Thursday, April 14, 2016

Oracle fundamentally disagrees with Judge Alsup on how to instruct jury on fair use in Google case

On a recent occasion, Judge Alsup has correctly stated what the Federal Circuit ruled to be copyright-protected in this case (unlike in an order earlier in the build-up to the May retrial). But when I read his draft jury instructions on fair use for the upcoming Android-Java copyright retrial, I couldn't believe my eyes because he basically portrayed the "fair use" determination as if copyright was anti-innovative and "fair use" was the way and the light. I'm now even more of a Judge Alsup skeptic than before, and depending on how this case continues, "skeptic" may soon be a gross understatement.

Oracle has just filed a document that makes it very clear that it deems those "fair use" jury instructions to be fundamentally flawed (to put it diplomatically) and between the lines it indicates that Judge Alsup only has two choices: he can either rephrase his instructions so as to be consistent with the Federal Circuit opinion (and mandate) or he can have another appeal, with a high likelihood of him ending up being the loser of the appeal once again. And if another one of his key decisions had to be overruled in this case, it could become a major embarrassment.

When judges ask for "critique" of a draft jury instruction, they expect and get extremely respectful responses. Oracle's response in this case is not disrespectful, but it voices such a fundamental disagreement with the judge that it's clear there's no love lost between them (this post continues below the document):

16-04-14 Oracle Response Re. Fair Use by Florian Mueller

I'm now just going to quote, and comment on, some passages of Oracle's filing that show just how much the Java copyright holder disagrees with the judge:

  • "The Court's instructions do not accurately and evenhandedly state the law on fair use."

    COMMENT: I don't want to read too much into that, still I can't help but interpret "not ... evenhandedly" as a way of saying "Your Honor, you want Google to get away with what it's done and short of saying so, you couldn't have made it any more obvious than with those proposed jury instructions."

  • "Taken as a whole, the effect of the Court's proposed instructions would be to eliminate the exclusive right to prepare derivative works (or to authorize others to do so)."

    COMMENT: Just like Judge Alsup's non-copyrightability ruling threatened to "vitiate" (that term showed up in the amicus brief of a former U.S. copyright chief) software copyright, Oracle now claims that Judge Alsup's misportrayal of "fair use" law would effectively mean that anyone who creates a derivative work is highly likely to benefit from the fair use exception. Oracle's lawyers' use of "to eliminate" is typical hyperbole and I wouldn't go that far. It's not like those jury instructions would make it absolutely impossible or even highly unlikely for Oracle to prevail in a jury trial, but Oracle would indeed face an uphill battle in a situation in which, also in light of the Federal Circuit opinion, it should normally be on the winning track.

  • "If the instructions are given in the proposed form, they would violate the Federal Circuit's mandate and constitute reversible error."

    COMMENT: This announcement of an appeal sounds so strong that I wouldn't even be surprised if Oracle tried to avoid a jury trial based on unacceptable jury instructions and take this to the appeals court as soon as possible. If I were in Oracle's shoes, I would certainly see no point in going into a second trial where the court's jury instructions on "fair use" are a huge issue. Maybe Judge Alsup thinks that this may up the pressure on Oracle to settle but that won't happen. It just won't happen unless Google makes Oracle an offer that is too good to refuse. What's going to happen is that Oracle will, if necessary, litigate for a few more years, or even many more years.

  • "The [introductory] instruction is too narrow and one-sided in favor of Google in characterizing copyright as protecting against 'plagiarism' and fair use as progress ('development of new ideas that build on earlier ones'). "

    COMMENT: Here, Oracle comments on the most outrageous part of various problematic parts of the proposed jury instructions. You may wonder why I haven't published Judge Alsup's proposed instructions but that's because I don't think anyone would do the world a favor by publishing them: potential confusion of whomever may read them clearly outweighs any potential benefits.

  • "The Court's statement of the 'policy' of fair use [...] is not accurate, contravenes the legislative history, and is unsupported. Copyright protection (not just fair use) 'promotes' progress."

    COMMENT: How can you critize a judge any more harshly than by saying (more diplomatically than how I'm now going to paraphrase it) that he's wrong on the law (this reminds me of a Federal Circuit judge wondering about just how much Judge Alsup got confused last time around), he's disrespecting lawmakers, and he's writing up things that are baseless?

  • "The instruction on commercial use should direct a finding in Oracle's favor that Google's use is 'purely' or 'entirely' commercial. The Federal Circuit found that Google copied 'for what were purely commercial purposes.' [...] Counsel for Google admitted at oral argument that Google's purpose was 'entirely commercial.'"

    COMMENT: I expected Oracle to demand that the jury instructions, beyond merely stating the law and the Federal Circuit decision on copyrightability, contain some clear findings. It's one of the (many) things I missed when I read those draft instructions.

  • "The Court's proposed definition of transformative use is incorrect. It is inaccurate to instruct that '[a] new use is transformative if it is productive ….'"

    COMMENT: In my observation Judge Alsup's draft instructions describe "fair use" in a way that would make the "fair use" exception almost seem to be a rule on an equal footing with copyright itself as far as software is concerned.

  • "It is also incorrect and confusing to refer to transformative use as any use 'adding value.' [...] It is not about 'added value.' A 'distinct' purpose is critical."

    COMMENT: Oracle is right on this one: just think of a case like Campbell, where a parody of a music song also changed the music style. The "distinct" purpose is a reasonably high hurdle, and Judge Alsup's proposed instructions make it appear a much lower one than it actually is.

  • "Additionally, 'little more than plagiarism' [...] is not the test for what is not transformative. Transformation requires 'real, substantial' modification of the original work [...]"

    COMMENT: I already disagreed at the time of the 2012 trial with how Judge Alsup described the concept of transformation to the jury, and things have only gotten worse since then...

  • "[R]eferences to SSO [...] must be accompanied by reference to the declaring code, otherwise the instruction misleadingly suggests that declaring code is not protected when the Federal Circuit 'conclude[d]' that both 'are entitled to copyright protection,' [...], and all agree Google copied the declaring code and the SSO."

    COMMENT: On this one, let me refer you once again to the posting I already linked to at the start of this post. The Federal Circuit indeed held both the declaring code and the SSO copyrightable, and Judge Alsup has on at least one recent occasion stated this correctly, so the jury instructions should also make it absolutely clear (as opposed to confusing jurors).

  • "[S]pecific reference to a 'computer program' as functional biases the instruction. Finally, as per the Federal Circuit's finding, the jury should beinstructed that 'it is undisputed here that the declaring code and the structure and organization of the API packages are both creative and original.'"

    COMMENT: As I wrote above, the proposed jury instructions suggest that software is always, no matter how creative or original, a second-class citizen in the realm of copyright law.

  • "It is incomplete to tell the jury it must 'decide how much weight to give each … factor[].” [...] 'The Supreme Court has said that this [fourth] factor 'is 'undoubtedly the single most important element of fair use.’'"

    COMMENT: The fourth factor is the effect of the infringement on the market for the infringed product (and authorized derivative products of the infringed product)

  • "Oracle respectfully notes that 5 pages are insufficient to raise all of its objections to these proposed instructions. Oracle hereby preserves all its objections, which include any deviations from Oracle's already proposed instructions, [...] and any additional objections by way of Oracle's proposed instructions and jury instruction briefing as per the Court's Standing Order."

    COMMENT: By contrast, Google only raised three pseudo-objections to the proposed instructions and had enough space to discuss each of them in excruciating detail.

It's disappointing that there is now a real risk of the retrial being made pointless. I have little hope that Judge Alsup will make major changes to his proposed instructions. I guess he'll do something, but I can't imagine that it would be anything but insufficient. And depending on how insufficient it is, the appeals court may get involved with this case again rather 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 and Google+.

Share with other professionals via LinkedIn:


Saturday, April 9, 2016

New Oracle filing might result in damages theory north of $10 billion in Google copyright case

Last month there was quite some talk on the Internet about Oracle seeking $9.3 billion in damages from Google in the famous Android-Java copyright case that will go to trial again in a month, with the bulk of that amount ($8.8 billion) being a claim for disgorgement of infringer's profits. Since the related documents are all heavily redacted, I'm far from sure that the disgorgement and reasonably royalty figures can be added up: those are more likely alternative theories. Either way, we're talking about roundabout $9 billion that Oracle wants. At those altitudes, a few hundred million dollars more or less don't matter.

Google naturally rejects Oracle's claim, but it has a problem: the court-appointed damages expert, Dr. Kearl, has come up with a very wide range that also includes Oracle's number as a plausible scenario. That's why Google brought a motion--without previously seeking permission from the court--to exclude Dr. Kearl's testimony from the trial. Judge Alsup, who usually appears to be very strict on case management matters, merely gave Google a slap on the wrist and allowed Google to break a couple of rules at the same time (limit of number of motions in limine, page limit) with that additional motion. In exchange, he also allowed Oracle to bring an additional motion in limine.

Oracle has just filed its response to Google's motion. It's not an opposition brief in all respects: Oracle had already opposed Dr. Kearl's renewed involvement with the case and would still like to see his testimony excluded. Oracle is being very consistent up to that point.

But what if Judge Alsup, whose decisions I've struggled to understand on more than one occasion (most recently, I was very surprised about how he dealt with Google's unauthorized motion) and the most important one of which (copyrightability) resulted in a disaster for him, still wants Dr. Kearl to testify? In that case, Oracle at least wants to prevent the "cherry-picking" it accuses Google of. Apparently, Dr. Kearl came up with three numbers, and Google would like to limit that testimony to a number that would amount to Oracle getting nothing.

In its opposition to such cherry-picking, Oracle first argues that a disgorgement analysis should not consider non-infringing alternatives ("NIAs"). Oracle may be right on the law, but as a matter of policy, I disagree with Oracle on this one. Anyway, here's the most interesting passage from Oracle's latest filing:

"Prof. Kearl's Oracle Number is consistent with Oracle's $8.8 billion, and is in fact much larger than Oracle's $8.8 billion if the Court excludes his improper use of NIAs. Presumably Rule 706 experts do not often offer a number larger than the Plaintiff intends to offer."

It would be unusual indeed, and it would be the result of the court agreeing with Oracle's legal position that non-infringing alternatives cannot be used to reduce a disgorgement figure.

I don't know what Oracle's lawyers believe to be the number resulting from Dr. Kearl's analysis if adjusted by means of removing NIAs, but I guess they wouldn't say "much larger" (emphasis in original) if we were talking about a 5% or 10% difference. A 5% or 10% difference would be "significantly" larger. Therefore, it is fairly possible that the proposed adjustment to Dr. Kearl's numbers would result in a claim well in excess of $10 billion.

Besides that information I wanted to share something else from Oracle's filing. Oracle refers to a 1985 Ninth Circuit decision, Frank Music Corp. v. MGM, Inc.. In that one, a disgorgement of profits from unauthorized performance of scene from a musical at the (old) MGM Grand was based on multiple revenue streams including increased hotel and casino revenues. While it turned out later that the MGM Grand was still doing well without continued infringement, evidence had been provided that the MGM Grand used shows such as the infringing one to bolster its hotel and gaming revenues. The Ninth Circuit wrote:

"Just because one element could be omitted and the show goes on does not prove that the element was not important in the first instance and did not contribute to establishing the show's initial popularity."

This has some important bearing on Oracle v. Google, where Google argues that it no longer needed the 37 Java API packages or that it could already have removed them in 2010. Oracle argues that "the causal connection here is far stronger, because the revenues Oracle seeks to recover were realized on the infringing work (Android), while the gaming and hotel revenues in Frank were earned separately from the infringing work (the show)."

Here's Oracle's filing:

16-04-08 Oracle Response to Google Motion to Strike Kearl Testimony by Florian Mueller

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:


Friday, April 1, 2016

Google joins the Fair Standards Alliance: discontinued the bad fight, now fighting the good fight

The following press release by the Brussels-based Fair Standards Alliance would have been nothing more than an April Fools' Day joke a few years ago (this post continues with commentary below the press release):

--- START OF PRESS RELEASE ---

Google Joins Fair Standards Alliance

Brussels, March 31st 2016 – The Fair Standards Alliance (FSA) announced today that Google has become its 19th member.

Launched in November 2015 and based in Europe, the Alliance seeks to promote the licensing of standards-essential patents (SEPs) on fair, reasonable and non-discriminatory (FRAND) terms.

"We are delighted in the vote of confidence that a company such as Google is showing in our growing coalition. FRAND licensing of SEPs is a critical part of ensuring that the innovation ecosystem for 5G and the Internet of Things (IoT) can flourish, and we relish the opportunity to work with Google to further our goals of fair and reasonable SEP licensing on a non-discriminatory basis," said Robert Pocknell, Chairman of the FSA.

Allen Lo, Deputy General Counsel for Patents at Google, said: "Google has joined the FSA to support the leadership that it has demonstrated in showing the way to a fair and principled result."

The FSA believes that the entire innovation ecosystem is threatened by unfair and unreasonable SEP licensing practices. Failure to honour the FRAND commitment that exists in most standardisation licensing creates barriers to market entry, threatens to stifle the full potential for economic growth across major industry sectors, and ultimately curbs consumer choice.

The FSA's member companies, who hold more than 160,000 patents and spend more than 32 billion euros per year on R&D and innovation, include: AirTies, BMW, Cisco, Dell, Fairphone, HP, Intel, ip.acess, Juniper Networks, Lenovo, Micromax, peiker acustic, Sierra Wireless, Telit, u-blox and Volkswagen.

--- END OF PRESS RELEASE ---

This is excellent news for the FRAND cause. Google's economic weight alone exceeds that of all the other FSA members, though they already had a pretty good membership base before. Moreover, I could imagine that this here increases the chances of other heavyweights--Apple and Samsung, I'm looking at you in particular--joining the same organization.

Google's decision to join the FSA comes about a year after it contributed 4G (LTE) patents to Via Licensing's pool, a move that I already credited to "the real Google--not the FRAND abuser." There is an unfortunate history of Google (through Motorola) having tried to gain undue leverage with SEPs, a fact for which a court ordered (and the Ninth Circuit affirmed) it owed Microsoft damages, and no one in the blogosphere fought Google (and Samsung) harder over that kind of behavior. I knew that it was all reactive: Google and Samsung didn't draw first blood (only Motorola did, but that was before anyone would even have imagined that Google would buy it). They just wanted patent peace and sought to protect Android. But the end doesn't always justify the means.

Two major players who have fought the good FRAND fight in court are not FSA members as we speak: Apple and Microsoft. I have consistently supported them and everyone else (also smaller players like India's Micromax) in this regard. I was, of course, disappointed when I saw Apple take positions on reasonable royalties in a non-SEP damages context (including the position it's still defending with regard to design patents) that are not just inconsistent but totally irreconcilable with some of the really good points it has made in connection with SEP royalty demands by others.

When Apple and Google/Motorola entered into a second-class settlement of their patent suits almost two years ago, they said they'd work together on certain aspects of patent policy. So maybe Google can persuade Apple that the FSA's efforts would benefit greatly from being supported by both of the world's most valuable companies.

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: