Tuesday, June 28, 2011

Oracle says it claims $2.6 billion in damages from Google, denies $1.4-6.1 billion range

The dispute between Oracle and Google has several interesting aspects. Damages claims are one of them, and that's what this blog post is about. But if you're interested in this dispute, I strongly encourage you to read my previous blog post as well -- it provides more accurate information on the status of the reexaminations of Oracle's patents-in-suit than previous reports by others and explains the significance of those reexaminations, which could affect the timing and the outcome of a trial as well as the future enforcement of a ruling.

Two weeks ago, Google filed what is called a Daubert motion, asking the court to exclude Oracle's damages expert and his report from the further process. I reported on it and concluded that Oracle's expert had apparently calculated damages amounting to billions of dollars. A day later, Oracle confirmed this, without stating a particular number (just "billions of dollars"). A filing made by Google on June 6 became available on June 18 in an unredacted form. It averred that Oracle's damages claims were in the range "between 1.4 and 6.1 billion dollars".

Today Oracle has responded to the June 14 Daubert motion, which it obviously opposes. In that filing, which I have uploaded to Scribd (as a PDF), Oracle denies the range stated by Google and says the actual amount of its claims is $2.6 billion (first bullet point on page 8 of the PDF document):

"Google falsely claims that Prof. Cockburn concludes that Oracle is owed anywhere from $1.4 to $6.1 billion in damages. (Motion at 1, 16, 19). He does not. His opinion is that the total damages that should be awarded to Oracle is $2.6 billion."

A paragraph later, Oracle outlines the payment schedule:

"[D]amages would be split between a $0.9–$1.4 billion up-front payment plus a revenue share of between 10 and 15 percent of Google advertising revenues on Android phones"

That percentage contrasts with the 50% "even split" a previous Google filing suggested, and the "over 20 percent" claim revealed by a more recent Google filing.

Anyway, it's important to understand these two things about Oracle's damages claims:

  • None of the Daubert-related filings indicate that the purported numbers reflect a possible tripling of damages if Google is found to have infringed willfully (something that Oracle asserts). Oracle consistently asserts that Google "ultimately chose to willfully infringe Oracle's intellectual property and release the Java-based Android platform", but the methodologies presented don't mention a tripling of any amount.

  • The total amount of Oracle's damages claims presupposes that Google continues to infringe at least some of the asserted rights "through 2021". However, Oracle also asks the court for an injunction, which might (based on the current case schedule) be issued before the end of the year.

    If Google is not found to infringe, there will neither be damages nor an injunction. Oracle would simply lose.

    But in the event Google is found to infringe, there may or may not be an injunction. In my opinion, it's highly likely that an infringement would in this case also result in an injunction, but it's not certain until the decision has been handed, and that's why we need to look at two possibilities:

    • Injunction awarded:

      In this case, Oracle says "only past damages need be awarded, in the form of the up-front payment plus revenue sharing from 2008 through the date of trial". Damages would accordingly be less than the total $2.6 billion amount, but the effect of an injunction would be that Google would then have to stop distributing any infringing technology unless it negotiates a license deal with Oracle. In that scenario, Google might have to pay even more -- going forward -- than Oracle's damages expert calculated. It might also have to make technical changes to its software to bring Android at least to some degree into line with the official Java rules and specifications.

      I just described what might happen in a scenario in which Google would either have to abandon its application platform or accept whatever terms Oracle might dictate from a position of incredible legal strength and economic leverage. Google wouldn't be able to force Oracle to offer a license on any particular terms, not even the ones found in the damages report by Oracle's expert. With an injunction in its hands, Oracle would have an absolute right to say no.

      Google would face a similar situation if it negotiated a settlement with Oracle in a situation in which Oracle believes it has a very good chance of prevailing and obtaining an injunction.

    • Injunction denied:

      In this scenario, Oracle would have to tolerate continued infringement and the damages calculated by Oracle's expert would effectively be a running royalty that Google would have to pay for as long as it infringes Oracle's rights.

      If the parties negotiated a settlement in a situation in which Oracle is very unsure of its ability to obtain an injunction, the royalty on which the parties would agree would probably not exceed whatever numbers the Cockburn report proposes. It could even be substantially less in that case.

Those were the most important pieces of information found in Oracle's filing. There are also some other details, which I'll discuss in the remainder of this blog post, but prior to that I'll invite you once again to read my previous blog post on the ongoing reexaminations of Oracle's patents-in-suit and their potential impact.

Additional details on Oracle's damages theories

Controversy over the Cockburn report

Most of the filing has to be read against the background of what the Daubert debate is about: it's about whether Oracle's damages expert (Professor Cockburn) and his report will be admitted to trial. Google wants to prevent that, or at least wants to call Professor Cockburn's credibility into question and force Oracle to reduce its demands ahead of trial by way of a modified Cockburn report. Oracle wants to defend its expert, the entirety of his report, and every single dollar it claims.

Today's filing by Oracle often says that Google may not like the Cockburn report but doesn't present legally relevant reasons for excluding the expert and his report. Instead, Oracle argues, Google might write a rebuttal report and try to convince the jury that it's right and Oracle's wrong.

Oracle essentially argues that Google fails to prove any mistakes by Professor Cockburn that would make him unreliable and unworthy of presenting his views to the jury. It is possible to exclude experts and their reports, but not just because the other party disagrees with him.

Fragmentation of the Java platform

Previous Google filings indicated that the Cockburn report added a substantial amount of money to Oracle's royalty claims on the grounds of "fragmentation of the Java platform". Google said that this concept alone corresponded to hundreds of millions of dollars.

While Oracle now says that "the dollar amount that Prof. Cockburn currently assigns to fragmentation is zero", it simultaneously admits that "the adverse consequences of fragmentation are important in informing Prof. Cockburn's analysis of how a reasonable royalty would be structured".

Based on the information that is available (which is still limited but much less limited than before), this particular disagreement doesn't seem too important to me. They could both be right at the same time: Google may be right that this fragmentation theory effectively adds hundreds of millions of dollars while Oracle may be right that the Cockburn report doesn't state a specific amount. Maybe we'll find out at a later stage (such as when Google responds to today's filing) who's right.

Global vs. only U.S. revenues

Google's criticism of the Cockburn report relates in part to the allegation that it "improperly includes revenues from international sales" while the asserted patents are enforceable only in the United States. Google would want Oracle's damages report to exclude "extraterritorial -- and therefore noninfringing -- activities". According to Oracle's filing, the part to be excluded would be "up to 88% of Cockburn's damages".

Oracle argues that some of its "four different theories of patent liability in this case" would entitle it "to damages that would include foreign sales." I must admit that I don't like the idea of U.S. patent law effectively resulting in damages including sales in other parts of the world. But Oracle cites cases in which foreign sales fell within the scope of U.S. patent law even though the patents themselves were valid only in the U.S.

In connection with damages based on foreign sales, Oracle also mentions that it asserts copyright infringement. Copyright is not country-specific in the way patents are. But Oracle doesn't emphasize copyright too much in this context. I have believed all along that the copyright part of Oracle's assertion primarily serves the purpose of portraying (and, if successful, exposing) Google as a "copycat". From an economic point of view, this really is mostly abuot patents. I don't want to go into too much detail on the copyright part of the case here because in little more than a month we will see one or two summary judgment motions related to copyright. At that stage it there'll be a much better basis for a detailed analysis of Oracle's copyright infringement assertions.

Other findings

I found many other interesting pieces of information in Oracle's filing, including the attachments. For example, Oracle tries to get some mileage out of statements made by Google's leadership to financial analysts about how big and profitable its Android business is. Some of the issues addressed in Oracle's filing in opposition to Google's Daubert motion may play an important role in the further process, and I'll be sure to address them at the appropriate points in time.

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.

Share with other professionals via LinkedIn: