April 21, 2003

Software Patents Reprint

FFII is planning a conference to stop the proposal for legalizing software patents in Europe on May 7 to 8.

For the occasion, I reprint something I wrote in March 2002 against that proposal. That text is also in my book "Grenzen des Patentwesens".

The version here is reduced: I don't include footnotes that point to texts which are not online.

Sinking the Proposal for a Directive on Software Patents

1. Sink the Proposal!

Being of the opinion that software patents should be abolished immediately, worldwide, I have not much sympathy for the recent European Commission proposal to change the long established rules and force software patents down our collective throats.

So I want to join others who are already fighting this proposal at full blast. We need to stop it as soon as possible.

This proposal is not like a seventy thousand ton aircraft carrier, but rather like a submarine surrounded by enemies. One torpedo should be enough to sink it for good. I am going to fire several torpedoes against it right now, and advise everybody riding on the proposal to get off the ship in time. These will have real warheads, aimed at destruction. Quite possibly they will all fail to hit the target; but I will try as hard as I can.

2. What do they want anyway?

The Commission says harmonisation is needed because there is some difference in the interpretation of the European Patent Convention between Member States. The Commission does not discuss the interpretation in all Member States; it limits its discussion to two Member States, Germany and United Kingdom; and it finds that the courts in these two Member States do not interpret the European Patent Convention in the same way.

However, the European Patent Convention is in force in all Member States, as the Commission correctly notes somewhere in the proposal. That means that already with the status quo there is one uniform legal text (Article 52 of the Convention) in force in all Member States. This obviously raises the question why the interpretation would become more uniform if instead of the uniform legal text in the Convention a new uniform legal text in a Directive gets adopted. If anything, having one text in the Convention and another in the Directive would mean that every court decision would have to address which of those two different texts it should follow, which would seem to lead to more confusion and differing opinions between Member States court decisions, rather than less.

This might be mitigated somewhat by the possibility of having the Court of Justice decide on the interpretation of a future Directive, which would seem to be the only real gain for the goal of having a uniform interpretation in all Member States.

It is very difficult to predict if the noble goal of avoiding confusion will be helped or rather damaged by introducing this legislation. We would probably have to wait and see. At least the legislator should have the benefit of doubt. But I think the following sentence is very much worth noting, and I want to call everyone's attention to it:

'Thus, patentees and the public at large who may be users of patentable matter currently lack certainty as to whether in the event of litigation patents which have been granted in this field will be upheld.'

This is what the Commission says in the 'need for harmonisation' part.

And this seems to be exactly what they want. I think that the current practice of the European Patent Office to grant software patents is clearly illegal under any conceivable understanding of Article 52 of the Convention. I think this is so, and that it is plain to see from the wording of that Article, but I do not intend to give any reasons for this opinion here, having done so in detail elsewhere (in German language). The point here is that no one has really tried to enforce these illegal software patents in Europe, since that would have meant fighting a real opponent in a real litigation, as opposed to discussing the best way to work around the wording of the Convention in collaboration with the European Patent Office in the quiet procedure before the Board of Appeal. That would have meant a real fight against someone hell-bent on proving that software patents are illegal, and indeed, the holder of a software patent would seem to 'lack certainty as to whether in the event of litigation patents which have been granted in the field will be upheld', and would therefore find his software patents unenforceable and worthless at the time (their usefulness for preventing walrus attacks notwithstanding).

So what the Commission wants is not harmonisation. What the Commission wants is not uniform interpretation. What the Commission really wants is that all software patents illegally granted already become enforceable by litigation. They want to open the box of Pandora.

So why do they say that the proposal is necessary for harmonisation, if that is not their real goal?

The answer is that they have to say that, so as to be able to make the proposal in the first place. They can't just get in the ring and say: 'Okay, we know that everyone hates software patents, but we want introduce them anyway, and we want to have lots of great litigation from all the software patents already granted to happen.' They need to say: 'We don't really want to change anything, but there are these horrible differences in interpretation threatening the freedom of movement of goods in the internal market. So we need to harmonise.'

This is not exactly telling the truth; this is only the superficial and pretended reason for the proposal, as opposed to the real reason. The Commission needs to pretend that it wants to harmonise when in reality it wants to change, because without pretending so, they would have no right to propose this legislation.

This reasoning is a noisemaker, intended to fool the torpedo looking for the submarine. We are not going to fall for that noisemaker. Instead, we are going to use our control wires to aim the torpedo right at the real submarine: This proposal is not about harmonisation. This proposal is not about legal certainty. It is about change. It wants to introduce software patents, which are illegal right now.

In the 'Frequently Asked Questions' file, my favourite answer consists of the following two sentences: 'In broad terms, nothing will be made patentable which is not already patentable. The objective is simply to clarify the law and to resolve some inconsistencies in approach in national laws.'

Okay, Commission, now take this first torpedo. If nothing will be made patentable which is not already patentable, then your proposal is doing exactly nothing at all and should be dumped because it adds to the confusion, rather than reducing it.

But we know that this isn't really true, don't we?

3. Democracy, Commission flavor

Next we will have a look at the way the Commission has dealt with the opinions presented in the consultations leading to this proposal.
The Commission says, in the section on the consultations:

'Thus although the responses in this category were numerically much fewer than those supporting the open source approach, there seems little doubt that the balance of economic weight taking into account total jobs and investment involved is in favour of harmonisation along the lines suggested in the paper.' (Actually they said 'fewer that' instead of 'fewer than,' I have taken the liberty to correct the obvious grammar error.)

This is great stuff. It just shows that the idea of listening to the numerical majority is stupid. Rather, people should listen to the 'balance of economic weight', that is to the voting power of the big wallets. Or so it might seem, if we follow the Commission's concept of democracy.

I was curious if that concept might be mirrored in the fundamental values of the Social Democratic Party and the Green Party, which happen to build the government coalition in Germany right now. So I had a look at their respective websites. Here is what I found.

The Leipzig Grundsatzprogramm (general principles) of SPD, says on its page 49:

'Demokratie bezieht ihre Lebenskraft aus der Gesellschaft und ihrer politischen Kultur. Sie wird durch die Ballung von wirtschaftlicher oder Medienmacht und durch die Anhäufung von Herrschaftswissen in privater oder öffentlicher Hand bedroht. (Democracy receives its vitality from society and its political culture. It is threatened by the concentration of economical or media power, and by the accumulation of ruling knowledge in private or public hands.)'

The draft Green Party Grundsatzprogramm (general principles) says on its page 68:

'Demokratie lebt vom Wettstreit der politischen Positionen und Konzepte. Deshalb halten wir es für falsch, wenn Positionen nicht mehr offen eingeführt und erstritten, sondern allein mit den großen Interessenverbänden ausgehandelt werden. Wir wollen die Rolle der Parlamente und der Abgeordneten im politischen Willensbildungs- und Entscheidungsprozess aufwerten. (Democracy has its foundation in the competition of political positions and concepts. Therefore, we think it is wrong if positions are not introduced and fought for openly, but rather only negotiated with big interest lobby groups. We want to strengthen the role of Parliaments and of representatives in the political opinion building and deciding process.)'

Maybe the German Social Democrats and the Greens need to reconsider their principles. 'Democracy threatened by accumulation of economic power', 'wrong if only negotiated with lobby groups.' Tsk, tsk. As the Commission tells us in the above sentences, politics is not about something superficial as numerical majorities, politics is about economic majorities.

Maybe. But I prefer to think that it is the Commission who needs to reconsider its understanding of democracy. Following the big lobby groups is not going to be a popular flavour of democracy, here and now (it might have had its place in the Nineteenth Century, but not in the Twenty-first). And since the procedure will follow Article 251 of the Treaty, at some point the proposal will need a majority in the European Parliament.

And I sure hope that the Parliaments idea of democracy is somewhat different from the Commission's concept.

To strengthen this point, maybe someone could set up a website collecting the opposition to software patents in a petition to the European Parliament. Quite possibly, our side might receive several thousand votes from citizens all over Europe for such a petition against software patents. No, wait, that has happened already, hasn't it? It will be interesting to see what the European Parliament does with that petition, with over one hundred thousand signatures, now that the Parliament's point of view will actually matter.

4. Where's the beef?

Lawrence Lessig asked some time ago that Europe should consider the economic consequences before legalizing software patents (he seemed to think that software patents in Europe are illegal right now, which is actually correct, even if most people at the European Patent Office would rather ignore that fact).

Of course he is right. If Europe is going to take the drastic step to push patent inflation by adding a vast area to the patentable domain, at the least one should expect that someone would have a look at the economic consequences first.
If nothing changes by the proposal, that would seem to be unnecessary, but we have already refused to be fooled by this noisemaker. The proposal is to introduce software patents; to abolish the long-standing prohibition against software patents in Article 52 of the Convention; to make all the illegally granted software patents enforceable.

So where is the economic justification for this complete reversal of European legislation?

The Commission's main point is that a study by the Intellectual Property Institute of London, published 2000, supports software patents. The problem with that study, however, is that it is contradicting itself, as is evident even in the citations of the study in the Commission proposal.

On the one hand, the Commission tells us:

'It (the study) finds that the patentability of computer program related inventions has helped the growth of computer program related industries in the States, in particular the growth of SMEs and independent software developers into sizeable indeed major companies.'

On the other hand, the Commission says:

'Any move to strengthen IP protection in the software industry cannot claim to rest on solid economic evidence.'

Only one of the above can be true. If the patentability of computer programs is the greatest thing since the internal market for the economic success of SMEs, then why is there 'no economic evidence' for a move to stronger IP protection?

Anyway, the study the Commission presents as its strongest witness has done no factual research of its own, limiting itself to quoting from the research of other people instead. More to the point, this study has done no factual research whatsoever on the situation in Europe, which would have been necessary.

In that respect, it contrasts with the much more serious study initiated by the German Federal Ministry of Economics and Technology, which was ignored completely by the Commission in preparing the proposal. The German study actually did some fieldwork, asking people questions, trying to get raw data, none of which however would seem to be sufficient proof of any beneficial effect of software patents.

So, what is the Commission telling us about the economic justification? They tell us two things. 'Our study gave us no economic evidence for a move to stronger IP protection,' and 'We are ignoring the German study.'

Both of which are certainly no sufficient reason for the proposal.

As a dedicated opponent of software patents, it is not really my most important goal in life to find some sufficient reason where the Commission fails to show it. But I am curious about how the proposal of legalising software patents might fit in with the general direction of European Union policy in information technology matters. Which is another question the proposal conveniently ignored.

At the European Council in Lisbon in March 2000, the European Union adapted as a 'strategic goal' for the next decade 'to become the most competitive and knowledge-based economy in the world.'

One might have expected the Commission proposal to explain why a vast expansion of government-granted monopoly rights contributes to the goal of becoming 'most competitive in the world'. Patents are the exact opposite of competition, and while it might be possible to argue that their harm to competition is less serious than their merits, it is impossible to argue that patent inflation increases competition.

One might note that the European Union policy as a whole is directed at killing off monopolies, and strengthening competition, especially in the telecommunications sector, which is of strategic importance for the above goal. Quite recently, the Commission said in a Communication on 'The Impact of the E-Economy on European Enterprises - Economic Analysis and Policy Implications' that there is a need to promote open standards and competition, which is correct.

So how is the legalisation of software patents going to help promoting open standards and competition? I think that it is doing the exact opposite; that it will damage open standards and competition; that it will be a crucial factor in preventing the 'most competitive in the world' goal from happening. And if we introduce monopoly rights on software ideas all over the place, we might as well go back to monopolies in the telecommunications sector as well.

5. Me too?

Last time I checked the European Union was not a colony of the United States. Rather, we reserve the right to disagree. The European Parliament is under no obligation whatsoever to adjust European legislation to American standards. One point of European integration is exactly that: having more weight in bilateral relations.

This means that the proposal to introduce software patents in Europe needs a better reason than that Americans like them, as Lawrence Lessig has pointed out quite convincingly. 'Me too' is no reason for patent inflation; at least no sufficient reason.

There seems to be only one place where the unfortunate situation in America is given as a reason for this proposal to introduce software patents. In the FAQ, as an answer to the question why patents are good for innovation, the Commission says, among other misguided observations:

'Other countries which are successful in the information technology sector (such as the US and Japan) also grant patents for computer-implemented inventions.'

There are two points which need to be addressed here, both conveniently left out in the proposal.

One point is the hard data readily available comparing research and development in Japan, Germany and the United States. For anyone who cares to look, the Bessen/Maskin study clearly indicates that research and development in the software sector went down in the United States as a consequence of introducing software patents. Everybody knows that, but the more important point is that no such reduction was to be seen in Germany (see Figure 2 on page 9 of the paper). Having no software patents is a competitive edge for Europe.

This is much like the goofy American policy requiring export licenses from software publishers, which contributes nothing to the goal of stopping the spread of cryptography, but adds a lot of red tape for American industry, and the suspicion that the American government makes sure that Microsoft product security can easily be compromised by the CIA, hurting sales to anyone interested in keeping their data for themselves. The Americans are shooting their own foot with that policy, which is great for the European industry, since it helps the Europeans to catch up on the market.

Exactly in the same way, the Americans are shooting their own foot by transforming their software industry in law firms with small software development departments, and having everyone sue everyone else in sight over some trivial idea some rookie at the Patent Office stamped with his seal of approval. This is great for the European industry. And the last thing we need is the Commission to copy American patent policy, giving up on the competition advantage we have right now, and which is clear to anyone who cares to look at the Bessen/Maskin data.

The other question is who would pay the license fees, as a balance, European industry or American industry. This is simply a question of who has more and more license-fee generating software patents. Either the Europeans or the Americans will come out ahead.

With the current dominance of world markets enjoyed by the American software industry, it would seem at first sight that any such balance would necessarily need to come out with large numbers in the red from an European perspective. In other words, any move in the direction of patent inflation in this field means massive money transfers from Europe to America, which is definitely not in the European interest.

This is not a theoretical consideration. Look at all the cases in which software patents cause real damage to European interests right now. In most of them American patent holders are harassing European software developers. So if the bold project of the Commission to legally accept software patents in Europe sails through, we will see more of the same, on a much larger scale. We will see American patent holders harassing, threatening, and cashing in on European software developers.

Probably the European software industry would be better off if the Commission just proposed a one percent tax on all sales of European software developers in Europe, to be paid out as a lump sum to the American BSA. That way we would at least avoid the hassle and the confusion.

But even the BSA couldn't write that in their proposal draft for the Commission, could they?

6. So what?

Now, what exactly would happen if all my attacks, and those of many others, would fail to sink this proposal?

The Commission wants us to believe that we need not worry. Contrary to the situation in America, there will be no patents on business methods, because with the proposal, every 'computer-implemented invention' needs to show a 'technical contribution'.

This means that the Commission actually listened to the consultation to a certain extent. Even those few answers that supported the Commissions position drew a line at the point of allowing business methods patents.

This also means that the current proposal is much less damaging than the proposal to abolish 'computer programs' from the list of not patentable subject matter in Article 52 of the European Patent Convention was. That would have been following the American example without even pretending not to do so.

What does that magical 'technical contribution' requirement actually do? I don't know. The Commission does not know either, as they say in their FAQ:

'It would not be possible for a legal text such as a Directive to attempt to spell out in fine detail what is meant by "technical", because the very nature of the patent system is to protect what is novel, and therefore not previously known (emphasis added). In practice the courts will determine in individual cases what is or is not encompassed within the definition. However, earlier court decisions have indicated that a technical contribution may arise if there has been some improvement in the way that processes are carried out or resources used in a computer (for example an increase in the efficiency of a physical process), or if the exercise of technical skills beyond "mere" programming has been necessary to arrive at the invention.'

Now, this is real magic. How can there be any 'earlier court decisions' on a law not yet adopted? And wasn't the whole point of the exercise to stop all those nasty court decisions from confusing the issue and have a clear and predictable regulation? At least that was what the proposal pretended to be, as we have seen above.

So we have to stick with the first part of that answer. We don't know. And we can't give details. We will have to wait and see what the courts will do with that new text, once they get a chance to apply it to real cases, and start the interpretation work.

But I know one thing. For close to 30 years, Article 52 of the European Patent Convention has said that 'computer programs' are not patentable subject matter. That is about as clear as it can get. But the courts chose to ignore this; the courts chose to work around this; the courts chose to grant thousands of software idea monopolies anyway. Just look at the Software Patent Horror Gallery, to see what good that exception has done for the opponents of software patents.

So, as a matter of course, I expect the patent lobby and the courts to do the same with any new text adopted as a result of the proposal. Don't worry? With that track record? The 'technical contribution' requirement will be interpreted as being fulfilled when the software program in question is running on a computer. If you think this is a joke have a look at the Japanese regulation of software patents, which takes exactly that position (easily confirmed by looking at the German study mentioned above on page 216). In Japanese law, you can get a software patent if the software is designed to run on hardware, which doesn't seem to have much of a restricting effect.

No, wait. There will be at least three different interpretations in court decisions. Most will see the requirement fulfilled if the software is designed to run on a computer, but others will take the bold step to require that a Central Processing Unit and a display is necessary to run it; and a third theory will be that any software making use of the Internet in some way contributes technically. All of which will mean: There will be no restricting effect whatsoever, and, of course, we will see patents on business methods and patents on Internet usage methods. And everything already granted illegally will be perfectly enforceable, which alone will be enough to catch up with the American state of craziness in the field.

So, yes, Commission, we are extremely worried that your proposal will remove any restrictions; that it will break the dam and bury the European industry in a flood of American software patents; that there will be no limit whatsoever. And we are especially worried that everything already granted will be enforced.

So all those opposing software patents might have to think about some other approach. If the proposal gets adopted, software patents will be like spam. Harassing, threatening to chill and choke all research and programming activity, restricting competition and open standards, and harming Open Source more than anything else, and they will be perfectly legal. And just as the sum of damages caused by spam is rising with the number of spam messages sent, the sum of damages caused by software patents will rise with the number of software patents granted and enforced.

And as more and more projects will be taken down all over the place, destroyed by some trivial software patent, people will get mad. When the first cases enforcing software patents hit the courts, the defendants will receive ample support in public, while the plaintiffs will need some pretty professional footwork to limit the PR meltdown their lawsuits will cause. Just look at British Telecom and the perfect joke they are making of themselves by trying to enforce their Hyperlink Patent in the United States. Everybody and their dog hate them for that, and laugh about them, and despise them. I for one have to thank them for their great contribution to the cause of software patent opposition, but that doesn't change much of the PR damage.

And just as people eventually will find some way to deal with spammers, even if the law does not stop them, the overwhelming majority of opponents to software patents will find a way to deal with corporations who try to use software patents, even if the law does not stop them.

So what if the Commission proposal actually succeeds in forcing software patents down our throats? We'll see about that when it happens. Right now, I hope that all of my attacks hit the proposal, and sink it for good.

Posted by Karl-Friedrich Lenz at April 21, 2003 05:50 PM | TrackBack
Post a comment