This is a talk about preventing software patentability in the EU
By Ciarán O'Riordan
Brussels Representative, Free Software Foundation Europe
Committee Member, Irish Free Software Organisation
Claimer: Most of the ideas came from other people but the writing of this is by Ciarán O'Riordan. The mistakes are his.
Copyright © 2005 Ciarán O'Riordan
Royalty-free permission is granted to redistribute verbatim.
Note: Copyright doesn't restrict ideas or information, so it goes without saying that the ideas and information contained can be copied, modified, and redistributed freely. Also, copyright doesn't prohibit copying of excerpts, so feel free to do that too.
I'm really glad to be here because it was just over two years ago at AFFS's first AGM that I gave my first talk about software patents. Since then, I've helped set up IFSO, Irish Free Software Organisation, and have moved to Brussels to work on EU directives. In January of this year, I was talking to Georg Greve of FSF Europe. He was surprised to hear I was living in Brussels, he said he couldn't find anyone that wanted to live there. And when I told him I was supporting myself by working as a trainee barman, he said “What a waste. We can pay you a trainee barman's wages!”. So in April 2005 I started working full-time for FSF Europe.
My main focus is preventing software patentability, but I also represent FSF Europe at meetings that happen in Brussels, I try to promote FSF Europe's Fellowship program, which I'll mention at the end of this talk, and I try to do the things I used to wish FSF Europe would do.
One thing that falls into the last category is to set up a wiki for the European free software community. So myself and Glenn Strong of IFSO are currently setting up a wiki for FSF Europe, which I hope will be useful for things like collaboration between IFSO and AFFS. One thing we might collaborate on is the new directive which is being proposed on intellectual property rights enforcement. This directive aims to introduce the threat of jail sentences for intentional copyright infringement, and would give the next SCO more investigative power than SCO had in the USA.
Another project I'm thinking about is a “What's GNU?” project. Not to promote the GNU project, but to promote the question: “What's GNU?”.
Use of GNU/Linux has greatly overtaken understanding of software freedom.
There are a lot of companies promoting GNU/Linux that don't want the users to know about free software, and there are a lot of individuals promoting GNU/Linux that don't see the importance of informing new users about software freedom. It seems a lot of people don't hear about software freedom until they ask the question “What's GNU?” - The answers are there on gnu.org, but people aren't asking the question. So I think we should find a way to promote that question.
In July, we prevented the pro-software-patent campaign from pushing through a directive which would have made software patentable.
They spent 10s of millions on there campaign, possibly over 100 million, and walked away with nothing. They won't want to do that again.
What we achieved is particularly amazing since this was the first big directive that was thrown at the EU that would damage software freedom. In 2000 the only experience we had in bad legislation was laughing about stupid US laws when reading Slashdot. Suddenly we had one on our own shore, but we pulled together a campaign strong enough to hold back the pro-software-patent camp.
But all we have is a pause, the battle's not over. The pro-software-patent camp will be examining our strengths and weaknesses. Any holes in our argument will have to be closed. For the next battle, we have to get our message right - so that's what this talk is about.
There's a pause right now in the software patents battle. We didn't win much in July, but we prevented our opponents from winning. We'll have to do better next time if we're to get a real victory, so I want to use this pause to pass on the useful things I've learned.
What I want to do is get people to be informed, help people to contribute in whatever way they can.
A lot of the detailed discussion happened on closed lists. This was necessary so that people could speak frankly without political backlash, speak secretly so our opponents don't hear, and speak away from the disturbance of others. But disturbing the active people is how new people learn what's going on - so the flow of information was broken, for a good reason, but now I want to try to summarise what happened, so we're all on the same level again.
There could be legislation at national level, but this is unlikely. Every national parliament knows this is a hot topic in the EU level. Anything done at national level would be purely symbolic and would be overwritten by EU level stuff some time in the future.
At the moment there's disagreement over whether software functionality is patentable or not. Interpretation is needed, and this needs a debate in every EU member state. The only place the debate counts is in the courtroom.
There could be case law. This quite likely. Anyone can bring a patent to court without needing support from others so it's almost inevitable that such cases will happen - whether the anti-software-patent or the pro-software-patent camp like it or not.
So should we bring such a case? Probably not. It would involve money and full-time lawyers: this is not where our resources lie. Plus, the precedent is at the national level only.
Some good news is that Hartmut Pilch is reporting that a UK court has recently rejected a patent litigation suit involving Haliburton, and one of the reasons given by the judge, according to Hartmut, was that the patent covers software and that's not valid.
We can ask the Parliament to pass a resolution, supporting our 21 amendments. This wouldn't be legally binding, but would put an end to the spin-doctoring of “the MEPs voted for the status quo”.
This would only be a temporary measure but it would also send a useful signal to other countries. We would have been very glad during our struggle if the USA's equivalent of a parliament had passed a resolution saying they wished software wasn't patentable.
Another directive, called the “Community Patent” directive, could be modified to address this issue. Unlike the Software Patents directive, the Community Patent directive doesn't modify the European Patent Convention, so it can't define what is and isn't patentable - but it could establish a patent appeals board, or worse, a patent appeals court. One that could give definitive decisions on patent validity. Where will they find the experts for this court? Who has a deep understanding of patents at the EU level - oh yeh, people from the EPO. The very ones that interpret “software as such” to not include software.
If Community-Patent-based solutions are to be based on EPO current practice, it's useful now to discredit EPO current practice. Our representatives should be shown patents that are harmful to society.
The Community Patent directive is being handled under a different procedure to the Software Patents directive. The European Parliament only has an advisory role in the Community Patent. Unanimity among the member states, which is the Council, is what's required. This won't happen quickly.
So I'm going to focus on legislative stuff. Not only is it the definitive answer to the question of what's patentable, but also because it's where our abilities lie, and because this won't be the last directive and I think we've seen as complete a picture of lobbying tactics as has happened in the short history of the EU Institutions. So we hope to get a new directive based on the parliament's first reading, or based on the amendments for the second reading.
We might be able to get the parliament to request a new directive. We'll have to look at the Rules of Procedure - maybe rule 39 can be used.
The European Commission: Does studies, drafts legislation. In 1997 it did a study which concluded that patentability criteria should be harmonised across the EU member states. In February 2002 it published it's proposal for the software patents directive.
The European Parliament: Represents the citizens. Comments on legislation. In September 2003 it amended the Commission's text to exclude software from patentability.
The Council of Ministers: Represents the national governments. Comments on legislation. In March 2004 it looked at the parliament's amendments and decided to discard them all.
So the text of directive usually bounces back and forth between the Parliament and the Council until their opinions start to converge and then their's consensus. This wasn't happening for this directive.
A month before the parliament's 2nd reading, the number of pro-software-patent lobby groups mushroomed. They gained a lot of ground, and the lobbying on both sides changed a lot.
In the month before the vote, one of the political parties in the European Parliament, the Greens/European Free Alliance, got a room for the anti-swpat lobbyists. So we had a room, 4 computers with Internet access, and a photocopier.
The only downside was that this room had no windows and the air conditioning level was clearly set on the assumption that the room would contain 4 people working 8 hour days. When there are 10 people working 12 hour days in such a room, the air starts to taste funny. But without this we would have been lost. So the Green group deserve thanks for that.
So we had an office, and each day we'd turn up for work at 09h00 and ambush an MEP assistant to ask them to sign us in.
During the day we would meet MEPs and prepare SME representatives for their meetings. And when the MEPs were gone home we would write mass letters, print a copy for each of the MEPs and before we went home we would drop a copy into each of the 730 mailboxes.
Our opponents also delivered such mass leaflets on a daily basis just before going home, but in the morning our leaflets were on the top of the mail piles because our opponents went home at 17h00. We went home at 22h00, 00h00, or 03h30.
The lobbying got intense. Our opponents started saying “this directive won't introduce software patents”. This was easy for them to say, and to refute it we had to show MEPs the text and explain our interpretation to them.
As Lawrence Lessig says: When you're explaining, you're losing.
Our opponents were professional lobbyists backed by well known brand names.
So we were very glad when SAP bought 4 full page ads in a prominent European Newspaper. From then on, when an MEP said “but this directive won't introduce software patents”, we could take out our photocopy of the SAP ad and say “Then why has this company, that makes nothing but software - a 'pure software' company - why have they spent 28,000 on this set of ads?”
It's nice to get something so clear, so understandable.
The three days before the vote were even more intense. So as not to waste time traveling from the parliament to a hotel, I slept on the floor in the parliament. I got out of bed on Sunday morning, and the next time I was in a bed was Wednesday night. We all knew we could sleep as much as we liked after the vote.
The parliament was covered in software patent lobbyists from both sides.
The day before the vote, there was a yacht in the river that divides the two parliament buildings. It was just driving around with a big pro-software-patent banner. This was visible from the bridge that joined the two parliament buildings, and from the two restaurants.
So some anti-software patent people hired two canoes, and paddled around the nameless yacht. This was a brilliant message that the MEPs understood. The yacht was soulless, expensive lobbying, and 4 guys with much smaller budget, but that were willing to put the effort into paddling a canoe around.
Of course, the guy in the yacht didn't react at all to the anti-software-patent canoes. He was just doing what he was paid to do. He didn't care.
But the day before the vote, the strangest thing happened. We had 21 amendments, and because of the voting system, each one needed a majority of all MEPs those present and those that were not. So in practice we needed a Yes vote from probably 65% of those that turned up. We didn't know if we'd get it.
Then the pro-software-patent camp decided it was going to push for rejection.
Rejection in the European Parliament is rarely asked for, but is pretty easy to obtain because there is a core of the MEPs that vote to reject every directive - The UK Independence Party is one example.
When we heard they were pushing for rejection, we saw an opportunity that mightn't come again. This needed serious thinking. Quickly.
The stakes of this gamble were very high: if we don't get 65% majorities for most of our amendments, we've lost everything, software would be patentable. And if we win, we have no guarantee that the Council won't simply discard our amendments like it did the first time.
So what we were gambling was massive, and maybe we were playing for nothing.
If the parliament's amendments could be ignored, it seemed that the only sure power the Parliament had was the power to reject directives. So the anti-software-patent camp pushed for rejection too. And it was rejected. FSF Europe's position was still that we wanted a directive. We we're one of the only groups that didn't move to rejection. We wanted a directive, we all do, but in the end this directive was rejected.
Most of the press misinterpreted this in two ways.
First they thought this to mean that things were neck-a-neck, but neck-a-neck would have been a clear victory for pro-software-patent camp because of the voting rules. The actual meaning was that the pro-software-patent camp estimated that the anti-software-patent campaign would get the 65% it needed for each or most of it's 21 amendments.
The other thing was that some journalists said that the MEPs voted for the status quo. In fact, this was a vote against one proposed solution. The status quo isn't good, we want a directive, just not that one.
So that directive is dead. And the MEPs never want to hear the words “software patents” ever again. But they will. Soon after that vote, the Parliament disbanded for holidays. Giving everyone a break, to forget about it, or to think about what's next.
Looking back, even the day before the vote, the many MEPs didn't understand the issue. Some didn't know the difference between software and hardware. Some didn't know that European patents were only enforceable in Europe. Some didn't know that European companies could get US patents already. From a democracy point of view, maybe it was right that they rejected it.
The depressing part is that the European Commission was one body that promoted these two misconceptions, and it even added one more. It said in it's press releases that the anti-software patent movement was lead by the “Open Source and Industry Alliance”. You're thinking: “Who?” well, the OSIA is a USA-based free software group. There's is no way the DG Internal Market could think the OSIA was the biggest lobby group on the anti-software-patent side. They were one group that provided panelists for some FFII events. They were picked because they were USA-based, they were free software motivated, and they were “open source”.
There are signs that the pro-software-patent's lobbying style was beginning to fall apart as the vote got closer. All those glossy colour brochures were stacking up. Their credibility about representing SMEs was getting more and more doubtful.
One time I was delivering leaflets. A one page summary of the main misunderstandings circulating about the directive. It was black ink printed on normal paper, available in 6 languages. While I was hand-delivering bunches, MEPs and assistants decided that my box of paper was a convenient place to throw away their junk mail. My box filled up with Campaign For Creativity's glossy brochures, and those of Phillips, and EICTA, and Siement, but I didn't get one copy of FSFE's page.
So I think the anti-software-patent camp created more lasting value. Even though it was hard to understand what we were talking about sometimes, at least we were trying to spread information.
The Commission's directive put the EPO's current practice forward to be confirmed and the directive was rejected by 95% of the parliament. When 95% are not happy with how you're working, your days are numbered. But this isn't a victory speech.
Further down the line is much harder. We can see in the USA.
Patent trolls will not be stopped. They've nothing we can counter attack them over. All they want is our money, or if we don't have money, they want to free up our market presence so that their investors can take it, or so that someone with money can take it and then they'll have someone they can profitably sue.
The contributors to such a patent pool will be our friends, so we weren't at risk of them suing anyway, and their patents can't be used to fight our enemies because cross-licensing agreements already exist between the friendly big patent holders and our enemies.
Won't work. This only works for invalid patents. If software can be patented, and the patent is valid, you can't get it thrown out. Being harmful to society is not enough.
And anyway, patents are churned out far faster then we can swat them. “Swatting” is a good term, because it can be compared to trying to wipe out malaria by swatting mosquitoes one-by-one.
We need to improve a lot of things. To get it out of the way, the first one I'll mention, is the press.
We need to help the press.
The pro-software-patent camp won their support base because they saw that one party accounted for 40% of the European Parliament. They found a leader in that party and pushed for solidarity.
This move was very effective at first. If they could secure 40% of the vote, we could never get the 65% that we needed. They found a leader, and he set the party line. Klaus-Heiner Lehne was the name of the pro-software-patent's champion. Lehne rallied the EPP to take a pro-software-patent stance. He also sat on the board of directors of a patent law firm.
When we found this out, we weren't sure how to use this. As soon as we mentioned it, the other side started saying “of course he's on the board of a patent law firm - he's a patent expert and that's most of why he should be listened to”.
Luckily, a European newspaper and a German magazine did pick up on this, but we should have been much more helpful.
Another thing is that the pro-software-patent camp told the MEPs to make the current practice of the EPO into law. There is a reason for having the EPO not writing laws. It's concerns are for itself. Law is to be drafted by those whose concerns are for the citizens. This concept, the separation of legislative and administrative power, has been a cornerstone of European democracies for centuries.
The members of the EPO are not elected by the people. Their motives are private. I'm not saying for a minute (in this sentence) that their motives are not in the public interest, they could be, but the legislator's task is to consider what should be law - irrespective of what any private entities are doing, including the patent office. The Parliament has to represent the European people that elected them.
This also means that their job is not to “strike a balance” between the two lobby campaigns. If the public had not become organised on this, and only Microsoft and Sun had lobbied this directive, should the Parliament strike a balance between Microsoft and Sun and ignore the public? Obviously not. The job of the members of the European Parliament is to do what's best for the people of Europe, and to do so with an eye to doing what the electorate asks for. So if there was only one lobby,
But talking about democracy wasn't something we did. First we weren't experts on European history, and second it was a tangential issue. We wanted to talk about software patents. This is where we should have used the press better.
When we see irregularities, distortions of the democratic process, we have to inform the press and let them do their job.
In my talk 2 years ago, I think I complained about the lack of action from FSF Europe on the software patents issue. I can clear that up now: In 2001, FSFE decided not to work publicly on the software patents issue. This was for two reasons: first, it would be better to build a broad coalition of both free and unfree software companies, users, and developers, and second, free software didn't have much political importance and wasn't so well understood back in 2001. It would be very harmful if the pro-software-patent camp could paint the anti-software-patent camp as being “just those few people that give their software away”.
Things have changed since 2001. The coalition has been built, free software is better understood, and it's importance is widely and recognised by politicians. FSFE has always been working behind the scenes, using it's contacts in the Commission when possible, but at the start of 2005 FSFE decided the time was right to take a more public stance.
Move from push to pull. Fewer rants, more "have you read this?"
Tell the MEPs who represents you on this: FSFE, FFII, AFFS, whoever.
For the most part, move to acting like a rating system for the MEPs. If they receive 100 letters by email, they can't read them all, but if they receive 100 emails saying “have you read this article, I think you should”, most will be duplicates. The MEPs will probably only receive between 5 and 10 unique links so it's practical for them to read and the quality will stay high.
The same goes for mentioning who represents you. If they receive a lot of emails saying “Talk to FSFE about this, they represent a position similar to mine”, then FSFE will be consulted.
If you want, tell them, there's a guy in Brussels called Ciaran O'Riordan who's working on this. Talk to him, or can I set up a meeting with him for you.
For national politicians, it's a little harder to get them to enter dialogue about European issues. We haven't figured out the way to do this properly yet, but the best idea I've heard is that we should tell our MPs that they have to listen to both sides, and that “the other side” is FFII, FSFE, and similar groups. Tell them "you've talked to EICTA, now talk to FFII and FSFE - they represent the anti-swpat position: my position".
One common mistake was that people often argued against the patent problem as a whole, much of which was out of scope for this directive. What I mean there is that people would point out flaws in the European Patent Office, but this directive is about legislation.
we're lobbying for an effective exclusion of software from patentability, so before making your arguments, you have to think: “is excluding software from patentability the only solution to my argument?”
People claimed that patents being granted were overly-broad. This may be true, but the permitted breadth of patents isn't fixed at legislation level, it's fixed in the EPO. If the EPO is granting overly-broad patents, maybe they lack staff, maybe their funding should be doubled.
But we're not campaigning to double the EPO's funding.
Broad applications are not the problem, broad coverage is the problem.
Microsoft could define an extremely narrow patent, make it part of the Microsoft Word file format, and it would cover almost every desktop computer in the world.
Silly patents are not the core problem (judging "non-obviousness" with a low barrier), think about the RSA encryption algorithm, it was a great innovation. Expensive, time-consuming, purposefully sought - but the coverage can be overly broad, and it's too abuse-able.
There is some benefit in pointing out flaws in the EPO: It can show MEPs that the system isn't working so well, and therefore they shouldn't be expanding it - at least until it's fixed. But there are plenty of people making the anti-EPO arguments. Informed people should stick to the anti-software-patent arguments.
I did mention earlier that right now there is value in discrediting the patents of the EPO, this is because the current situation risks proposing a solution based on enshrining the EPOs process via the Community Patent. So there is some value right now but this is a special time and even now this is less important than convincing our representatives that legislation that makes the software exclusion effective is required.
We're working at the legislative level to prevent software patentability. The international agreement called “TRIPS” requires us to make patents available for inventions in fields of technology, that have industrial application.
So the legally sound way to prevent software patents is to define software as not being a field of technology.
Software is already excluded from patentability. The European Patent Convention says that “Software will not be patentable, as such”. Unfortunately the European Patent Office have decided to read this in the narrowest way possible. They have decided to apply “as such” to the word “software”, not to the concept of software.
So, software is not patentable “as software”, but it is patentable, as “a method involving a computer program, stored on a carrier, that when executed does ...”.
So there's already an exclusion, but it has proved to be ineffective. That's why I say we're working for an effective exclusion of software from patentability; we're lobbying for a clarification of the legislation to prevent this interpretation that we find ludicrous.
Our opponents have a different task. They want to “preserve the status quo”. And by that they mean preserve the current practice of the European Patent Office.
They have a nice line: they say that defining what is technical is out of scope for this directive. Of course, TRIPS forces the definition word “technical” to be the line which defines patentability, it can't be out of scope: it's the requirement of the directive. Anyway, our definition of technical defined technical only “for the purpose of patent law”.
The other hindrance we have is that calling software non-technical sounds like counter sense to most people. Sometimes an MEP begins a meeting by saying: “I find this too technical, can you simplify it”, and we have to reply “actually, software is not technical at all”. So we have to watch our language. Software isn't “technical”, whatever that word means, software is “complex”.
Books and their plots can be described as technical, but they're not covered by patents. Those books are “complex”.
So those are the two positions: clarifying the legislative exclusion, and preserving the status quo.
We never faced an amendment which said “Software shall be patentable”. Here's four of the main examples of how they wanted to preserve the status quo:
They wanted directive were the difference between the innovation and the invention (where should the 4 criteria lie?). This causes the combination of two unpatentable things to be patentable: software, even if it has no technical character, and a computer, even if you didn't innovate anything in it.
They wanted program claims being allowed IF, where they if clause is always true.
They tried to insert wording about solving technical problems – instead of being a technical solution to a problem. Technical problems can be solved by non-technical means - you can rearrange your schedule and solve a technical problem.
The other thing was “applied exact science”, which sounds like our wording, which was “applied natural science”, but actually covers even mathematics.
So we have to know: Why are we doing this?
We talked a lot about SMEs, and the other side countered by shipping SMEs into the parliament. Microsoft has a “development partners” program, or something like that, so it brought some of it's SME development partners to the parliament to lobby for software patents. Microsoft even accompanied their SMEs when they were meeting the MEPs.
If we keep pounding on the SMEs point, they'll draft a compromise that helps SMEs in some way. Such solutions have already been mentioned - when Ireland held the presidency, our representative to the Council of Ministers suggested that national governments could set up patent defense funds for SMEs. Basically a tax-payer supported jackpot that the Mega Corporations can litigate for.
That never received wide support, but if we pound on the SMEs problem, we'll be offered an SMEs solution, and free software will be excluded. It will be what we have been asking for, and it will be what we were campaigning against.
So: ask for what you want, or you won't like what you get. We want the right to write software. Freely.
Every company that makes it's website is a software developer. Every company that automates their network, scripts their database. They're all developers. Computer science students are developers.
Patentability creates liabilities as well as assets. They make software writing carry a liability. Avoiding the liability is not possible. Obscurity is usually your best defense. Never publishing your software is a good defense too, but we shouldn't be encouraging software to stay obscure or secret.
We know the liability can't be avoided because Cisco and IBM have stated that they don't do clearance searches anymore, it's not cost effective for them. What about everyone smaller than them?
Software patents are land mines for software developers.
The barrier to participation is so low, that you get millions of producers and 100s of millions capable of distribution. There's a potential for massive scale development - and therefore massive scale liability.
In traditional patentable areas, such as pharmaceuticals or machinery, patents are an industrial regulation. This is not true for software. I can write it, and distribute it. We need to be sure that we own what we write. This is significant because the only commercially viable reason to develop pharmaceuticals is for commercial gain, not for personal utility. For software this isn't true. If you're developing pharmaceuticals, you're doing it for profit. If one branch of development is blocked by patents, you go somewhere else. Software is usually developed to get a job done - writing software to do a different job will not suffice.
We want everyone, including ourselves, to be permitted to develop software for us so that our use of computers, both business and personal use, won't be restricted by the patents of others.
Patents, in software, specifically block development of generally useful software. This happens because software's utility relies heavily on compatibility. When writing a word processor, if it doesn't read and write Microsoft's Word format, or if it doesn't look like existing word processors, it won't be generally useful.
Some people proposed amendments to say that use of a patented idea for interoperability purposes was not be patent infringement. This would be nice, but there were two problems: first, it's not enough, we can solve the whole problem, so lets. Second, would this survive the remainder of the legislative process. Even if we get the parliament to support an amazing interoperability amendment, how do we know that afterwords some body wouldn't insert RAND, or some other anti-free-software terminology like that.
There could also be GPL-incompatibility problems. If I write code to read a patented file format, can I distribute it under the GPL if I know people can't modify it to do some other purpose? This is a grey area of copyright law.
In other fields, development and distribution is only undertaken by those with commercial interests. When looking to make money, if one area is locked down by patents, you look somewhere else. In contrast, software is usually written to get a specific task done. In that case, doing something else will not achieve your goal.
So we have to prepare our arguments to be precise, but we also have to have a broad range of arguments. Different MEPs are convinced by different topics, and worse, different MEPs are confused about different topics. Dealing with an MEP that thinks they understand the issue, but is actually very mixed up - is quite difficult.
Felipe Wersen had a nice technique. He'd go to the MEPs with a US software patent and ask if it deserved to be granted. The MEP would say “no, but that's a US patent”. Felipe would turn the page and there's the EPO equivalent. We have to dispel this myth that European software patents will be better than US software patents.
Monopolies. Creating monopolies is bad. And we have enough monopolies in the computer market: one. (Monopolies usually come in ones.)
The pro-swpat camp is trying to make software development require due diligence patent searches, and patent litigation defense. This makes the software development models of most software developers no longer economically viable.
Company development models are shot. If you don't have decent market share, or cash, you're probably safe - there's no value in selecting you to be litigated against. If you make popular software, or if you are commercially successful, you become a target. Venture capital injections will inspire litigious desires.
Competition. Without patents – without "protection" – people with ideas have to develop products and get them to the market. Quickly. Otherwise someone will eat their lunch. This is not a bad thing. With 20 year patents. You can have an idea, patent it. Think about a business model for a year or two. Look for funding, spend a few years going around trying to gather venture capital. Seven years pass, OK, give product development a try. 3 years go by. The business isn't that great, you decide to license the patent. Or the business goes kaput, and an intellectual property firm buys your patent for low cost (because bankruptcy is like that). Now there's an enforcer on the loose. They get first-to-market advantage.
Patent holders are holding back. They don't want the legislators to see how big a mistake it is. They want the EU to make a decision before they
The area of computers, which is new in terms of democracy, has started off down a bad course. GNU has been working on the fix, you, as computer specialists, have a duty to help.
The first of two cases that the pro-software-patent camp made very well.
The automotive industry lobbied very well based on Automatic Breaking Systems (ABS). They would put up a chart with 3 boxes connected by lines. He'd say: “Here's sensors, a processor, and some breaks”. If we invent an improved way of breaking, that doesn't require any changes in the sensors or in the breaks, then the innovation lies solely in the software. This is a software patent. The automotive industry needs software patents.
This is very simple, understandable, and it sounds reasonable. This works for MEPs. This is good lobbying.
The response is that, the innovation is implemented in software, but the innovation is a new way of breaking. It's a new way of using breaks. This is still patentable because it's a new teaching in applied natural science. So you can patent it, but you have to patent a system for breaking, not software.
ABS is not even a “computer-implemented invention”, it's a brakes-implemented invention. The achievement, the technical effect, of the invention happens in the breaks. It can be computer controlled. That's all.
Can it be done without brakes? no. Can it be done without computers? Yes, it would be stupid, but you could do it manually.
Siemens would say: “We make medical equipment. To scan bodies for cancer, people are put through 3-d x-ray machines that generate data. We develop software that converts this data to another form of data which is an image. This is data processing, pure data processing, and should be patentable”.
Hmmm, that sounds reasonable too. And it can't be reconciled with applied natural science. Siemens is correct: this wouldn't be patentable with our amendments.
So does this deserve a patent.
Siemens invention is already protected from copying. Nobody can steal the idea because the software never leaves their computers, and it's covered by copyright and trade secret.
But do Siemens deserve a monopoly? To compensate them for their research?
If such a monopoly is necessary to get this medical software written, maybe they do, but there are other ways to give Siemens a monopoly. The hospitals could give this monopoly to someone if they found it was necessary. So we can leave it to the customers of the software to decide if they want to deal with a monopoly or not. Siemens wants to speak on behalf of it's customers, but patents are not required for development of this, or any other, data processing system.
They say they need patents because copyright doesn't protect the idea. But the idea is hidden in the software. Only the input and the result is visible, and these things were already known. The idea is in the software, that's copyrighted, compiled, trade secreted. Independent research and development is the only way someone else will get that idea.
They want to have a patent to get a monopoly position: it's not necessary, and it's not requested. Suppliers do not speak on behalf of their customers.
As well ask speaking on behalf of their customers, companies liked to speak on behalf of their employees. We're a big company and we employ 1,000 people, therefore we represent those people.
This needs some critical thinking. Is the want of a large organisation always in the interests of it's employees? Well, large companies like to off-shore jobs to India - but this is the exact opposite of the wishes of their employees.
Off shoring is something I think we should talk more about. Many lobbyists for the other side talked about patents keeping jobs in Europe because we could use patents to exclude low cost economies from producing software for Europeans.
They talked about keeping jobs in Europe, but never mentioned that off-shoring is a growing phenomenon in the USA, and they never mentioned that according to the Gartner Group, IT jobs were down 16% in the last 3 years in the USA.
Software patents were introduced briefly in India, but that move was quickly overturned. Free Software Foundation India played a role in getting them overturned, and I'd like to congratulate FSF India, but I can't help feeling they had a slightly easier battle than the 6-year tooth-and-nail battle we're having here in the EU.
Why aren't the Mega Corporations aggressively pushing for software patentability in India? Could it be that they need a safe haven for their software development activity?
When you think about it, patents are quite useful for people that want to off-shore their staff to a low cost economy. Being intangible, there's no cost in moving the value of a patent from India to The West. There's no cultural contribution, so an Indian innovation is as good as a European one.
So they can pay lower salaries and get the same value. And they can lay off staff in The West and have the security that those staff can't compete with them because the knowledge of the staff is all locked up in patents owned by the Mega Corporation.
Another hitch was when MEPs said to us: “You're predicting doomsday, but the USA has had software patents since 1986, and I don't see the doomsday there”. This was a problem for us, simply because we weren't prepared. We all knew that bad stuff was happening in the USA, but the details didn't interest us because they were just that - implementation details: stories to be memorised and we didn't have time.
100 cases go to court per year in the US. 2,500 cases are filed though. The majority, however, don't reach even that stage. They're settled privately and we'll never know their number or outcome.
But we have something to go on. The Electronic Frontier Foundation in the USA have a patent hit list. They list the 10 harmful patent holders, along with information about their behaviour. This is a good source of information for the doomsday question.
Acacia Research, and Intellectual Ventures: are worth a look. Entrepreneurs in the USA have realised that knowledge of law and patents is more profitable in the software industry than developing software is.
M$ announced an increase in patenting from 2,000 to 3,000, but they didn't announce a 50% increase in developers. They're not making more, they're just going to slice the current stuff thinner.
I want to finish up by talking about the Fellowship of FSFE is a program. You sign up and pay 120 a year and this supports FSFE. I used to donate to FSF, so now that FSFE donations are being given to me as my wages, I try to use the money the way I hoped FSFE employees used the money when I was on the donor-end.
FSF Europe pays me, and it works at the UN level, the WIPO level, and other levels (WSIS, WGIG) to prevent the next TRIPS. We can't keep fighting these things at legislative level, we need to fix the source.
FSFE do a lot of stuff that doesn't get seen. The software patents directive, as well as the copyright directive, and other directives past and future, they come from international treaties. So while I and others are fighting these EU directives, FSFE is active at the WIPO level, the WSIS, WGIG, and UN level, stopping the source of these directives. A lot of this isn't reported in the media, and gets little recognition even in the free software community. While some are making a lot of noise about SCO, FSFE are doing far more important work and getting little or no credit for it.
By the way, while our resources are tied up in the software patentability battle, a very bad European Copyright Directive became legislation. The next really big threat is Trusted Computing. We need to increase our resources to fight that.
And of course the GPLv3 is going to take up a lot of resources.
The FSFE SWPAT page, with my notes for this talk: Become a Fellow of FSFE: The FFII website for information is: Here's some good books in print: And the work of Eben Moglen, best collected at Wikipedia:
Information Feudalism, Peter Drahos with John Braithwaite
Free Culture, Lawrence Lessig
Free Software: Free Society, Richard Stallman
The FSFE SWPAT page, with my notes for this talk:
Become a Fellow of FSFE:
The FFII website for information is:
Here's some good books in print:
And the work of Eben Moglen, best collected at Wikipedia:
We did something pretty amazing in the last two years. When the next battle comes, we'll need to do even better. So I hope what I passed on just now can help us work together and win this next time.