Thursday, April 29, 2010

FATal patent ruling in Germany?

Last week the Federal Court of Justice of Germany upheld a Microsoft patent related to the Windows file system named File Allocation Table (FAT). As H-Online mentioned in this context, European software patent critics dread the notion that rulings such as that one could elevate the status of software patents in Europe, where different national courts have different approaches to how to interpret the European Patent Convention and its exclusion of patents on "programs for computers [as such]".

Impact of decision yet unclear

Whether this ruling represents an endorsement of software patents in the largest European country remains to be seen. The actual decision has not yet been published. The court only issued a short (German-language) press release.

That announcement does not mention whether or not the validity of the patent had been called into question by a third party on the grounds of patentable subject matter. The press release only makes it clear that a certain CD-ROM-related protocol was not deemed to be relevant prior art since it solved a different problem than the one addressed by Microsoft's patent.

This is an important legal detail to know: the Federal Court of Justice of Germany is an appeals court and it is not even allowed to speak out on aspects of a case that are not formally put before it as part of an appeal. This means that even if the judges look at the patent as a whole and may believe that it could be invalidated for some other reason, they must not say what the outcome would have been if a different kind of question had been asked.

In this case, the previous instance (the Federal Patent Court of Germany) had declared the patent invalid, so it's a safe assumption that Microsoft itself filed the appeal in order to defend its patent against invalidation. If Microsoft's appeal related to the prior art question only, then the appeals court had to rule inside of that particular box. That's the way it works. An appeal to that court can have a very narrow scope.

Even though the press release begins with the result for this particular case -- the appeals court deemed the patent valid --, one has to look more closely to see whether there is a precedent that will make other software patents more likely to be upheld. When the full text of the decision is published, I will post an update.

FAT patents prove indestructible

This is not the first time that someone fails with an attempt to have one of Microsoft's FAT patents invalidated. Here's a CNET story from 2006 on the failure of a US-based organization (formally a non-profit but certainly designed to keep some lawyers busy) with a similar attempt.

There's an important conclusion to draw from this. Even though the FAT solution for storing longer file names in addition to shorter ones (8 bytes plus extension of 3) would seem "obvious" to most of us in a colloquial sense, this qualifies as "inventive" under patent law on both sides of the Atlantic.

If it were up to me, the legal test would have to be considerably higher. Instead of requiring just some limited evolution over existing technologies I believe the question would have to be whether a 20-year monopoly on a particular solution is justified. I for my part would deny it in a case like this, and it's by far not the worst. I'm hesitant to call it a "trivial patent" because I've seen so many software patents that are far worse. For a software patent, this is -- I regret to say so -- definitely not below the average level of quality.

With many patents comes great responsibility

Given the way the system works, I can't blame any company for filing such patent applications. Some large corporations like IBM and Microsoft file thousands of them every year now. I can also understand that if a third party then tries "patent busting", the patent holder doesn't want to lose what he's got, so he'll defend the patent against any invalidity claims put before patent offices and courts. That's understood.

The key issue is how patent holders, especially "mega patent holders", exercise the rights that the patent offices of the world grant them. If those organizations strike cross-licensing deals with their peers, it doesn't affect smaller companies and the FOSS community (who will never own a huge patent arsenal). If they use those patents as a measure of innovative capacity, one may disagree that it's the appropriate criterion (quantity instead of quality), but again doesn't hurt the rest of us.

Where things become problematic is when such patents are used to shut out competitors from the market, especially if a market is in dire need of more competition and customer choice.

If you'd like to be updated on patent issues affecting free software and open source, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents.