Sunday, July 8, 2012

German court doubts that Android infringes Apple's touch event model patent

On Friday (July 6, 2012), the Mannheim Regional Court held two trials involving Apple's touch event model patent (EP2098948). In the first one, the defendant was Google subsidiary Motorola Mobility; in the second one, the defendant was Samsung. There had already been an Apple v. Samsung trial over this patent in April, but the proceedings were reopened. Later this year, HTC will also have to defend itself against this patent in the same court.

Two days before those hearings, the London-based High Court of Justice had declared this patent both invalid and not infringed by HTC. On this occasion, I have a couple of errata to report. In that report on the UK ruling, I correctly listed the different parts of the decision in the hierarchical bullet-point list at the start, but I confused two patents in my analysis of the further implications. The result of my analysis is unchanged, but I had mistakenly stated i it that the touch event model patent was the only one to be deemed valid (that was, however, the photo gallery page-flipping patent). I've corrected that post. Also, I've corrected a couple of posts that mentioned a letter from six United States senators to the ITC. I had called all six signatories Republicans. Five of them were, but Senator Herb Kohl, Chairman of the Antitrust Subcommittee, is a Democrat. The fact that the letter was signed by senators from both sides of the aisle makes it even more significant. I apologize for these little oversights and wish to thank the two attentive readers who flagged them to me.

The part of the UK ruling that related to the touch event model patent was obviously mentioned in Mannheim, but Judge Andreas Voss ("Voß" in German) reminded the parties that his court is not the appeals court for Justice Floyd's decision: just like Justice Floyd reached his conclusions independently of a previous decision in The Hague, the Mannheim court will makes its own decision. The Mannheim Regional Court, especially the chamber over which Judge Voss presides, is very technically-savvy. I don't want to compare its competence to that of the UK court since I know much less about Justice Floyd (other than a couple of decisions that I've read), but for a court whose judges don't have an engineering background, this Mannheim panel has an extremely good grasp of technical issues.

It became clear that four key elements of the UK decision on the touch event model patent may not be adopted by the Mannheim court in its rulings on these Apple v. Motorola Mobility and Apple v. Samsung cases:

  1. Neither Google (Motorola Mobility) nor Samsung argued at these trials that the patent should be deemed invalid for lack of a technical contribution (or to put it in very simple terms, because it's a pure software patent and, as such, prohibited by the European Patent Convention). Since HTC made that argument in the UK, it probably does so in the Mannheim case as well, but that one hasn't gone to trial yet.

  2. Judge Voss stressed his court's reluctance to base an assumption of invalidity on the concept of obviousness over common general knowledge. No one I asked knows of any example in which the Mannheim court did so. Judge Voss said that the patent has been granted and it is binding on the court unless the judges are quite convinced that it's invalid. They want to see prior art. Some prior art was referenced, but based on the discussion, I didn't get the impression that the court doubts the validity of this patent on that basis.

  3. The first reason for which the UK court believed that Android doesn't infringe this patent is that there isn't a separate touch flag associated with each user interface object. The purpose of an exclusive touch flag is to suppress touches on other items. Judge Voss mentioned as an example that a user shouldn't simultaneously touch a Play and a Rewind button in a music player app. The London ruling concluded from the wording of the patent claims that there must be a "per-view granularity" (a "view" is a user interface element, or UI control, in the context of this patent), while Android has such a flag only at the window level (but it then relates to each view contained within that window). The Mannheim court is not going to interpret the claim language as rigidly as its counterpart in London did. Nevertheless, a more functionally-oriented analysis may very well result in a finding of non-infringement, as I'll explain further below

  4. The second non-infringement argument concerns the selective sending of touch events to the program code associated with a given view. This one succeeded in London but has much less traction in Mannheim.

I believe the Mannheim court would be absolutely right to reject the UK court's thinking with respect to item 3 (the "associated with each view" thing), and probably right with respect to item 4 (selective sending). Even before the Friday trial I told different people in private conversations that I thought the UK court's understanding of the association of flags with views was too literal. As far as the second item (obviousness over common general knowledge) is concerned, it's unfortunate that German patent lawsuits are bifurcated, i.e., the system allows an infringement court to order an injunction long before a separate, specialized court has concluded its analysis of the validity of a patent, and gives undue strength to patents that should never have been granted in the first place but get enforced anyway. The Mannheim court didn't create that flawed system; lawmakers did.

Even though the three Android device makers Apple is suing in Mannheim over this patent won't get as much mileage out of the UK ruling as they probably hoped to, Apple is still not very likely to prevail on this patent in Germany. After the first Samsung trial in April I thought that Apple had a chance of approximately 40% of prevailing. I now think it's closer to 20%. The hurdle that is hard for Apple to surmount here is that the Mannheim court essentially said this: "OK, we accept that there doesn't have to be a separate flag for each view; a window-level flag might do the job. But does it do the job? Can it do the job at all? Or is the net effect that there won't really be the exclusive touch flag the patent discloses?"

That functional approach is hands-down superior over the literal one adopted by the London court. Apple's primary infringement contention relates to a scenario in which there are two triangular, movable text selection markers, each of which is a one-view window, on top of an app that has a number of other views (buttons etc.). And at the April hearing Judge Voss noted that a single instance in Android in which the patent is infringed would result in a finding of infringement. But even Apple's preferred example is not too likely to persuade the court. There is doubt about whether there really is an exclusive flag or whether touches on other elements than the two selection markers are still processed. One of Apple's lawyers had to concede that Apple doesn't have access to all of the source code and therefore don't know which of these items are windows and which ones are (at least in more recent Android versions) view groups.

The exclusivity issue is fairly likely to result in an overall finding of non-infringement. If there is a finding of infringement, it will most likely be limited to particular situations in which there is only one user interface control per window with an exclusive touch flag. Such a narrow finding would still cause significant problems to Android in terms of erratic behavior, probably including crashes, by an unknown number of apps until those apps are rewritten. But in order to cause huge problems, Apple needs an infringement finding of broad scope, and that one is not likely at this stage.

The court will make a decision on the Motorola case on August 31, and on the Samsung case on September 21.

Apple is having a hard time winning anything in Mannheim, but I wouldn't blame it on the court. I think Mannheim would be a great venue for the '263 "realtime API" patent, the '647 "data tapping" or the '604 "Siri (unified search)" patent, but Apple doesn't appear to have European versions of those patents. Sooner or later, it will identify some European patents on which it can prevail. It has already prevailed over two European patents (slide-to-unlock and photo gallery page-flipping, both against Motorola in Munich and the latter one also in the The Hague against Samsung), but those were too easy to work around.

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: