Yesterday Google filed its motion for summary judgment on count VIII -- the copyright part -- of Oracle's amended complaint. Google wants to have all copyright infringement allegations dismissed before the patent infringement claims in this case go to trial (currently scheduled to begin on Halloween).
Google's motion proposes September 15 as the hearing date on this issue. According to a schedule on which the parties agreed in May (and which the judge approved), Oracle now has until August 19 for its opposition brief. Google will then have until August 29 to file a reply defending its motion.
Oracle could have asked the court for permission to file its own summary judgment motion on copyright. However, Oracle didn't even ask (at least not formally; maybe they asked the judge in private and he discouraged such a motion).
These are the possible outcomes and their implications:
If Google's motion succeeds in its entirety, no copyright infringement claim will be put before the jury if this case goes to trial. This would turn the original patent and copyright case into a patent-only case. Getting the copyright assertions stricken would increase the likelihood of a partial or complete stay of the case in the event that the judge and Oracle cannot agree on how many patent claims the jury should evaluate. It would also avoid an impression of Google being a copycat -- an impression that would make the jury more likely to believe that Google actually (and, in particular, willfully) infringed some of Oracle's patents. Even though the jury is likely going to draw the obvious conclusion from some of the other evidence, such as internal Google emails that strongly suggest willful infringement, the first-hand impression of copying would probably influence jurors in Oracle's favor.
Google very much wants this motion to be granted and has gone to extreme lengths in putting together this motion. The motion comes with three supporting declarations, each of which has many documents attached. All in all, this summary judgment request has a volume of several hundred pages. A particularly long one of all those documents is a 153-page "opening expert report" by Duke University Professor Owen Astrachan.
If Google's motion is denied, the case continues to be a patent and copyright dispute, and the jury will look at Oracle's accusations of copying. But by having its motion denied, Google wouldn't lose much other than the opportunity described in the previous bullet point. A denial of the motion doesn't necessarily mean that Google is a copyright infringer: the judge might just conclude that the reasoning presented by Google doesn't warrant summary judgment (no matter how likely or not a jury may be to agree with Google).
The legal basis for summary judgment is that there's only an issue of law for the judge to decide, not a dispute of fact that must be put before the trier of fact. Summary judgment is possible if a judge can take a decision "as a matter of law" because there wouldn't be a different outcome even if the party opposing the motion was right with all of its factual representations.
The motion could also be granted in part and denied in part. Such a mixed result wouldn't be too surprising. The copyright infringement allegations at issue here can really be categorized into two groups, and the judge might throw out one of them and keep the other one alive.
This isn't Google's first attempt to have Oracle's copyright infringement assertions thrown out. Almost six months ago, Google already asked for permission to file such a motion and stated the following:
"These [copyright] issues are ripe and can be decided without further discovery as a matter of law, thereby simplifying the issues and conserving resources."
The judge denied that request and said that discovery should continue before Google tries again to tackle those copyright claims on a summary judgment basis.
Also, on October 4, 2010, Google filed a "motion to dismiss claim VIII of [Oracle's original] complaint or, in the alternative, for a more definitive statement". Judge Alsup agreed with Google that the original complaint wasn't sufficiently specific on the copyright side and gave Oracle an opportunity to amend it accordingly. That was the "alternative" proposed by Google.
Now that the procedural context of this motion has been outlined, I'd like to discuss the substantive question. In order to understand the further process, it's essential to make a clear distinction between the two kinds of copyright infringement allegations at issue.
Two kinds of infringement allegations
Oracle alleges that Google infringed Java-related copyrights because it
uses (significant parts of) the allegedly copyrighted Java Application Programming Interface (API) in terms of interface specifications found (according to Google's representation of Oracle's position) in thirty-seven Java language API packages in Android, and
copied 12 Java source files (including the eight files I already listed in January and identified not only in the Android code base but also in the official source availability packages of some major Android device makers)
Those two categories of allegations have fundamentally different characteristics. The second group is a very simple and clear case of unauthorized copying and distribution of material that I consider to be undoubtedly copyrightable. However, Google has apparently already fixed that problem and replaced or removed those files, so this isn't the kind of leverage with which Oracle could stop Google from distributing Android going forward. The indirect effect of the impression this makes on the jury (as mentioned further above) would be the only noteworthy benefit here. By contrast, if Oracle succeeded with the first group of allegations, Google and the entire Android ecosystem would be in fairly serious trouble, but it's far harder to win with that group than with the second one.
Google's argument against group 1: not copyrightable, or otherwise fair use
Google argues that the API information it used is purely functional and dictated by the need for Java compatibility, and therefore not protected by copyright.
Even though I've read the motion in detail and taken a close look at the declarations and their numerous attachments, it's not exactly clear to me from Google's filings what Oracle's position on copyrightability is. That's why I prefer not to take a position on this issue until I've seen Oracle's opposition brief, which is due on the 19th.
Apart from who's right or wrong on this, I think Google's lawyers had a pretty clever idea for how to convince the judge of the idea that function names shouldn't be copyrightable. They presented tables that show how Java itself actually uses keywords that other programming languages (like C and C++) used before it, and they also show how the spreadsheet program that's part of OpenOffice.org and StarOffice uses function names that are in many instances identical or at least very similar to those used in Excel.
Google's argument against group 2: any copying is de minimis
This is what the introductory part of Google's motion has to say about group 2:
"[...] minimal alleged copying, concerning only 12 files out of over 50 thousand in Android, and material of no qualitative significance. A reasonable jury could only conclude that any such similarities are de minimis and thus not actionable.
I personally have a problem with any de minimis (too small to be legally relevant) argument in this respect. Google argues that those files represent just a tiny part of the overall Java codebase, "comprising less than 750 lines of comments and code out of 2.8 million lines of code in the Asserted Works as a whole, and 12 files out of more than 9,500 files in the Asserted Works as a whole."
Google also highlights (in a different part of the document) the size of Android's codebase as a whole: "Android contains over 50 thousand files and over 11 million lines of code, and is built on an underlying Linux kernel."
In my view, the really important question is whether those "less than 750 lines" are copyrightable. I've seen a large part of them and I believe they should be. In that case, copying such a quantity of protected material should not be protected by a de minimis rule.
If one agreed with Google, it would be possible to steal similar portions of code from any large program, while stealing the same code from a smaller program would not be allowed. That doesn't make sense to me. And I would be even less willing to accept the notion that a company with a large codebase can just incorporate copyrightable parts of other people's code because the stolen code is such a little part of the larger codebase into which it's incorporated.
Google also argues that those directly copied files weren't really important -- most of them just served test purposes. But at the time when Oracle filed its original lawsuit, there was an infringement, and that's why Oracle is -- in my view -- entitled to a decision on this issue, even if the economic consequences of this may not be huge.
In two-and-a-half weeks we'll see the reasoning with which Oracle will oppose Google's motion.
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: