Thursday, January 26, 2012

Judge ups the ante for Oracle with respect to two patents asserted against Google

Against Google's objections, Judge William Alsup of the United States District Court for the Northern District of California gave Oracle a third chance to present a damages report, and Oracle accepted his proposal, which may (unless any new obstacles come up) pave the way for a mid-April trial date. The judge has meanwhile acknowledged Oracle's willingness to bear the costs associated with this incremental effort.

But he also had bad news for Oracle with respect to two of the six patents Oracle is asserting at this stage of the litigation (out of the original seven), in the form of a non-final supplemental claim construction order. It's non-final in the sense that the court reserves the right to modify it if any evidence shows up closer to trial that warrants second thoughts. "Counsel, however, may not ask for modification", writes the judge, knowing that Oracle would otherwise keep fighting for more favorable interpretations of one key term for each of the two patents the order relates to.

These are the two new claim constructions:

  • U.S. Patent No. 6,192,476 relates to a dynamic security method for determining access privileges. Oracle and Google couldn't agree on the meaning of the term "computer-readable medium" in the context of this patent. They had disagreements over that term and closely-related terms in connection with six patents, but since the interpretation of a phrase depends on the specifics of a given patent, the judge determined that "individualized attention to the intrinsic evidence of each patent" would be needed. Yesterday's order interprets the term only with respect to the '476 patent.

    Google has, for the time being, succeeded in persuading the judge that the term should be interpreted broadly so as to include signals over physical and wireless networks. If a defendant in a patent case advocates a broad claim construction, the agenda is, as it is in this case, invalidation: for a broad patent it's easier to find prior art, or to argue obviousness over prior art. Oracle, by contrast, wanted the term to be construed as "a storage device for use by a computer".

    It's now going to be quite difficult for Oracle to convince the jury that the '476 patent should be presumed valid. As I stated in my recent update on the ongoing reexaminations, the USPTO has issued a "final" (though still appealable) Office action declaring the only asserted claim (claim 14) invalid. The jury will be informed of that (appealable) decision. And if Judge Alsup adheres to the claim construction he ordered yesterday, the jury will be even more inclined to hold the patent invalid.

  • U.S. Patent No. 6,910,205 relates to increasing
    execution speed by replacing bytecode with instructions to access faster native code. With respect to this one, Google's strategy is to argue that it doesn't infringe, so it seeks a narrow interpretation. And it succeeded on this one, too: Judge Alsup agreed that "at runtime" must be confined to the execution of the virtual machine instructions, while Oracle wanted any execution of the virtual machine (including initialization) to be covered. The judge adopted a construction that is materially consistent with Google's proposal except for a difference that appears to be merely editorial.

    The '205 patent, like most of Oracle's patents-in-suit, is also under major reexamination pressure. A first Office action by the USPTO held the asserted claims invalid. A final (though appealable) Office action might still come down in time for the trial.

Last year the judge had already construed five terms from some of the patents-in-suit. At that time, Oracle won by 4-1. I discussed the first claim construction order in September when I looked at the state of Google's affirmative defenses at the time. On the second one, it's 2-0 for Google. That's 4-3 in total, with a number of additional constructions to be decided on. And again, yesterday's order need not be the final outcome, so a 2-0 victory for Google could, for example, still become a 1-1 draw.

When it comes to claim construction, defendants mostly try to narrow the scope of patents so as to avoid a finding of infringement. In the cases I watch, that's what they try about 4 out of 5 times, while they seek to broaden a claim only about 20% of the time. When a defendant seeks a narrow construction, it's sometimes a "Plan B" to argue that if his proposed narrow interpretation was rejected, the plaintiff's broader interpretation would raise validity issues.

But Google's defense against Oracle is different. Google has clearly placed more emphasis on invalidation than any other defendant in a comparable case that I've seen so far. And based on where things stand today, Google has made remarkable if not phenomenal headway with this part of its defense. Of course, fortunes can change and it's possible that at a later stage Oracle will resurrect some of its patent claims, but BPAI (Board of Patent Appeals and Interferences) appeals, followed by potential CAFC (Court of Appeals for the Federal Circuit) appeals, take time, and Oracle needs and wants -- and is now very likely to get -- a near-term trial.

Oracle knows that it's not going to be easy to prevail on its patent infringement claims. That's why its near-term hopes are mostly based on its copyright infringement allegations. However, the judge rejected Oracle's proposal to put the patent infringement claims on the back burner. He wants the case as a whole resolved swiftly and efficiently. Oracle is now, presumably, scrambling to put together a new damages report in order to stay on schedule.

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: