Thursday, May 12, 2011

Oracle vs. Google: judge proposes three-week patent trial in October, summary copyright decision in August

The US District Court for the Northern District of California is pressing on with Oracle vs. Google. In his latest filing the judge reaffirms his intentions to have this case resolved within a matter of months. A stay of the case pending reexaminations of Oracle's patents-in-suit by the USPTO now appears to be more unlikely than ever in this process. But it's not yet clear whether Oracle will accept to narrow its claims to the extent that this kind of schedule can work out.

The parties have until noon on next Wednesday (May 18, 2011) to reply to the judge's proposal. Those filings could be interesting, although it's also possible that neither party will oppose the judge's proposal aggressively even though Google would definitely prefer to stall the case if it could.

Under the tentative plan,

  • discovery would be completed in July (which would be consistent with the schedule that is already in place);

  • Oracle's copyright infringement allegations against Google would be decided by summary judgment (i.e., by the judge himself without being put before the jury) following a motion to be filed in August;

  • the judge wants the parties to commit to refrain from other summary judgment motions (such as on patent infringement issues);

  • and a three-week jury trial of "all issues" (I'll explain which ones) would take place in October.

By "all issues" the judge means whatever issues are left to be decided come October. At a minimum those outstanding issues would be the patent infringement allegations (including such questions as whether the patents are valid, and the related damage claims if they are deemed valid and infringed). If summary judgment is such that Google is deemed to have infringed Oracle's copyrights, the question of damages (what amount of money Google owes Oracle for such copyright infringement) would in my understanding also be put before the jury.

I have previously expressed and defended my belief that there is an infringement of Oracle copyrights. I don't know whether all of Oracle's infringement allegations are true, but I've seen enough to be sure that there's at least some infringement. I would really be very surprised to find the judge say that there's none at all. We may know the answer in a matter of months.

Possible implications of summary copyright judgment for the patent trial

I've repeatedly said that the potential economic impact of the patent issues is far greater than of the copyright-related claims, but I've also pointed out a risk that Google faces now under the proposed schedule: assuming the judge puts in a summary judgment that Google infringed some Oracle copyrights, the jury will look at the patent issues against the background of Google already having been found to be an infringer, or in other words, a "copycat". In psychological terms that would hurt Google significantly.

In the other event, if all of Oracle's copyright infringement claims were unexpectedly thrown out, Oracle would look like a party that claims rights even where it doesn't have any, but if they interpreted copyright too aggressively there could still be an infringement of patents. By contrast, if Google is considered to have "stolen" code, it gets a very unfavorable reputation concerning intellectual property.

In terms of that unfavorable "copycat" reputation, it wouldn't help Google too much to claim that any direct copying of code was limited and, as some Google apologists previously pointed out, the related files may have been used only for testing purposes. It's all about whether Google gets exposed as a company that fundamentally disregards other companies' intelletual property.

A summary judgment that Google copied some Java code without authorization would hurt Google not only in connection with the Oracle case but could also affect dozens of other Android-related intellectual property lawsuits. Any claims, such as by Apple, that Android is a "rip-off" might get a significant credibility boost.

Stay pending reexaminations apparently won't happen

The judge's tentative case plan also contains this important item:

"No stay pending re-examination, for Oracle will take risk that claims selected for trial will be cancelled or modified during re-examination."

Google argued vehemently for such a stay in one of its recent filings (as I reported in this recent blog post). It now seems that Google's related hopes have been thrashed.

Apparently, the judge is willing to resolve this case now on the basis of the assumption that patents granted by the USPTO are valid unless proven otherwise, instead of waiting three years or so for the USPTO to finalize its reexaminations.

Oracle will have to narrow its case and focus on a subset of the 132 originally asserted patent claims. Since there's no report on the Wednesday meeting, I don't know whether the number of patent claims to go on trial has already been agreed upon, but we will know soon and it's certain that the number won't be 132 (nor do I think it will be as low as 3, which was just a bargaining position the judge built last week).

The "risk" for Oracle that the judge mentioned in what I just quoted from the tentative case plan is that Oracle may pick particular patent claims to go on trial and those may be invalidated or narrowed a few years down the road as a result of reexaminations, while some of the claims Oracle chooses not to pick might be in better shape after reexaminations.

So what is the practical risk? It's that Oracle might have a ruling in its hand according to which Google was deemed to infringe certain patent claims deemed valid by the jury at trial time but if, for instance, those claims didn't survive reexamination, Google would be free to practice the claimed inventions since an invalid patent claim can't be infringed anymore. Oracle might then regret not having picked claims for trial that would, with the benefit of 20/20 hindsight, have been more useful. Whichever claims Oracle doesn't pick for trial in this case will not be assertable against the same accused products (only against new ones) in the future, so Oracle wouldn't be able to file a new lawsuit and then assert those claims that survived reexamination.

Why can Oracle apparently live with this? I think Oracle is probably quite confident that it can prove an infringement of a variety of patent claims and that it can either defend all of them in reexamination, or at least most of them. Also, all patent holders always have the risk of their patents being reexamined with consequences such as the ones described here. But at some point they want to get decisions, and I believe Oracle is probably confident that the benefits of a swift resolution of the current case far outweigh the potential future downside of not having known the outcome of reexaminations at the time of selecting the patent claims that will go on trial in the current case.

From a business point of view, if Google loses the current patent case (in whole or in part), it will have to work out a solution with Oracle because it wouldn't really be a viable option for Google to base its plans and its technology roadmaps on the assumption that reexamination will change the situation fundamentally within a few years.

Tentative schedule suits Oracle, but narrowing of claims could still be a poison pill

All in all, the judge's tentative case plan is pretty attractive to Oracle, and Google may now decide not to fight the judge over that schedule. But what's not known is whether Oracle will on the bottom line prefer to proceed on the basis of the schedule offered by the judge if the price to pay is a massive reduction of the number of patent claims to go on trial.

On Friday (May 6, 2011) Oracle and Google filed their replies to Judge Alsup's tentative scheduling order for narrowing the claims to go on trial. The judge then summoned the parties to his office for a private case management conference on Wednesday (May 11, 2011).

Since neither party requested the presence of a court reporter, there will never be an official record of that meeting other than the judge having subsequently filed a notice that it indeed took place. According to the filing, it lasted 30 minutes (as expected) and was attended by the parties' lead litigators (Michael Jacobs of Morrison & Foerster for Oracle, Scott Weingaertner of King & Spalding for Google) as well as one corporate representative per party (Oracle sent its patent counsel Matthew Sarboraria, Google dispatched its associate litigation counsel Renny Hwang).

While the content of the conversation won't become known, we will know soon what number of claims will go on trial, and whether Oracle's request to make that final determination only at a later stage of the process will be granted. For Oracle it would be quite valuable to have the opportunity to prove that it has a number of non-duplicative claims that it can rightfully assert. Setting a hard limit of X number of claims would be much harder for Oracle to accept unless that number suits Oracle's needs.

So it's still not certain that Oracle agrees to what the judge wants to do. By proposing the tentative case plan I discussed further above, the judge is basically dangling a carrot in front of Oracle, but the question is what price Oracle will be prepared to pay (in terms of dropping claims) in order to get that carrot. We should know soon.

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: