Thursday, April 19, 2012

Motorola under pressure in Germany from Microsoft patent on event architecture

Today I went to a trial at the Munich I Regional Court and saw Motorola, for the first time ever, in a sitation in which it was highly uncomfortable with a patent dispute in Germany, its preferred jurisdiction. While Microsoft was first to sue Motorola in the U.S., Motorola was first to sue Apple and Microsoft in Germany. Both of them struck back in Germany a few months later, and Apple has already scored two partial victories, but as I wrote after Apple's second win, the enforced patents weren't "killer patents". The Microsoft patent at the heart of today's trial, EP0651328 on an "event architecture for system management in an operating system", is a patent of considerable technical depth that can also be worked around (it's not essential to any particular industry standard), but if this one is successfully enforced, significant changes to Android and, potentially, to many if not all existing Android applications will be needed.

Motorola currently doesn't have a huge market share in Germany, but this is a major market and it would be more than just awkward for Google to have to choose between a very sighificant reengineering effort for Android on a worldwide basis (since it wouldn't make sense to modify only the German version in such a fundamental way as it would be necessary in this case) and leaving the market. Of course, the best choice for Google/Motorola would be to strike a license deal with Microsoft, and it's no secret that this is Microsoft's objective, but Google/Motorola could have settled this dispute a long time ago if it had opted for licensing rather than litigation.

There are several reasons for which I am convinced that Motorola is worried about the potential outcome of this case. I have seen a number of trials here involving Motorola, and this is the first one at which there was really tension in the air that one could feel. The court's inclination prior to today's trial was that none of Motorola's non-infringement arguments appeared overly convincing, though it's also a fact that the court still saw a need for quite some clarification. And the court's preliminary position was that this patent is not particularly likely to be invalid. At one point, the Judge Dr. Peter Guntz even felt forced to reproach Motorola's lead counsel for laughing out too loud while Microsoft's lawyer was speaking. The judge also asked Microsoft's lawyers not to provoke such emotions through dismissive rhetoric, but this really didn't occur at any of the other Motorola trials I saw in recent months.

The trial lasted almost three-and-a-half hours, which is very long by German standards where most patent trials take only approximately two hours. And this time was spent almost exclusively on claim construction and infringement issues, without an in-depth discussion of Motorola's invalidity contentions. Even though those contentions don't have much traction so far (Motorola's two primary prior art references may not even be eligible because they are standardization documents that were password-protected), I felt that Motorola's counsel rightly asked the court to consider a separate hearing on Motorola's invalidity defense -- and the court agreed. Motorola's interest in being heard in connection with its invalidity contentions is, however, not principled. If Motorola truly supported the idea of a reasonably strong invalidity defense, it would not have capitalized on the highly problematic approach to a combined FRAND and invalidity defense that a Mannheim judge chose in February (in a trial at which Microsoft had to defend itself against two Motorola patents). But Munich isn't Mannheim, and Motorola may still get an extra hearing (if it takes place, then probably in May) for an in-depth discussion of its invalidity arguments.

There were parts of the discussion that I was able to follow because I have quite some experience with object-oriented programming, but there were als parts at which the parties discussed infringement claim charts that I don't have access to. Still, it's a safe assumption that Motorola needs to prevail on at least one of its key claim construction arguments in order to have a decent chance of avoiding an infringement finding. With respect to two fundamental claim constructions issues that Motorola raised, with the help of a computer science professor who was sitting next to Motorola's counsel, it was easy to understand why the court was, at least at the start of the trial, rather reluctant to adopt Motorola's attempts to narrow the scope of the patent.

For the programmers reading this blog, I'll quickly explain the two technical issues on which I'm convinced the court is right to approach Motorola's argument with skepticism. Both of these issues relate to the difference between abstract classes and interfaces.

Motorola would like the patent to be interpreted as being limited to (standardized) interfaces in a very narrow sense that excludes (standardized) interfaces for abstract classes in order to argue that Android's C++ code uses abstract classes to achieve the same goal. From my own programming experience I know that the difference between an abstract class and an interface in the narrow sense of the word is a subtlety that may play a role in connection with a very few patented inventions but is pretty irrelevant to an event architecture. I have often started designing base classes that had abstract members, but at some point I felt that an interface was the more elegant solution. The only fundamental advantage that interfaces (in the narrow sense of the word) have in some programming languages over abstract classes is that one class can implement multiple interfaces while multiple inheritance is rather uncommon (it raises too many issues). At the level of the object code, one can achieve identical results anyway. The court was right that what it really comes down to is whether the standardized functions the patents refers to are available.

Motorola furthermore argued that diferent classes inherited from the same base abstract class have their own constructors, and I don't deny that constructors can come in all shapes and forms, but for the purposes of this event management patent there won't really be a difference (apparently all those constructors need just one parameter for this purpose, though I guess additional parameters are always an option, which doesn't mean that the interface isn't standardized).

A key item of discussion on the claim construction side was the term "interface pointer". The court appeared inclined to agree with Microsoft's lawyers that this term cannot be interpreted in only one way but probably needs to be construed based on what the patent (the claims and the description) says about the intended meaning of the term. Motorola insists that the term had a unique meaning to a person skilled in the art of computer programming at the time of filing. Theoretically, this could require the court to appoint an independent expert, but Judge Dr. Guntz indicated that this would be a last resort, given that the interpretation of a given patent is usually a matter of law, not fact. Without knowing what the implications of different definitions of "interface pointer" are (with a view to the infringement analysis), I can't take a definitive position here, but generally speaking, a "pointer" is something that computer programs often pass around without a developer even thinking about it too much. That is particularly true of higher-level languages, and such languages clearly existed when this patent was applied for in the early 1990s. To the extent that I was able to follow the discussion of this item, I felt that a functional understanding of the term madem ore sense than anything based on extrinsic evidence.

At today's trial, Motorola tried to tackle this patent from different angles, as opposed to focusing entirely on just one attack vector. And toward the end, there was quite some wrangling over procedural questions. Motorola tried hard to raise issues that could result in a delay of the ruling. That's not unusual, but the way in which this effort was made added to my impression that of the three cases Microsoft originally brought against Motorola in Munich, this is the one Motorola is most concerned about.

This doesn't mean to say that Microsoft has won. The court was open-minded and gave Motorola every opportunity to dispute infringement (and will possibly hold an additional hearing only on Motorola's invalidity arguments). A decision will come down on July 26, 2012, unless the court reaches its conclusions ahead of schedule (Judge Dr. Guntz said they would try to decide earlier if possible, but pointed to their busy patent docket).

If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents and Google+.

Share with other professionals via LinkedIn: