Wednesday, May 16, 2012

Scope of damages phase of Oracle-Google trial will likely depend on copyrightability decision

[Update] The parties have just reached an agreement and put it in writing (a stipulation) that I'll discuss in a new post. [/Update]

In this blog post (published on Sunday), I already expressed my opinion that Oracle's insistence on disgorgement of infringer's profit with respect to the rangeCheck function and eight decompiled files makes sense if one considers, beyond the current state of the infringement findings, the possibility of an appeal and/or retrial and the necessity to determine remedies for Google's infringement of the 37 asserted Java APIs. In that scenario (which presupposes a finding of copyrightability and a unanimous vote by a future jury against Google's "fair use" defense or a judgment as a matter of law to that effect), the parties will have to deal with the disgorgement question anyway. Now that the court is holding a trial and that the parties have prepared their evidence and their arguments, it would be relatively easy to talk about which costs Google wants to deduct from its Android-related gross revenues prior to a disgorgement of copyright infringer's profits.

Judge Alsup tried hard to dissuade Oracle from insisting on disgorgement with respect to the infringements for which Google is liable so far, and Google brought a motion that would preclude Oracle from pursuing disgorgement for the time being. But it appears that Oracle actually is entitled to disgorgement even based on smaller infringement findings, and by showing Google's Android-related gross revenues, Oracle can shift the burden of proof and require Google to prove its deductions. Google argues that Oracle is entitled to statutory damages only.

Neither the case law nor the relevant numbers are on Google's side. Courts have ruled in some cases that the burden of proof had not shifted to certain defendants, but in each of those cases the owner of the infringed copyrights just proposed an unacceptably broad revenue base despite having more reasonable options. In Oracle v. Google, the gross revenues proposed by Oracle as the starting point are as narrow as they could be under the circumstances. There can't be an argument that there's a smaller identifiable revenue base. The judge has said that Oracle won't get hundreds of millions of dollars for rangeCheck and the decompiled files, but that's not the issue: Oracle's counsel confirmed that he understands (and he also said he knows that electing actual damages and disgorgement, instead of statutory damages, could result in less, or possibly zero, on those counts.

Google's Android business is so huge now that even a relatively small percentage of its revenues that might be attributed to those kinds of infringements would dwarf the maximum amount of statutory damages ($150,000).

It appears to me that the judge can't help Google with respect to the burden of proof for deductions. Google argues that Oracle has failed to show causation, and the judge basically argues that the theoretical outcome of a disgorgement analysis would be unreasonably high, but under the law, infringed copyrighted material either is de minimis (too small to matter) or it's not. In this case, the jury found that rangeCheck is not de minimis, and the judge didn't overrule it. The judge did overrule the jury on the eight decompiled files. No matter how much the judge and Google would like to create a separate de minimis standard for disgorgement than for infringement, the law doesn't seem to provide for it. Oracle cites some interesting cases according to which the unlicensed use of a tiny amount of copyrighted material (ten notes) in a commercial required a car maker to prove deductions from the sales of that car in the United States during the launch campaign.

Within the U.S. copyright system, disgorgement of infringer's profits is a very important concept because it's meant to ensure that infringement doesn't pay off. Without the concept of disgorgement, there would be scenarios in which an infringer gets away with tiny amounts of damages compared to the benefits he derives from infringement. It's a fundamental question of law and order to disincentivize violations.

Yesterday, Oracle counsel David Boies offered a solution to the court that makes me believe even more strongly in my original theory (mentioned in the first paragraph of this post) that what Oracle described as a "principle" does have some connection with the "37 APIs" issue. Law.com, ZDNet, Wired and The Verge report that Oracle would be prepared to drop disgorgement with respect to rangeCheck and the eight decompiled files in the event Judge Alsup rules against the copyrightability of the structure, sequence and organization (SSO) of the 37 APIs, but if the SSO of the APIs are deemed copyrightable, Oracle will still pursue disgorgement.

This is consistent with Oracle's proposal to postpone the damages phase of the trial until all liability questions, including the "fair use" defense to the infringement of the API SSO, have been adjudicated. So far, it appeared that Judge Alsup wanted to go ahead with Phase Three (damages and disgorgement) immediately after Phase Two (patent liability). Phase Two could theoretically end any moment now: yesterday, the parties made their closing presentations on patent infringement (I will comment on that part later), and the jury commenced its related deliberations. I believe Phase Two will take a few days and is very likely to add additional infringement findings to the list, but whatever comes out of Phase Two is unrelated to the disgorgement issue.

By insisting on the disgorgement analysis if the API SSO are deemed copyrightable, Oracle underscores what a major inefficiency it would be to hold the damages phase now rather than later -- but even if the damages phase took place immediately, it would at least be more productive (with a view to the future, even if not from a very short-term perspective) to clarify some of the general disgorgement issues now.

At any rate, there can be no doubt about the following: Oracle isn't trying to get disproportionate compensation for infringements of limited scope and scale. It just has a consistent position on its copyright damages claim.

Even if Judge Alsup held the API SSO non-copyrightable, the Federal Circuit, applying Ninth Circuit law, may very well decide otherwise. In that case, some of the issues (such as "fair use") will be remanded to the district court. And if Judge Alsup holds the API SSO copyrightable, Google will certainly appeal -- absent a settlement. It may take a while before we know whether Google is liable for the API SSO (infringement has already been identified, but "fair use" needs to be decided one way or the other). The fastest way to get there would be for Judge Alsup to rule that the API SSO are copyrightable, to postpone the damages phase, and to hold a retrial at the earliest opportunity.

One of the reasons for which I had waited since Monday morning before doing another post on Oracle v. Google is that I hoped the judge would rule, or speak out, on some of the outstanding questions. There's some more clarity now. Judge Alsup appeared to consider Oracle's proposal to make disgorgement dependent on the API copyrightability question at reasonable and encouraged the parties to work out a written agreement on this. At the time this post goes live, the parties's lawyers are discussing this in court and trying to work out a deal on this basis. I will update this post as soon as there's more clarity concerning the way forward.

[Update] The parties reached an agreement and put it in writing (a stipulation). I discussed that stipulation and order in a new post. [/Update]

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: