Who stole my software?

This is a (expanded and modified) version of a speech I gave to the Management Consultants and Information Technology section (12,000 members) of the Austrian Chamber of Commerce on 17 November, 2006.

Introduction

Last week I was at an "IP policy" workshop sponsored by Microsoft. I heard economists claim that free and open software and open standards were bad for business. I heard a computer scientist claim that open source was not significant, and that it went against a strong intellectual property system. I asked him if he wrote software. He said, "well, I'm a computer scientist". I asked again, "but have you written software recently?" "It's been a few years," he admitted. "Have you ever written any free software," I asked. "No", he shook his head. "So," I said, "you're like a virgin making theories about sex."

The context of the software patent debate is vast. To understand it fully we need to look at globalisation, at the impact of cheap internet-based communications on service industries over the last decades, at political competition between major trading blocks and countries, at free and open software, at the rise in IP "licensing" as a major business.

To put it very simply, the internet is changing our world in fundamental ways, and the software patents debate divides those who want to live in the new world from those who want to keep hold of the old rules.

Overview of the debate

In the EU, the public software patent debate started about six years when the EPO - the European Patent Office - began granting patents on software and business methods. This was the start of a new spurt of growth in the European patent industry. Since that time the patent industry has been trying to get its main new products - what I call 'soft' patents - past the regulators. And for six years, the European Parliament and national high courts have systematically said, "no, we don't want software and business method patents". The industry has squirmed and argued, trying to say, "OK, we don't want to patent software, just inventions that run on a computer." Still, today, most of the hundred thousand-plus software patents that the EPO has granted are not valid in the EU member states.

The EU Commission has been a constant friend to the patent industry. It backed the failed Software Patents directive of 2005, and it is now backing an initiative called EPLA, the European Patent Litigation Agreement. EPLA is an EPO plan. It is a complex proposal, it takes months to decrypt and understand, but basically it means that software patents are back on the agenda, as I'll explain later.

Before I do that, let me give you a concrete example of a fairly typical EPO software patent. In June 2005, the EPO granted a patent application from a Belgian company. The patent's central "invention" is to take a mobile phone number, looks it up in a database, and returns an email address, to which material can be sent.

As one journalist wrote, of this patent, "what is non-obvious or technically innovative about a database that collects email addresses via SMS and sends them files (magazines, audio, video)? No one knows, except apparently the EPO and the proud parents."

I know about this patent because my firm built a mobile applications product from 2002-2005. This product did many things, and one of the things it did was to lookup phone numbers in a database and return an email address that was then used to send an email.

When the company (who's name I will not speak, because they do not deserve the press) announced that the EPO had granted their patent (pending 9 months of opposition), they did a media campaign to convince the mobile applications industry to "license their technology". The firm made no products. But they had money, and one of the best patent lawyers in Belgium.

While the phone-to-email patent was just an applicatiobn, this firm had successfully sued a firm for selling an "infringing" product, in Belgium, to a newspaper. They had the best patent lawyer, the opponents did not prepare their case properly, and the judge was a little confused. As anyone who has gone to court knows, this is fairly typical. Justice is not automatic - money and aggression are more important than right and wrong.

In early 2005 I organised a workgroup to react against this patent, and software patents in general. It taught me a lesson. We had eight firms in this taskforce, most of the large Belgian mobile application firms. We discussed the patent. We had meetings. We spoke to the mobile phone operators. We discussed. Time passed. Eventually no-one wanted to commit to any action, and the patent opposition window closed. There was anger, fear, passivity. But there was no understanding that this patent could probably be stopped simply by going through the proper opposition process and saying, "this is a pure software patent, it is not legal".

Most sectors, when confronted by a patent claim that is defended by a top patent lawyer and a deep pocket, will not fight back. They will try to compromise, do a deal, and eventually end up paying a percentage. One out of ten firms may actually fight back. My company, tiny, did not.

I asked our insurance company about patent infringement insurance. They said "no", and when I asked again, they thought for a while and said, "HELL NO!" We took legal advice, and then in 2006 we cancelled our SMS product, fired the team, and took a major loss.

The patent was granted about the same time the FFII first asked me to come and speak about software patents in Brussels. They tell me that my first speech on behalf of the FFII was somewhat angry, but effective.

What and who are the EPO?

The European Patent Organisation

Let's look at the EPO again. What is this organisation, and why are they granting software patents that make such a mess of normal business? From the EPO website1:

The European Patent Organisation is an intergovernmental organisation that was set up on 7 October 1977 on the basis of the European Patent Convention (EPC) signed in Munich in 1973.

The Organisation currently has 31 member states, comprising all the member states of the European Union (except Malta) together with Bulgaria, Iceland, Liechtenstein, Monaco, Romania, Switzerland and Turkey.

The two organs of the Organisation are: a) the European Patent Office, located in Munich with a branch in The Hague and sub-offices in Berlin and Vienna, and b) the Administrative Council.

In fact there is both an 'organisation' and an 'office', but for most purposes "EPO" covers the whole quite accurately.

The EPO is a treaty organisation like the World Postal Union, or the United Nations. It has 6,000 employees, many of whom have diplomatic immunity and don't pay any taxes. It earns more than Euro 1.25bn each year from patent examinations. It occupies ten buildings the size of the European Parliament. It has permanent lobbying offices in Brussels, Munich, and all other European capitals. It is not an EU institution. Local law stops at the EPO gates. It is run by officials who come from the different national patent offices. It is a bureacrat's dream. Work hard in a national patent office, follow the party line, say the right things, and you too could get a slice of the EPO cake.

The EPO is effectively the European head quarters and main organ of the global patent industry. For every Euro earned from patent examination, the patent industry earns another 5-6 Euro from filings, translations, and above all, litigations. And this ignores the profits to be made from licensing patents, which run into the hundreds of billions, even trillions of Euro.

Industries come and go. The patent industry has struck gold in the past from ship building, from electronics, from pharmaceuticals. But the internet revolution makes a single new technology a key to controlling entire markets. That technology is software. Own the patent for phone-to-email lookup, and you could earn billions per year from the mobile phone market.

You can see why the patent industry wants software patents so badly. It is a huge new market that has worked well in the States. It is so frustrating to have people like the FFII preventing the natural and profitable growth of this market in Europe.

Who wants software patents?

There are three main groups who want software patents. First, the patent industry, as I explained. For these people, patents are just good business, and software patents - like gene patents, biotech patents, marketing method patents, business patents - are very profitable. The patent industry includes patent trolls like the phone-to-email firm. It's an easy step to make from filing patents for other people to file them for yourself, and make money from them if possible.

Second, the software monopolists, meaning mainly Microsoft, SAP, Nokia, Phillips, and sometimes IBM. The first four are especially aggresive in claiming new software patents at the EPO and trying to establish these at the national level.

Last, the Commission. To be honest, I don't really understand why the Commission should have fallen for the propaganda of the (mostly US) patent industry, but they have, hook line and sinker. The DG Internal Market, especially, is pushing software patents like they are the cure for old age.

However, it's taboo to ask for software patents since the Parliament - the democratic part of the EU - voted massively to kick out the software patent directive in 2005.

So in 2006 the patent industry, the software monopolists, and the Commission came with a new plan, one that has been baking in the oven for some years, and which cleverly side-steps the whole software patent debate and focusses on something entirely different. At least, that is what we're being told. This plan is EPLA and I will now explain what that is, and why EPLA is all about software patents.

Proposal for a new patent court

EPLA is an EPO plan to create a new patent-only court system2. Very roughly, this is the way EPLA would work:

  1. The countries that have signed the EPC agree to set-up a new international organisation called the European Patent Judiciary (EPJ).
  2. The EPJ consists of a European patent court (multiple courts of first instance, and a court of appeal), and an Administrative Council.
  3. The Administrative Council is composed of representatives of the EPC countries. That is, the same patent officials who run the EPO.
  4. The Administrative Council runs the Court, appoints judges (for fixed terms, not life), and manages budgets.
  5. The European Patent Court would take over jurisdiction from national courts over a period of years.

EPLA is mainly designed to make life easier for large patent owners who want to litigate. It's cheaper and more effective than the current system of national courts.

So what's wrong with this system, apart from the obvious problem that it's been conceived entirely by and for the patent industry, rather than via the normal democratic legislative system?

  • The European Court of Justice would have no jurisdiction. Thus, EU citizens would be tried and judged by a court system that allows no appeal to the highest EU court.
  • The European Parliament would have no oversight. Thus EU citizens would be subject to a legal regime that allows no input from democratic processes.
  • EPJ judges would be appointed by the same people who run the EPO, a serious conflict of interests.
  • EPJ judges would be appointed for fixed terms, so they would be forced to follow EPO practice.
  • The EPJ would be financed by patent litigation, creating a second international organisation that makes money from patents.

Very simply, the (same people who run the) EPO - which already grants software patents on a huge scale - would get full judicial control over these patents. As Mark Webbink, Deputy General Counsel for Red Hat, notes, "EPLA's supporters are experts in obtaining patents, not innovation."

Instead of solving Europe's patent problems, EPLA casts them into concrete. It guarantees that any patent reform would be delayed by at least ten, and probably twenty to fifty years.

The US Federal Circuit Experiment

The USA has experimented with something like EPLA. In 1982, the US Congress put all patent litigation into a single court called the Court of Appeals of the Federal Circuit (CAFC). This is how the US CCIA (Computer & Communications Industry Association) described the effects of CAFC3:

The Federal Circuit has:

  • Lowered the threshold standard of patentability
  • Encouraged extortionate demands and settlements
  • Eliminated virtually all limits on patentable subject matter
  • Endowed issued patents with an unjustifiably high presumption of validity

Regarding EPLA, the CCIA said:

[CAFC] is not only a questionable model for Europe to follow but the proposed European system would be far more insular in the following respects:

  • No legislative oversight. In the United States, Congress is directly empowered to change patent law and patent institutions. Neither national legislatures nor the European Parliament would have this power.
  • No judicial oversight. In the United States, the Supreme Court can review Federal Circuit decisions, and it has granted review frequently in the last two years as concerns about the malfunctioning of the patent system have grown.
  • No independent judiciary. EPJ judges would be appointed by the EPO Administrative Council for limited terms, whereas Federal Circuit judges are appointed by the President for life.
  • Wholly specialized. EPJ would be exclusively focused on patents. Although the Federal Circuit has made its mark on patent policy, it has other responsibilities as well.

The CCIA concludes that:

[CAFC] has made patents more potent, easy to get, easy to assert, and available for a virtually unlimited range of subject matter. As shown in the 2002 joint Department of Justice/Federal Trade Commission (FTC) hearings, this has led to over-patenting, portfolio racing, opportunism, extortionate settlements, and failure of the public disclosure function. As the final FTC report, To Promote Innovation: The Proper Balance of Competition and Patent Law and Policy, makes clear, the ill effects have been borne disproportionately by the ICT sector.

This is what EPLA promises for Europe.

The CCIA, meanwhile, has taken its complaints to the US Supreme Court. No such remedy would exist under EPLA.

How much did it cost to prepare EPLA?

Something as dangerous and controversial as EPLA cannot be introduced offhandedly, especially when people are starting to question the sanity of the CAFC system. Someone is going to ask difficult questions. So, the EPO, working though its friends in the Commission, has prepared the ground by getting all the answers into place before the Commission officially endorses EPLA.

Preparing the ground for EPLA was a massive and expensive project. From the outside, we can only guess at the size of the budgets involved, but they are significant. We estimate that each Euro we spend on activism is matched by a hundred Euro of commercial lobbying. The FFII spent something like Euro 1m in contributed time to understand EPLA and prepare counter-proposals. So we can estimate that about Euro 100m was spent on preparing it.

Does that sound a lot? A top lawyer or lobbyist earns Euro 1m per year.

In fact our Euro 100m estimate is possibly too low. It might be Euro 250m, or more. That does not really matter. Software patents represent a large, even majority slice of the Euro 6bn a year earned by the patent industry. The bill for the EPLA project is a small investment. And the ticket was split between the software monopolists and the EPO, and a large part was provided by the tax payer. That's the nice thing of getting the Commission to work for you. Great value for money!

But what was the money spent on? This are the visible aspects:

  • Documentary and legal support. EPLA is a complex design that depends on some very delicate manoeuvers, such as the EU becoming a signatory of the European Patent Convention. So the EPO has spent a lot of money on preparing the legal ground work and producing web sites, reports, studies, and other materials. Most of this work was done before 2004.
  • Media events like breakfast meetings, conferences, workshops. Speakers are brought from across the world, they are lodged, fed, and assisted.
  • Public relations. This means paying professionals to get the right stories in the news.
  • Public opinion 'research'. The outcome of such "let's ask the people" programmes is always the same. The people's opinion is generally discounted and ignored, while special interests get a broad canvas on which to paint their dogma. The Commission (some DGs more than other) is an expert in conducting fake research.
  • Direct lobbying. This means paying professionals to attend meetings, convince MEPs, speak in workshops, etc. Lobbyists are expensive and the EPO has teams in every major capital, but especially Brussels and Munich.

It's the fake public research that annoys me most, because it wastes so much public time and effort on an empty exercise. The Commission launched a complex and badly-documented consultation on patent policy. The FFII and many IT firms responded. The Commission discarded these responses and helped hundreds of patent-holders to provide answers instead. This technique of legitimising a pro-forma answer by wasting other people's time is used over and over again. The FFII is starting to boycott such exercises because participation makes them more credible.

The pro-EPLA propaganda

A major part of the EPLA preparation was the creation of a solid propaganda piece, a nice mix of truth and lies that leads the unaware reader to believe that EPLA is indeed a cure for old age, the end of oil, and a bad back. We read the propaganda piece over and over again, in reports, studies, conferences.

It goes like this:

  • Europe is falling behind in economic performance
  • Europe needs to be more innovative
  • Innovation depends on a strong intellectual property regime
  • This means we need stronger and more reliable patent system
  • Litigation in Europe's current patent system is too expensive
  • The Community Patent is the ultimate solution
  • EPLA plus the London Protocol are a good interim solution

This explanation comes back in many places. To debunk it, we can note that:

  • The US is actually falling behind in economic performance (from 1st to 6th place last year, according to the OECD)
  • Innovation depends on many things, such as taxation, market size, and fair competition
  • Intellectual property is many things, not just patents
  • The main problem with patents is not litigation, but bad patents
  • Yes, an EU patent system is a desirable goal
  • EPLA would block, not promote, an EU patent system

Alternatives to EPLA

The obvious alternative to EPLA is an EU patent system. There has been long discussion about a Community Patent, but all proposals have had big flaws. Generally speaking, any patent system that is designed by the patent industry will be flawed.

The most radical alternative to EPLA is full privitisation of patent examination, so that the patent system becomes nothing more than an online registry of claims and nullifications.

The patent industry has lobbied against EU-based patent systems. It does not want a democratic patent system for the simple reason that such a system will stop the flow of cheap patents.

It all comes back to two things. One, the extraordinary value of software patents to the patent industry. Two, the decision of the European Parliament in 2005 to kill the software patents directive. If the EP had legitimised software patents, we would never have heard of EPLA, but instead be looking at a community patent proposal that would turn software patents into a nice, harmonised growth sector.

EPLA was in fact the patent industry's back-up plan in case democracy failed to deliver the desired results.

About the FFII

It was mainly the FFII that helped democracy to do its work and kick out the software patent directive in June 2005. The FFII is not a typical organisation. It is born from free software ideology but it has nothing (directly) to do with free and open software. It is known for its professionalism and the quality of its research, but its internal organisation is a mess. It is kept running by the volunteer efforts of a network of activists who struggle to find time for their normal lives, but who will spend hours and days discussing arcane subjects by email.

For all its faults, I admire the FFII enormously. A careful reading of the founding document, the statutes, shows that the original idea of Hartmut Pilch, its founder, was to apply the techniques of free software development to the development of knowledge. To Hartmut, this is obvious and needs no explanation. To me, it's a brilliant concept.

Free and open software is not a business model, it's a technology, a set of techniques that improve the quality and accuracy of the software produced. I know this because for fifteen years, I've been using it for my own software. Small teams cannot afford to make highly polished big products - this is very expensive. But they can afford to take part of a big problem, and solve that very, very well. And because most of the work is done pro-bono, no-one can afford to spend time on things that the marketing department asks for. In a living free and open software system, every line of code does something that someone, somewhere, desperately needed. This makes the software accurate. All that's needed to make the whole thing work is cheap communications that lets dozens, hundreds of small teams work together.

And it's the same for the FFII. We don't produce highly-polished brochures or web sites. Try to become a member, it's shamefully difficult. But to look at the FFII and search for pretty logos is to miss the point. What the FFII produces, systematically and in huge volumes, is accurate and detailed knowledge.

This is why journalists and MEPs like us. It's why we can speak to a whole spectrum of opinion, from free software hobbyists to global corporations, with accuracy and conviction. It's why we can maintain a message like "all software patents, without question, are bad" over years, while many other groups try to compromise, and find themselves lost. It's because we solved the patent problem a long time ago, and the conclusion "ban software patents, they are a negative quantity" came out of the equation very rapidly.

The FFII does three things. First, we monitor and research policy development. This means tracking hundreds of events, workshops, committees, and initiatives. We do this through a large network of ears and eyes. Second, we develop policy. This means writing amendments, frameworks, whitepapers that can be introduced into the formal or informal discusison. Lastly, we lobby, which means convincing people to adopt our policies.

Any industry think tank does the same. The main difference is that the FFII does not do this for money. We do it because we believe we are right, and because we have no choice. Conviction is a powerful lobbying weapon, it is infectious, unbreakable, and keeps people going far beyond the normal limits. It's why the FFII can fight and beat highly-paid professionals. We are real people, speaking from the heart, and that amplifies our message a hundred times.

Towards a New European Patent System

The obvious question is how the FFII is bringing its weight onto the problem of EPLA. I can answer this quite simply.

Starting on November 25, we're organising a series of conferences called "The European Patent Conference"4. These conferences will bring together different opinions, data, case studies, and experts, with the goal of constructing the foundations of a new European patent system. We don't think the EPO is capable of doing this.

If you want to participate or help, you can come to one of the EUPACO conferences, or help us organise one in your city.

The Tip of the Iceberg

The debate over software patents is the tip of an iceberg. It is easy to demolish the arguments of the patent industry, who exploit the misconception that in a capitalist society, all property is good. Adam Smith showed that wealth comes from the division of labour and from trade. Some property forms, such as private ownership of business, encourage specialisation. Some, like public ownership of transport routes, encourage trade. But some property forms, like private ownership of essential infrastructure, create monopolies which damage the market.

The best metaphor for software patents is that your local government decides to raise money by selling off the roads and streets around your house to private owners who can then tax you to use them. EPLA is this flawed economic model taken to its logical, but insane, extreme. The private road owning industry decides it has trouble enforcing its tolls, so it has convinced government to set-up a court, run by the road owners, that will try and convict trespassers.

There are good objective reasons why you can buy a house but not a street. Similarly, there are good objective reasons why software copyright is good, and software patents are not.

But the real problem is not software patents. The real problem is a broken patent system that is exploited viciously by the patent industry, and no amount of insight will cause this system to magically heal itself, and the patent industry to vanish in a puff of smoke.

It is, similarly, insane to log and burn a forest on which communities depend, yet this is standard practice. When a powerful industry is unregulated, it can commit acts that would be considered criminal in a sane world, but are praised as "sound capitalism" in a world that looks no further than the quarterly balance sheet.

The sight of cities under water makes people stop and think, "why did we chop down all the trees?" It can take a long time for insane - but profitable - policies to translate into communal disaster. In pharmaceutics, software, medicine, telecoms, the lifecycles from research to product can be very long. Many software patents only "bite" after a decade or more of sitting on the shelf.

The impact of laws like the 1980 Bayh-Dole act5, which turned US universities from centers of research into patent factories, are only starting to be felt today. The impact of software patents on the US IT sector, is only today having a visible effect on productivity and innovation.

It's the slowness of this response that gives the global patent industry breathing space to promote its message across the world. This is why the FFII's delaying action is so vital. It gives Europe the chance to avoid making the same errors as the US and thus to gain a significant advantage.

But the fundamental question in the European context is deeper, uglier, and much harder to solve. How can an industry like the patent industry, which lives off others' work, gain such a foothold in the EU structures that it can drive the Commission and member states to consider a structure like EPLA?

The problem is that the European political process is, right now, bent, and near to breaking. We failed to get a constitution, for good reasons. We have a Commission and Council that lends itself to special interests. We have a Parliament that lacks authority. When Parliament fails to make the "right" decisions, it can be circumvented, ignored, and discredited. And this makes a mockery of the EU, surely to the delight of Washington.

Conclusions

The solution to the software patents problem in Europe lies in the creation of a proper and democratic political process which ensures that parasitical industries cannot buy and lobby their way to owning public property, and enforcing that through self-constructed kangaroo courts.

Democracy, a powerful Parliament, and proper processes. These are the cures for a broken patent system.

It's a massive challenge, and it goes far beyond the scope of an organisation like the FFII.

However, the FFII is not alone, and challenge exposes opportunity. There is a vacuum in the heart of Brussels. A new generation of politicians is emerging, who understand the difference between profit and wealth, who believe that democracy is more than a pallative for the middle classes, and who derive their mandate from the digital majority, not special interests. It is these politicians that will shape Europe. They are already doing so.

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License