Thursday, September 29, 2011

Oracle wants jury in Google lawsuit to look at 26 patent claims: 15 distinct ones plus 11 'mirrors'

Tomorrow (Friday, September 30) there will be a third and presumably last pre-trial mediation talk between Oracle and Google in court. I'm not too optimistic about that one, but I think we may see a settlement shortly before trial. For now, a trial is scheduled on a purely tentative basis to start on Halloween.

Google's proposed acquisition of Motorola Mobility is now under in-depth regulatory scrutiny, and I think Google probably won't settle as long as it hopes to obtain clearance for that acquisition prior to a final decision in the Oracle case. But if the Halloween trial date was confirmed definitively, Oracle would be reasonably likely to obtain an injunction well ahead of the conclusion of the merger review, and then Google may realize that it has to settle to prevent the worst.

Meanwhile, preparations for a trial -- even though its date is uncertain -- progress unabatedly. Oracle just filed a case management statement requested by Judge Alsup, the federal judge presiding over this case. The judge wanted Oracle to specify which patent claims (a patent can consist of multiple claims, and software patents typically consist of dozens of claims) it still wants to assert at trial.

The tricky thing for Oracle is that if it insisted on putting too many patent claims before the jury for the judge's taste, the judge might use his discretion to stay the case in whole or in part, given that the ongoing reexaminations of Oracle's patents-in-suit by the USPTO could lead to the invalidation of some claims (so far there have only been preliminary positions taken by the USPTO, which don't count), which would make the case more efficient (the jury wouldn't have to waste time on claims that aren't valid). But if Oracle asserted too few just in order to maximize its chances of a near-term trial, it would greatly increase the risk of losing the case just because a small number of asserted claims leads to a result that is much less predictable. Oracle wouldn't be able to use those claims in the same context (meaning: against Google in connection with materially the same infringement allegation) in a subsequent lawsuit. Any claims that Oracle drops now it drops forever with respect to the allegations at issue in this litigation.

Oracle's narrowing of claims doesn't mean that there's anything wrong with those claims. It's all about case management: the judge wants to make sure the trial isn't overly burdensome on the court and the jury, so Oracle has to focus. Let's put today's proposal of 26 claims (effectively 15 different ones plus 11 mirrors into the context of how these numbers of claims evolved.

Oracle's reduction of asserted patent claims from 132 to (now) 26

In its original complaint, Oracle asserted seven patents without listing the asserted claims. A few months later, when Oracle presented infringement claim charts (tables specifying its patent infringement contentions), it asserted 132 claims from those 7 patents. In this blog post I published a list of those 132 claims along with their reexamination status (as of a few months ago).

In the spring, there was some misinformation spread by propagandists and and picked up by various media, suggesting that Oracle had to drop 129, or 98%, of those 132 claims. In this blog post I commented on that misinformation and referred to earlier posts in which I explained on this blog that there would be a reduction, which is normal, but hardly to 3 claims, which was only an idea tossed out by the judge -- a tentative order to start a discussion, not a definitive one.

In mid-May, Oracle and Google filed a joint memorandum on case management. They agreed on some items and disagreed on others. They failed to agree on the number of claims to be asserted at trial. Google proposed a "compromise" of 10-14 patent claims, but Oracle insisted on 21. What they did agree upon included an intermediate step of Oracle narrowing its claims from 132 to 50 -- basically creating a shortlist from which to select the ones to be asserted at trial -- by June 1, 2011.

That shortlist wasn't made public right away. But it showed up in a filing in July, and at the time I published a table jointly submitted by Oracle and Google to the court, reporting the reexamination status of the claims from the shortlist of 50 claims.

A mid-August filing by Google indicated that in the meantime there had been another reduction of the shortlist from 50 to 41 claims.

Today, Oracle proposed to assert 26 patent claims, but selected in a way that some can be looked at by the jury as a group, which is why Oracle says that the "effective number of claims to be tried" comes down to only 15, which "improves upon [Oracle's compromise proposal of 21 claims] by more than 25%, ensuring that the case is ready for trial".

Oracle would be happy to put just 15 distinct claims before the jury if Google accepts that if one claims of a given technical scope is infringed, its "mirrors" are also infringed. By "mirrors" Oracle means claims that repeat the language of another claim except that the original claim might, for example, be a "method" claim, while one mirror claim is an "apparatus" claim and another one possibly a "computer-readable medium" claim.

Oracle also drops one patent -- the '447 patent -- entirely.

Here's Oracle's list of claims to be asserted at trial:

"Applying the mirrored claim approach, Oracle proposes to try the following 26 claims from 6 (of the originally 7) asserted patents:

'104 patent – 3 effective claims: Claims 11, 27, and 40, and mirrored Claim 29 (a mirror of Claim 27) and Claims 39 and 41 (mirrors of Claim 40).

'205 patent – 2 claims: Claims 1 and 2.

'702 patent – 3 effective claims: Claims 13, 15, and 16, and mirrored Claims 1 and 7 (mirrors of Claim 13) and Claims 6 and 12 (mirrors of Claim 15).

'520 patent – 3 effective claims: Claims 1, 8, and 12, and mirrored Claim 20 (a mirror of Claim 8).

'720 patent – 3 effective claims: Claims 1, 6, and 21, and mirrored Claims 10 and 19
(mirrors of claim 1) and Claim 22 (a mirror of Claim 21).

'476 patent – 1 claim: Claim 14."

By continuing to insist on a significant number of claims to be tried, Oracle sends out an unequivocal signal of strength on the day before the third mediation conference. The judge had once proposed three claims on a tentative basis and has repeatedly threatened to stay the case unless Oracle greatly reduces the number of claims it seeks to assert. If Oracle was under pressure to take this case to trial very quickly, it would probably have gone down to 6 or 7 claims. But by standing pretty firm on the number of claims to be asserted, Oracle shows that it isn't worried about a stay. Oracle is clearly more interested in a favorable outcome than in a quick resolution of the case.

It's possible that the judge agrees with Oracle's logic of "grouping by technical subject matter" and that he welcomes this reduction. There could still be some more bargaining ahead of a trial. But the immediate effect is that Google has to negotiate with a plaintiff who's determined to win, even if it still takes another year or two.

Part of Oracle's consideration could also be that besides the patent part of the case it has a pretty good chance of success with its copyright infringement claims. Those claims wouldn't be affected by a partial stay of the case since copyrights can't be reexamined by the patent office. Just winning on the copyright count might already give Oracle enough leverage to force Google to settle.

There are complicated tactical considerations behind all of this, and for observers like me it's difficult to figure out what the parties want since only parts of the information enter the public record. However, the public filings provide an indication as to the judge's agenda and the strategies of the parties.

Google is under immense pressure. Samsung's decision to pay patent royalties on Android to Microsoft shows that Android has serious intellectual property infringement issues, and Oracle is another major industry player who wants to make money on each and every Android device sold. Apple is also enforcing its rights (more aggressively than anyone else, in fact), and there are numerous smaller patent holders (smaller operating companies as well as non-practicing entities) lining up to be paid. Google's "free" Android strategy worked very well in the past but isn't sustainable, which is why Google now wants to acquire and become a device maker. Oracle's royalty demands play an important role in this development, too.

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: