decoration decoration
Stories

GROKLAW
When you want to know more...
decoration
For layout only
Home
Archives
Site Map
Search
About Groklaw
Awards
Legal Research
Timelines
ApplevSamsung
ApplevSamsung p.2
ArchiveExplorer
Autozone
Bilski
Cases
Cast: Lawyers
Comes v. MS
Contracts/Documents
Courts
DRM
Gordon v MS
GPL
Grokdoc
HTML How To
IPI v RH
IV v. Google
Legal Docs
Lodsys
MS Litigations
MSvB&N
News Picks
Novell v. MS
Novell-MS Deal
ODF/OOXML
OOXML Appeals
OraclevGoogle
Patents
ProjectMonterey
Psystar
Quote Database
Red Hat v SCO
Salus Book
SCEA v Hotz
SCO Appeals
SCO Bankruptcy
SCO Financials
SCO Overview
SCO v IBM
SCO v Novell
SCO:Soup2Nuts
SCOsource
Sean Daly
Software Patents
Switch to Linux
Transcripts
Unix Books
Your contributions keep Groklaw going.
To donate to Groklaw 2.0:

Groklaw Gear

Click here to send an email to the editor of this weblog.


To read comments to this article, go here
Oracle v. Google - Google: Why the Cox and Leonard Supplemental Reports Should Stand As Is
Monday, March 05 2012 @ 04:30 PM EST

As promised, here are some of the additional documents from last week. Specifically, we provide text versions of document 759 [PDF; Text]), Exhibit A to Document 760 [PDF; Text]), and 762 [PDF; Text]). These represent Google's opposition to Oracle's motion to exclude portions of the supplemental expert report of Dr. Cox, selected portions of Dr. Cox's supplemental report, and Google's opposition to Oracle's motion to exclude portions of the supplemental expert report of Dr. Leonard, respectively.

The bottom line in these responses is that, in Google's view, Oracle is merely objecting without providing any rational argument for their objection. Google refers to it as Oracle's tit-for-tat objections. As Google points out, when the court granted Oracle/Dr. Cockburn a third attempt at a damages report, the court offered Google (through Drs. Cox and Leonard) carte blanche to respond to any new material introduced by Dr. Cockburn.

As Google points out, Cockburn, in his third attempt, introduced two new damage methodologies: "group and value" and "independent significance." In response, Dr. Cox offers a "lost profits" approach and an "infringer's profits" approach.

Oracle argued that Cox was improperly apportioning the lost profits. Google responds that the court had already allowed Cox's approach on apportionment, and that this did not constitute a change, merely an application of that apportionment approach to the lost profits method.

One of the more important aspects of the Cox supplemental report is his apportionment of infringer's profits between the allegedly infringing material and all other factors. As Google points out, where the infringement plays only a small role in generating profits, such an apportionment need not be perfect, nor does it need to be supported by evidence. Cox asserts that little or none of Google's profits are attributable to the alleged infringement. The value of Android, according to Cox, is primarily driven by its open source licensing, its open application platform, a higher sharing of revenues with application developers (than Apple), fewer block applications (than Apple) and Google's strong brand presence, and almost nothing to do with the allegedly infringing APIs. In Exhibit A to the Cox report (see Text) we see more of the explanation behind the approach Cox takes and the degree of documentation and evidence supporting that approach he provides.

The challenges posed by Oracle to the Leonard supplemental report are a bit different. Oracle attacked Leonard in part for excluding some of the claim citations (used to show reliance by others on the claims and, therefore, the value of the claims). But the claims Leonard excluded are either claims from earlier patents in the '104 chain (predecessor patents) that are not included in the '104 patent or they are citations to claims that were disallowed during the reexamination that resulted in the reissue of patent '104. Oracle just can't seem to get this concept that it is the claims that are important, not the patents themselves.

Oracle also argues that Leonard's use of forward citation analysis, i.e., how many successor patents cited back to the asserted patent's claims, is not a useful way to measure the relative value of the patents-in-suit. Yet, one of the studies Oracle itself introduced, the Harhoff study, and on which Dr. Cockburn relied concluded that "the number of . . . citations a patent receives [is] positively related to a patent's value." So Leonard is merely relying on Oracle-introduced evidence to support his approach. By the way, the Harhoff report is not an outlier; the forward citation valuation approach is widely accepted.

In the end, the court went out of its way to provide Oracle and Dr. Cockburn one last chance to put forward a proper damages report. In so doing, the court granted Google wide latitude with which to respond. Oracle now wants to have its cake and eat it too by trying to narrow the scope of the Cox and Leonard responses while seeking wide latitude for the Cockburn third attempt. We can only hope the court stands by its earlier position.


**************

Documents

759

KEKER & VAN NEST LLP
ROBERT A. VAN NEST, #84065
[email]
CHRISTA M. ANDERSON, #184325
[email]
MICHAEL S. KWUN, #198945
[email]
[address]
[phone]
[fax]

KING & SPALDING LLP
DONALD F. ZIMMER, JR. - #112279
[email]
CHERYL A. SABNIS - #224323
[email]
[address]
[phone]
[fax]

KING & SPALDING LLP
SCOTT T. WEINGAERTNER
(Pro Hac Vice)
[email]
ROBERT F. PERRY
[email]
BRUCE W. BABER (Pro Hac Vice)
[address]
[phone]
[fax]

IAN C. BALLON - #141819
[email]
HEATHER MEEKER - #172148
[email]
GREENBERG TRAURIG, LLP
[address]
[phone]
[fax]

Attorneys for Defendant
GOOGLE INC.

UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF CALIFORNIA
SAN FRANCISCO DIVISION

ORACLE AMERICA, INC.,
Plaintiff,
v.
GOOGLE INC.,
Defendant.

Case No. 3: 10-CV-03561-WHA

GOOGLE'S OPPOSITION TO
ORACLE'S MOTION TO EXCLUDE
PORTIONS OF THE SUPPLEMENTAL
EXPERT REPORT OF DR. ALAN J. COX

Dept: Courtroom 8, 19th floor
Judge: Hon. William Alsup


TABLE OF CONTENTS

   
Page
I. INTRODUCTION 1
II. ARGUMENT 1
A. Dr. Cox's alternative calculations of lost profits and infringer's
profits are directly responsive to changes Dr. Cockburn made in
his third expert report
1
B. Dr. Cox's alternative calculation of lost profits is admissible 2
C. Dr. Cox's alternative calculations of` wrongful infringer's profits
are admissible
4
1. The evidence in support of apportionment of an infringer's profits can be a reasonable
approximation
4
2. Dr. Cox's alternative analyses of` wrongful infringer's
profits rely on evidence that "may rationally be used as a
springboard" for apportionment
6
III. CONCLUSION 9

i


TABLE OF AUTHORITIES

Pages(s)
Federal Cases
Amini Innovation Corp. v. Anthony Cal., Inc.
439 F.3d 1365 (Fed. Cir. 2006)
4
Cream Records, Inc. v. Jos. Schlitz Brewing Co.
754 F.2d 826 (9th Cir. 1985)
4-8
Fiskars, Inc. v. Hunt Mfg. Co.
279 F.3d 1378 (Fed. Cir. 2002)
4
FEDERAL STATUTES
17 U.S.C. § 504(b) 2, 6, 7

ii


I. INTRODUCTION

In response to Dr. Cockburn's third expert report, Dr. Cox offered several supplemental opinions that directly responded to new material in Dr. Cockburn's latest report. Oracle seeks to exclude some of Dr. Cox's supplemental opinions, because they relate to lost profits and infringer's profits, while Dr. Cockburn's revisions concerned damages for an alleged lost license fee. But the Court granted Google's experts leave to supplement their prior reports in response to new material in Dr. Cockburn's latest report—it did not limit the manner in which Google's experts could respond to that new material. Dr. Cox's entire supplemental report is squarely within the scope allowed by the Court.

Oracle also argues that Dr. Cox's supplemental opinions are without foundation and unreliable. Dr. Cox, however, has explained the bases for all of his opinions, including the evidence upon which he relies and the methodology he uses to reach his conclusions. His supplemental opinions are well-founded and admissible.

Oracle's motion should be denied.

II. ARGUMENT

A. Dr. Cox's alternative calculations of lost profits and infringer's profits are
directly responsive to changes Dr. Cockburn made in his third expert report.
In granting Oracle leave to serve a third damages report from Dr. Cockburn, the Court also allowed Google's damages experts to revise their opinions. The Court ordered that "[o]nly revisions directly responsive to new material by Dr. Cockburn will be allowed." Order Conditionally Allowing Third Report [Dkt. 702] at 2. Oracle now argues that because Dr. Cockburn's third report does not revise his lost profits or wrongful infringer's profits numbers, Dr. Cox was not allowed to supplement his opinions on those topics, either.

Oracle is wrong. The Court did not limit Google's experts to supplemental opinions regarding the categories of damages for which Dr. Cockburn offered new opinions. Instead, the Court allowed revisions directly responsive to "new material by Dr. Cockburn." Id. (emphasis added). In his prior report, Dr. Cox twice made use of Dr. Shugan's analysis—which Dr. Cockburn had himself relied upon—to explain why Dr. Cockburn's calculations overstated

1


damages for lost profits and infringer's profits. Declaration of Meredith Dearborn in Support of Oracle America, Inc.'s Motion to Strike Portions of Alan J. Cox's Supplemental Report [Dkt. 735] ("Dearborn Decl."), Ex. A. (excerpts from revised Expert Report of Dr. Alan J. Cox ("Cox Rep.")) at 38-42, 58. Now in his new report, Dr. Cockburn offered two entirely new methodologies to calculate damages in this case, which he terms the "group and value" approach and the "independent significance" approach. Dr. Cox directly responded to these newly disclosed methodologies, in particular by incorporating material from Dr. Cockburn's new "group and value" approach to offer alternative methods of correcting Dr. Cockburn's calculations of damages, including damages based on lost profits and infringer's profits. Dearborn Decl. [Dkt. 735], Ex. B (Supplemental Expert Report of Dr. Alan J. Cox ("Supp. Cox. Rep.") ¶ 5. Accordingly, Dr. Cox's supplemental opinions regarding lost profits and wrongful infringer's profits are directly within the scope of revisions allowed by the Court. See Order Conditionally Allowing Third Report [Dkt. 702] at 2.

B. Dr. Cox's alternative calculation of lost profits is admissible.
Oracle argues that the Copyright Act does not allow apportionment of lost profits. Mot. at 12. But the Copyright Act allows the copyright owner only to recover "actual damages suffered by him or her as a result of the infringement." 17 U.S.C. § 504(b) (emphasis added). Dr. Cox's apportionment method merely seeks to limit any claimed damages to those allegedly caused by the alleged infringement, rather than allowing all damages that were allegedly caused by all of Android.

As an initial matter, Oracle's direct attack on apportionment is untimely. In his prior report, Dr. Cox performed precisely the same type of apportionment that Oracle now claims is improper. Dearborn Decl. [Dkt. 735], Ex. A (Cox Rep.) at 58. ("Applying this apportionment percentage results in a revised Java ME lost profits damages estimate . . ."; "if I were to . . . attribute the failure of Project Acadia all to Android . . . and apportion it to the alleged copyright infringement, the revised Project Acadia lost profits damages estimate . . ."). The Court granted Oracle leave only to challenge any "revision" made by Dr. Cox. Order Conditionally Allowing Third Report [Dkt. 702] at 3. Dr. Cox's choice to apportion lost profits is not a revision; the only

2


thing that Dr. Cox did was to offer an alternative way (in direct response to the new methodologies advanced by Dr. Cockburn in his third report) of calculating that apportionment. Dearborn Decl. [Dkt. 735], Ex. B (Supp. Cox Report) ¶ 44. To the extent that Oracle argues that any form of apportionment of lost profits is improper, that argument comes too late.

Further, the alternative methodology that Dr. Cox discloses in his supplemental report rests on firm foundation. Oracle argues that Dr. Cox has not explained why using the "group and value" figure overstates the value of the API copyrights. Mot. at 10-12. He has. Dr. Cox's supplemental approach makes use of the lower bound of Dr. Cockburn's apportionment figure from his new "group and value" methodology. Dearborn Decl. [Dkt. 735], Ex. B (Supp. Cox Report) ¶ 44. Dr. Cox selected the lower bound for the range because Dr. Cockburn relied on four assumptions, each of which overstates the value of the allegedly infringed APIs. Id. ¶¶ 15-20. "Because all of [Dr. Cockburn's] errors have the effect of overstating the value of the API copyrights, to the extent that his range is meaningful at all, I conclude that only the low end of that range is meaningful." Id. ¶ 21.

Oracle also accuses Dr. Cox of relying on the same reasoning that the Court rejected in striking portions of Dr. Cockburn's second report. Mot. at 9-10. But, unlike Dr. Cockburn, Dr. Cox explained why the approach he used was conservative, and thus why the "group and value" apportionment figure, which according to Dr. Cockburn is the percentage of the 2006 negotiation bundle that is due to the API copyrights, can be used to apportion lost profits in a manner that, if anything, favors Oracle. The "group and value" apportionment "purports to represent the percentage of the value of the intellectual property that Sun proposed to license in the 2006 proposal that the API copyrights represent." Dearborn Decl. [Dkt. 735], Ex. B (Supp. Cox Report) ¶ 32. The rights that Sun proposed to license in 2006, however, are only some of the inputs that created Android. See id. This means that, if anything, using the "group and value" percentage to apportion lost profits overstates the lost profits attributable to the alleged copyright infringement. See id.

This is a far cry from Dr. Cockburn's conclusory assertion that a calculation was "conservative." See Order re Google's MIL No. 3 [Dkt. 685] at 8 ("It is no answer to say, as Dr.

3


Cockburn did in footnote 327, that his approach is 'conservative.'"). Dr. Cockburn's "bewildering" footnote, id., did nothing more than state an unsupported conclusion. In contrast, Dr. Cox—in several paragraphs rather than a single "turbid" footnote, id.—has explained why his approach is conservative. Dearborn Decl. [Dkt. 735], Ex. B (Supp. Cox Report) ¶¶ 15-21, 32. If Oracle disagrees with that reasoning, that is a matter for cross-examination.

C. Dr. Cox's alternative calculations of wrongful infringer's profits are
admissible.
Oracle argues that the alternative calculations of wrongful infringer's profits in Dr. Cox's supplemental report are so unreliable that they should be stricken. Mot. at 8-10. Unlike patent damages in this case, which generally are governed by Federal Circuit law, see Fiskars, Inc. v. Hunt Mfg. Co., 279 F.3d 1378, 1381 (Fed. Cir. 2002), copyright damages are governed by Ninth Circuit law. Amini Innovation Corp. v. Anthony Cal., Inc., 439 F.3d 1365, 1368 (Fed. Cir. 2006). The Ninth Circuit has held that where "it is clear . . . that not all of the profits are attributable to the infringing material," there is a duty to apportion damages so long as "'. . . the evidence suggests some division which may rationally be used as a springboard'" for apportionment. Cream Records, Inc. v. Jos. Schlitz Brewing Co., 754 F.2d 826, 828-29 (9th Cir. 1985) (quoting Orgel v. Clark Boardman Co., 301 F.2d 119, 121 (2d Cir. 1962)).

1. The evidence in support of apportionment of an infringer's profits can
be a reasonable approximation.
In Cream Records, the Ninth Circuit addressed an appeal by the plaintiff from a judgment that awarded wrongful infringer's profits of $5,000, out of approximately $5 million in profits—one-tenth of one percent. 754 F.2d at 828. The jury found infringement based on Schlitz's use of a short passage from "The Theme from Shaft," to which Cream Records owned the copyright, in a beer commercial. Id. at 827, 828. The parties agreed that the court would decide damages. Id. at 827. The district court found that the infringement was "minimal," that the passage did not substantially add to the value of the commercial, but that the commercial must have sold some beer, and the music must have played some role in those sales. Id. at 828. The court concluded,

So I have to find some profit of the defendants which is allocable to the infringement, but, as I say, I think it's miniscule. I have interpolated as best I can. They made a profit of $5 million. One-tenth of 1 percent is $5,000, so I will add

4


that ....
Id. (quoting district court). The district court reached this conclusion even though the defendant offered no evidence on apportionment. See id.

The Ninth Circuit affirmed this portion of the judgment. Quoting the Honorable Learned Hand, the Ninth Circuit explained:

But we are resolved to avoid the one certainly unjust course of giving the plaintiffs everything, because the defendants cannot with certainty compute their own share. In cases where plaintiffs fail to prove their damages exactly, we often make the best estimate we can, even though it is really no more than a guess (Pieczonka v. Pullman Co., 2 Cir., 102 F.2d 432, 434), and under the guise of resolving all doubts against the defendants we will not deny the one fact that stands undoubted.
Id. at 829 (quoting Sheldon v. Metro-Goldwyn Pictures Corp., 106 F.2d 45, 51 (2d Cir.1939), aff'd 309 U.S. 390 (1940)). The Copyright Act does not require perfection in apportioning profits. "As to the amount of profits attributable to the infringing material, 'what is required is . . . only a reasonable approximation.'" Id. (quoting Sheldon v. Metro-Goldwyn Pictures Corp., 309 U.S. 390, 408 (1940)).

The Cream Records standard—under which the Ninth Circuit affirmed a 0.1% apportionment that was based on no apportionment evidence—not only is binding law, but good policy. Apportioning profits with precision between infringement and other factors will generally be the most difficult where, as Dr. Cox opines is the case here, the infringement played only a small role in generating profits. In such a situation, the margin of error could sometimes be larger than the actual value one is trying to calculate. Recognizing this issue, Cream Records requires apportionment even where the evidence in support arguably allows only an imperfect calculation of damages. Here, Dr. Cox has opined that little or none of Google's profits are attributable to the alleged infringement. And, relying on evidence that Oracle has itself sponsored, he has offered alternative ways of calculating the profits attributable to the alleged infringement. Dr. Cox's evidence and methodologies easily meet the Cream Records standard, under which the Ninth Circuit affirmed an apportionment of 99.9% of profits to noninfringing factors that was based on no apportionment evidence at all, but only a common-sense "interpolation" by the finder of fact.

5


2. Dr. Cox's alternative analyses of wrongful infringer's profits rely on
evidence that "may rationally be used as a springboard" for
apportionment.
Dr. Cox's alternative analyses comport with Cream Records. Dr. Cox has opined that none or close to none of Google's profits are attributable to the alleged copyright infringement, and Oracle never moved to strike that analysis. Dr. Cox's supplemental report includes alternative methods of apportionment that are based on evidence that meets the Cream Records standard, and Dr. Cox's opinions about those methods are thus admissible.

In his first expert report, Dr. Cox opined that essentially none of Google's Android-related profits are attributable to the alleged copyright infringement:

The evidence demonstrates that the success of the Android architecture is almost entirely, if not entirely, due to Google. At the very least this evidence, weighed against the evidence provided by Dr. Cockburn, indicates that Dr. Cockburn's measure of the contribution of the API claim is too speculative to merit an award of damages.
Declaration of Reid P. Mullen In Support of Google's Opposition to Oracle's Motion to Exclude Portions of the Supplemental Expert Report of Dr. Alan J. Cox ("Mullen Decl."), Ex. A (Cox Rep.) at 16. Dr. Cox explained that the open source nature of Android was a key driver of its success. Id. at 17-21. Moreover, in contrast to Apple's "app" market approach, Google gave developers a higher percentage of app revenues, and blocked fewer applications, which also drove Android's success. Id. at 19. Dr. Cox also opined that Google's strong brand was a significant reason for Android's success. Id. at 21-22. The APIs at issue, meanwhile, are only a small part of Android. Id. at 27-28. For these reasons, Dr. Cox concluded,

the contribution of the material covered by Oracle's API claim provided little in value compared to the elements contributed by Google to the success of the platform. Consequently, a low or zero damage for the alleged copyright infringement is appropriate.
Id. at 28. As mentioned above, Oracle did not move to strike this analysis.1

If the jury agrees with Dr. Cox, then it will award no damages for wrongful infringer's

6


profits (i.e. "profits of the infringer that are attributable to the infringement"). 17 U.S.C. § 5 04(b).2 In case the jury concludes that some of Google's profits are attributable to the alleged infringement, however, Dr. Cox has also offered alternative analyses the jury could use "as a springboard" for apportioning wrongful infringer's profits. Cream Records, 754 F.2d at 829.

Two of the alternatives that Dr. Cox discussed in his first expert report made use of Dr. Shugan's conjoint study, on which Dr. Cockburn had himself relied in other aspects of his own (second) report. Just as Oracle did not try to exclude Dr. Cox's analysis explaining why little or none of Google's profits are attributable to the alleged infringement, Oracle did not move to exclude these analyses in Dr. Cox's first report, which were based in part on Dr. Shugan's conjoint study.

In his supplemental report, Dr. Cox offered two further alternative approaches in direct response to, and making use of, Dr. Cockburn's new "group and value" approach. By making use of Dr. Cockburn's "group and value" range, Dr. Cox's alternative analyses use Dr. Shugan's study only to estimate the relative value of the copyright material at issue and the patents-in-suit. Dearborn Decl. [Dkt. 735], Ex. C (Cockburn Dep. Tr.) at 30:10-23 (explaining the limited role of the conjoint study in the "group and value" approach). As Dr. Cox explained in his supplemental report, by artificially limiting the universe of features tested to seven, Dr. Shugan's C 18 conjoint study "likely overstates the absolute value of those factors." Dearborn Decl. [Dkt. 735], Ex. B (Supp. Cox Rep.) ¶ 18. But in Dr. Cox's opinion, "[d]espite the problems with Dr. Shugan's analysis the conjoint analysis can offer some insight into the relative importance of having many applications versus applications that launch quickly. That is, Dr. Shugan's analysis may provide some support for the conclusion that having a large number of applications is less important that ensuring that applications launch quickly." Dearborn Decl. [Dkt. 735], Ex. B (Supp. Cox Rep.) ¶ 18.

First, Dr. Cox repeated one of the analyses from his prior report that made use of Dr. Shugan's conjoint study, but substituted the low end of Dr. Cockburn's "group and value" range

7


for the percentages Dr. Cox previously derived from Dr. Shugan's conjoint study. Dearborn Decl. [Dkt. 735], Ex. B (Supp. Cox Report) ¶ 31. As Dr. Cox explained, the apportionment from Dr. Cockburn's "group and value" range represents a different ratio than the one from Dr. Shugan's study. Id. ¶ 32. However, as noted above in connection with Dr. Cox's lost profits analysis, Dr. Cox explained why using the "group and value" approach is acceptable in this context, and if anything favors Oracle. Id. ¶¶ 15-21, 32.

Second, Dr. Cox offered an alternative analysis that did not directly substitute the "group and value" apportionment for numbers from Dr. Shugan's study. Under this approach, Dr. Cox analyzed the 2006 proposal that Dr. Cockburn has relied upon, and concluded that the 10% revenue sharing term of that proposal is the aspect of Sun's proposal that most closely resembles a running royalty. See id. ¶¶ 38-40. Although apportioning profits based on this 10% figure does not directly measure the exact portion of Google's profits that are attributable to the alleged infringement, Dr. Cox explained that this approach allows him to "estimate the proportion of the revenue attributable to the relevant APIs." Id. ¶ 37 (emphasis added). This is true because even though the 2006 bundle of intellectual property rights is not the same as the actual 2011 Android platform, id. ¶ 36, that does not mean that there is no relationship between them. Given Dr. Cox's opinion that most or all of Google's profits are attributable to factors other than the alleged infringement, he is entitled to rely on this evidence, because it "may rationally be used as a springboard" for an apportionment. Cream Records, 754 F.2d at 829.

Dr. Cox's analyses are based on far more reliable evidence than the district court's "interpolation" in Cream Records. Dr. Cox's supplemental opinions are admissible.

/ /

/ /

8


III. CONCLUSION

For the foregoing reasons, Google respectfully requests that the Court deny Oracle's Motion to Strike Portions of the Supplemental Expert Report of Alan J. Cox.

Dated: March 2, 2012

KEKER & VAN NEST LLP

BY: /s/ Robert A. Van Nest
ROBERT A. VAN NEST

Attorneys for Defendant GOOGLE INC.

9


1 Oracle did move to exclude a single factor considered by Dr. Cox—non-infringing alternatives to the APIs at issue—and the Court did strike that discussion, while denying the rest of Oracle's motion to strike. See Order re Motion to Exclude Portions of Leonard and Cox Expert Reports [Dkt. 632] at 7. Oracle did not, however, move to exclude the portions of the analysis discussed above, or Dr. Cox's ultimate opinion that the APIs at issue contributed little if anything to Android's success.

2 That is not to say that in that case the jury would award no damages at all. If the jury finds copyright infringement, and awards no damages for wrongful infringer's profits, it would still need to determine actual damages. See 17 U.S.C. § 504(b).


760, Exh. A

EXHIBIT A


UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF CALIFORNIA
SAN FRANCISCO DIVISION

ORACLE AMERICA, INC.,
Plaintiff
v
GOOGLE, INC.,
Defendant.

Case No. CV 10-03561 WHA

Revised October 21, 2011 and November 25, 2011

HIGHLY CONFIDENTIAL
SUBJECT TO PROTECTIVE ORDER


Even if I were asked to assume there was a lost license fee, I disagree with Dr. Cockburn's conclusions regarding the value of that license. I provide a more complete analysis of the licensing landscape below.

In the following sections, I will address damages in this matter due Google's alleged infringement of material covered by Oracle's API claim, since I have already determined that evidence indicates that the material covered by the file claim were easily replaced or worked around and therefore had no value. I describe the relative importance of the material covered by the alleged API claim with the contribution of Google's efforts and resources. I also consider other factors that had an impact on Google's success. Given the overwhelming importance of these other factors, I find that a reasonable remedy for Oracle's API claim is also an award of zero damages. However, I review Dr. Cockburn's copyright damages. Without accepting that a damages award is appropriate, I make adjustments to his calculations that lead to much more reasonable, though still high, damages.

I will address each of the three damages theories on which Dr. Cockburn has opined in his report — Google's allegedly wrongful profits attributable to the purported infringement, Oracle's actual lost profit damages, and Oracle's lost license fee. For each damages theory, I will address both my opinions on damages as well as my critique of Dr. Cockburn's calculation of damages.

A. Reasons for the Success of the Android Platform

I now review some of the elements of Android's success that are due to Google's contributions. The evidence demonstrates that the success of the Android architecture is almost entirely, if not entirely, due to Google. At the very least this evidence, weighed against the evidence provided by Dr. Cockburn, indicates that Dr. Cockburn's measure of the contribution of the API claim is too speculative to merit an award of damages.

16

HIGHLY CONFIDENTIAL
SUBJECT TO PROTECTIVE ORDER


1. Google's Decision to Make Android Open Source Was an Important
Contributor to the Success of the Platform.
A very important contributor to Android's success is Google's decision to make Android available as open source software.54 By making the source code available and establishing the Open Handset Alliance, Google allowed companies with a stake in an efficient mobile operating system with particular features to participate in the development of the operating systems. By participating in the development of the operating system, members generally improved the chance that the system would meet their specifications, would continue to improve, and would allow features that might be important to that company. Google described these benefits: "Our primary purpose is to build an excellent software platform for everyday users. A number of companies have committed many engineers to achieve this goal, and the result is a full production quality consumer product whose source is open for customization and porting."55

Sun recognized the benefits to open-sourcing when it decided to open source its Java SE and Java ME platforms. According to an interview with James Gosling, Vice President and Sun Fellow, he expected that open sourcing would improve collaboration among stake-holders in the development of these programs. He recognized the interest that programmers had in finding and fixing bugs. He also mentioned the ability of contributors to add new features.56 These observations are consistent with those found in the academic literature.57

17

HIGHLY CONFIDENTIAL
SUBJECT TO PROTECTIVE ORDER


These predictions were borne out. Google itself stated, "[w]e credit Android's rapid adoption to the fact that we made it available under an open source license."58 Outside observers also attributed at least part of the success of Android to open sourcing the platform. An analyst for Jeffries & Company made the point that open sourcing Android facilitated Google entering into partnerships with multiple handset manufacturers and carriers.59 Android offered handset manufacturers such as Motorola, Samsung, LG, and HTC a way to better compete with Apple's iPhone.60 Handset manufacturers were also able to customize Android for their own purposes, thereby differentiating their product offering from other manufacturers' products and providing a greater diversity of alternatives to consumers.61

The resulting adoption of Android by many mobile handset manufacturers and carriers has, in turn, accelerated Android's growth.62 The mobile handset manufacturers and the carriers have furthered this growth by making a wide range of Android handset models available to consumers, with some at low prices, encouraging first time smartphone purchasers.63

Google also set up the Android Market to facilitate distribution of applications. Many of the major applications for the iPhone are also now available for Android.64 Android rivals the iPhone's developer base because Android is less costly for developers and the open source nature of the Android platform is attractive to developers.65 Indeed, a Bank of America Merrill Lynch analyst quoted, "Google charges a lower fee at $99/year than Apple ($299/year) for developers

18

HIGHLY CONFIDENTIAL
SUBJECT TO PROTECTIVE ORDER


to deploy their apps. Fewer hurdles to write apps for the Android platform can potentially lead to quicker uptake among developers, in our opinion."66 Android's Applications Framework (which is largely programmed in C++) also provides an easy-to-work-with environment for developers.67

The development environment Android has created for its developers has been well received. A review from one mobile developer summarizes, "Android's platform and developer tools are excellent."68 Another review confirms, "Android has been gaining the lion's share of interest." It explains, "Android is offering developers a bigger slice of the financial pie and will also block fewer applications, a practice which is winning Apple no friends. If Android can build a solid developer network in the next year [2010], it has a real chance of success."69

Android's managers correctly predicted that an open source operating system would be an attractive feature to application developers. In August 2005, the Google Wireless team wrote, "the plan is to beat Microsoft and Symbian to volume by offering an Open Source handset solution."70 In 2009, when Android saw its market share rise from 0.3% in 2008 to 4.1% in E 200971, an industry analyst commented, "open source will be key to the growth of mobile platforms, in terms of opening up users to new applications and keeping costs down for manufacturers."72 The New York Times compared Android to Apple's iOS, noticing, "Unlike Apple, Google has eschewed a review process, allowing any developer to publish an application to the Android Marketplace, its version of the App Store, instantly."73 Google later commented in a 2010 Android Quarterly Review presentation:

19

HIGHLY CONFIDENTIAL
SUBJECT TO PROTECTIVE ORDER


We credit Android's rapid adoption to the fact that we made it available under an open source license ... Because Google was historically seen as a threat to operators, giving up control was a key component of operators adopting Android. This is one reason Android is considered one of the most commercially successful Linux distribution.74
Dr. Cockburn often ignores discussions of the importance of the open source feature, even in the material that he quotes. For instance, in paragraph 475 of the Cockburn Report, he quotes Andy Rubin as saying: "There is no purpose of building an open platform other than to attract third-party developers to it. So anything that we would do to jeopardize the support of third-party developers would be bad for the success of the platform."75 Dr. Cockburn also cites the following statement by Mr. Rubin: "Third-party developers contribute to the success of a platform by having their companies invest in the platform by basing their businesses on the platform. It was my intention to create an independent third-party developer ecosystem ..."76 While Dr. Cockburn claims that Mr. Rubin is talking about Java when he makes the statements, it appears clear that Mr. Rubin is talking about the concept of preserving advantages of the open source format. From my conversations with Mr. Rubin and other Google employees, it is clear that they believe that it would have been as easy for Google to use other programming languages to develop the Android architecture and that there was a very large community of programmers able to write applications in those alternative languages. This is borne out by the fact that most of the most popular applications are written largely in C or C++, as I show below. The use of the Java programming language does not appear to have enhanced the creation of an open source community of programmers building applications for the Android.

Outside commentators also pointed to the importance of the open source feature to Android's success. For instance, an analyst at Merrill Lynch commented:

While the proprietary operating systems of Apple and RIM have been very successful, trends suggest that the open source system could have leading consumer market share. Android delivers value by providing: 1) effective integration between hardware, software and applications, 2) a mass market platform that reaches across devices and carriers, [3] a large developer network with growing barriers to entry. With wide adoption including more than 20

20

HIGHLY CONFIDENTIAL
SUBJECT TO PROTECTIVE ORDER


manufacturers producing more than 50 phones (up from just three models a year ago), we expect Android phones to continue to gain market share."77
Another industry publication "Informa Telecoms & Media" was quoted as saying: "The analyst firm said in a review of the state of the mobile phone operating system market that open source will be key to the growth of mobile platforms, in terms of opening up users to new applications and keeping costs down for manufacturers."78

It is also important to remember that the concept of open source software for mobile handset operating systems was a new and unaccepted concept for carriers and therefore a risky proposition for Google: "The concept of a true ecosystem, in which the roles of the various members remain in collaborative balance to achieve mutual benefit by the overall health of the system, is alien to this traditional mobile value system."79 Android was taking a considerable risk in developing its technology, a risk that was not borne by Sun or Oracle and which would have been the same had Google used C, C++, or Objective C as the programming language used in the Application Framework or the applications.

2. Google's Own Brand Name Created a Great Deal of the Excitement and Public
Acceptance around Android.
In addition to the appeal of being open source, the success and adoption of the Android platform also benefited from the Google brand name. For example, one industry analyst stated: "Google has proved that its own brand carries enough leverage to garner support from developers without relying on the Java brand. Google is one of only a handful of technology companies that have the marketing muscle to use this strategy successfully."80 Another industry analyst commented on the fact that Android handsets were still appealing despite the higher "price tag": "Consumers (and professional users demanding fancier phones from their IT

21

HIGHLY CONFIDENTIAL
SUBJECT TO PROTECTIVE ORDER


The evidence that other programming languages represent a very acceptable non-infringing alternative is very strong. Languages other than Java provide important applications for the Android platform, Android was written in C++ in its early stages, Java represents a small part of the Android architecture and other smartphone operating systems are written in languages other than the Java programming language. Dr. Cockburn, on the other hand, does not appear to be able to find evidence to support his claim that the Java programming language was important to developing a community of application developers, except to quote Dr. Mitchell.102 At one point Dr. Cockburn even quotes a single page of a document for the proposition that Java is important to Google.103 The quotation he cites refers to the rapid growth of the application store, not to Java. In fact, in the entire 40 page document from which Dr. Cockburn extracted that quotation, the word Java does not appear.

If Google had chosen C++ or another alternative to the Java programming language, it would have had to create tools for applications developers.104 On the other hand, it would not have had to create the Dalvik virtual machine. The incremental cost and time to Google of going with C++ or another programming language other than the Java programming language would not have been significant.105 In light of this non-infringing alternative, Google would not have been willing to pay much to obtain a copyright license to the APIs. As a result, this reduces the benefits, if any, that could be considered attributable to Google's alleged copyright infringement.

5. The APIs at Issue Contribute a Small Part of the Functionality of the Dalvik
Virtual Machine
The APIs that Oracle claims are derivative provide only a small part of the functionality of the Dalvik virtual machine, which in turn is only one part of Android, as was illustrated above. All of the code that enables the Dalvik virtual machine to function is Google code, written by Google (with the exception of the handful of copied files that are so inconsequential and unimportant that they have been removed or rewritten). All of the code for the Dalvik virtual machine is written in C. Similarly, the vast majority of Android that is not the Dalvik virtual machine was developed independently by Google or acquired elsewhere (and in any event is not

27

HIGHLY CONFIDENTIAL
SUBJECT TO PROTECTIVE ORDER


accused of infringement). I understand that Android phones use the Dalvik virtual machine only about a third of the time, when they are running the Java-language applications that have been converted to bytecode. Applications written in native code, which I have shown are the majority of the most popular applications, use technology that is not accused of infringing. Similarly, making phone calls, or browsing the Web, or any number of other activities, do not use technology that is accused of infringing Oracle's API claim. Moreover, the 37 API packages at issue are only a portion of the 150 API packages in the Android core libraries. Many of the other API packages—the ones not at issue—address important functions, such as interacting with the user via a touch screen.

B. Google's Allegedly Wrongful Profits Derived from the Android Platform

1. Overview
The previous review of just some of the large amount of evidence indicates that Google's efforts, business decisions, and brand drove the success of the Android platform. This evidence also demonstrates that the contribution of the material covered by Oracle's API claim provided little in value compared to the elements contributed by Google to the success of the platform. Consequently, a low or zero damage for the alleged copyright infringement is appropriate.

In light of this finding, I now turn to a review of Dr. Cockburn's damage analysis of Google's allegedly wrongful profits attributable to the purported infringement, Oracle's actual lost profit damages, and Oracle's hypothetical license fee.

Dr. Cockburn states that, when calculating allegedly wrongful profits attributable to the purported infringement damages in a copyright infringement case, it is the burden of the plaintiff to prove only the alleged infringer's revenue as the basis of a damage claim and that it is up to the alleged infringer "to prove his or her deductible expenses and the elements of profit attributable to factors other than the copyrighted work."106 Dr. Cockburn claims, in his determination of Google's allegedly infringing revenues, the entire amount of what Google reports on its Android Profit and Loss statements as Gross Android Ad Revenues, Nexus Phone

28

HIGHLY CONFIDENTIAL
SUBJECT TO PROTECTIVE ORDER


are considered, the lost license fee would be $24.1 million, but if the analysis is restricted to sales of accused models only, the lost license fee would reduce to $3.1 million.

(signature)
Alan J. Cox

Dated: October 3, 2011

(Revised October 21, 2011 and November 25, 2011)

65

HIGHLY CONFIDENTIAL
SUBJECT TO PROTECTIVE ORDER


54 I understand that although Java and Android have both been released under open source software licenses, Java is licensed under the GNU General Public License version 2 (with an exception known as a "linking exception"), and most of the Android platform (other than the Linux kernel) is licensed under the Apache License, version 2.0. The GPL is a so-called "copyleft" license that requires any modifications in distributed products to be made available in source code form under GPL terms. The Apache license is "permissive," which allows any developer (such as a handset manufacturer) to incorporate the software in products that are licensed under licensing terms of its own choice. Most handset manufacturers include significant proprietary software in their products, so the more flexible and permissive approach of the Apache 2.0 License helped reduce the intellectual property licensing issues that manufacturers faced in adopting the Android platform. Also, the Apache 2.0 License corresponded to a standard "Contribution License Agreement," which Google also adopted for the Android project. This standard contribution agreement gave handset manufacturers comfort when analyzing the intellectual property licensing terms that applied to contributions they made to the Android project. This, in turn, resulted in robust cooperation among various manufacturers.

55 Ex. AE - About the Android Open Source Project _ Android Open Source.pdf.

56 James Gosling on Open Sourcing Sun's Java Platform Implementations, Part 1.

57 Lerner, Josh, and Jean Tirole, 2002. "Some Simple Economics of Open Source," Journal of Industrial Economics, 52(2), 197-234 and West, Joel and Siobhan O'Mahony, "The Role of Participation Architecture in Growing Sponsored Open Source Communities," Industry & Innovation, 15, 2 (April 2008): 145-168, Lakhani, Karim R. and Eric von Hippel, (2003) "How Open Source Software Works: "free" User-to-user Assistance" Research Policy 32 (2003) 923-943.

58 "Google Android, OC Quarterly Review — Q4 2010," October 12, 2010, GOOGLE-01-00053552-591 at 562.

59 "Android on Steroid: Google Enters Mobile Market with a Splash; Main. Buy," Jefferies & Company, Inc., September 24, 2008, p. 4.

60 "Android: On a Bender, Telecom Equipment," Arete Research Services LLP, July 14, 2010, GOOGLE-01-00049780-784 at 780.

61 "NARRATIVE: Mobile + Android, last updated: 03/18/08," GOOGLE-23-00000001-027 at 003. Also see "Android 101: An Introduction to Android and Android Partnerships, Last Updated: December 2008," GOOGLE-00298438-484 at 476.

62 "Android Poised to Overtake Symbian as Smartphone Leader," VNUNet United Kingdom, January 13, 2011; "Google's Android Mobiles Overtake Global iPhone Sales," Financial Times, August 12, 2010; "Finding Value in Android," Bank of America Merrill Lynch, September 9, 2010, GOOGLE-01-00048436-484 at 442.

63 "Android Poised to Overtake Symbian as Smartphone Leader," VNUNet United Kingdom, January 13, 2011; "Android; On a Bender, Telecom Equipment," Arete Research Services LLP, July 14, 2010, GOOGLE-01-00049780-784 at 780.

64 "Android is Shaking Up the Market," Sunday Business Post, October 24, 2010.

65 "Making Sense of a Fragmented World: Mobile Developer Economics 2010 and Beyond, Insights and Analysis from the Definitive Mobile Developer Survey Plus Benchmarks on the Platform Development Experience," VisionMobile Ltd, July 2010, p. 11.

66 "Finding Value in Android," Bank of America Merrill Lynch, September 9, 2010, GOOGLE-01-00048436-484 at 471-472.

67 Interview of Tim Bray; "NARRATIVE: Mobile + Android, last updated: 03/18/08," GOOGLE-23-0000000]-027 at 003. Also see "Android 101: An Introduction to Android and Android Partnerships, Last Updated: December 2008," GOOGLE-00298438-484 at 466.

68 David Green, "Android vs. iPhone Development: A Comparison," Javalobby, July 6, 2009.

69 Iain Thomson and Shaun Nicholos, "Top 10 highs and lows of 2009," VNUNET United Kingdom, December 24, 2009.

70 Google Engineering Operation Plan for Q3-2005, Q4-2005, Q1-2006, GOOGLE-01-00062240-48 at GOOGLE-01-00062241.

71 "Global Smartphone Sales Forecast by Operating System and Region," Strategy Analytics, January 2011.

72 Iain Thomson, "Android tipped to overtake iPhone by 2012," VNUNET United Kingdom, March 6, 2009.

73 Jenna Wortham, "Apple's Game Changer, Downloading Now," The New York Times, December 6, 2009.

74 Google Android, OC Quarterly Review — Q4 2010," October 12, 2010, GOOGLE-01-00053552-591 at 562

75 Cockburn Report, ¶ 447 Citing Rubin 4/5/2011 Dep. 91:19-23.

76 Ibid. Citing Rubin 4/5/2011 Dep. 24:21 — 25:2.

77 "Finding Value in Android," Bank of America Merrill Lynch, September 9, 2010, GOOGLE-01-00048436-484 at 444.

78 "Android Tipped to Overtake iPhone by 2012," VNUNet United Kingdom, March 6, 2009.

79 "Google's Android Ambition is to Reshape the Mobile Industry, Report Says; But Android Faces Big Problems, Tight Deadlines, Says Report," Network World Fusion, January 3, 2008.

80 Adam Leach, "Smartphone platform profile: Google Android," Ovum, October 9, 2009.

102 See, for instance, Mitchell Report, ¶ 446.

103 Footnote 602 of the Cockburn Report.

104 Interviews of Andy Rubin, Dan Bornstein.

105 Interviews of Andy Rubin, Dan Bornstein, Brian Swetland.

106 17 U.S.C. § 504(b)


762

KEKER & VAN NEST LLP
ROBERT A. VAN NEST, #84065
[email]
CHRISTA M. ANDERSON, #184325
[email]
DANIEL PURCELL, #191424
[email]
[address]
[phone]
[fax]

KING & SPALDING LLP
DONALD F. ZIMMER, JR. - #112279
[email]
CHERYL A. SABNIS - #224323
[email]
[address]
[phone]
[fax]

KING & SPALDING LLP
SCOTT T. WEINGAERTNER
(Pro Hac Vice)
[email]
ROBERT F. PERRY
[email]
BRUCE W. BABER (Pro Hac Vice)
[address]
[phone]
[fax]

IAN C. BALLON - #141819
[email]
HEATHER MEEKER - #172148
[email]
GREENBERG TRAURIG, LLP
[address]
[phone]
[fax]

Attorneys for Defendant
GOOGLE INC.

UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF CALIFORNIA
SAN FRANCISCO DIVISION

ORACLE AMERICA, INC.,
Plaintiff,
v.
GOOGLE INC.,
Defendant.

Case No. 3:10-CV-03561-WHA

GOOGLE'S OPPOSITION TO
ORACLE'S MOTION TO EXCLUDE
PORTIONS OF THE SUPPLEMENTAL
EXPERT REPORT OF DR. ALAN J. COX

Dept.: Courtroom 8, 19th Floor
Judge: Hon. William Alsup


TABLE OF CONTENTS

     
Page
I. INTRODUCTION 1
II. ARGUMENT 2
A. Dr. Leonard's forward citation analysis is reliable 2
B. Dr. Leonard's use of Oracle's estimation of the fair value of
Sun's "core technology" assets is reasonable
7
C. Dr. Leonard's economic, not legal, opinion concerning what
would have been important in the Sun-Google negotiations is
admissible
11
III. CONCLUSION 12

i


TABLE OF AUTHORITIES

Pages(s)
Federal Cases
Hammes v. Yamaha Motor Corp. USA., Inc./i>
CIV. 03-6456 2006 WL 1195907 at *6 (D. Minn. May 4, 2006)
8
ID Sec. Sys. Canada, Inc. v. Checkpoint Sys., Inc.
249 F. Supp. 2d 622 amended, 268 F. Supp. 2d 448 (E.D. Pa. 2003)
7
In re TMI Litig.
193 F.3d 613 (3d Cir. 1999) amended, 199 F.3d 158 (3d Cir. 2000)
Ruff v. Ensign-Bickford Indus., Inc.
168 F. Supp. 2d 1271 (D. Utah 2001)
5
Ruiz-Troche v. Pepsi Cola of Puerto Rico Bottling Co.
161 F.3d 77 (1st Cir. 1998)
5
FEDERAL STATUTES
35 U.S.C. § 251 3
OTHER AUTHORITIES
Elizabeth Webster & Paul H. Jensen, Do Patents Matter for Commercialization?, 54 J. L. &
Econ. 431 (2011)
4

ii


I. INTRODUCTION

Once again, Oracle has filed a tit-for-tat Daubert motion purporting to challenge aspects of the analysis of Google's damages expert, Dr. Gregory Leonard. In its 10-page attack on Dr. Leonard's 13-page Supplemental Report, Oracle attempts to discredit three points made by Dr. Leonard. Each of these challenges is baseless.

First, Oracle's criticism of Dr. Leonard's use of forward citation analysis in rebutting Dr. Cockburn's valuation of the patents-in-suit is completely off-base. Oracle's primary criticism is that Dr. Leonard rigged his analysis by omitting citations to the predecessors of the patents-in-suit. This is nonsense. The only one of the patents-in-suit Oracle identities specifically as having predecessors is the twice-reissued '104 patent, and Oracle neglects to mention that the claims in the predecessors are entirely distinct from the claims in the '104. Indeed, not one of the claims from the predecessors remains in the '104. The '104 therefore covers a different range of technology, and counting the citations from those patents would distort the value of the '104. More generally, Oracle cannot—and does not—dispute that forward citation analysis is a widely recognized method of estimating the economic value of a patent, regardless of whether Dr. Cockburn chose to use it in his own report. Oracle also criticizes Dr. Leonard for failing to adjust citation counts for the age of each patent, but this is simply untrue. As even Dr. Cockburn recognized, Dr. Leonard did, in fact, explicitly control for the age of each patent by comparing their forward citation counts to other patents in the same technology class that were issued within three years before or after the patent in question (i.e., of approximately the same age). To the extent that Oracle quibbles with this method for controlling for age, that is at most a matter for cross-examination. Moreover, any quibbles Oracle has are baseless. The conclusion of the forward citation analysis—that the patents-in-suit are in the middle of the pack—does not change if a one- or two-year window is used instead of a three-year window, or if an alternative adjustment for patent age is made.

Second, as it did in its attempt to strike Dr. Leonard's earlier report, Oracle attempts to turn a Daubert proceeding into an evidentiary dispute. Oracle argues that Dr. Leonard was not entitled to rely on a document [REDACTED]

1


[REDACTED] This argument is a disagreement about what evidence Dr. Leonard should have relied on, and the Court should reject it. Similarly, because the document was prepared by Oracle as part of a submission to federal regulators and investors, and valued Sun's intellectual property, including the IP at issue in this action, this document is very different from the Sun v. Microsoft settlement.

Third, Oracle misconstrues Dr. Leonard's economic opinion about the considerations that would have driven Google in a negotiation as a misstatement of patent damages law. Dr. Leonard is not a lawyer and was not purporting to state (and did not cite to) any legal rule. Dr. Leonard's opinion is only that, in a hypothetical negotiation, Google would have evaluated how much to pay based on its own projected gains or losses, because no businessperson would ever make a deal based on whether the other party would make out well. Dr. Leonard made no statement excluding other evidence from consideration in resolving the hypothetical negotiation, and indeed evaluated all the appropriate evidence under the Georgia-Pacific factors in his initial expert report.

Finally, the Court should dismiss Oracle's suggestion that the Court must apply a higher level of Daubert scrutiny to Dr. Leonard's report because Oracle is not permitted to take a second deposition of Dr. Leonard or offer further rebuttal testimony from Dr. Cockburn. Motion to Strike at 2. That Oracle would even raise this complaint shows chutzpah. Oracle has already deposed Dr. Leonard once, and had a full opportunity to rebut Dr. Leonard's initial report. As it knows, it has only itself to blame for limitations on further depositions or rebuttal, because Dr. Leonard has only offered a supplemental report as a result of Oracle's and Dr. Cockburn's repeated failures to identify a viable damages methodology. That is no reason to punish Google or its experts. Oracle will have the opportunity to cross-examine Dr. Leonard at trial, where it can question Dr. Leonard concerning any of the issues in its Motion to Strike.

II. ARGUMENT

A. Dr. Leonard's forward citation analysis is reliable.
In discussing why the three patents-in-suit that Oracle's engineers ranked among the top in the entire Sun Java mobile portfolio (the '104, '205, and '720 patents) are not the three

2


most valuable patents among those 22, Dr. Leonard considered each patent's "forward citation count"—the number of prior-art citations each of those 22 patents received from future patents. Based on the number of forward citations received by each of the 22 patents, Dr. Leonard concluded that the three patents-in-suit "are not the three most valuable patents of the 22, as Dr. Cockburn assumes in determining the top of the range for his apportionment percentage? Declaration of Beko Reblitz-Richardson in Support of Oracle America, Inc's Motion to Strike Portions of Gregory Leonard's Supplemental Report ("Richardson Decl.") EX. A (Supp. Leonard Rep.) at 7. Instead, he concluded that the "patents-in-suit are worse than the middle of the set of patents. This, in turn, suggests that only the lower bound on Dr. Cockburn's range has any support." Id. Dr. Leonard thus used these studies for the limited purpose of questioning Dr. Cockburn's assumption that the three patents-in-suit are the three most important patents of the entire set of 569 examined by Dr. Reinhold and his team.

Oracle offers three objections to Dr. Leonard's analysis, none of which has merit.

First, Oracle offers the astonishing argument that Dr. Leonard rigged the game by failing to include forward citations to predecessors of the patents-in-suit, thereby undercounting total citations and undervaluing those patents. But the only one of the patents-in-suit Oracle calls out as purportedly undervalued is the twice-reissued '104 patent. And as to the '104, as Oracle must have known, there is a very good reason why Dr. Leonard did not include citations to the '104's two predecessors—neither the first patent in the chain (U.S. Patent 5,367,685) nor the second (U.S. Patent RE36,204E1) included any of the claims of the '104 patent that are asserted against Google in this case. Compare Declaration of David Zimmer in Support of Google's Opposition to Oracle America, Inc.'s Motion to Strike Portions of Gregory Leonard's Supplemental Report ("Zimmer Decl.") Ex. A ('104) with Zimmer Decl. Ex. B ('204) and Zimmer Decl. Ex. C ('685). Not only is the '104 different in scope from the '685, but it is a disputed issue in this case as to whether the '104 is sufficiently broader than the '685 such that it is invalid under 35 U.S.C. § 251.1 See Order Denying Leave to File Motion for Summary Judgment on '104 Patent [Dkt.

3


No. 327]. In other words, Oracle is making the same mistake it has made throughout the damages phase of the case, by thinking in terms of the patents instead of the specific claims of those patents. In fact, Sun itself forfeited the ten claims in the '685 patent as part of the reissue proceedings. Whatever Sun's reason for forfeiting those claims, Oracle cannot now argue that the value of the '104 is based on citations to ten claims that are nowhere to be found in the '104. Not only did Dr. Leonard analyze the '104's citations correctly, it would have been an error for him to do as Oracle now demands, by counting citations to other, irrelevant patents that include claims abandoned by the patent holder and that do not include the claims that Google is accused of infringing.

Second, Oracle argues that forward citation analysis is not a "useful" way to measure the relative value of the specific patents-in-suit, Motion to Strike at 3, but there is a large body of literature using exactly that analysis for this precise purpose. For example, as Dr. Leonard noted in his report, the Harhoff study on which Dr. Cockburn relied in his apportionment analysis, and which had the goal of "generat[ing] an assessment of the value of patent rights from publicly available data," concluded that "the number of . . . citations a patent receives [is] positively related to a patent's value." Zimmer Decl. Ex. D (Harhoff Study) at 1344, 1345. Numerous other published articles have reached similar conclusions. See Elizabeth Webster & Paul H. Jensen, Do Patents Matter for Commercialization?, 54 J. L. & Econ. 431, 453 n. 17 (2011) ("A number of studies have found that both technological significance and commercial value are correlated with U.S. Patent and Trademark Ofhce (USPTO) forward citations.").2

4


Oracle attempts to get around this widely-held understanding that forward citation counts are relevant to a patent's value by pointing out that Dr. Cockburn chose not to use forward citation analysis, and by citing a single article that to some extent critiqued the predictive value of forward citation analysis. However "Daubert neither requires nor empowers trial courts to determine which of several competing scientific theories has the best provenance." Ruiz—Troche v. Pepsi Cola of Puerto Rico Bottling Co., 161 F.3d 77, 85 (1st Cir. 1998); see also In re TMI Litig., 193 F.3d 613, 682 (3d Cir. 1999) amended, 199 F.3d 158 (3d Cir. 2000) ("However, this apparent dispute between the two collaborators does not render Gunckel's methodology unreliable. The dispute goes to the weight to be afforded Gunckel's expert opinion, not the reliability of his methodology."); Ruff v. Ensign-Bickford Indus., Inc., 168 F. Supp. 2d 1271, 1283 n. 11 (D. Utah 2001) ("The Court need not choose between competing theories to satisfy its gatekeeping function."). Oracle cannot prevent Dr. Leonard's use of forward citation analysis from going to the jury just because Oracle can identify one article disagreeing with the numerous studies concluding that forward citations are correlated with economic value. And Oracle certainly cannot prevent the jury from hearing the results of Dr. Leonard's forward citation analysis because its own damages expert, who this Court has previously found guilty of "overreach[ing] in multiple ways [to] compound[] damages ever higher into the billions," Dkt. No. 230 at 15, made a decision to value the patents through subjective metrics less reliable than forward citation analysis.

Third, Oracle's accusation that Dr. Leonard "makes no effort to control for the obvious effect of time," Motion to Strike at 5, is just false. In addition to ranking the 22 alleged top patents based on the simple total of forward patent citations, Dr. Leonard also noted in his report that "[e]ven more informative is the number of forward citations for a patent relative to the average number of forward citations for a patent in the same class as the patent in question." Richardson Decl. Ex. A (Leonard Supp. Rep.) at 7. Under this analysis, Dr. Leonard explicitly

5


accounted for the age of patents in his analysis of forward citation counts by only comparing each patent to patents in the same class that were issued either 3 years prior to or 3 years afer the date of issuance of the patent being analyzed. He then calculated the percentage of patents within this comparison group that had an equal or greater number of citations. lt is disturbing that Oracle would tell the Court that Dr. Leonard did not perform such an analysis, particularly when Oracle attaches as Exhibit C to its Richardson Declaration Dr. Leonard's own chart showing this time-adjusted ranking. Richardson Decl. Ex. C. Even Dr. Cockburn discussed Dr. Leonard's relative rankings in his Declaration supporting Oracle's motion. Declaration of Iain M. Cockburn in Support of Oracle America, Inc.'s Motion to Strike Portions of Gregory Leonard's Supplemental Report ("Cockburn Decl.") ¶¶ 8-10. Despite knowing that Dr. Leonard conducted this analysis, and attaching it to its filing, Oracle falsely asserts in its brief that the analysis never happened. Motion to Strike at 5.

Dr. Cockburn's criticisms of Dr. Leonard's relative citation count ranking in his declaration, which appear nowhere in Oracle's brief, are largely incoherent. Dr. Cockburn's argument appears to be that within each comparison group, older patents have more citations than newer patents. Cockburn Decl. at ¶¶ 8-10. But this is why each patent is compared to a group of patents that stretches three years forward and backward from that patent's issuance date—in order to mitigate exactly that effect. Declaration of Gregory Leonard in Support of Google's Opposition to Oracle's Motion to Strike Portions of Gregory Leonard's Supplemental Report ("Leonard Decl.") ¶ 4. By comparing each patent's citation count to the counts of patents in a group of older and newer patents, Dr. Leonard was able to control for the effects of time on citation count. Id. Oracle is free to raise this issue on cross-examination, but it is not a methodological flaw implicating Daubert.

Oracle also states, without any evidence or supporting analysis from Dr. Cockburn, that Dr. Leonard's alleged failure to adjust the citation counts for patent age "has a substantial impact on the outcome of the ranking." Motion to Strike at 5. As already discussed, Dr. Leonard did adjust for patent age, as Oracle knows. But in any event, Dr. Leonard's conclusion that the three patents-in-suit "are worse than the middle of the set of 22 patents," Richardson

6


Decl. Ex. A (Leonard Supp. Rep.) at 7, is completely resilient to different methods of accounting for the patents' age. For example, if the size of the window used in Dr. Leonard's original method is reduced from three years before and after the patent issuance date to two years before and after that date, the rankings of the patents-in-suit are 9th, 10th, and 16th, for an average ranking of 11.7 out of 22. Leonard Decl. ¶ 4. If the size of the window is reduced even further to one year before and after the patent issuance date, then the rankings of the patents-in-suit are 9th, 10th, and 17th, for an average ranking of 12 out of 22. Id. Similarly, using an alternative approach of dividing each patent's forward citation count by the number of years since the patent was issued, the rankings of the patents-in-suit are 10th, 13th, and 15th, for an average ranking of 12.7 out of 22. Id. ll 5, Ex. A. The fact is that whether ranked based on unadjusted citation counts, any of Dr. Leonard's comparison groups, or by citation counts divided by age, the patents' forward citation counts strongly suggest that the three patents-in-suit are in the middle of the pack, or worse, with respect to the 22 patents that Oracle contends were the most valuable in Sun's mobile Java portfolio.3 Id. ¶ 6.

B. Dr. Leonard's use of Oracle's estimation of the fair value of Sun's "core
technology" assets is reasonable.
Just as it did in its original motion to exclude portions of Dr. Leonard's report [Dkt. No. 585], Oracle attacks Dr. Leonard for relying on evidence that does not favor Oracle's view of the case. The Court should again reject Oracle's attempt to turn a Daubert motion into an evidentiary dispute. Dkt. No. 632 at 3-4; see also, e.g., ID Sec. Sys. Canada, Inc. v. Checkpoint Sys., Inc., 249 F. Supp. 2d 622, 655 amended, 268 F. Supp. 2d 448 (E.D. Pa. 2003) ("That Dr. Asher's theory rested on a particular interpretation of a disputed fact in this case does not mean that his testimony, based on a methodology that the court previously found to be reliable, should

7


have been excluded from evidence."); Hammes v. Yamaha Motor Corp. USA., Inc., CIV. 03- 6456 MJD/JSM, 2006 WL 1195907 at *6 (D. Minn. May 4, 2006) ("Yamaha's primary objection to this testimony is based upon Berke's interpretation of the evidence that Yamaha received notice of throttle problems in 2000. This is primarily a factual dispute over which experts could reasonably disagree and does not undermine the reliability of Berke's opinion.") [REDACTED]

[REDACTED]

8


[REDACTED]

Oracle's citations to other evidence that it prefers at most create a factual dispute. This is what trial is for—to resolve disputes just like this, about which evidence deserves the most weight. And, in any event, the evidence Oracle cites is hardly compelling. The selections from Douglas Kehring's deposition almost entirely concern the allocation methodology of Oracle's 10-K filing, see Richardson Decl. Ex. E (Kehring Dep.) at 142:9-147:4, and Kehring was not even sure whether the allocation methodology in the l0-K was based on the similar analysis in the Report, instead replying "I don't know" when asked that question. Id. at 148:15-18. Kehring's testimony does nothing to discredit the use of the Report in valuing Sun's intellectual property holding. And even to the extent Kehring's testimony is relevant, Dr. Leonard was not required to credit the interested testimony during litigation of an Oracle executive over pre-litigation statements in Oracle's financial reporting documents. Oracle may use the other valuations of Java—two of which were made by Oracle officers outside the context of financial reporting—on cross-examination, but an evidentiary dispute is not grounds for disbelieving or excluding the Report.

Second, the Report is nothing like evidence of the Sun v. Microsoft settlement, which the Court excluded earlier in the proceedings. There, Oracle was offering the settlement valuation as an affirmative matter to bolster its weak case for injunctive relief, whereas here Google is offering the Report as Oracle's own evidentiary admission, as one of many pieces of evidence that rebuts Oracle's damages case. Moreover, the Court struck Dr. Cockburn's discussion of the Sun v. Microsoft settlement because "[d]amages experts cannot use noncomparable licenses, with little relationship to the claimed invention or parties-in-suit, as a basis for calculating reasonable royalties." Dkt. No. 685 at 11 (citing ResQNet, Inc. v. Lansa, Inc., 594 F.3d 860, 870 (Fed. Cir. 2010)) (emphasis added). Here, the Report on which Dr. Leonard relied is a valuation, by Oracle, of Sun's entire intellectual property portfolio, whereas most of the money Microsoft

9


paid to settle its litigation with Sun was related to alleged antitrust violations or other issues having nothing to do with the value of the mobile Java technology at issue here. Id. at 11-12. It would be reasonable for the jury to decide that Oracle's own valuation of a set of things that includes the patents and copyrights at issue in this case and many other valuable items, should place a cap on damages. Dr. Leonard is entitled to testify to that effect, as a matter of economics, without converting his opinion into a legal instruction.

Third, Oracle misrepresents the use that Dr. Leonard makes of the Report. Dr. Leonard used the Report to demonstrate the weakness of Dr. Cockburn's patent portfolio mapping methodology. He first noted that [REDACTED] Richardson Decl. Ex. A (Leonard Supp. Rep.) at 9. He also noted that Sun owned over 14,000 patents. Id. Dr. Leonard assumed the allegedly three most valuable patents-in-suit were the three most valuable of all 14,000 patents, as had Dr. Cockburn. Id. He then applied Dr. Cockburn's patent apportiomnent method of mapping the distribution of the 14,000 patents onto the distribution of the PatVal data and multiplying the resulting apportiomnent by the entire bundle's value. Id.

Oracle claims Dr. Leonard cannot be "[u]sing Dr. Cockburn's own methodology" because Dr. Cockburn's "methodology does not rely on any accounting valuation of Oracle's 2010 acquisition of Sun at all." Motion to Strike at 5-6. This makes no sense. Of course Dr. Leonard had to change some aspect of Dr. Cockburn's methodology in order to test how robust it was; otherwise he would have been performing the same operations, in the same order, with the same data. In applying Oracle's admissions in the Report to the analysis, Dr. Leonard grew the size of the portfolio from 569 to 14,000 and also increased Oracle's (or Sun's) valuation of the relevant intellectual property, [REDACTED] Richardson Decl. Ex. A (Supp. Leonard Rep.) at 9. When Dr. Leonard mapped the distribution in the PatVal study to the altered portfolio, and applied the rest of Dr. Cockburn's methodology, the patents-in-suit returned a much lower value. Id. If Google is able to convince the jury that the numbers in the Report reflect reality, as Google is entitled to try to do, Dr. Leonard's analysis would be helpful in calculating damages.

10


Oracle also claims Dr. Leonard used the Report as "an absolute limit on the reasonable royalty Google and Sun would have agreed to in 2006." Motion to Strike at 5-6. This is wrong. Dr. Leonard did write that the "reasonable royalty in this case must be less than the total value of the patents-in-suit." Richardson Decl. Ex. A (Leonard Supp. Rep.) at 9. But, again, he is testifying to an economic opinion, not purporting to state a legal rule. Dr. Leonard is entitled to testify to his opinion that certain results make more economic sense than others. That is all he was doing here by using the Report to cast doubt on Dr. Cockburn's apportionment analysis.

C. Dr. Leonard's economic, not legal, opinion concerning what would have been
important in the Sun-Google negotiations is admissible.
Oracle's third argument takes one stray sentence of Dr. Leonard's Supplemental Report, gives it a meaning contrary to common sense, and again asks the Court to believe Dr. Leonard is attempting to instruct the jury on the appropriate legal rule to apply. Dr. Leonard began his description of Dr. Cockburn's apportionment methodology by writing:

Dr. Cockburn starts with the value that he claims Sun would have expected to generate from the proposed agreement with Google, apparently under the assumption that the parties would have expected to receive equal value from the proposed agreement. I note that there is no reason to believe this was necessarily the case; Sun may have expected to receive greater value (e. g., through Project Armstrong) than Google was expecting to receive. It is the value that Google was expecting to receive that matters for the reasonable royalty analysis.
Richardson Decl. Ex. A (Supp. Leonard Rep.) at 2. Dr. Leonard went on for several additional pages describing and criticizing Dr. Cockburn's apportionment methodology.

Out of all this text, Oracle takes the last sentence of the above paragraph as an attempt by Dr. Leonard to instruct the jury as a matter of law about what evidence it can and cannot consider in calculating a reasonable royalty. Motion to Strike at 8-9. For some reason, even though it only objects to the last sentence, it asks that the Court strike the entire paragraph. Motion to Strike at 10. Again, Dr. Leonard is an economist, not a lawyer. Dr. Leonard's point was that, standing in Google's shoes at the time of the hypothetical negotiation, Google would have decided how much it was willing to pay Sun based on its expectations of the value of the partnership to Google, not the value to Sun—particularly since the purported value of the deal to

11


Sun was contingent on Sun's unproven ability to start up and grow an entirely new business line. Even in a hypothetical negotiation, every party has a bottom line. Dr. Leonard is stating only that a party to a business deal decides what terms it ultimately will accept based on its own interests, not the interests of the party across the table.

Google has no intention of arguing, and Dr. Leonard has no intention of testifying at trial, about any issue of law, including the extent to which the jury can consider Sun's anticipated lost profits. That issue will be governed by this Court's instruction and the Georgia-Pacific factors. Indeed, in his first report, Dr. Leonard considered all the evidence Oracle now uses to support its calculation of Sun's lost profits, and offered an opinion under the Georgia-Pacific framework as to the impact that evidence should have on the reasonable royalty. Dr. Leonard is entitled to state his opinion, as a matter of economic logic, about the factors that would have guided Google in the hypothetical negotiation, including the maximum amount Google would have been willing to pay as a royalty. The Court should not strike any part of the paragraph.

III.CONCLUSION

For all of these reasons, Google respectfully requests that the Court deny Oracle's Motion to Strike Portions of Gregory Leonard's Supplemental Report.

Dated: March 2, 2012

KEKER & VAN NEST LLP

By: /s/ Robert A. Van Nest
ROBERT A. VAN NEST

Attorneys for Defendant
GOOGLE INC.

12


1 There are numerous distinctions between the technology covered by the ten claims in the '685 and the 31 claims in the '104. One important distinction between the '104 and the '685 is that both independent claims of the '685 patent require "a computer system comprising a program in source code form" and thus relate strictly to a combined compiler-interpreter apparatus. Zimmer Decl. Ex. C ('685) Claims 1, 6. By contrast, the source code requirement is not found in any of the claims of the '104, which implicates precompiled bytecode—the basis of Oracle's infringement theory in this case. Zimmer Decl. Ex A ('104) Claims 11-41.

2 See also, e.g., id. ("For example . . . Trajtenberg (1990), Hall, Jaffe, and Trajtenberg (2005), Allison et al. (2004), Guellec and van Pottelsberghe (2000), and Lanjouw and Schankerman (1997) establish a link between forward citations and commercial value."); Zimmer Decl. Ex. E (Manuel Trajtenberg, A Penny for Your Quotes: Patent Citations and the Value of Innovations, The RAND Journal of Economics 172 (1990)) at 172 ("The central hypothesis is that patent citations (i.e., references to patents appearing in the patent documents themselves), long presumed to be indicative of something like technological importance, may be informative of the economic value of innovations as well. Indeed, patent counts weighted by a citations-based index are found to be highly correlated (over time) with independent measures of the social gains from innovations in Computed Tomography.").

3 Moreover, not all analysts agree that adjusting citation counts for patent age is either necessary or appropriate. One study, concerned that "it could be that older patents are cited more often simply because they have had more opportunities to be cited," tested this question statistically and concluded that "neither age nor truncation could possibly account for the observed distribution of citation counts" and that "the null hypothesis that older patents received more citations just because of the passage of time is rejected by a wide margin." Zimmer Decl. Ex. E (Trajtenberg study) at 176-77.



  View Printable Version


Groklaw © Copyright 2003-2013 Pamela Jones.
All trademarks and copyrights on this page are owned by their respective owners.
Comments are owned by the individual posters.

PJ's articles are licensed under a Creative Commons License. ( Details )