
was provided, some amusing theatre on the subject of the unfair competition
between large companies and small ones was performed and banners
displayed. Prizes for the silliest patents, and the shortest infringing
program were also handed out.
I'd like to thank the organisers of this conference, especially
Hartmut Pilch, as it was extremely educational.
I hope I've persuaded European DDs that they need to act and
hassle/inform their MEPs. Particularly anyone who can present
themselves as a proprietary software author - currently people
complaining about this tends to be described (and perhaps dismissed) as
'Open Source advocates'. On this point proprietary and free programmers
have no disagreement and must stand together.
If you want to read more about this stuff then go here:
you can sign a petition supporting the ffii amendments to the proposed directive here:
This system also lets you find your MEP and contact them so you can point
out your opinions. Try to keep it simple: e.g. that you want to be able to
write software without having to worry about patents, and that in general
the existing swpats lok trivial to you, offend the sense and ethics of
software developers, and do not protect investments in such development.
Wookey, Debian Developer, Chief Nerd, Aleph One Ltd.
Aleph One Ltd, Bottisham, CAMBRIDGE, CB5 9BA, UK Tel +44 (0) 1223 811679
work: http://www.aleph1.co.uk/ play: http://www.chaos.org.uk/~wookey/