In what seems to have been a shock to all concerned, the European Commission yesterday said last week that it would ban software patents in the EU in the upcoming community patent legislation – and would be binding upon the European Patent Office.
Some background. Everyone agrees that anything to do with patents in the EU is ridiculously expensive. Every EU country has its own national office that can grant patents, and every country has courts that can invalidate them.
There are so many weird anomalies in European patent law that it's unreal. For example, the whole mess is subject to the 1973 European Patent Convention. But the EPC isn't an EU treaty; it applies to the countries that have ratified it. Most of those are EU members, but not all. The European Patent Office (which is headquartered in Germany), on the other hand, while it was formed under the EPC isn't subject to EU control – it's not an EU institution. The national patent offices are subject to their national laws. Under the treaty, you don't need to apply to each separate country; instead, you can list the countries you want on your application to one office, and if the patent is granted it will be recognized in each place. Patents granted by the EPO are valid in all the EPC countries.
Any patent can still be ruled invalid by a court of law. But – and this is the really expensive part – action against a patent can currently only be taken at the national level. So the system is hugely unbalanced toward granting patents and keeping them valid: you can apply in one place to get a patent that, if awarded, will be valid everywhere in the EU, but you must bring an action in every single national court if you want it overturned. It is incredibly expensive and time-consuming, and probably by the time you'd gotten through all those national courts the patent would have expired anyway. But what's your other choice? Wait 20 years to pursue your idea?
On the other hand, if you're the small business with a great idea that competitors are trying to grab, it's just as expensive to defend yourself. Hence the comment a struggling inventor made to me a couple of years ago: "Patents are only as valuable as the money you've got to litigate to protect them."
OK. You're the EU and you want to compete with the US, which grants 18.75 patents per hour (2004 figures; that rate has undoubtedly gone up since then). You want – or ought to want – a patent system that inventors respect, that promotes progress, and that confers whatever economic benefits you believe patents confer. You therefore ought to want a system that rewards genuinely new inventions and discoveries while not rewarding bad ones. It needs to be efficient, inexpensive enough for small businesses to use (but not so cheap as to make it as easy for patent trolls as it is for spammers), and responsive. Bad patents should go quickly; they are detritus clogging up the system.
Obviously harmonization is important; in fact, it's been on the EU agenda since 2000, and about once every couple of years someone tells you, "I think it's really going to happen this time." But the announcement removes one of the biggest stumbling blocks.
The stumbling block for the last few years has been software patents. Germany, according to anti-software patent campaigners, tends to take the most robust view, declining to patent anything that doesn't have an effect on the physical world. The UK and the EPO, according to the same source, tend to be the most liberal (and US-like) about awarding patents on software. Even if that's not true (as the UK Patent Office has said), opponents of software patents remain deeply suspicious of any move toward harmonization; they remain convinced that software patents are the hidden agenda. The 2005 defeat of the European Computer-Implemented Inventions Directive, after all, was more because of an internal power struggle between the European Parliament and the European Commission (at least, according to Florian Mueller's book) than because of the substantial grass roots opposition. If the new law is forced upon the EPO, that could go a long way toward changing practice in the EU. The key will be in defining the appeals process, the route by which a European patent can be overturned. You want independent judges and a streamlined system.
A couple of weeks ago, I ran into an MEP who was active on the pro-software patent side of that struggle. The political will, he said, is not there now to get software patents through. But harmonizing European patents and making litigation cheaper to make the system more usable for small businesses are, very much so. Taking those off-the-cuff comments and yesterday's announcements together, it looks as if proponents of software patents are willing to sacrifice them, at least for now, in the interests of getting harmonization through. Software patents can – and I'm sure will – be revisited later, at which point they may be easier to legalize. I can't believe Microsoft, Nokia, and all those other large companies are just going to shrug their shoulders and give up. But that's a way off.
Wendy M. Grossman’s Web site has an extensive archive of her books, articles, and music, and an archive of all the earlier columns in this series. Readers are welcome to post here, at net.wars home, at her