Tuesday, February 22, 2011

Oracle alleges Google derived code from specifications for hundreds of Oracle's copyrighted Java files

Just before Washington's Birthday weekend, U.S. District Judge William Alsup gave Google short shrift, ruling on Friday that he does not even want to give consideration at this stage to an envisioned motion for summary judgment against Oracle's copyright infringement allegations. Google had requested a leave (i.e., an opportunity to make such a filing) and claimed that those "[copyright] issues are ripe and can be decided without further discovery as a matter of law". But the District Court for the Northern District of California came down on Oracle's side. I'll quote and explain the two most important sentences:

"Having considered both parties' submissions, this order finds that good cause has not been shown to engage in a summary judgment battle at this time."

This means: Google has failed to convince the court that Oracle's copyright infringement claims lack substance to such an extent that there's no need for further scrutiny. Whether Oracle will ultimately prevail is another question, but there's no way that its allegations could be viewed as bogus claims at this early stage. No surprise here.

"Google's request is DENIED without prejudice to renewal after a more complete evidentiary record has been developed through discovery."

The second half of this means that the earliest point in time at which Google may get another bite at the apple will be when there's more evidence on the table as a result of continued discovery. While the wording could be interpreted as any additional production of relevant evidence potentially enabling such a motion, I guess this means in practical terms that Google will have to wait for completion of discovery, at least of the part of discovery that's related to those copyright infringement allegations. And even then it will be a long shot, considering that the judge sided with Oracle's position within hours of receiving it and that the decision comes across as a no-brainer.

Oracle claims tons of evidence (Java code found in hundreds of files)

Oracle's successful letter in opposition to Google's motion is, in my opinion, the single most interesting document in this whole lawsuit to date as far as the copyright part of the case is concerned (on the patent side, the detailed claim charts that fill approximately 400 pages are no less impressive).

Here are some quotes concerning the amount of evidence believes to have found so far:

"Google derived its Android code from the specifications for hundreds of Oracle’s copyrighted Java files."

In the court filing, "hundreds" is italics. I boldfaced it here because my quotes are all in italics, so I needed to set the word apart. And it really is a bold statement, figurately speaking.

"Oracle has identified fourteen registered copyrights--not two--that Google has infringed"

While copyrights can be registered in the US, registration isn't mandatory the way it works with, for example, patents. So the total number of allegedly infringed copyrights can be higher if non-registered ones are included. Also, a copyright registration can contain many source code files, so this number isn't inconsistent with the allegation that hundreds of Java files were exploited by Google.

"the demonstrable fact that at least eight Android source code files are actually decompiled Oracle object code"

If you read my blog post on evidence that I found (about a month ago) concerning this case, you may remember that I showed (among other things) seven source files that I juxtaposed to decompiled Oracle/Sun code. One of those files had been presented by Oracle in its amended complaint; the other six were found by me in an adjacent directory. Now Oracle arrived at the same conclusion concerning the genesis of those files.

Why do they find one more file -- 8 vs. 7? There are two possibilities. Either "No. 8" is in a directory I didn't look at (there were too many for me to look at all of them) or they mean a file named AclEnumerator (which I also saw and which looked like decompiled code to me, but which I didn't publish because I couldn't produce the same striking similarities that the other seven files showed).

In that blog post I had also mentioned a ZIP archive containing 37 files with Sun copyright notices. What Oracle's letter says doesn't indicate that those files have been processed in discovery so far. Maybe we'll find out much later. But the decompiled files were the more important ones anyway.

At the time there was some discussion over whether that code was actually shipped with any Android devices. I commented on that debate in a follow-up post: interestingly, some official source availability packages of major Android OEMs such as Samsung, LG and Motorola included material I had identified as presumably infringing.

The fact that Oracle successfully opposed Google's motion and the statements made by Oracle in that context certainly show that I had a point about the copyright part of the case. That said, the patent part is the one that Google must be more worried about. As I explained in a previous post, there are ways in which losing the copyright part could also affect Google's credibility with a view to the patent part.

Oracle expects to discover more evidence as the process carries on

Certain passages of Oracle's letter explain why Oracle believes that more evidence may still be found as discovery unfolds:

"Oracle’s copyright infringement contentions [...] represent only a snapshot of our case taken at the time they were submitted."

"Google has not yet produced all of its source code--and none of its proprietary code. Nor has Google produced the requested change log for its source code repository."

"Moreover, when we take depositions of Google’s developers, we anticipate uncovering the full scope of Google’s copying. Much of it may be disguised copying, which our source code comparison to date may not have uncovered."

"Some of Google’s Android developers previously worked for Sun or had access to proprietary Oracle materials. A plaintiff may prove copying by showing that a defendant had access to the copyrighted work and that the accused work is substantially similar to it. Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435, 1442 (9th Cir. 1994). 'Where a high degree of access is shown,' the Ninth Circuit 'require[s] a lower standard of proof of substantial similarity.' Swirsky v. Carey, 376 F.3d 841, 844 (9th Cir. 2004)."

Decision made it harder for Google to stall the process

Some believed that Google's attempt to have the copyright part of the suit thrown out was part of a strategy to stall.

Without the copyright part of the suit continuing (as it does), Google's recent requests for reexamination of some of the patents asserted by Oracle (by now, such requests have been put in with respect to five of those seven patents and might still be filed for the remaining two) could have served as an argument in favor of staying the case pending reexamination. But as Scott Daniels, the lawyer who first reported on those reexamination requests, pointed out, "reexamination could not resolve the copyright allegations", a fact that Scott considers one of a couple of reasons for which a Google motion to stay might not be granted.

While it may be true that Google has an interest in playing for time, I'm not a big believer in that strategy.

In my view, the best strategy in such a patent dispute is to countersue and cross-license, but Google's patent portfolio appears to be too weak for that.

Also, I believe that even if Google somehow managed to stall the current court case, Oracle could always step up the pressure in some other ways. For example, Oracle could file ITC complaints that might result in import bans against major Android device makers within 16 to 18 months.

Furthermore, Oracle could elect to sue Google in other jurisdictions, just like Apple and Nokia are duking it out not only in the US but also in three European countries. Some of the patents Oracle asserts against Google have Chinese, Japanese and European counterparts.

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: