Wednesday, April 25, 2012

Oracle 'intends to assert [revived] patent in Phase Two', asks court to confirm -- Google opposes

While the current part of the Oracle v. Google trial is a copyright-only trial, the parties are in parallel having a dispute over whether or not a key Oracle java patent revived by the United States Patent and Trademark Office late last week has already been dismissed with prejudice. Late on Tuesday by local time, Oracle filed a "motion for clarification regarding [the] '702 patent", asking the court to confirm that it may, as it officially intends, assert this patent at the patent trial (Phase Two). A couple of hours later, Google filed its opposition, which was expected in light of what Google told reporters about the fact that the supposedly "final" rejection of the '702 patent wasn't truly final.

Oracle's motion is very respectful. Oracle explains the reasons for which its motion should be granted and it's always a given that the denial of a motion could be appealed, but this motion is not presumptuous. For example, Oracle notes that "the Court has characterized Oracle's conditional withdrawal differently at different times". This here is not about Oracle picking a fight with the court -- it's all about a need for clarification that has arisen from the USPTO's change of mind combined with a record that falls far short of a definitive withdrawal.

If there had been a straightforward, definitive withdrawal, Google would have pointed to it. Tellingly, after subtracting the title and signature pages of both pleadings (which in both cases don't contain any part of the actual arguments), Google's opposition brief is twice as long (four pages of argument) as Oracle's motion (two pages of argument). Those numbers are, in and of themselves, a clear indication that the legal situation here is not as trivial as Google's succinct statement to the media suggested. Regardless of which party one believes is right, at the very least this serves to show that Judge Alsup could, in his discretion, allow the assertion of the '702 patent.

Oracle's motion highlights the technical and economic (and, therefore, legal) relevance of the '702 patent:

"The evidence shows that Google has significantly benefited from its use of the '702 patent. Testing shows that Android application files are between 1.45 and 3.33 times smaller than they would be if the patented technology were not used, which results in a variety of additional performance benefits. [...] An injunction against Google's continued infringement is warranted, particularly because the '702 patent does not expire until October 2017."

In the following I'll juxtapose the parties' positions, firstly on how to interpret the pleadings and orders, and subsequently on the practical implications of adjusting to the fact that the presumed-final rejection of the '702 patent wasn't truly final.

Interpretation: Oracle consistently denies having dismissed the '702 patent -- Google claims Oracle's promise to withdraw was not "open to interpretation"

In my previous post on the '702 patent I already outlined my interpretation of Oracle's March 9 promise and the court's March 13 order. Basically, what happened is that the judge stated his "reliance upon" Oracle's promise, and Oracle's promise related to a future dismissal. That's why there's no such thing here as a statement by Oracle according to which it would actually have dismissed the patent. Nor is there anything in the record according to which the judge dismissed any patent against Oracle's will. Therefore, the primary question here is how to reasonably interpret Oracle's conditional withdrawal.

It's a little bit like contract interpretation. The judge and Oracle made a deal: early trial in exchange of dismissal of patents standing rejected. Judge Alsup's acceptance of Oracle's offer started with the following words:

"In reliance on Oracle's withdrawal with prejudice of the '720, '205, and '702 patents, given the final rejections by the PTO examiner, and [...]"

Having looked at this sentence again I now find that the passage "given the final rejections by the PTO examiner" is yet another reason to admit the '702 patent at this stage. The judge assumed that those rejections were final. Now one of them turns out not to have been final. Had the judge known at the time that it wasn't final, he might not even have listed the '702 patent in this context. Why should the court not have the benefit of better information now, in the interests of justice? I think it comes down to questions of prejudice to the parties and case management considerations. The '702 patent can be asserted if there's a will to assert it. Google naturally doesn't have that will to make it happen -- everything else would constitute malpractice by its lawyers.

Oracle's motion also quotes a court order that I didn't address in my previous post on this issue. On March 15, the judge wrote:

"Another three [patents] rejected by the PTO examiner were withdrawn if the trial is held before the administrative appeals are completed, a withdrawal whose effect will be considered below. Therefore, there is a strong possibility that only the '104 and '520 patents will be asserted at trial, and this order will only address issues pertaining to these two patents, without prejudice to revisiting objections specific to the withdrawn patents if they later arise."

While the first sentence, if taken out of context, could be seen as reflecting the judge's understanding that the patents were dropped, the final part of the second sentence opens the door to "revisiting objections [...] later". Again, it all comes down to what Oracle actually promised and the judge relied upon.

The briefs also show how the parties acted after that conditional withdrawal. Here's a quote from Oracle's brief:

"Both Oracle and Google kept their '702-related exhibits on the trial exhibit list. (When considering whether and what exhibits could be dropped, Google asked if Oracle had indeed dropped the '702 patent with prejudice. Oracle said 'no.')"

This doesn't answer the question of whether the beginning of the copyright trial or the beginning of the patent trial (or a time ahead of the patent trial that gives both parties enough of an opportunity to prepare) should be the latest point in time at which a reconsideration by the patent office can be considered. But it confirms that Oracle never dismissed the '702 patent with immediate effect.

There were some case management (streamlining) discussions in which both parties simply referred to the patent part of the case as a two-patent case. Google notes that even Oracle at some point made reference to "a trial on two software patents". I wouldn't attach too much importance to reference to a two-patent trial because all those pleadings have page limits, and if a party doesn't make a distinction between a patent trial involving a minimum of two patents versus on involving exactly two patents, it still doesn't state a definitive withdrawal -- it just avoids a more cumbersome way to state its position.

Google's opposition brief states that Oracle very much wanted a quick trial and was willing to make concessions. That's true, but the question is just how those concessions can be reasonably interpreted, and how any interpretation can be reasonably applied.

Furthermore, Google recalls that Oracle once made a proposal to postpone the patent part of the case and that the judge rejected this, calling it a "piecemeal approach". This, too, is true per se, but once again the actual question is whether it's reasonably doable to assert the '702 patent now that the court and the parties know the "final" rejection wasn't final -- the patent office itself recognized it as an error. That's different from Oracle's original proposal of revisiting the patent issues in the fall (at the earliest).

Practical implications of an assertion of the '702 patent at trial

When there's at least room for interpretation, a decision can depend, in part, on practical considerations. This is also a question of prejudice to Google. The prejudice to Oracle in a scenario of not getting to assert a patent that wasn't truly rejected on a final basis is clear, and no one can deny that this kind of prejudice (also in light of the importance of the '702 patent that I mentioned further above) far exceeds any prejudice to Google. But Judge Alsup has repeatedly shown that he wants this case to be managed efficiently, and chances are that case management considerations will again be relevant to him -- though they need not, and in my personal opinion should not, tilt the scales in the end. Also, Judge Alsup more than once noted that the parties and their external law firms have substantial resources. Several of the reporters observing the trial have made note of the "armies of lawyers" involved. The judge knows that resource constraints won't prevent the parties from having everything in place for a patent trial involving the '702 patent.

Google points out that it has been preparing for a two-patent trial, and that the admission of the '702 patent "would require a revised report from [court-appointed damages expert] Dr. Kearl and potentially further expert discovery". Given that damages will be discussed only in Phase Three, this is a weak point, and that's why Google quickly says that "leaving aside the prejudice to Google in upending what has been the established scope of the case, there are other concerns with the proof that Google could present in the imminent patent phase of this trial". This is sort of a prejudice argument, but focused on issues involving evidence. Google raises the following issues:

  1. "[...] Google did not subpoena Nedim Fresko, the named inventor of the ’702 patent, and has no idea whether he would be available to testify at trial. Mr. Fresko is a former Sun employee, and not under the control of either party to the case."

  2. "[...] Google's invalidity expert on the ’702 patent relied on Oracle’s withdrawal in setting his own schedule for the coming weeks". Google lists some his commitments during the next several weeks and the fact he has to prepare an expert report in another case. "With all of that work filling his schedule, Google would not be able to prepare Dr. Levine for trial testimony in this case." But Oracle doesn't believe there will be a need for any invalidity discussion. Google raised the same invalidity contentions in its reexamination request that it would present a trial, and in connection with the other affirmed patent, the '520 patent, Judge Alsup already suggested, as per Oracle's request, that Google withdraw those, as it then did. If the judge told Google that it shouldn't complicate the process now with an insistence on failed, doomed-to-fail-once-more, invalidity contentions, then I think this point here is going to be moot -- unless Google wants to upset the judge at this stage.

  3. "Oracle has made statements during prosecution of the reexamination, and the Examiner has responded with a Statement of Reasons for Patentability. This additional information, now part of the intrinsic record of the patent prosecution, will be relevant to the interpretation of the claims and affect how, if at all, it can claim infringement of those claims." I concur with Google on relevance (I'm not saying that the infringement analysis will change or have a different outcome, but yes, Google needs to look at whether anything outcome-determinative has changed). However, I think Google draws the wrong conclusion from this by claiming that "[i]t would not be fair to require Google to analyze the statements made in prosecution and, in the middle of trial, adjust its case [...]". There's a sufficient amount of time before the start of the patent trial that this can be done, especially thanks to both parties' vast resources.

The judge usually spends some time with the parties' counsel to discuss case management and substantive issues at times when the jury has a break or has gone home. I guess the '702 patent will come up later today.

[Update] My Twitter feed from the courtroom shows that the matter was discussed early today. If the '702 patent was added, this would take the number of patent claims to be tried up to 18. Oracle indicated its willingness to drop claims from other patents in order to make room for this one. Judge Alsup said he'd think about this motion. [/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: