Friday, March 30, 2012

Munich trial: Motorola's Android devices may infringe Microsoft patent on multilingual programs

Yesterday I attended a second hearing, effectively a trial on Microsoft's assertion of EP0669021 on "multi-lingual computer programs", at the Munich I Regional Court. In mid-December, the court held a first hearing on this patent (and two other patents). I reported on the first hearing and said that this patent appeared to be difficult for Microsoft to prevail on, though I wouldn't write it off.

This is a rather old patent. Its priority date is November 11, 1992. The claimed invention -- keeping multiple language versions of a text string pool available and loading only the required messages -- was apparently made in connection with an early version of Microsoft Excel. At the time, it represented an improvement over existing approaches to multilingual software primarily in the sense that it avoided recompilation after translation and allowed the distribution of a single version of the executable part of the product.

Judge Dr. Peter Guntz, who is presiding over this and a number of other smartphone-related patent infringement litigations in Munich, made it pretty clear that none of Motorola's invalidity contentions are likely to persuade him to stay the case pending the resolution of a parallel nullity action. I take this as an indication that Microsoft was indeed first to come up with this particular and efficient solution for easily localizable software. Leaving aside my personal perspective on software patents (especially since Motorola doesn't even challenge this patent on the grounds of patent-eligible subject matter), it appears, within the existing legal framework, that Microsoft was indeed entitled to a scope of protection that appears broad from today's perspective -- almost 20 years later.

The challenge is to come up with, and apply, claim constructions that reflect both Microsoft's innovative accomplishment at the given time and the evolution of hardware and software technologies since then. For example, a term like "to load into memory" has a different meaning in the age of virtual machines than it did back in the early 1990s.

The court identified four issues. In each of those four cases, a broader interpretation favors Microsoft's infringement theories (without making Motorola's invalidity contentions any more availing), while Android won't be deemed infringing -- at least not by the Munich I Regional Court -- if only one of the four terms is interpreted as narrowly as advocated by Motorola. With respect to at least one of the terms, it was hard to draw the line, just based on the discussion at trial, between claim construction in a narrow sense and the infringement theories specific to Android.

Microsoft's lawyers (from Bardehle Pagenberg, which was just named Germany's leading IP litigation firm by Managing Intellectual Property magazine) made significant headway during the course of yesterday's trial. At the outset, it appeared that the court had major doubts about the proposed interpretation of at least two of the four disputed claim elements. But Microsoft presented arguments that the open-minded panel of judges was fairly receptive to. As a result, Motorola's lead counsel, Quinn Emanuel's Dr. Marcus Grosch, felt compelled to dwell on these issues much longer than he intended to (he broke his promise to keep it brief) and usually does.

It's very hard to predict the outcome because the court appears torn between Microsoft's perspective, which is based on very strong technical arguments that reconcile each of the disputed claim limitations with the original purpose of the invention, and Motorola's argument that a combination of multiple broad interpetations, some of which appear to blur the line between particular terms, results in an overly broad scope on the bottom line.

As a computer programmer, I really agree with Microsoft's lawyers on this one. The solution for multilingual apps (allowing language-switching on the fly) that Android implements is undoubtedly the one that Microsoft invented some 20 years ago for Excel. I have been thinking very hard about this and I heard a total of almost five hours (in the aggregate of both hearings) of oral argument. Whoever wants to accuse me of bias can and will do so, but I have reached my conclusion by asking myself the following test question:

Which one of the following three answers is most accurate?

  1. Android uses Microsoft's solution

  2. Android uses one of the solutions that existed prior to Microsoft's patent

  3. Android uses a new, innovative solution that doesn't make substantial use of Microsoft's invention

Possibility #2 can be ruled out entirely. The mere fact that Android apps can switch languages on the fly but don't require redistribution of multiple separately-compiled program packages shows that Android's way is not "the old way".

Possibility #3 does not apply either. Yesterday, Microsoft's lawyers made an entire infringement argument based on the representations of Motorola's own expert report. It was crystal clear that the problem-solving strategy employed by Android is the one of Excel, just applied to today's virtual machines.

This leaves the question of whether answer #1, which is the right one from a technical point of view, can be reconciled with the logic of patent law. The court actually knows that each of the four issued identified can be decided in Microsoft's favor. The outcome is, without a doubt, a reasonably broad patent. But Motorola has, at least based on how I understood the court, not been able to identify prior art that would render it invalid -- even though it's a safe assumption that Google (since this is about Android at large) also contributed a lot to the quest for prior art.

The decision will be announced on May 24, 2012. We will see if the court overcomes its unease over the effective (but apparently legitimate) breadth of this patent.

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: