Monday, April 30, 2012

Apple and Motorola move for summary judgment in Wisconsin litigation over FRAND issues

Microsoft's U.S. lawsuit to enforce Motorola's FRAND licensing promises and obligations is pretty well-known by now, at least since the temporary restraining order that Judge James L. Robart recently entered against Motorola Mobility's near-term enforcement of a German patent injunction it may win the day after tomorrow (May 2). For a long time, only a few people were interested in that case. So far, there's also been little interest by third parties in Apple's similar lawsuit against Motorola Mobility. While Microsoft filed its FRAND enforcement action in the Western District of Washington (back in November 2010), Apple brought its comparable lawsuit (filed in March 2011) in the Western District of Wisconsin, which at some point used to be the center of gravity for the dispute between these two parties. Even though three FRAND-pledged Motorola patents are in play in the case over which Judge Posner is presiding in the Northern District of Illinois, which was pending in Wisconsin for some time, the FRAND case wasn't transferred to Chicago. In my opinion, Apple's proposal for a transfer was a good idea, but it was denied. Judges enjoy significant discretion in this regard.

The Apple-Moto Wisconsin FRAND case is now entering a stage at which it's getting really interesting. On Friday, Motorola Mobility filed a motion for summary judgment that would do away with most of Apple's claims if granted, and Apple moved for partial summary judgment on some elements of its claims that it believes are ripe for being decided as a matter of law. Judge Barbara B. Crabb today set a briefing schedule: opposition briefs are due by May 30, 2012, and then the parties get to reply in support of their motions until June 11, 2012. Thereafter there will almost certainly be a summary judgment hearing, and then a decision. The Chicago action will go to trial in June, but Apple's FRAND defenses will be resolved after the infringement and invalidity issues, so Judge Crabb's decisions on the parties' summary judgment motions may still come down in time to have an impact on the Chicago case. The transfer of the large infringement case (originally 15 Apple and 6 Motorola patents) from Wisconsin to Illinois was the result of close coordination between Judge Posner and Judge Crabb, so I guess they will again try to keep things efficient for both courts.

At this stage, pretty much everything concerning these motions, except for lists of the claims addressed by them, is sealed. Different courts and different judges have different rules. Judge Alsup, who is presiding over Oracle v. Google in the Northern District of California, definitely would have objected to such extensive sealing.

This is the list of all of the issues Motorola's summary lists:

  • Count I - Equitable Estoppel

  • Count II - Breach of Contract - ETSI/3GPP

  • Count III - Breach of Contract to Which Apple is a Third Party Beneficiary – ETSI/3GPP

  • Count IV – Breach of Contract to Which Apple is a Third Party Beneficiary – IEEE

  • Count V – False F/RAND Commitments and Deceptive Acts in Violation of Section 2 of the Sherman Act

  • Count VI – Unfair Competition and Unlawful Business Practices in Violation of Cal. Bus. & Prof Code para. 17200, et seq.

  • Count VII – Declaratory Judgment That Motorola's Offers have not been on F/RAND Terms

  • Count XI – Declaratory Judgment of No Entitlement to Injunctive Relief

  • Count XII – Declaratory Judgment of Patent Misuse

  • Count XIII – Interference with Contract

If some of the terms or acronyms weren't familiar to you, let me just sum it all up by saying that this is a mix of contract, equitable and antitrust claims, and all of this is about Motorola's FRAND obligations, its alleged breach, and the effect this has on Motorola's patent enforcement. Some of these claims would also entitle Apple to damages from Motorola for its alleged conduct. ETSI/3GPP refers to the standard-setting organizations in charge of 3G/UMTS and some other wireless standards. IEEE is a standard-setting organization in the United States. The IEEE standard at issue between these parties is 802.11 (WLAN, or WiFi).

Apple's summary says that the company "moves for partial summary judgment to establish elements of its breach of contract, antitrust, unfair competition, and patent misuse claims". That's a subset of the number of issues Motorola tries to do away with. Apple has generally shown a lot of focus in its patent litigations recently, such as by quickly dropping claims that weren't too promising or by suing two Android device makers, Motorola and HTC, simultaneously in the same Miami-based court over a set of six patents. By raising fewer issues in its partial summary judgment motion than Motorola does in its summary judgment motion, Apple will get to discuss each item in more detail in its pleadings.

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:


Oracle v. Google: definitive copyright-related jury instructions and special verdict form

I started writing this quick post while Judge Alsup was reading the first part of the copyright-related jury instructions to the jurors who will later today, and possibly for more days to follow, decide on the copyright liability issues in Oracle v. Google. Here's the document -- I'll provide some comment below:

12-04-30 Oracle v Google Copyright-related Jury Charge and Verdict Form

Earlier today I blogged about Oracle's and Google's objections to the jury instructions and the verdict form. Judge Alsup has not granted many of the parties' requests for edits, and especially not changed anything over which the parties had a serious dispute.

These are the changes (other than simple grammatical changes such as replacing "virtual identity" with "virtually identical" that I have identified:

  • Instruction 27

    This is one of various instructions that mention the de minimis (too small to matter) defense. Here, the judge inserted a clarification that this defense relates only to the rangeCheck function and about a dozen files, but not to the structure, sequence and organization of the 37 asserted APIs:

    "To be clear with respect to a different issue. The parties are in agreement that the structure, sequence, and organization of the API packages is more than de minimis."

    Oracle didn't propose this particular context, but it wanted this clarification.

  • Instruction 29

    In connection with the term "work as a whole", the judge accomodated Oracle's request not to overemphasize the word "all" in a couple of instances.

  • Instruction 30: public dedication

    Google said that the question of whether Sun dedicated the Java APIs to the public (something that Google claims to have occurred, though I'm unaware of any hard evidence) is still an issue the jury needs to consider. Accordingly, Judge Alsup deleted the statement that "Google makes no such contention in this trial".

  • Special Verdict Form, Question 3: verbatim copying

    Judge Alsup inserted the parentheticals "infringing"/"not infringing" below the Yes/No answers. This is a clarification on which the parties agreed.

  • Special Verdict Form, Question 4: issues relevant to Google's equitable defenses

    The judge granted Google's request to refer to "Questions 4A and 4B", for the sake of clarification, as opposed to "these questions" or "these interrogatories".

I later double-checked if there were any other edits worth reporting on, but didn't find any. My first version of this blog post already listed all of the relevant ones.

In other news relating to this case, the father of Java, James Gosling, stated that "Google slimed Sun" and that "Oracle is in the right" with this lawusuit. He says then-Sun CEO Jonathan Schwartz only decided to put on a happy face, which annoyed many people at Sun.

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:


Microsoft and Barnes & Noble settle Android patent litigation -- Motorola increasingly isolated

Patent settlements typically come at times when nobody predicts them rather than as a result of court-ordered mediation.

Today, Microsoft and Nook maker Barnes & Noble announced a strategic partnership relating to e-reading. Microsoft invests $300 million in exchange for an equity position of approximately 17.6% in a Barnes & Noble spin-off. The press release also says this about the Android-related patent litigation that was pending between the parties:

"Barnes & Noble and Microsoft have settled their patent litigation, and moving forward, Barnes & Noble and Newco will have a royalty-bearing license under Microsoft's patents for its NOOK eReader and Tablet products. This paves the way for both companies to collaborate and reach a broader set of customers."

The term "royalty-bearing" is a real blow to Google, which tells Android device makers to refuse to pay but doesn't do what it takes to properly address Android's intellectual property infringement issues.

The fact that Barnes & Noble partners with Microsoft proves, if anyone still needed any proof, that its mostly antitrust-related "patent misuse" allegations against Microsoft, which the ITC threw out even ahead of trial and refused to reconsider, were bogus claims borne out of desperation (for lack of patents that could be used to bring counterclaims). Barnes & Noble may have had very bad advice from certain lawyers who self-servingly raised totally false hopes in Barnes & Noble's management as to what they could achieve by shouting "antitrust! antitrust!" Look at it this way: if you really thought someone was an anticompetitive abuser of patents, would you form a joint venture with him? There you have it.

An important effect of this settlement is that Motorola Mobility is now the only Android device maker to be embroiled in litigation with Microsoft. Just four days ago, Microsoft announced yet another Android- and Chrome-related patent license agreement (with Pegatron, a major Taiwanese company that manufactures devices for others). More than three months ago Microsoft already said that more than 70% of Android devices sold in the U.S. market have a license to Microsoft's patents reading on Android. Those device makers notably include Samsung, HTC and LG -- companies that defend themselves in court against patents all the time if they think there's a point in doing so in a given context.

Motorola's strategy is different. Motorola hopes to gain leverage, not only but primarily from abusive assertions of standard-essential patents, in federal court, at the ITC and in Germany, where a decision is scheduled to come down on Wednesday morning (May 2). That conduct has already triggered two EU antitrust investigations and has led a federal judge in the Western District of Washington to enter a temporary restraining order. But it's quite possible that Motorola, too, would have settled the dispute with Microsoft a while ago if Google had not offered to buy the company -- a transaction that hasn't been closed yet.

At any rate, the good news is that there's one Android-related dispute less as a result of the agreement Microsoft and Barnes & Noble just announced.

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:


Oracle and Google object to 'final' jury instructions and verdict form

After 10 PM local time on Sunday, Google filed its objections to Judge Alsup's "final charge to the jury (Phase One) and special verdict form". At 3 AM local time on Monday morning, Oracle also filed objections. In my previous blog post I discussed what it means that the instructions say names (which are not copyrightable on a standalone basis) are copyrightable as part of a larger structure. Now that both parties have filed their objections to the "final" jury charge, I believe the so-called "final" version may have to be replaced with a "final-final" one.

In the following I will show the disputed passages of the document and sum up the parties' concerns (and one clarification on which they actually agree). I will do this in the order of the sections of the docuemnt, but I would like to point out that the definition of fair use in section 26 of the jury instructions appears particularly important -- and in need of corrections -- to me.

Instruction 16: definition of copyright infringement

The "final" April 29 version of the jury instructions states the following:

"As stated, the owner of a copyright has the exclusive right to make copies of all or more than a de minimis part of the copyrighted work, subject only to the right of anyone to make fair use of all or a part of any copyrighted material, all as will be explained below."

While Oracle agreed to this wording at the charging conference on Friday, it finds that the jury charge as a whole now "lacks a clear and affirmative definition of infringement", which was contained in the original draft:

"As stated, the owner of a copyright has the exclusive right to make copies of all of [sic] part of the copyrighted work. If someone else does so without consent from the owner, then there is infringement (except in certain circumstances I will describe below."

Oracle notes that "Google's affirmative defenses of fair use and de minimis defenses are mentioned or explained in 10 of the 18 substantive instructions" but the document still needs "at least one clear definition of infringement, either in this instruction or elsewhere". I think Judge Alsup does strive to be balanced, and there's no reason why he couldn't put back the definition of infringement contained in his first draft.

Instruction 17: copyrightability and ideas/expression dichotomy

"Final" version of April 29:

"The copyright confers ownership over the particular expression of ideas in a work but it never confers ownership over ideas themselves. For example, if a book describes a strategy for playing a card game, the copyright prevents anyone (but the owner) from duplicating the book itself but everyone is still free to read the book and to use the strategy, for the idea set forth in the book, that is the strategy, is not protected by copyright. And, everyone is entitled to write their own book about the same game and the same strategy so long as they do not plagiarize the earlier book. Again, the main point is that the copyright protects the particular expression composed by the author.

Another statutory limitation on the scope of a copyright is that copyright never protects any procedure, process, system, method of operation, concept, principle, or discovery. Possibly such things can be claimed under the patent system or by trade secret laws but they may not be claimed by copyright. For purposes of your deliberations, I instruct that the copyrights in question do cover the structure, sequence and organization of the compilable code."

As I mentioned in my previous post, Google is unhappy about the final sentence and wants it to state the opposite. That is not going to happen -- the judge will decide on copyrightability at a later point. Google simply preserves the record with a view to an appeal.

Oracle would be fine with the final sentence if it stood on its own but objects to the remainder of that instruction. Oracle sees no point in presenting to the jury the ideas/expression issue since this will be decided by the judge himself, so all it can do to the jury is create confusion and "invite[] the jury to deliberate over whether the works at issue are protected by copyright".

Instruction 24: burden of proof for de minimis defense; virtual identity test for API documentation

"Final" version of April 29:

"Oracle must also prove that Google copied all or a protected part of a copyrighted work owned by Oracle and that the amount of copying was not de minimis. So, there are two elements Oracle must prove to carry its burden on infringement, namely copying of a protected part and the part copied was more than de minimis when compared to the work as a whole. These are issues for you to decide.

There are two ways to prove copying. One is by proof of direct copying, as where the copyrighted work itself is used to duplicate or restate the same words and symbols on a fresh page.

The second way is via circumstantial evidence by showing the accused had access to the copyrighted passages in question and that there are substantial similarities or, in certain instances, virtual identity between the copyrighted work and the accused work. The virtual identity test is used when the subject under consideration is a narrow one and we would expect certain terms and phrases to be used. This is in contrast to, for example, a fictional work in which there will be a broad range of creativity, in which case it is necessary only to prove substantial similarity. In this trial, you should use the substantial similarity test for all such comparisons except for those involving the API documentation, in which case you should use the virtual identity test. This is because the documentation for the API packages describe narrow technical functions and it is to be expected that some of the same words and phrases would likely be used.

To determine whether the copyrighted work and the accused work are substantially similar, or where appropriate, virtual identity, you must compare to the works as whole. I will define the works as a whole in a moment.

However, in comparing to the works as a whole, you cannot consider similarities to unprotectable elements of Oracle's works. I have instructed you about the protectable and unprotectable elements of Oracle's work."

The final part about non-consideration of unprotectable elements is much less relevant in light of the clarification that the names are copyrightable as part of a larger structure, even if they are not on their own.

Oracle raises two objections against instruction 24:

  1. Oracle reminds the court of the fact that de minimis is an affirmative defense raised by Google, and Google has the burden of proof. At any rate, with respect to the 37 APIs the judge already said last week that "[n]o reasonable jury could find that the structure, sequence, and organization is [de minimis]", which means that Oracle is entitled to judgment as a matter of law on this particular defense. I guess there's a good chance that the judge will rule on de minimis, at least with respect to the 37 APIs, and modify the jury instructions accordingly. That would be the most efficient way to provide more clarity, and to avoid confusion.

  2. Oracle says that if the court actually intended to have the virtual identity test applied to (quoting the definition provided in jury instruction 18 as it currently stands) "all content--including English-language comments as well as method names and class names, declarations, definitions, parameters, and organization--in the reference document for programmers", this would be "incorrect" because it's too broad a definition.

Instruction 25: comparison of copyrighted work and accused work

"Final" version of April 29:

"To determine whether the copyrighted work and the accused work are substantially similar, or where appropriate, virtual identity, you must compare to the works as whole. I will define the works as a whole in a moment.

However, in comparing to the works as a whole, you cannot consider similarities to unprotectable elements of Oracle's works. I have instructed you about the protectable and unprotectable elements of Oracle's work."

Google notes that "virtually identical" (as opposed to "virtual identity") would be "necessary for proper grammar" in this context.

Furthermore, Google wants "the works as whole" (at the end of the first sentence) to be replaced with "the works as a whole".

Oracle refers the court to the objections it already stated yesterday, in a separate brief. Oracle's position is that this instruction shouldn't be delivered at all and has the following shortcomings:

  1. "the proper frame of reference for the 'work as a whole' should be Oracle's work, not Android;"

  2. "the instruction does not include the fact that the comparison to the work as a whole should be both qualitative and quantitative, as even a quantitatively small amount of copying can have significance; and"

  3. "the instruction should explicitly state that there is no need to engage in the substantial similarity analysis if there is evidence of direct copying of the kind Google has admitted to here."

All of the issues raised above are supported by applicable case law. This jury instruction is almost as problematic as the one on fair use (the next one).

Instruction 26: fair use

"Final" version of April 29:

"Now, I will explain the law governing Google’s defense based on the statutory right of anyone to make 'fair use' of copyrighted works. The public may use any copyrighted work in a reasonable way under the circumstances without the consent of the copyright owner if it would advance the public interest. Such use of a copyrighted work is called a 'fair use.' The owner of a copyright cannot prevent others from making a fair use of the owner's copyrighted work. For example, fair use may include use for criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research.

Google has the burden of proving this defense by a preponderance of the evidence.

In determining whether the use made of the work was fair, you should consider the following factors:

  1. The purpose and character of the use, including whether such use is of a commercial nature, for nonprofit educational purposes, and whether such work is transformative (meaning whether Google’s use added something new, with a further purpose or different character, altering the copied work with new expression, meaning, or message). Commercial use cuts against fair use while transformative use supports fair use;

  2. The nature of the copyrighted work, including whether the work is creative (which cuts against fair use), functional (which supports fair use), or factual (which also supports fair use);

  3. The amount and substantiality of the portion used in relation to the copyrighted work as a whole. The greater the quantity and quality of the work taken, the less that fair use applies; and

  4. The effect of the use upon the potential market for or value of the copyrighted work. Impairment of the copyrighted work cuts against fair use.

All the factors should be weighed together to decide whether Google's use was fair use or not. It is up to you to decide how much weight to give each factor but you must consider all factors. If you find that Google proved by a preponderance of the evidence that Google made a fair use of Oracle's work, your verdict should be for Google on that question in the Special Verdict Form."

Google preserves the record with respect to its demand for a fifth factor, which would turn the fair use analysis into something that has no boundary at all. After it came up, I said that the court would probably reject it, and that's what happened. Other than that, Google is fine with the fair use instruction -- which comes as little surprise because the way Judge Alsup wrote up that one is exceedingly Google-friendly, not in all respects (for example, it now does reflect four suggestions Oracle made, one for each of the four factors) but on balance.

Oracle has two major issues with this instruction:

  1. The passage on the second factor defines "transformative" too broadly by considering a "further purpose or different character, altering the copied work with new expression, meaning, or message" sufficient. The fair use rule was not intended to let anyone get away with infringement as long as they add some stuff of their own. Oracle notes that prior to Android there were already a billion phones in the world running Java -- Oracle stresses this fact, which shouldn't even have to be talked about, because Google seriously claims that Android's way of putting Java on a smartphone was "transformative". Oracle requests that the definition be changed to "meaning whether Google's use of copyrighted material was for a distinct purpose unrelated to the function and purpose of Oracle's original material". I believe this is the only way to open the floodgates to a "transformative" fair use defense for pretty much every kind of infringement.

  2. Oracle says the term "functional" (in the explanation of the first factor) makes it too easy for Google to mislead the jury. The problem I see here is that program code, including API code, is always functional in a certain way, but still the rule is that it's protected and that using and especially redistributing it without a license is an infringement rather than fair use.

Instruction 28: de minimis

"Final" version of April 29:

"Copying that is considered 'de minimis' is not infringing. Copying is 'de minimis' only if it is so meager and fragmentary that compared to the work as a whole the average audience would not recognize the appropriation. You must consider the qualitative and quantitative significance of the copied portions in relation to the work as a whole. The burden is on Oracle to prove that the copied material was more than de minimis.

The relevant comparison is the copied portion contrasted to the work as a whole, as drawn from the copyrighted work, not contrasted to the accused infringer’s work as a whole. For example, if an infringing excerpt is copied from a book, it is not excused from infringement merely because the infringer includes the excerpt in a much larger work of its own."

In addition to reiterating that Google has to prove its de minimis defense, Oracle wants the instruction "clearly state that the defense of de minimis is only available to excuse Google's use the 11 Java source and object code files copied verbatim, and is not a defense to Google's use of the structure, sequence, and organization of the compilable code of the API packages or for the documentation of the API packages". This appears to be merely a request for a clarification of what the judge also thinks.

Instruction 29: work as a whole

"Final" version of April 29:

"In your deliberations, you will need to make certain comparisons to the 'work as a whole.' It is my job to isolate and identify for you the 'work as a whole.' You must take my identification as controlling if and when this comes up in your deliberations. This issue arises when (1) comparing Oracle’s work and Android's work for similarity under both substantial similarity and virtual identity standards, (2) deciding whether Google copied only a de minimis amount of Oracle's work, and (3) evaluating the third factor of fair use: the amount and substantiality of the portion used in relation to the copyrighted work as a whole.

Although you have seen that the copyright registrations cover a large volume of work, the entire registered work is not the work as a whole for these purposes. This may seem odd to you, so let me give an example. An entire magazine issue may be copyrighted but a specific article advertisement or photograph may be the relevant work as a whole, depending on what was allegedly copied.

For purposes of this case, I have determined that the 'work as a whole' means the following: For purposes of Question No. 1 in the Special Verdict Form, the 'work as a whole' constitutes all of the compilable code associated with all of the 166 API packages (not just the 37) in the registered work. This excludes the virtual machine. Similarly, for the purposes of Question No. 2 in the Special Verdict Form, the 'work as a whole' means the contents (including name, declaration and English-language comments) of the documentation for all of the 166 API packages (not just the 37) in the registered work. For purposes of Question No. 3, the 'work as a whole' is the compilable code for the individual file except for the last two files listed in Question No. 3, in which case the 'work as a whole' is the compilable code and all the English-language comments in the same file."

Oracle seeks to avoid a comparison of "the structure sequence and organization of the 37 packages with all of the compilable code in the 166 packages". Instead, the structure, sequence and organization of one work should be compared to the structure, sequence and organization of another.

Furthermore, Oracle asks the court to deemphasize the word "all" in front of "of the 166 API packages (not just the 37)". Oracle attributes the italicization of that word to the evolution of this document: in the original draft, the italicized word "all" was meant to contrast particularly strongly with the equally-italicized passage "that individual API package", but Oracle no longer asserts individual API packages, so there's no more need to emphasize the word "all".

Instruction 30: sublicense, public dedication

"Final" version of April 29:

"Unless you find fair use, de minimis, or non-infringement in Google's favor, Google had no right to copy any elements of the Java platform protected by copyright unless it had a written license to do so from Sun or Oracle or had a written sub-license to do so from a third party who had a license from Sun or Oracle conferring the right to grant such sub-licenses. The burden would be on Google to prove it had any such express license or sublicense rights. But in this trial it makes no such contention. Put differently, if Google claims a license from a third party, Google has the burden to prove that the third party itself had the proper right and authority from Sun or Oracle as to any of the copyrights owned by Sun or Oracle and used by Google, for Google could acquire from the third party no greater right than the third party had in the first place. Similarly, if Google contends that Oracle or Sun had dedicated elements protected by copyright to the public domain for free and open use, the burden would be on Google to prove such a public dedication but the parties agree that there is no such issue for you to decide. Again, Google makes no such contention in this trial and this statement of the law regarding licenses is simply to put some of the evidence you heard in context."

Google objects to this instruction in its entirety, as it previously did. And it says that "Sun dedicated the Java language, including the APIs, to the public", so there is now a jury issue (and not just an issue for the court to decide) in connection with the concept of "public dedication".

Special Verdict Form Question 3: infringement by direct copying of source code, comments or decompiled code

This question is whether "Oracle [has] proven that Google’s conceded use of the following was infringing, the only issue being whether such use was de minimis", then listing the rangeCheck method, eight decompiled files, and the English-language comments in two other files.

Oracle reiterates that the burden of proof for de minimis is on Google. The parties agree, however, that it would be useful to make perfectly clear to the jury what a Yes or No answer means by placing parentheticals below the answers ("infringing", "not infringing").

Special Verdict Form Question 4: issues relevant to Google's equitable defenses

Question 4 consists of two questions that the jury is asked to answer only if it previously identifies an infringement of the compilable code for the 37 APIs:

"A. Has Google proven that Sun and/or Oracle engaged in conduct Sun and/or Oracle knew or should have known would reasonably lead Google to believe that it would not need a license to use the structure, sequence, and organization of the copyrighted compilable code? [Yes/No]

B. If so, has Google proven that it in fact reasonably relied on such conduct by Sun and/or Oracle in deciding to use the structure, sequence, and organization of the copyrighted compilable code without obtaining a license? [Yes/No]

Your answer will be used by the judge with issues he must decide. These interrogatories do not bear on the issues you must decide on Questions 1 to 3."

Google wants to avoid a possible misunderstanding by the jury regarding what its answers to other questions will be used for. Therefore, it asks to specifically reference "Questions 4A and 4B" in a couple of contexts.

Oracle already objected in a separate pleading yesterday to these questions. Today Oracle again expressed concer that "including these interrogatories will improperly focus the jury on Google's equitable defenses, which are for the Court to decide". Also, Oracle says these equitable defenses have additional elements that the proposed questions "do not capture".

Judge Alsup is an early riser. There could be a new jury charge and verdict form very soon. There are serious issues involved, and there are pretty clear solutions. I'll report as soon as I see a revised jury charge and verdict form.

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:


Final Oracle-Google jury instructions confirm that names are copyrightable as part of larger structure

Late on Sunday, Judge Alsup filed the final jury instructions and special verdict form (all of this limited to the current trial phase, which is focused on copyright). He adopted various proposals from both parties, but neither party will be completely satisfied with it. Depending on which party wins the verdict, we'll almost certainly see an appeal unless there's a settlement. Google has already filed some objections and I will comment later when I see both parties' objections. In other Sunday news, the parties made various filings with respect to their motions for judgment as a matter of law. I will address those at a later stage as well, with one exception (copyrightability instruction) that I'll talk about further below.

The jury will primarily have to decide on infringement and, if any infringement is found in any given area, the fair use defense. It won't get to decide on copyrightability (the parties agreed a while ago that this is a legal, not factual, question), and it won't render a complete verdict on Google's equitable defenses, though the form contains a couple of questions that the judge wants the jury to answer with a view to his analysis of those equitable defenses.

Even though the jury doesn't decide on copyrightability, what the final jury instructions say in this regard is important for the following four reasons (in no particular order):

  1. The stronger Oracle's claim of copyrightability appears, the weaker Google's story looks in the eyes of the jury. Originally, the judge said he was going to instruct the jury that the structure, sequence and organization of the APIs are protected by copyright, a plan that Google opposed because it didn't want the jury to think that it has already lost on a key disputed question. The judge's first draft, which the judge filed on Thursday, stated that the jury "must assume" the copyrightability of the structure, sequence and organization of the APIs for the purposes of its deliberations. The final version says the following:

    "For purposes of your deliberations, I instruct that the copyrights in question do cover the structure, sequence and organization of the compilable code."

    Google's first objection relates to this. Google says the jury should be told the opposite. Google can't claim prejudice because either the instruction is right or Google is right. If the instruction is right, Google isn't prejudiced (if anyone is prejudiced in that case, it's Oracle because the statement would be even clearer otherwise). If Google is right, it wins the case regardless of the jury verdict.

    I believe the final instruction on copyrightability is considerably more favorable to Oracle than the "must assume" version, though there may very well be people who are more impressed by being told they "must assume" something.

  2. The instructions provide at least a hint as to where the judge stands on the key issues.

    I have already expressed my belief that the judge wouldn't even be holding this jury trial if Oracle didn't have a reasonably good chance of prevailing on copyrightability. I have read the decisions on which the parties base their arguments and those are clearly in Oracle's favor, especially the Ninth Circuit decision in Johnson Controls v. Phoenix Control Systems.

  3. The scope of what is copyrightable for the purposes of this trial obviously influences the likelihood of Google being found to infringe. In this regard, Oracle won an important insertion:

    "While individual names are not protectable on a standalone basis, names must necessarily be used as part of the structure, sequence, and organization and are to that extent protectable by copyright."

    This is actually just consistent with what the judge wrote in an order denying a Google motion for summary judgment on Oracle's copyright claims last summer. That sentence is so very important because it's really the names (a massive number of them, in fact) that the jury can easily identify when looking for virtually identical or substantially similar elements. The structure, sequence and organization would also be identifiable, but that's a much more abstract concept if one has to assume at the same time that the names themselves are not protected.

  4. The infringement questions (one of which relates to the compilable code while the other relates to the Java documentation) are binary (yes or no) questions on the verdict form, but with a view to the fair use issue, the extent of any infringement identified will play a major role. I believe Google will most likely lose out on the infringement side (this is so obvious it shouldn't even be put before the jury in the first place) but the fair use verdict is not realistically predictable, especially since I think the jury instructions are, at least, misleading in some ways that may (but need not) help Google. Anyway, one of the fair use factors is related to the scope and scale of infringement. Let me quote from the final jury instructions:

    "The amount and substantiality of the portion used in relation to the copyrighted work as a whole. The greater the quantity and quality of the work taken, the less that fair use applies[.]"

    If the judge hadn't adopted Oracle's proposed insertion regarding the copyrightability of the names as part of the structure, sequence and organization, Google would have had a better chance (relatively speaking) of the jury considering the extent of the infringement to be limited. Now, the use of such a massive number of names that are protected as part of a larger structure will very likely lead the jury to consider the infringement to be quite substantial, which in turn weights against a finding of fair use.

    In another slight departure from my plan to talk about the parties' objections to the final form later, I'd like to mention that Google is very unhappy that the judge didn't allow the open-ended definition of fair use that Google proposed. Google wanted the judge to tell the jury that it can consider any factor it wants in connection with fair use. Oracle objected, and even before Oracle made its related filing, I wrote that "I would be surprised if the court granted this request" -- and indeed, the court rejected it.

Today is an important day: the parties will make their copyright-related closing arguments. I will definitely write about this case again shortly. For now I wanted to highlight what I thought was the most important aspect of the final jury instructions.

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:


Sunday, April 29, 2012

Google to court: Galaxy Nexus code is closed just like that of the iPhone

Apple and Google routinely have discovery disputes. Apple's second California action against Samsung, which includes a motion for a preliminary injunction against the Galaxy Nexus, is already the eighth U.S. lawsuit in which Apple subpoeaned Google, and it's no surprise that Apple would like to receive more information than Google is willing to provide. Typically, the court will resolve such a dispute by narrowing the discovery requests. But what's particularly interesting this time (even if not completely new) is that Google puts its attitude toward "open source" on display.

One of Apple's requests that Google opposes is a demand for documents and testimony regarding "[t]he differences, if any", between the code actually used on the Galaxy Nexus and "the Android 4.0 Ice Cream Sandwich code publicly available from https://android.googlesource.com/platform/manifest, or through the process described at http://source.android.com/source/downloading.html".

Google's first objection is that it "has already agreed to produce source code underlying" the Android software running on the Galaxy Nexus. Google disputes the relevance of of the open-source code because it "is not loaded on Samsung's Galaxy Nexus device and cannot be relevant to Apple's infringement claims against Samsung generally or the Galaxy Nexus in particular".

Then Google goes on to complain that "Apple's request would reveal critical Google trade secrets". Yes, the Galaxy Nexus source code is described as a "trade secret":

"Although Google releases some versions of Android through the Android Open Source Project, the internal functionality of Android running on the Samsung Galaxy Nexus is Google's trade secret."

In other words, Google will publish whatever it publishes on open-source terms, but the software running on the lead device for the current generation of Android is a secret.

Google suspects that Apple made this request "for one purpose only – to design claims for other litigation". Whether or not that is Apple's motive is a separate issue. Someone who claims to be "open" can't have his cake and eat it. Of course, being open means that, potentially, someone will look at code to identify patent and copyright infringements. But under open-source rules the one who wants to do so has the same right to inspect the code as someone who wants to fix a bug (possibly a critical security vulnerability), add a feature, or translate the thing to the Klingon language. Open source rules don't discriminate against anyone who wants to look at the source code.

In its effort to claim trade secret protection for the code running on the Galaxy Nexus, Google then cites a discovery ruling from an iPhone antitrust case (In re Apple & AT & TM Antitrust Litig.):

"Because the iPhone source code is a trade secret, plaintiffs have the burden to establish that it is both relevant and necessary."

Google indeed wants to have it both ways. It wants to keep Android closed like the iPhone in some contexts but at the same time demands legal privileges, including the privilege to infringe anyone's intellectual property, because Android is supposedly "open". Google's claim that the Galaxy Nexus source code is just as much of a trade secret as the code running on the iPhone contrasts with what Google told the ITC two months ago in connection with Apple's complaint against Motorola:

"Exclusion of Motorola's handsets and tablets would also threaten Android, the only freely adaptable, open-source mobile platform developed and distributed in the U.S. Android's open, generative platform will facilitate the next round of innovations in mobile computing and fuel the U.S. technology sector [...]"

"Unlike Apple's much less flexible iOS, generative platforms such as Android encourage innovation, promote self-expression, and protect individual freedom, providing consumers and industry with substantial value."

It would be nice if Google could either come clean and admit that it's not really open or (ideally) honor its open-source promise. But if Google buys Motorola, I'm afraid we'll see more "trade secret" claims than ever.

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 28, 2012

List of 50+ Apple-Samsung lawsuits in 10 countries

Apple and Samsung will meet in San Francisco on May 21 and 22 to talk settlement. Settling this massively-escalated, planet-spanning dispute will be a Herculean task:

All in all, the parties have filed approximately 50 lawsuits against each other, but that's just the number of initial filings. After an initial filing, lawsuits can be consolidated (which effectively happened to the first two lawsuits the parties filed against each other in California) or, which happened far more frequently, split up into multiple lawsuits by the courts. For example, German courts usually look at only one patent per litigation, and sometimes assign more than one case number per patent if there's a delay with serving the complaint on one of two or more defendants. Also, these parties tend to appeal each and every decision, and the list below doesn't contain cases before appeals courts. Furthermore, I'm sure that thre are more nullity actions before the German Federal Patent Court than the ones on the list I obtained (Samsung definitely brought several such actions). Counting all of those additional cases not listed below, the number of cases between the parties is far above 50 -- it might even be closer to 100.

Here's a list of all initial filings the parties brought against each other between April 2011 (when the dispute started) and April 25, 2012 (the date of the last update I received), to the extent that I've been able to find out about them, and sorted by country and venue:

  • United States

    • U.S. District Court for the Northern District of California

      • Apple Inc. v. Samsung Electronics Co., Ltd. et al. (case no. 5:11-cv-01846, filed on 04/15/11; includes substance of formerly separate case no. 5:11-cv-2079, filed on 04/27/11)

    • U.S. International Trade Commission

      • In the Matter of Certain Mobile Electronic Devices (Samsung v. Apple; investigation no. 337-TA-794, complaint filed on 06/28/11)

      • In the Matter of Certain Electronic Digital Media Devices and Components Thereof
        (Apple v. Samsung; investigation no. 337-TA-796, complaint filed on 07/05/11)

    • U.S. District Court for the District of Delaware

      • Samsung Electronics Co., Ltd. et al. v. Apple Inc. (case no. 1:11-cv-00573, filed on 06/28/11; mirrors simultaneous ITC complaint)

  • Germany

    • Landgericht Mannheim
      (Mannheim Regional Court)

      • Samsung Electronics GmbH v. Apple, Inc. and Apple GmbH (case no. 7 O 247/11, filed on 04/21/11)

      • Apple Inc. v. Samsung Electronics GmbH (case no. 7 O 166/11, filed on 06/17/11)

      • Apple Inc. v. Samsung Electronics GmbH (case no. 7 O 70/12; filed on 11/28/11)

      • Samsung Electronics Co. Ltd v. Apple Inc. (case nos. 7 O 19/12, 7 O 54/12, 7 O 53/12, and 7 O 18/12; filed on 12/16/11)

    • Landgericht Düsseldorf
      (Düsseldorf Regional Court)

      • Apple Inc. v. Samsung Electronics GmbH (case no. 14c O 194/11; filed on 08/04/11)

      • Apple Inc. v. Samsung Electronics GmbH (case nos.  14c O 219/11 and 14c O 219/11, filed on 09/02/11)

      • Apple Inc. v. Samsung Electronics GmbH (case no. 14c O 255/11, filed on 10/07/11)

      • Apple Inc. v. Samsung Electronics GmbH (case nos. 14c O 291/11, 14c O 293/11, and 14c O 294/11; filed on 11/25/11)

    • Landgericht München I
      (Munich I Regional Court)

      • Apple Inc. v. Samsung Electronics GmbH (case no. 21 O 26022/11; filed on 11/28/11)

    • Bundespatentgericht
      (Federal Patent Court, based in Munich)

      • Apple GmbH v. Samsung Electronics Co. Ltd. (case no. 5 Ni 49/11, seeking invalidation of EP1005726; filed on 08/05/11)

      • Apple Inc. v. Samsung Electronics GmbH (case no. 5 Ni 51/11, seeking invalidation of EP1114528; filed on 08/26/11)

      • Apple GmbH v. Samsung Electronics Co. Ltd. (case no. 5 Ni 55/11, seeking invalidation of EP1188269; filed on 09/16/11)

  • United Kingdom

    • UK High Court of Justice, Chancery Division, Patents Court

      • Samsung Electronics Co. Ltd v. Apple Retail UK et al. (case no. HC 11 CO 2180, filed on 06/29/11)

  • France

    • Tribunal de Grande Instance de Paris

      • Samsung Electronics Co. Ltd and Samsung Electronics France v. Apple France et al. (case no. 11/10464; filed on 07/08/11)

      • Sasmung Electronics Co. Ltd. et al. v. Apple Inc. et al. (case no. 11/58301; filed on 10/05/11)

  • Italy

    • Tribunale di Milano

      • Samsung Electronics Italia s.p.a. v. Apple Inc. (filed on 06/29/11)

      • Samsung Electronics Co. Ltd. et al. v. Apple Inc. et al. (case no. 59734/2011; filed on 10/05/11)

  • The Netherlands

    • Rechtbank 's-Gravenhage
      (The Hague District Court)

      • Apple Inc. v. Samsung Electronics Co. Ltd. (case nos. KG 11-730 and KG 11-731; filed on 06/27/11)

      • Apple Inc. v. Samsung Electronics Co. Ltd. (filed on 11/25/11)

  • Spain

    • Community Designs Court, Alicante
      (EU-wide jurisdiction)

      • Samsung Electronics Co. Ltd. and Samsung Electronics Iberia, S.A.U. v. Apple Inc. (case no. unknown, seeking declaratory judgment with respect to Community design no. 00018607-0001; filed on 09/08/11)

      • Apple Inc. v. Samsung Electronics Co. Ltd. (case no. 755/2011-C; filed on 02/27/12)

    • Office for Harmonization in the Internal Market (OHIM)

      • Samsung Electronics Co. Ltd., Samsung Electronics Benelux B.V., Samsung Electronics Europe Logistics B.V. and Samsung Electronics Overseas B.V. requested invalidation of five Community designs held by Apple; filed on 08/09/11

  • South Korea

    • 서울중앙지방법원 (Seoul Central District Court)

      • Samsung Electronics Co., Ltd. v. Apple Korea Ltd (case no. 2011 Kahap 39552; filed on 04/21/11)

      • Apple Inc. v. Samsung Electronics Co., Ltd. (case no. 2011 Gahap 63647; filed on 06/22/11)

  • Japan

    • 東京地方裁判所 (Tokyo District Court)

      • Samsung Electronics Co., Ltd. v. Apple Japan, Inc. (case nos. 2011 (Yo) No. 22027 and 2011 (Yo) No. 22028; filed on 04/21/11)

      • Apple Inc. v. Samsung Japan Corp. (case nos. 2011 (Yo) No. 22048 and 2011 (Yo) No. 22049; filed on 06/17/11)

      • Apple Inc. v. Samsung Japan Corp. (case nos. 2011 (Yo) No. 27781 and 2011 (Yo) No. 27941; filed on 08/23/11)

      • Apple Japan, Inc. v. Samsung Electronics Co. Ltd (case nos. 2011 (Wa) 30843 and 2011 (Wa) 38969; filed on 09/16/11)

      • Samsung Electronics Co. Ltd. v. Apple Inc. (case nos. 2011 (Yo) 22079, 2011 (Yo) 22080, 2011 (Yo) 22081, and 2011 (Yo) 22082; filed on 10/17/11)

      • Samsung Electronics Co. Ltd v. Apple Japan, Inc. (2011 (Yo) 22098M; filed on 12/06/11)

  • Australia

    • Federal Court of Australia

      • Apple Inc. v. Samsung Electronics Co. Ltd. (case no. NSD1243/2011; filed on 07/28/11)

      • Samsung Electronics Co. Ltd. v. Apple Inc. (case no. NSD1792/2011; filed on 10/26/11)

Furthermore, the European Commission is formally investigating Samsung's enforcement of patents essential to wireless industry standards against Apple (case no. COMP/C-3/39.939 – Enforcement of standard essential patents in the mobile telephony sector within the meaning of Article 11(6) of Council Regulation No 1/2003 and Article 2(1) of Regulation No 773/2004; formally launched on 01/30/12).

Let me remind you: this list is not exhaustive. But as far as I can see, it's the most comprehensive one of its kind to have been published.

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:


Apple and Samsung CEOs to meet on May 21 and 22 for settlement talks in San Francisco court

On April 17, I already reported on the fact that Apple and Samsung agreed, under soft pressure from Judge Lucy Koh of the United States District Court for the Northern District of California, to meet for court-moderated settlement talks at the level of the CEOs and their chief lawyers. Judge Koh wanted the parties to meet within 90 days. The judge who will oversee this settlement effort, Magistrate Judge Joseph C. Spero, has now scheduled this settlement conference for May 21-22, 2012, starting on each day at 9:30 AM.

The meetings will take place in a San Francisco courthouse, while the litigation itself is before the San Jose division of the court. With Oracle v. Google, it was just the opposite: the case is currently being tried in San Francisco, but court-ordered settlement talks took place in San Jose. It makes sense to have parties discuss settlement in front of a judge who is not personally involved with the lawsuit. For example, one of the things Magistrate Judge Spero wants the parties to do is to provide a settlement statement until May 9 including, among other things, "a candid evaluation of the parties' likelihood of prevailing on the claims and defenses" [emphasis NOT mine]. I think this is wishful thinking because the parties won't really say that any of their claims are legally weak, no matter in front of whom these talks take place, but there's no way they would ever express even the slightest skepticism over any of their claims in front of a judge involved with the actual litigation.

These parties have disputes pending in ten countries. For a long time I said that they were suing each other in nine countries and also had a purely defensive Samsung action (against several Apple design-related rights) pending in Spain -- but in a recent filng I saw that on February 27, 2012, Apple sued Samsung in the Community Designs Court of the European Union in Alicante, Spain, for infringement, apparently seeking a decision of Europe-wide effect with respect to its design-related rights. The Alicante court has the authority to impose a Europe-wide ban on products infringing Community designs (EU-wide design rights) even on a Korean company like Samsung Electronics Ltd. (and not only on its local subsidiaries) regardless of whether or not the worldwide parent company of the Samsung Electronics group is considered to have "establishments" in a given European country, an issue that was disputed between the parties in Düsseldorf, Germany. Anyway, with Apple's infringement action in Spain, it's a fact that offensive lawsuits between the two players are now pending in ten countries, but by another count, the number is actually -- don't be shocked -- 31. Here's why: Apple's lawsuit in Alicante will relate to the entire European Union, which currently has 27 member states. Previously, there were infringement lawsuits in five of those countries (Germany, UK, France, Italy, Netherlands) and in four non-EU countries (United States, Japan, South Korea, Australia). 27 EU member states + 4 non-EU-member-states = 31.

The next post will list more than 50 lawsuits the parties have filed against each other in little more than a year.

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 27, 2012

Just discovered: Apple hit Motorola and HTC with six new patents in Miami last month

With all that's going on, it's hard to stay on top of all smartphone-related patent lawsuits. I'm doing my best but today I found out about six new patent assertions by Apple against Motorola and HTC that were discoverable, and weren't actually discovered by anyone else who would have written about them, since March 12.

In the Southern District of Florida, Apple responded to a Motorola complaint fild in January 2012, targeting new products and services (the iPhone 4S and the iCloud) over six patents previously asserted against older Apple technologies. Motorola originally wanted to bring those infringement allegations in a Miami litigation involving those same patents that has been going on since October 2010, but the judge presiding over the case at the time didn't allow this. Meanwhile a new judge has taken over and vacated that denial, and wants to explore the possibility of consolidating this cases. But at the time Apple had to respond to Motorola's new Florida lawsuit, that possibility wasn't on the horizon. Apple decided to pay back in kind by asserting its own patents from the first Florida case against some newer Motorola products and to go for further escalation by adding six patents not previously asserted against Motorola. And Apple decided to simultaneously assert them against HTC as well, for the sake of sweeping but efficient litigation.

As a result, the claims and counterclaims in that lawsuit are asymmetrical: in addition to Apple and Motorola asserting against each other old patents targeting new products, there are six new patents in play against which HTC has to defend itself in the same case. This is the first time for Apple to bring claims against two independent Android device makers in the same lawsuit.

By bringing counterclaims, Apple ensured that Motorola wouldn't be able to quickly extend a potential infringement finding against older Apple products in the first Miami case. Its counterclaims definitely slow down that second Miami action and ensure that whenever Motorola may win any favorable ruling there, Apple also has a chance of winning something.

The six new patents are quite young. The "oldest" one was granted in December 2010, while the other five patents were granted between September 2011 and January 2012. It's a clear message to the Android camp that Apple's patent portfolio is getting stronger by the day.

Four of those patents were previously asserted against Samsung, while two of them didn't previously show up in litigation. These are the six patents Motorola and HTC now have to deal with:

The next important step in Miami will be a decision on how to organize those two cases. It would probably make sense for the court to consolidate the parts that assert old patents against new products. But in that case, the parts remaining in the second Florida action would just be a six-patent Apple lawsuit against Motorola and HTC. Apple's disputes with those parties have different centers of gravity (Illinois and Delaware). I'm sure Apple would want to stay out of Delaware, where a district court stayed all of Apple's pending claims against HTC in a decision that was unusual and unfair to Apple. I guess Apple would rather litigate over those patents in the Southern District of Florida, which is reasonably fast by U.S. standards (though it would be slow by German standards), or alternatively go to the Northern District of California (a district to which HTC once wanted to move its litigation with Apple, a transfer that would have been much better for Apple than staying in Delaware). Illinois is rather unlikely because Judge Posner will take the Apple v. Motorola case pending there to trial in June, so it would be too late for a case transferred out of Florida to have any synergy effects with litigation pending in Chicago.

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:


Oracle and Google comment on (copyright-related) draft jury instructions and verdict form

On Thursday afternoon, Judge William Alsup filed his "proposed charge to the jury": draft jury instructions and a draft special verdict form for the current, copyright-focused, trial phase. This (Friday) afternoon, he will discuss these documents with the parties' lawyers, who already identified their priority issues in filings made at around midnight local time.

Since it's possible that this material still undergoes significant change (even though the judge "believes [it] adequately and fairly covers all issues actually still in play"), it's too early to discuss it in detail. I'll just focus on the items the parties addressed before and after the judge presented his proposal.

Jury "must assume" copyrightability of APIs

This is what Judge Alsup now intends to tell the jury about copyrightability:

"It is my job to decide whether or not the copyrights on the computer programs do or do not extend to protect the structure, sequence and organization of the code of the programs. I will not be able to decide this question until after your verdict. For purposes of your deliberations, you must assume that the copyrights do cover the structure, sequence and organization of the code."

He was previously inclined to simply say that the structure, sequence and organization of the API packages is copyrightable. Google objected strongly because this would lead the jury to think that Google lost on a key point.

This means Google avoided a representation of the legal situation that it didn't like. At the same time, "must assume" is probably the most that the judge could do short of saying that copyrightability is a given.

Oracle's midnight filing doesn't address this matter at all. Nor did Oracle reply in any separate pleading to Google's argument about the assumption of copyrightability. I don't know if Oracle will still raise this issue, such as at the hearing this afternoon, but at least for now it appears that Oracle can live with "must assume".

Definition of "work as a whole" currently favors Oracle

Google wants the definitio nof "work as a whole" to be as broad as possible. It made a filing dedicated to this issue even before the proposed jury instructions were presented, and two of the three priority items of its critique of the judge's proposal relate to this matter. Google complains that "[a]cross the three questions that involve the “work as a whole” concept, the Court asks the jury to apply four different works as a whole". It wants the entire J2SE platform to be the work as a whole "for all purposes" (substantial similarity, fair use, and de minimis). While Google's interest in this is understandable (the broader the work as a whole is defined, the easier it is to argue that Android's APIs are different or that Google's use of such material was allowed), at least one of its arguments doesn't make sense: "Oracle claims that it does not allow subsetting of the API libraries" -- that's simply a rule to ensure full compatibility and doesn't mean that subsets can't be relevant in an infringement context.

As far as the copyrighted original material is concerned, Oracle insists that "[t]he structure, sequence, and organization ('SSO') of the API packages exists within the API documentation as well and Google should be liable for copying it", arguing that "[r]egardless of whether the SSO is expressed in the compilable code or the API documentation, it is protectable expression in both cases".

Furthermore, Oracle reminds the judge of his holding, at the time he threw out Google's motion for summary judgment against Oracle's copyright claims, that there's a "possibility that the selection or arrangement of those names [which are uncopyrightable on their own] is subject to copyright protection". That doesn't mean Oracle could sue anyone for using a simple word like "print", but without Oracle's requested clarification, the concept of structure, sequence and organization would be too abstract -- and it's a fact that elements that are non-copyrightable on their own can form part of a larger copyrightable work (in a way, every copyrightable work consists of non-copyrightable elements).

Fair use: both parties request clarifications based on past decisions supporting some of their positions

Either party requested additional instructions with respect to Google's fair use defense. While Oracle wants those instructions "supplemented", Google objects rather strongly to the current proposal.

Google wants an open-ended definition of fair use, asking the court to insert this sentence: "You may consider whatever additional factors you believe are appropriate, on the facts of this case, to assist in your determination of whether Google's use is a fair use." While the case law on fair use is reasonably inclusive, it's not open-ended. I would be surprised if the court granted this request.

Another Google proposal is much more reasonable: "All of the fair use factors must be weighed together. No single factor compels a conclusion of fair use or no fair use." This particular wording may not be adopted, but something along those lines might be.

Google wants the assessment of "the purpose and character of the use" to include "whether such work is transformative (meaning that it adds something new, with a further purpose or different character, altering the original with new expression, meaning, or message)". A criterion like "adds something new" is a very low hurdle.

Regarding the assessment of the nature of the copyrighted work, Google wants to place some emphasis on "whether the work is fictional or mostly functional or factual". Obviously computer programs are more functional or factual than a novel or a song -- nevertheless they are copyrightable.

Oracle proposes the inclusion of part of the preamble from Section 107, the fair use section of U.S. copyright law:

"[F]air use of a copyrighted work . . . for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright."

I think this is useful context for the question of what kinds of use are fair use, even though some of this typically isn't relevant to computer programs.

Oracle also proposes a reference to one Ninth Circuit (the circuit the Northern District of California belongs to) decision for each of the four fair use factors:

  1. "Commercial use weighs against a finding of fair use."

  2. "If the original copyrighted work is creative in nature, this cuts against a finding of fair use."

  3. "The greater the quantity and quality of the work taken, the less that fair use applies. Copying may not be excused merely because it is insubstantial with respect to the infringing work."

  4. "When the defendant's use of the copyrighted work competes with the copyrighted work, then it is less likely a fair use"

Three of these four cases involve high-profile parties. Items 1 and 3 are based on a case in which Elvis Presley Enterprises was the plaintiff, and item 2 is a case in which Napster was the defendant.

Those requests for clarification may be easier to sell to the court than fundamental changes such as with respect to what the "work as a whole" should be.

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:


Former Sun chief about Google: 'immune to copyright laws, good citizenship, they don't share'

The ongoing Oracle v. Google trial continues to bring interesting material to light on a daily basis. In a single document made public by Oracle yesterday, which contains Sun-internal communication about Google in connection with Android and Java, I found several interesting statements on Google's attitude toward copyright and licenses.

On page 4 of the PDF file, Sun co-founder and long-time CEO Scott McNeely wrote to his successor Jonathan Schwartz on March 8, 2007 (you can click to enlarge the image or read the text below the image):

Here's the text again (in case the device on which you're reading this doesn't display the image in a legible resolution):

"The Google thing is really a pain. They are immune to copyright laws, good citizenship, they don[']t share.

They don[']t even call back."

In yesterday's testimony, Jonathan Schwartz tried to do as much damage to Oracle's case as he could. But that doesn't mean he always felt good about what Google was and is doing. On March 26, 2008, in correspondence with former MySQL CEO Marten Mickos (who after Sun's $1 billion acquisition of MySQL served as senior vice president of Sun), Schwartz wrote to Mickos, in the context of Google's business strategy of taking without paying, that Google also "take[s] Java for Android, without attribution or contribution."

The email thread that culminated in that statement is found on pages 18 and 19 of the PDF file I linked to above. Since everyone replied at the top (the customary way), I'm now going to show the most interesting parts in chronological order. The thread started on March 25, 2008, with Schwartz writing to ensure that Mickos had seen a CNET article on an IBM investment in EnterpriseDB (Postgres Plus). Like MySQL, Postgres is an open-source database, but under a different license (BSD, not GPL). After Mickos listed other Postgres-related facts and events, Schwartz replied with a story about what a "Google buddy" told him about how Google's senior managers internally communicated their view of different open source licenses (you can click on the image to enlarge or read the text below the image):

Here's the text again:

"Jonathan Schwartz wrote:

...was with my Google buddy over the weekend, and we got to talking abuot licenses. He made some pretty interesting comments about their internal (as communicated by senior mgrs) view of licenses.

They hate GPL, they like Apache, and they love BSD.

Just like Microsoft..."

Concerning the last part, in recent years I haven't heard anything derogatory from Microsoft about the GPL. The official position appears to be that they "disagree" with it. Anyway, this here is about Google, not Microsoft.

The GPL has a contribute-back mechanism that Google seeks to avoid. That's why it doesn't use the Java code that Oracle/Sun made available under the GPL. Unlike the GPL, the Apache license allows software published under it to be incorporated into proprietary (closed-source) software. And the BSD family of licenses are tantamount to an author of a program saying, "I hereby dedicate this work to the public domain".

Mickos then described Google's business model, including its use of open source software, in the following way (you can click on the image to enlarge or read the text below the image):

Here's the text again:

"It's funny with Google. They take (without paying):

  • the [Free and Open Source Software] code 10 million developers

  • the web contents of 100 million websites

  • the searches of 1,000 million web users

and add some magic of their own after which they sell ads on this to some 0.1 million companies. And everyone is happy."

I actually have a somewhat more favorable view. I don't see Google generally "taking" the content of websites by indexing it and generating traffic, except in some cases in which content (such as from vertical search portals) gets integrated into the search engine in a way that Google strengthens its own services at the expense of other companies' websites. I also do respect, to be clear on this, what Google itself does in terms of developing great technologies of its own. I really like many of their services (this blog is hosted by Google's Blogger service), and Android. At the same time, there are intellectual property and antitrust and even sometimes data privacy issues, and I call them out on some of that.

Schwartz had a less balanced perspective than mine. He totally agreed with Mickos and said they all (at Sun) agreed with his assessment, and went on to mention Android and Java (you can click on the image to enlarge or read the text below the image):

Here's the text again:

"I so totally agree with you. We all do.

They also take Java for Android, without attribution or contribution."

Finally, with respect to Google's attitude toward copyright and licenses, page 24 of the PDF file mentioned further above shows an email reply that Schwartz sent, on an off-the-record basis, to a New York Times reporter (click o the image to enlarge or read the text below the image):

Here's the text again -- but in chronological order, so let's start with the journalist's questions:

"On Nov 6, 2007, at 8:22 AM, John Markoff wrote:

Hi Jonathan,

how the heck is Java going to be part of the Apache distro that the Android software is being given away under? Is this a legal issue between you and Google? How come they are using Java and you aren't part of their Alliance?"

This clearly shows that the reporter was fully aware of the incompatibility of the GPL and the Apache software license as well as the fact that Sun never allowed the Apache Software Foundation to provide licensed Java APIs under the Apache license. 20 minutes later, Schwartz replied:

"Off the record..."

God knows. They didn't want us to open source Java - they've already made some stupid comments about the GPL (the license of both Java and Linux - the foundation of what they're building).

As for how they avoid those licenses, I don't know - they've shown a frankly stunning naivete about free software - even their 'alliance' seemed all over the map, with second tier hardware guys and carriers. They appear to be believing their own hype - in my mind, Apple got mindshare because they had a great product. Google's presuming their brand will carry to a phone... in my view, it actually has to *be* a phone, not an alliance. And they showed nothing...

But they did a great job of hyping Andy Rubin :-)"

That answer shows Schwartz underestimated Android at the time. He's right about what worked for Apple, but what Google did with Android also worked (in terms of appealing to application developers).

I think Schwartz was being overly diplomatic by referring to Google's approach to free software as naiveté. Scott McNealy's statement that they are "immune to copyright laws, good citizenship" was a more accurate description. In my opinion, a very consistent disregard for other people's and companies' intellectual property rights, whether it's in connection with Linux, some non-Linux GPL software, Java or most of YouTube's content, can't be attributed to naiveté. There comes a point when someone who's always naive when it benefits him is simply being reckless and possibly arrogant.

Anyway, those emails probably helped Oracle to demonstrate to the jury that Schwartz didn't genuinely believe Google had the right to make use of certain Java-related intellectual property the way it did and continues to do.

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: