The court heard the oral arguments on the Daubert motion last Thursday and rendered its decision [PDF] on the same - back to the drawing boards, Oracle. But we shouldn't be too quick to draw a blanket conclusion in favor of Google from Judge Alsup's order. Rather, we need to parse it and consider what the order says.
Judge Alsup "largely" grants Google's motion to exclude Prof. Cockburn's damages report. What the court doesn't conclude is that the report can't be repaired and resubmitted. In fact, the court invites a new report. At the same time the court clearly instructs what is and is not wrong with Prof. Cockburn's original report, to wit:
- The Oracle patents do not cover all of Java and do not cover all of Android. They only cover, if at all, select elements of each.
- Based on quoted email, Google appears to have made a conscious decision to proceed with Android, including the Java-like elements, despite the fact that they did not have a license from Sun that would have, presumably, included the patents at issue. The paper trail also evidences the fact that these technologies purportedly covered by the patents were critical to the development of Android. Those points could prove problematic for Google if it is found infringing.
- No supporting evidence was submitted with Prof. Cockburn's report to establish that the allegedly infringed claims covered the most valuable elements of Android. As a result, the report is flawed in that all of the damages relate to the entire system, not just those components purportedly covered by the patents.
- Oracle will not be allowed to apply the "entire market" rule in its revised damages report.
- The damages determination must be conducted as of the time the infringement first occurred, not at the time Android was coming to market. This only adds to the complexity of the determination, but ultimately it is not likely to impact the result tremendously. However, the important point here is that the hypothetical license negotiation to determine a reasonable royalty would be between Sun and Google, not Oracle and Google.
- Oracle may argue the point of increased damages because of potential fragmentation of the Java platform by the alleged Google infringements, but it must provide evidence of such a likelihood.
- There is a relationship between Google's ad revenues and the value of Android.
- Damages must be limited to infringement occurring within the U.S.
- The prior negotiations between Google and Sun do not establish a cap on damages.
- The damages must be determined under two alternatives: without a permanent injunction issuing and with a permanent injunction issuing. Damages would be different in each of those scenarios. Some have chosen to emphasize the court's statement: "There is a substantial possibility that a permanent injunction will be granted in the event
infringement is found, especially if willfulness is proven." We caution not to read too much into this. A substantial possibility means more likely than not, but it isn't an absolute certainty, especially in light of the eBay ruling of a few years ago.
- Prof. Cockburn's use of the Nash Bargaining Solution is not going to fly.
- The $100 million license proposal from Sun is a good starting place for the hypothetical license negotiations.
So Oracle now goes to work fixing its report. In all likelihood the damages claims will be substantially less than the numbers previously bandied about. On the other hand, there is some increased likelihood that Google could face a finding of willful infringement and an increased possibility of an injunction. None of that is certain at this time.
The revised damages report will now be due on or about September 12, so Oracle has some time. Prior to that date Oracle also has to cut down the number of claims to be tried to conform with the court's order. That should give Oracle plenty of time, but if they cannot meet that schedule, the trial will be postponed.
Similarly, Judge Alsup has made no determination at this juncture about a stay pending the outcome of the reexaminations. A stay remains a possibility. I will say, however, I would expect a stay in this case to be less likely than, for example, the Interval Licensing case because this case involves a single plaintiff and single defendant. It does not have the same complexity of issues of infringement as the Interval case.
In a separate determination [PDF], the magistrate has concluded that Google's non-mobile revenues may be relevant to a determination of damages and has, consequently, directed Google to provide that information to Oracle.
Here are the documents:
228 -
Filed & Entered: 07/21/2011
Order
Docket Text: MINUTE ORDER re [215] Letter filed by Oracle America, Inc.
Signed by Magistrate Judge Donna M. Ryu on 7/21/2011. (dmrlc2, COURT
STAFF) (Filed on 7/21/2011)
229 -
Filed & Entered: 07/21/2011
Order
Docket Text: ORDER re [215] Plaintiff's Request for Additional
Depositions and [228], granting in part and denying in part Request.
Signed by Magistrate Judge Donna M. Ryu on 7/21/2011. (dmrlc2, COURT
STAFF) (Filed on 7/21/2011)
230 -
Filed & Entered: 07/22/2011
Order on Motion for Miscellaneous Relief
Docket Text: ORDER GRANTING IN PART MOTION TO STRIKE DAMAGE REPORT OF
PLAINTIFF EXPERT IAIN COCKBURN by Judge Alsup granting in part and
denying in part [171] Motion to Strike Expert Report (whalc1, COURT
STAFF) (Filed on 7/22/2011)
231 -
Filed & Entered: 07/22/2011
Transcript
Docket Text: Transcript of Proceedings held on 7/21/11, before Judge
William Alsup. Court Reporter/Transcriber Sahar Bartlett, CSR, RPR,
Telephone number (415) 626-6060/sahar_bartlett@cand.uscourts.gov. Per
General Order No. 59 and Judicial Conference policy, this transcript may
be viewed only at the Clerks Office public terminal or may be purchased
through the Court Reporter/Transcriber until the deadline for the
Release of Transcript Restriction.After that date it may be obtained
through PACER. Any Notice of Intent to Request Redaction, if required,
is due no later than 5 business days from date of this filing. Release
of Transcript Restriction set for 10/20/2011. (Bartlett, Sahar) (Filed
on 7/22/2011)
232 -
Filed & Entered: 07/22/2011
Order
Docket Text: ORDER re [227] Letter filed by Oracle America, Inc. Signed
by Judge Donna M. Ryu on 7/22/2011. (dmrlc2, COURT STAFF) (Filed on
7/22/2011)
And here are the two orders, beginning with Judge Alsup's order regarding the expert report and testimony:
*************************
IN THE UNITED STATES DISTRICT COURT
FOR THE NORTHERN DISTRICT OF CALIFORNIA
________________
ORACLE AMERICA, INC.,
Plaintiff,
v.
GOOGLE INC.,
Defendant.
___________________
No. C 10-03561 WHA
ORDER GRANTING IN PART
MOTION TO STRIKE DAMAGE
REPORT OF PLAINTIFF
EXPERT IAIN COCKBURN
_______________________________
INTRODUCTION
In this patent and copyright infringement action involving features of Java and Android,
defendant moves to exclude the report and testimony of plaintiff's damages expert. For the
following reasons, the motion is largely GRANTED.
STATEMENT
The claims asserted in this action relate to the Java software platform, which was
described in the claim construction order (Dkt. No. 137). The seven asserted patents purportedly
cover inventions that improve the efficiency and security of Java. The copyright claim concerns
the allegedly expressive elements of source code for Java class libraries. Java was developed by
Sun Microsystems, Inc., in the 1990s, and it has become one of the world's most popular software
platforms. By using a "virtual machine," Java enabled software developers to write programs that
were able to run on a variety of different types of computer hardware. Java is commonly used on
desktop computers to facilitate compatibility with application programs distributed through the
internet. A more recent "micro edition" of the Java platform known as Java ME is used in mobile
computing devices.
The accused product in this action is Android, a software platform that was designed
specifically for mobile devices and that competes with Java in that market. Both Java and
Android are complex platforms comprising virtual machines, programming languages,
development and testing kits, software libraries, and other elements. Significantly, only part of
Java and part of Android are said to embody the asserted claims. For example, the virtual
machine concept underlying Java's "write once, run anywhere" solution is not covered by the
asserted claims and indeed was part of the prior art that predated Java. And, it is undisputed that
the Java programming language is in the public domain and anyone was free to use it without
charge, as Android does. The asserted patent claims purport to disclose only incremental
improvements to the efficiency and security of the Java system. For its part, Android uses the
Linux kernel and has many non-Java elements as well.
Google acquired Android, Inc., in August 2005 and soon began discussing the possibility
of taking a Java license from Sun for use in Android. The Android project sought to include a
virtual machine that used Java technology in an open-source format -- but for mobile
applications. Historically, Sun never "refused to license any of the Java technologies," and "the
proportion of total Java licensing costs as against total software revenues [for Sun's Java
licensees was] de minimis" (Weingaertner Exh. H at 64-65). Sun, however, seemed reluctant to
authorize an open-source implementation of Java technology, possibly for fear that it would
decrease other Java licensing revenue (Norton Exh. D).
In October 2005, following "discussions with Sun regarding Android's Open Source VM
strategy," Google's then Senior Vice President Andy Rubin remarked in an e-mail, "If Sun
doesn't want to work with us, we have two options: 1) Abandon our work and adopt MSFT CLR
VM and C# language - or - 2) Do Java anyway and defend our decision, perhaps making enemies
along the way" (ibid.). Google and Sun continued to negotiate over the next several months, but
they were unable to reach a deal.
2
In January 2006, Google internally discussed a possible co-development partnership deal
under which Java technology would become an open-source part of the Android platform. The
deal was projected to cost Google 25-50 million dollars, plus a negotiable share of revenue from
"platform-enabled mobile ads" (Norton Exh. C at GOOGLE-14-00042248). The record,
however, contains no evidence that Google actually proposed this idea to Sun.
The following month, Sun supposedly "proposed [to Google] a deal that would include
both a payment of $20 million per year for three years plus 10 percent of the revenue generated
by Google on handsets running the open source platform, capped at $25 million" per year
(Weingaertner Exh. A at ¶ 294). No documentation of Sun's proposal is in the record, but
Oracle's economist summarized it as "a proposed licensing of Java for use in Android" in an
official, conforming, Java-compatible implementation (id. at ¶¶ 294, 328, n.370). Google rejected
Sun's offer. Google supposedly made at least one more counteroffer, but the negotiations
ultimately failed (id. at ¶ 295).
With no Java license from Sun, Google evidently forged ahead with the second of
Mr. Rubin's fallback options -- "Do Java anyway and defend our decision, perhaps making
enemies along the way." Android was publicly announced in November 2007, and the first
Android phones went on sale the following year (Weingaertner Exh. S). Android-based
mobile devices rapidly grew in popularity and now comprise a dominant share of the
United States market.
In January 2010, Oracle Corporation acquired Sun and renamed it Oracle America, Inc.
Seven months later, Oracle America ("Oracle" herein) filed this action, asserting 132 claims in
seven patents as well as copyright infringement allegations. The 132 claims in suit have since
been reduced to fifty, and Oracle says it will select a fewer "triable" number for the jury trial set
for October. In August 2010, Andy Rubin received an internal email stating that the technical
alternatives to using Java for Android "all suck" and stating, "We conclude that we need to
negotiate a license for Java under the terms we need" (Oracle hearing Exh. at 74,
3
On May 21, 2011, one day after the deadline for serving "all of its damages report,"
Oracle served the expert report of Dr. Iain M. Cockburn (Weingaertner Decl. ¶1; Dkt. No. 56
at ¶¶ 8-9). Dr. Cockburn, who is a professor of finance and economics at Boston University, provided an opinion on the damages to Oracle due to the alleged patent and copyright
infringement asserted in this action (Weingaertner Exh. A at ¶¶ 1, 4). Using the Nash bargaining
solution and other economic analysis, Dr. Cockburn purported to calculate what a reasonable
royalty for the alleged infringement might be. He opined that the fair market value of a license
reached through hypothetical negotiations at the time infringement began would be "at
least $1.4 billion" and "could be as much as $6.1 billion," "depending on different assumed fact
scenarios." He further opined that "the most likely hypothetical license negotiation outcome in
this case would have been a total royalty with a net present value of approximately $2.6 billion,"
structured as "an up-front payment of $0.9 billion to $1.4 billion" plus "a share of revenues
attributable to Android . . . between 10 and 15 percent" (id. at ¶¶ 9.5-9.6). Google now seeks to
exclude Dr. Cockburn's damages testimony.
ANALYSIS
An expert witness may provide opinion testimony "if (1) the testimony is based upon
sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and
(3) the witness has applied the principles and methods reliably to the facts of the case." FRE 702.
District courts thus "are charged with a 'gatekeeping role,' the objective of which is to ensure that
expert testimony admitted into evidence is both reliable and relevant." Sundance, Inc. v.
DeMonte Fabricating Ltd., 550 F.3d 1356, 1360 (Fed. Cir. 2008).
Here, Google seeks to exclude the economic opinion testimony of Oracle's damages
expert. The parties contest a myriad of items in addition to Google's several basic challenges.
This order cannot address every point of contention. Instead, this order focuses on some of the
fundamental flaws in Dr. Cockburn's expert report that render his opinion testimony unfit for
the jury.
4
1. LACK OF RECORD SUPPORT.
Oracle now accuses Google of infringing fifty claims of seven patents. Only a triable
subset of the patent claims will be presented to the jury, and some of those may be cancelled by
the United States Patent and Trademark Office following the re-examination proceedings now
underway. But even all fifty claims certainly do not cover all of Java or all of Android.
Dr. Cockburn's analysis, surprisingly, was not tied to the specific patent claims at issue. He did
not even identify a single patent claim. Instead, he used "Java" and "Android" as large-scale
bargaining units (e.g., Weingaertner Exh. A at ¶¶ 196, 286). This approach was defective,
because Oracle has provided no basis for equating the asserted claims with either all of Java or all of Android.
Dr. Cockburn recited a litany of facts to which "other witnesses will testify" which, if true,
might support a conclusion that the asserted claims are important to Java or Android (id. at ¶¶ 19-24). But these facts are not in the sworn record. Oracle was required to serve "all of its damages report" two months ago, including any necessary support (Dkt. No. 56 at ¶9).
Supposedly forthcoming testimony that was not timely disclosed is not a proper basis for the
damages analysis. Accordingly, Dr. Cockburn's opinion was not based on sufficient facts.
2. ASSERTED CLAIMS VERSUS ALL OF JAVA.
Apart from evidentiary shortfalls, Dr. Cockburn's focus on Java as a whole runs afoul of controlling law. The reasonable royalty to be calculated is "a reasonable royalty for the use made of the invention by the infringer." 35 U.S.C. 248 (emphasis added). Java was not the invention.
Only the claims asserted were the invention.
Therefore, the hypothetical license must be limited to the asserted claims -- excluding the rest of the Java platform. That "[t]he intellectual property at issue here was generally bundled into the technologies that Sun licensed on a portfolio basis" because "Sun's practice was to license Java, not to license individual patents," does not change this statutory requirement
(Weingaertner Exh. A at ¶132). An opinion that the hypothetical negotiation would have
5
resulted in a Java license simply fights the hypothetical. In postulating a full-blown "Java license," Dr. Cockburn evaded the question he was statutorily charged with answering (id. at ¶286).
3. ASSERTED CLAIMS VERSUS ALL OF ANDROID.
Turning to the Android side of the calculus, Dr. Cockburn once again should have limited
his gaze only to the features of Android alleged to infringe the specific claims to be asserted at
trial. Instead, he assumed the entire value of Android was due to the claims asserted.
"The entire market value rule allows a patentee to assess damages based on the entire
market value of the accused product only where the patented feature creates the basis for customer demand or substantially create[s] the value of the component parts." Uniloc USA, Inc. v. Microsoft Corp., 632 F.3d 1292, 1318 (Fed. Cir. 2011) (internal quotations omitted). As
already stated, Dr. Cockburn provided no analysis showing either that the allegedly infringing features of Android create the basis for all value of the Android platform or that they substantially create the value of the non-infringing portions of the Android platform.
The fact that Java may be a critical component of Android does not justify application of
the entire market value rule. Wheels are critical to an automobile, but no one would apportion all
of the demand for a car to just the wheels.
It seems highly unlikely that Android derives its entire value from a small set of infringing
features, given its breadth. The next report must apportion the total value between the specific
infringing features versus the rest of Android.
4. DATE OF THE HYPOTHETICAL NEGOTIATION.
A reasonable royalty is derived by "envision[ing] the terms of a licensing agreement reached as the result of a supposed [bargain] between the patentee and the infringer at the time infringement began." Minks v. Polaris Indus., Inc., 546 F.3d 1364, 1372 (Fed. Cir. 2008) (emphasis added). Direct patent infringement includes making, using, selling, offering to sell, or
importing any patented invention in the United States without authorization. 35 U.S.C. 271(a).
Dr. Cockburn acknowledged that the hypothetical negotiation must take place "as of the
date the alleged infringement began," but he timed it "just before the launch of Android on
6
October 22, 2008, the date on which the first Google phone shipped" (Weingaertner Exh. A at ¶¶ 110, 119). This timing was incorrect because it ignored alleged infringement that would have taken place bringing Android to market. It also ignored the problem that determining the
date of first infringement requires a claim-by-claim analysis. The jury may well find that infringement of some patent claims began at earlier dates while infringement of other patent claims began at later dates. These variations must be taken into account.
Oracle argues that "the date of the hypothetical negotiation is the date that the infringing product was first offered for sale in the United States" (Opp. 18). This is not the law. The decisions cited by Oracle do not state otherwise. The hypothetical negotiation must be scheduled as of the time infringement began. Uniloc, 632 F.3d at 1312. Yes, in some cases, such as where a
foreign defendant expands its sales of an accused product into the United States market, the date the product is first offered for sale in the United States may in fact be the date infringement began. Where an accused product is developed and tested here in the United States, however, "use" and therefore infringement will almost always begin well before the first sale. Here,
neither Oracle nor its expert has provided support for any negotiation at the unlikely time of the first sale of an Android device. Dr. Cockburn was simply resorting to wishful thinking.
5. FIGHTING THE HYPOTHETICAL.
Upon a finding of infringement, the patentee is entitled to "damages adequate to
compensate for the infringement, but in no event less than a reasonable royalty for the use made of the invention by the infringer." 35 U.S.C. 284. "Two alternative categories of infringement compensation are the patentee's lost profits and the reasonable royalty he would have received through arms-length bargaining." Lucent Techs., Inc. v. Gateway, Inc., 580 F.3d 1301, 1324
(Fed. Cir. 2009) (emphasis added).
A reasonable royalty typically is determined from the "hypothetical results of hypothetical negotiations between the patentee and infringer (both hypothetically willing) at the time infringement began." Mahurkar v. C.R. Bard, Inc., 79 F.3d 1572, 1579 (Fed. Cir. 1996). "This hypothetical construct seeks the percentage of sales or profit likely to have induced the hypothetical negotiators to license use of the invention." Minco, Inc. v. Combustion Eng'g, Inc.,
7
95 F.3d 1109, 1119 (Fed. Cir. 1996). Certain assumptions built into this thought experiment may not be discarded in favor of the parties' subjective preferences and history. Dr. Cockburn's analysis improperly fought the hypothetical.
First, as stated, the hypothetical negotiation takes place "between the patentee and infringer . . . at the time infringement began." Mahurkar, 79 F.3d at 1579. At the time the alleged infringement began, Sun was the patentee, not Oracle. Dr. Cockburn erred in hypothesizing a negotiation between Google and Oracle, instead of one between Google and Sun.
Oracle and Sun were different companies with different interests. Injecting Oracle into the bargaining room was wrong.
Second, the license reached in the hypothetical negotiation must compensate the licensor for the actual infringement that took place -- no more and no less. See 35 U.S.C. 284; ResQNet.com, Inc. v. Lansa, Inc., 594 F.3d 860, 869 (Fed. Cir. 2010) ("At all times, the damages inquiry must concentrate on compensation for the economic harm caused by infringement of the claimed invention."). The hypothetical license therefore must be tailored to the amount and type of alleged infringement that actually occurred.
If, for example, the hypothetical license were to cover only ten patent claims, it might not portend fragmentation of the Java platform, and many of the large-scale considerations in Dr. Cockburn's analysis might simply fall away. The reasonable royalty must compensate for the infringing features, but not for non-infringing ones. It must be limited to the time period(s) and geographic region(s) in which infringement took place.
On the other hand, Google wrongly criticizes Dr. Cockburn for assuming "that Sun and Google would have negotiated an expensive, Java-incompatible license rather than a cheap, Java-compatible license" (Br. 20). Android is not Java-compatible, or so a jury could reasonably find. The hypothetical negotiation therefore may include an assumption that Google's use of the licensed rights is not limited to Java-compatible products.
6. FRAGMENTATION.
In evaluating the patentee's hypothetical bargaining position, Dr. Cockburn considered a foreseeable possibility that Google's use of such an unrestricted license would "fork" or
8
"fragment" the Java platform. For example, he explained that fragmentation would entail "[t]he potential for reduction in the size and loyalty of the Java developer community" as former Java developers chose instead to create Android applications, which could be written in the familiar Java programming language but which would not be compatible with Java systems (Weingaertner Exh. A at ¶ 159). According to Dr. Cockburn, the looming specter of fragmentation would have
provoked the patentee to demand a higher royalty and a larger up-front payment.
Google attacks this reasoning, arguing that fragmentation is not a proper element of a royalty analysis (Br. 11). Google is incorrect. At the time of the hypothetical negotiation, the parties reasonably could have foreseen the possibility of Java fragmentation as a consequence of a license for use in implementations not subject to Sun's approval. Fragmentation therefore would have been taken into account in the hypothetical negotiation, or so plaintiff is entitled to argue.
The weight of these fragmentation concerns and the degree to which they would have impacted
the negotiation back then are questions of fact.
7. ADVERTISING REVENUE.
The accused product is Android, but unlike a typical infringing product, Android is not sold. Instead, Google profits from Android indirectly. Any valuation of Android must take this business model into account. Put differently, the question is "what is the market value of Android -- what could it be bought or sold for -- and to what extent do the infringing features contribute to that overall value?"
While Google is correct that we must ultimately isolate the infringing features' contribution, Google is incorrect in asserting that the overall value of Android is irrelevant and is further incorrect in asserting that advertising revenues have nothing to do with the overall value. Of course, they do have something to do with the overall value. There is evidence, for example,
that users with Android phones "search twice as much" as users with other types of phones, increasing the advertising revenue derived from Google's search service. Additionally, for searches on Android devices, Google must share its revenue only with the device operator and not with any other intermediary (Norton Exh. G at 9). The relationship between Android and Google's advertising revenues would have been known to the parties at the time of the
9
hypothetical negotiation. Plaintiff may base its damages model on then-expected advertising revenue so long as apportionment is done.
8. WORLDWIDE REVENUE.
The parties dispute the propriety of considering international revenue in valuations of Java and Android, but the law on this point is clear (Br. 17; Opp. 17-18). United States patents are infringed only in the United States. 35 U.S.C. 271. Patent infringement damages, therefore, must compensate only for the consequences of domestic activities. The features in question can
infringe only when used in the United States. The analysis, therefore, must apportion worldwide revenue to isolate the part attributable to the features used in the United States, as those revenues would have been reasonably foreseen at the time of the hypothetical negotiation. The amount of
such then-expected revenue is a question of fact.
9. GOOGLE'S SOVIET-STYLE NEGOTIATION.
Google emphasizes that negotiations between Sun and Google regarding a Java license for the mobile space "never got anywhere near the billion-dollar level." Google recounts that "Google rejected a proposal made by Sun to pay Sun $60 million over three years plus an additional amount up to $25 million per year in revenue sharing" and suggests that the total value
of this offer -- "at most, a figure around $100 million" -- should serve as a ceiling for the reasonable royalty reached in any hypothetical negotiation (Br. 19-20). In other words, since Google rejected the offer, the rejected offer must serve as a ceiling for the hypothetical negotiation, or so Google argues.
This would be a Soviet-style negotiation: "What's mine is mine and what's yours is negotiable." The test is not what the infringer actually bargained for but what reasonable parties would have negotiated. Google may have simply been brazen, preferring to roll the dice on possible litigation rather than to pay a fair price. Moreover, by law, the hypothetical negotiation
presupposes infringement and validity of the claims in suit, whereas back in 2006 Google was entitled to a discount based upon the risks of litigation over validity and non-infringement. Google is wrong on this criticism.
10
10. PROSPECT OF PERMANENT INJUNCTION.
Google accuses Dr. Cockburn of "straight-up double counting" because his calculation included future damages even though Oracle seeks an injunction (Br. 16). Damages, of course, will be awarded only for infringement that actually occurs and possibly provable future damages.
There is a substantial possibility that a permanent injunction will be granted in the event infringement is found, especially if willfulness is proven. To account for this scenario, any damages report should address both the assumption that an injunction will be granted and the assumption that an injunction will not be granted. The hypothetical license, therefore, should be
structured as a series of yearly payments with no additional lump sum to be paid up front. This
structure should be adopted as an assumption in any damages analysis, so that the jury can more easily divide the amount of damages between past and future infringement. If and when a
permanent injunction is entered, the damages award will then be adjusted accordingly.
Furthermore, any projection of future damages must take into account the varying expiration dates of the asserted patent claims.
11. NASH BARGAINING SOLUTION.
The Nash bargaining solution is a mathematical model that purports to define the most mutually beneficial outcome of a two-party bargaining scenario. After identifying the profits each party could expect without a deal and the surplus created by their cooperation, the Nash model allocates the value of the deal in two steps: each party first receives the same profits it
could expect without a deal, and then the remaining surplus is divided evenly between them. The Nash bargaining solution relies on "a few general assumptions" that "idealize the bargaining problem." John F. Nash, Jr., The Bargaining Problem, Econometrica, Apr. 1950, at 155. Like
any mathematical model, the Nash solution cannot describe real-world behavior unless the conditions on which it is premised are satisfied in the real world.
It is no wonder that a patent plaintiff would love the Nash bargaining solution because it
awards fully half of the surplus to the patent owner, which in most cases will amount to half of the infringer's profit, which will be many times the amount of real-world royalty rates. There is no anchor for this fifty-percent assumption in the record of actual transactions.
11
Dr. Cockburn purported to use the Nash bargaining solution to "project[] bargaining
outcomes and calculat[e] the resulting payments and royalties" in the hypothetical negotiation he postulated between Google and Oracle (Weingaertner Exh. A at ¶ 281). He did not, however, adequately explain this method or tie it to facts in the record. In particular, Dr. Cockburn glossed over the axioms underlying the Nash solution without citing any evidence to show that those
assumptions were warranted in the present case (id. at ¶ 282). In this respect, his analysis was not based on sufficient facts. Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 591 (1993) ("An
additional consideration under Rule 702 -- and another aspect of relevancy -- is whether expert testimony proffered in the case is sufficiently tied to the facts of the case that it will aid the jury in resolving a factual dispute."). Because Dr. Cockburn did not provide an adequate justification for applying the Nash solution to the facts of this case, allowing him to testify to his results would
risk misleading the jury as to the soundness of the foundation for his conclusions.
The Nash bargaining solution has never been approved by a judge to calculate reasonable royalties in litigation, at least in the face of objection. This is despite the fact that for decades it has been lurking in the field of economics.*
The Nash bargaining solution involves complex mathematical formulas and equations that
would surely be incomprehensible to the average juror. For example, an article demonstrating how to apply the model to a reasonable royalty analysis includes much math, of which the following excerpt is illustrative:
12
See William Choi & Roy Weinstein, An Analytical Solution to Reasonable Royalty Rate
Calculations, 41 IDEA 49, 56-60 (2001).
No jury could follow this Greek or testimony trying to explain it. The Nash bargaining solution would invite a miscarriage of justice by clothing a fifty-percent assumption
in an impenetrable facade of mathematics. See Uniloc, 632 F.3d at 1312-15 (holding that
the 25 percent rule of thumb is inadmissible under Daubert and the Federal Rules of Evidence for
determining a baseline royalty rate in a hypothetical negotiation). Such confusing evidence will be excluded under FRE 403. No expert testimony based on the Nash bargaining solution will be
admitted. Instead, the normal Georgia-Pacific factors, which have been approved by the Court of
13
Appeals for the Federal Circuit and which are more understandable to the average fact-finder, will guide our reasonable royalty analysis. See Uniloc, 632 F.3d at 1317 ("This court has sanctioned the use of the Georgia-Pacific factors to frame the reasonable royalty inquiry.").
12. THE HYPOTHETICAL NEGOTIATION.
What we must do to stage the hypothetical negotiation is (i) to transport two fictitious persons back to an imaginary bargaining table on the date of first infringement, one person being a representative of the patent owner and the other being a representative of the infringer, (ii) to
imbue both of them with reasonableness and objectivity, (iii) to require them to be mindful of the reasonably relevant real-world business considerations then known to both sides (to the extent allowed by Georgia-Pacific), (iv) to inform them of all reasonable non-infringing alternatives that
would have been available in the absence of a license and, therefore, might have guided the negotiators, including their pros and cons, and (v) to instruct them then to negotiate in good faith a royalty fee that would fairly and reasonably reflect both sides' relevant business interests and bargaining positions so as to cover the actual infringement that eventually took place, while
taking as a given the validity and enforceability of the patent claims.
In deciding how reasonable negotiators would have acted, the trier of fact may consider "comparables" including even comparables unknown to the litigants back at the time in question. While the patents must be presumed valid, the negotiations should not be based on any premise
that the patent holder had the infringer "over the barrel" due to infringement that later occurred and, therefore, could extract a premium. Instead, the infringer must be presumed poised to start infringement, but still blessed with the option to pause and to pursue non-infringing alternatives.
Given the presence in this case of a real-world "comparable" close on point -- the last Sun offer in 2006 -- the Court is strongly of the view that the hypothetical negotiation should take that $100 million offer as the starting point and adjust as follows:
- Adjust downward for the fact that Java covers far more
than the claims in suit and therefore the 2006 offer covered
far more than the claims in suit.
- Adjust downward for the fact that Android covers far more
than the claims in suit.
14
-
Adjust upward due to the necessary assumption that all claims asserted must be deemed valid and infringed
(whereas in 2006 this was uncertain).
- Adjust upward or downward, as the case may be, for any
further changes known to the parties between the date of
the offer and the date of first infringement.
- Adjust upward or downward for other comparables and
other Georgia-Pacific factors.
-
Adjust downward for the fact that the offer included a
copyright license, an issue not addressed herein (and which
will not be addressed until the final report is done).
This order does not absolutely rule out all other formats. The foregoing would seem to be a clear-cut starting point.
CONCLUSION
On reflection, the Court erred in inviting the damages report to be submitted earlier than normal. The Court's express thinking was that allowing an earlier report would leave time to vet the analysis and then to adjust for the final report without trial delay. Instead, however, the patent owner here simply served a report that overreached in multiple ways -- each and every overreach compounding damages ever higher into the billions -- evidently with the goal of seeing how
much it could get away with, a "free bite," as it were. Please be forewarned: the next bite will be for keeps. If the next and final report fails to measure up in any substantial and unseverable way, including ways this order did not have time to reach, then it may be excluded altogether without leave to try yet again. While this order has not reached every possible criticism, Oracle must be
aware that some of the unaddressed criticisms seem, at least on one reading, to have merit, so it
should proceed with caution before overreaching again.
If Oracle believes that it cannot craft a report until it figures out what patent claims (of the many currently asserted) it will actually assert at trial, that is a problem of its own making. If Oracle needs to postpone the October trial date until it settles on which claims it truly believes are triable, then it should bring a prompt motion to do so; otherwise, Oracle's revised damages report
limited to the claims actually to be tried will be due 35 DAYS before the final pretrial conference
15
and any responsive defense report will be due FOURTEEN DAYS before the final pretrial conference. The author of the report must, of course, sit for another deposition.
With respect to a possible stay pending re-examination, the Court is yet unwilling to give
up on the October trial date. This will be revisited in due course, and in any event at the final pretrial conference. A factor will be whether this case will be truly trial-ready.
IT IS SO ORDERED.
Dated: July 22, 2011.
[signature]
WILLIAM ALSUP
UNITED STATES DISTRICT JUDGE
___________________
* Oracle cites no decision approving the use of the Nash bargaining solution to calculate a reasonable royalty in litigation. An exhaustive search has revealed only one decision that comes close. Amakua Dev. Co. v. Warner, No. 05 C 3082, 2007 WL 2028186, at *20 (N.D. Ill. July 10, 2007)
("Defendants have not challenged the reliability of Nash's theories, and the assessment of whether the
theory persuasively can be applied in the context of this case is for the jury. Again, the Court finds that the
proper method of challenging this testimony is on cross-examination."). The Nash bargaining solution is
the subject of another Daubert challenge currently pending in this district. Like defendants here, the defendants in that action argue that use of the Nash bargaining solution to calculate a reasonable royalty
runs afoul of guidance in the Federal Circuit's recent Uniloc decision. Mformation Techs., Inc. v. Research
In Motion Ltd., No. 08-CV-04990 (JW), Dkt. No. 416, at 4-6 (Mar. 28, 2011).
16
***********************************
***********************************
UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF CALIFORNIA
ORACLE AMERICA, INC.,
Plaintiff,
v.
GOOGLE INC.,
Defendant.
______________________
No. C-10-03561-WHA (DMR)
ORDER RE PARTIES’ JULY 1, 2011
JOINT DISCOVERY LETTER
Before the court is the parties’ July 1, 2011 joint discovery letter (“Letter”), in which
Plaintiff Oracle America, Inc. asks the court to order Defendant Google Inc. to produce “non-mobile
data and projections responsive to Request for Production ('RFP') Nos. 168-174.” (Letter at 1.) On
July 22, 2011, Judge Alsup granted in part Defendant’s motion to strike Plaintiff’s expert damage
report. See generally Oracle Am., Inc. v. Google Inc., No. C-10-3561-WHA (N.D. Cal. July 22,
2011) (order granting in part motion to strike damage report of plaintiff expert Iain Cockburn).
Despite Judge Alsup’s findings of fault within the report, and assuming that Plaintiff properly tethers
a revised report to its infringing claims, it remains clear that Plaintiff may take into account at least
some of Defendant’s non-mobile, American-based businesses in formulating its damages
assessment. See id. at 5-6, 8-10.
Consequently, the court hereby ORDERS that the parties shall meet and confer to resolve the
discovery disputes set forth in the Letter. The court notes that Plaintiff appears to request significant
amounts of discovery. In light of the proportionality requirements mandated in Federal Rule of
Civil Procedure 26(b)(2)(C) and the fast approaching end of discovery, the parties shall use
Plaintiff’s proposed compromise in the Letter as a starting point from which Plaintiff’s discovery
requests shall be reduced. Moreover, any non-mobile business that Plaintiff seeks to account for in
its damage report must be supported by the evidence and not be purely speculative. For example,
Judge Alsup noted that “[t]here is evidence . . . that users with Android phones ‘search twice as
much’ as users with other types of phones, increasing the advertising revenue derived from Google’s
search service.” Order at 9. Thus, Plaintiff will be allowed to seek reasonable discovery with
respect to Defendant’s search services advertising revenue. Other non-mobile discovery should be
similarly supportable. If the parties still cannot resolve their disputes, they shall file a joint
discovery letter with the court no later than July 28, 2011.
IT IS SO ORDERED.
Dated: July 22, 2011
[signature, seal]
DONNA M. RYU
United States Magistrate Judge
2
|