Thursday, October 4, 2012

Oracle gives notice of appeal of ruling in Google patent and copyright case

At close of business on Wednesday, Oracle filed with the United States District Court for the Northern District of California a notice of appeal of Judge William Alsup's ruling in the patent and copyright litigation over Android's use of Java material beyond the keywords of the language. Here's the text of the notice (click on the image to enlarge or read the text below the image):

Notice is hereby given that Plaintiff Oracle America, Inc. (“Oracle”) hereby appeals to the United States Court of Appeals for the Federal Circuit from the Final Judgment (Dkt. No. 1211) entered in this action on the 20th day of June, 2012, and from any and all other orders and rulings adverse to Oracle.

After Judge Alsup's decision holding many thousands of lines of Java API material to be uncopyrightable, Oracle had already announced its intent to appeal. Even without an announcement, all observers would have expected it to happen.

The notice was signed by Michael Jacobs of Morrison & Foerster, who was one of the top three or four attorneys on Apple's external litigation team in the Samsung case that went to trial in the same district. The two cases -- Oracle v. Google and Apple v. Samsung -- have even more in common. At both trials, evidence of the intentional use of intellectual creations was presented. But there are also some important differences concerning the issues, the approaches of the judges, and the inclinations of the juries (while the jury foremen in both cases, based on what they later told the media, supported the plaintiffs, the foreman in Apple v. Samsung was much more successful in building support for his positions than the one in Oracle v. Google, which is also why one jury spent far more time on deliberations than the other).

The overlaps between those two cases also serve as a reminder of something else: Google cannot solve Android's IP issues by fending off one case or even several cases, or by delaying their resolution. Courts around the globe have found Android to infringe a total of 17 different Apple and Microsoft patents, 6 of them over the last three months, while Google's Motorola Mobility and its independent hardware partners are enforcing only a single patent against Apple (and none against Microsoft) at this stage (by comparison, nine of the 17 aforementioned patents apparently violated by Android are being enforced as we speak, a couple even in more than one country at the same time).

While there is empirical evidence that Android's success in the market is in no small part built on the use of concepts and material actually developed and created by others than Google and its partners, I obviously don't mean to say that any assertion against Android must be presumed meritorious only because there are so many other cases in which Android's infringement has been established. Some of those lawsuits will ultimately succeed, some won't. But the combination of all of these official infringement findings (which were made in different legal systems) and the mountain of evidence of willful conduct that was presented at the Oracle v. Google trial (of which the "Lindholm email" is only one example) does suggest that when all is said and done, some more significant parts of Oracle's intellectual property than small snippets of code -- may very well have to be added to the list of third-party IP infringed by Android.

Ten days ago I commented on Apple's and Samsung's Rule 50 ("overrule-the-jury") motions and, since I don't think Judge Koh is likely to make more than some limited adjustments to the August 24 jury verdict, the prospects of their subsequent appeals to the Federal Circuit. At the start of that recent post I listed some Android-related cases that have already been appealed to the Federal Circuit and could result in some further infringement findings (in those earlier cases, plaintiffs, especially Apple, had a lower hit rate than Apple just had in the Samsung trial). And in this context I already mentioned that there are reasons for which Oracle's upcoming -- and now-notified -- appeal is actually more likely to yield a major breakthrough for the appellant than Samsung's future appeal. Now that Oracle has given notice of appeal, I'd like to quickly elaborate on this a little bit more.

What will be quite doable for Samsung is to get some adjustments. It will probably get some from Judge Koh herself. It may get even more from the appeals court. Apple, too, may get some adjustments, such as on the Galaxy Tab 10.1 design patent issue, but at this stage Apple is more interested in preserving the verdict than Samsung, which would prefer a new trial. I don't rule out that Samsung may get the damages award reduced below to an amount below a billion-dollars. By my own count, there's at least one striking misapplication of the law by the jury in connection with the Galaxy Prevail phone, and that item alone would take the amount below a billion dollars, though Apple is also pushing for willfulness enhancements and supplemental damages to the tune of more than $700 million. But neither Judge Koh nor the appeals court will modify the verdict to the extent that Samsung is cleared of all the infringements identified. The jury held Samsung to infringe six patents (three design patents, three software patents) and one trade dress. Some of these intellectual property rights account for a greater amount of damages than others, and some would be more strategically important than others if an injunction was granted over them. But all in all, Samsung is fighting against seven IPRs, and if it overturns the verdict with respect to one or two or even three of them, it still faces a major infringement problem. There's no single issue on which Samsung needs to prevail to fundamentally change the picture. Even if it proved jury misconduct (an issue I'm not going to follow in detail until it's resolved), it would merely get a new trial, at which it would once again face the same Apple patents and trade dresses, and would once again be unable to explain away all of the evidence of intentional copying.

By contrast, Oracle is two steps away from winning by far and away the most important part of the case: the API copyright liability part. In order to prevail on the API copyright liability part, Java API copyrightability is an issue on which Oracle absolutely must prevail at the Federal Circuit or, possibly, at the Supreme Court if the highest court agrees to look at this matter (if Oracle wins at the Federal Circuit, Google will appeal to the Supreme Court, and vice versa, and the issue might be important enough for the Supreme Court to grant certiorari). There's no single potentially game-changing issue like Java API copyrightability in Apple v. Samsung. And even the 10 most important items on which Samsung needs to prevail are different from the API copyrightability question in terms of raising issues of a more factual than legal nature. The jury will only be overruled if it found on a given item what no reasonable jury could find. A judge doesn't get the same level of deference, especially not if an issue is more legal than factual, or purely legal. And a district court gets pretty much no difference at all if an appeals court determines that the district court answered a legal question of first impression. In that case, the appeals court can easily substitute its own opinion for that of the lower court. The appeals court just has to arrive at a different conclusion, and use its authority to decide. That's just so different from the requirement for overruling a jury verdict that the appeals court must have an "abiding conviction" that the jury got it wrong.

Now let's assume, for the sake of the argument, that the asserted Java API specifications are ultimately deemed copyrightable. What else would it take Oracle to win on the API copyright count? It would still take something, but not really much.

The issue at the trial was not that the jury thought Google's use of the material in question was fair use. A majority of jurors may have thought so based on what one or more jurors told the media afterwards, but the jury simply couldn't agree. The jurors who were supporting Google's fair use defense were reportedly led to believe that Android's use of Java was "transformative" -- and I criticized the jury instructions on this fair use factor even before jury deliberations began because I felt that Judge Alsup's instructions suggested, to a layperson jury, that there wasn't a reasonably high threshold for finding in favor of fair use only because someone combines infringing and non-infringing material in one product.

That said, if Oracle prevails on copyrightability, it still needs to defeat Google's fair use defense. The jury did find Android to infringe the asserted Java APIs -- but it couldn't reach unanimity on the fair use defense, and even if it had, Judge Alsup's finding against copyrightability would have made that finding irrelevant for the purposes of the district court proceedings.

If Oracle gets back in the game because of a reversal of the (un)copyrightability decision, there are three ways in which it can subsequently solve the "fair use" problem:

  1. The best result for Oracle would be a finding that Google's fair use defense has to fail as a matter of law. In that case, the jury's infringement finding could be combined with a legal determination that there was no fair use, and Oracle has won.

    Many courts have decided "fair use" issues as a matter of law. This includes a "fair use" case Google frequently cited. "Fair use" is an issue of law and fact that is called "an equitable question of fact" -- since equitable determinations are normally made by judges, "fair use" is particularly amenable to judgment as a matter of law.

  2. In a second-best scenario for Oracle, the infringement finding from the first trial stands even though a retrial is needed on "fair use". Google argues that the two issues can't be separated, but Oracle says they can. A retrial focused only on "fair use" would be a new opportunity for Oracle to prevail on what would then be the last missing piece. A jury trial is always a lottery, but instructions do play a very important role. One of the things that I'm sure Oracle will try to win from the appeals court is a finding that the jury instructions on "fair use" at the first trial were prejudicial to Oracle, especially with a view to transformative use.

  3. If Google succeeded in connecting the API infringement and fair use issues to the effect that both would have to be retried together, there would be a new copyright trial. But the jury would no longer be told that it merely has to presume for the purposes of its deliberations that the asserted API material is copyrightable: it would be told that this is the case. Oracle would have to prevail on infringement again, and win a unanimous decision against fair use for the first time. So in this scenario, Oracle would have to prevail on two questions, not just one, after a reversal of the (un)copyrightability finding. But the infringement part is not that hard for a jury to establish. By looking at the accused program code and the asserted copyrights, there are enough similarities to see that a substantial amount of material was copied. Fair use was a bigger problem at the first trial because of some misconceptions about what kinds of transformative use justify what would otherwise constitute infringement.

What I'm advocating here is a rational approach to all those appeals of smartphone IP cases. An appeal is never easy to win, and no party likes to come from behind after the district court (or ITC) proceedings. But there are different standards of review (more deference or less, depending on the kind of question and original decision-maker), and there are differences in terms of how a reversal of a single finding can change the overall situation in an instant. As long as Oracle's appeal hasn't been adjudicated by the final judges, Google and its Android ecosystem still face considerable risk. And that's why this appeal is going to be one of the most important ones this industry has ever seen. I can't predict the outcome, but for the reasons I outlined above, I can say with certainty that Oracle v. Google isn't over till it's over, and Samsung's post-trial challenge is clearly the steeper one.

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: