Saturday, October 15, 2011

Trial briefs filed: Oracle wants an injunction unless Google brings Android into full compliance with the Java standard

Even though a postponement is likely, the Oracle v. Google is still formally scheduled to begin in about two weeks (on Halloween). Late on Friday, the parties filed their trial briefs and proposed jury instructions.

The trial briefs make it perfectly clear what the parties' key factual representations and legal arguments are at this decisive stage. I have uploaded the trial briefs to Scribd (Oracle's brief, Google's brief), but I'll also summarize them in this post for all those who don't have time to read a total of 60 pages.

Oracle's story: Google capitalizes on Java and causes irreparable harm through fragmentation

Oracle opens its brief with accusations of reckless infringement:

"Oracle will prove at trial that Google deliberately chose to base its Android software platform on Java technology, seeking to develop and deploy Android rapidly and to capitalize on the large community of Java software developers. [...] Google chose to take its chances and push forward with Java, helping itself to Oracle’s intellectual property without a license. Oracle will prove at trial that Google's Android software platform infringes (1) Oracle's copyrights in its Java software platform, and (2) six of Oracle's Java-related patents."

I can't help but think that the copyright part of this case has gained considerable weight during the course of this litigation. In Oracle's original complaint, it almost looked like an afterthought. The copyright infringement assertions in that one were so unspecific that the judge ordered Oracle back to the drawing board. Oracle's amended complaint fixed that deficiency. But now, in the trial brief, Oracle devotes a lot of space to the copyright part of the question and starts with the copyright part, while the original complaint discussed patents before copyright.

I attribute this to several factors. Google's motion for summary judgment on Oracle's copyright claims failed for the most part. The judge didn't make a determination as to whether Google does infringe valid copyrights (since Oracle didn't ask for that decision), but his reasoning for dismissing Google's request for a free pass certainly contained statements that Oracle will consider encouraging. Google doesn't deny the use of Oracle API material and code: it only says that it was allowed to, and that the related material wasn't protectable. But Google's own expert on this and Joshua Bloch, a Google engineer and API designer, said some things that appear to buttress Oracle's copyrightability claim. Additionally, Bloch practically admitted to copying. Oracle would ideally like the jury to see such evidence in order to enhance the credibility of its patent infringement assertions. Another important consideration at this stage is that all of Oracle's patents-in-suit except for one (the '520 patent) are under significant reexamination pressure, which might result in a partial stay, and in that case the first trial might be mostly, if not exclusively, a copyright trial.

With respect to all of the intellectual property rights involved, Oracle claims willful infringement:

"Google's infringement was willful from the very beginning. [...] Indeed, even after Oracle's legal team presented patent infringement charts to Google's lawyers on July 20, 2010, Google continued to make Android available on its website and even created new versions of the platform. Management repeatedly signed off on Android releases notwithstanding Oracle's patent assertions. [...]"

Oracle emphasizes the extent to which Google benefits from the alleged conduct:

"Google's damages expert says that Android is losing money. But the truth is that Android is wildly successful, and is one of the most popular technology products in the world today. 550,000 new Android devices are activated every day. Google Chairman Eric Schmidt has publicly touted Android's profitability, telling the market that Android revenues are 'enough to pay for all of the Android activities and a whole bunch more.' [...] Google estimates that the real value of Android could eventually become up to $10 billion per year."

The 550K figure of daily Android activations is already a few months old. At the time, Android chief Andy Rubin claimed week-on-week growth of 4.4%. The number may be substantially higher now, but maybe the Oracle trial is part of the reason why Google hasn't publicly talked about the latest figures in a while. Another possible explanation is that growth may have flattened, but even if so, I think the number must have gone up quite significantly.

Oracle sums up Google's benefits from Android in the following way:

"Google developed Android with the expectation that it would be widely adopted and profitable for Google by: (1) generating significant advertising revenue streams from Google search and other Google applications on Android devices; and (2) reinforcing Google's dominant position in search and advertisements by ensuring its access to and position in the increasingly important mobile marketplace."

Oracle raises Google's dominance in search and online advertising only because it seeks additional damages based on a network effects theory. Independently from Oracle's obvious agenda, regulators around the world have become increasingly concerned about a possible strengthening of Google's market dominance in search and online advertising because of Android's success and the related business model.

As I said on several previous occasions, the big question in this case is not the exact amount of damages that may be ordered but the possibility of an injunction. Oracle's brief argues that money is not enough to compensate for the damage Android is causing, and that argument is based on fragmentation (Android does not adhere to the official Java standard):

"Android’s incompatible forking of Java has caused irreparable injury to Oracle, and monetary damages are inadequate to compensate Oracle for the injury Android has caused to the value of the Java platform. An injunction is necessary to prevent Google from further fragmenting the Java platform and undermining Oracle's and others' investments in Java."

In a footnote (and elsewhere in the brief), Oracle clarifies that it would drop its request for an injunction as soon as Google agrees to bring Android into compliance:

"Oracle will forego its injunction request only if Google commits in writing, in a form prescribed by Oracle, to ensure that Android will be Java-compatible and comply with all applicable Java licensing and payment conditions."

I think this is a snart move by Oracle. The judge appeared reasonably sympathetic to the fragmentation argument. In his decision on Google's summary judgment motion, he wrote that "[o]n the present record, a reasonable fact finder could disagree with Google's rosy depiction of Android's impact on the Java market". Oracle has to prove irreparable harm, and fragmentation is the argument on which it's betting -- and it's a strong one (provided that Oracle can prove that Google infringes valid intellectual property rights).

By tying its request for an injunction to the issue of fragmentation, Oracle demonstrates that it's only defending a legitimate interest (one that Google is also pursuing in connection with its Android platform) and not just seeking an injunction for the sake of harming Google's business.

Bringing Android into compliance with the Java standard wouldn't be an easy decision for Google. It would require changes not only to Android's Dalvik virtual machine and development tools (and the related documentation) but also to most -- or possibly all -- existing applications. But Google can't blame Oracle for the fact that its (Google's) own decisions have led to the current situation.

By the way, Oracle also mentions the new Amazon Kindle Fire tablet as an example of Android's success and the growing fragmentation problem facing Java.

Google's story: an injunction isn't warranted and damages (if any) would be closer to $50 million

Google's lawyers basically try to portray Oracle as a loser and a troll:

"This lawsuit is an attempt to obtian through the courts what the company formerly known as Sun Microsystems [...] failed for years to achieve through innovation, negotiation, or even costly corporate acquisitions--a viable path to competing int he mobile computing market. Oracle missed that boat and now wants the benefit of Google's initiative and hard work, both on its own and with numerous partners."

While it's true that Java never became as popular on other mobile platforms as on Android, I think few people would seriously accuse Oracle of suing Google as an alternative to other forms of competition. One may or may not agree with Oracle's infringement allegations in detail (for example, I think a significant number of their asserted patent claims are invalid), but the quoted passage is clearly inconsistent with some of the documents in the case file. Such documents as the Lindholm email show that Google recognized a need to take a license, but decided not to take one.

Google's trial brief has many strong parts. There's no doubt that Oracle will have to overcome some significant hurdles in order to win. But the introduction tells a story that has zero credibility.

Google defends itself against Oracle's patent claims (26 claims of six patents) by asserting that those claims are not infringed, and if they are infringed, then they are not valid. On the copyright side, Google claims that the allegedly infringed material is simply not copyrightable. Google failed to win summary judgment on the basis of those arguments, but the judge didn't preclude Google from re-raising those arguments at trial.

With respect to both types of intellectual property rights at issue (patents and copyright), Google argues that even if it infringed any valid rights, it shouldn't be liable. Google raises a variety of defenses, such as the praise former Sun CEO Jonathan Schwartz heaped on Android and the passage of time between the publication of Android and Oracle's lawsuit. I didn't find anything in the public record that suggests the judge is overly impressed with those kinds of defenses.

Google denies that what it was trying to negotiate with Sun was a license agreement of the kind that Oracle claims Google needs. Instead, Google says, it was about "a partnership role". But even in that context, Google admits that the deal would have given the parties access to some of each other's intellectual property.

Google disagrees with Oracle's damages claims and says that Oracle's demands are based on "a flawed econometric model and a rigged consumer study". Knowing that the judge wouldn't accept a zero damages theory (he said at a hearing that "zero is ridiculous" and not going to happen), Google says that "[i]f Oracle is entitled to any damages at all, that amount is closer to $50 million than billions".

Then Google again accuses Oracle of litigating instead of innovating:

"Much has been said and written about this lawsuit; but in the end, all Oracle will prove is that eleventh-hour threats and delayed litigation are no substitute for innovation."

The one thing Google absolutely doesn't want to happen is, not surprisingly, an injunction:

"F. No matter what, Oracle will not be entitled to an injunction.

Even if Oracle were to prevail on both its patent claims and copyright claim, it would not be entitled to injunctive relief."

Google points out that willful infringement -- which it calls a "fiction" in this case -- does not necessarily result in an injunction. The legal standard is eBay v. MercExchange, which reaffirmed what is called the four-factor test. However, the judge warned Google (at a hearing) against the implications of a willfulness finding for an injunction. Both are right at the same time. Willfulness is neither a substitute for nor a part of the four-factor test. Google is also right that the purpose of an injunction is not to punish an infringer. However, willfulness certainly has the potential to affect the analysis of at least some of the relevant criteria. For example, a willful infringer has a harder time making a public interest argument. And Oracle's primary argument for an injunction is, as I explained further above, the fragmentation of Java. (Google's argument against fragmentation is that there were other incompatible Java forks before Android.)

Google reiterates its suggestion that the question of willfulness be addressed only at the second trial (on damages), not at the first trial (the liability part). I already mentioned this in a recent post and said that a reexamination expert I talked to considers this a very valid suggestion.

This lawsuit isn't a cakewalk for Oracle, but I still believe Google should settle rather than take the risk of losing at trial.

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: