Oracle America, Inc. v. Google Inc.

Filing 171

MOTION To Exclude Expert Opinions and Testimony (Daubert) filed by Google Inc.. Motion Hearing set for 7/21/2011 09:00 AM before Hon. William Alsup. Responses due by 6/28/2011. Replies due by 7/5/2011. (Sabnis, Cheryl) (Filed on 6/14/2011)

1
2
3
4
5
6

DONALD F. ZIMMER, JR. (SBN 112279)
[e-mail]
CHERYL A. SABNIS (SBN 224323)
[e-mail]
KING & SPALDING LLP
101 Second Street – Suite 2300
San Francisco, CA 94105
Telephone: [phone]
Facsimile: [phone]

IAN C. BALLON (SBN 141819)
[e-mail]
HEATHER MEEKER (SBN 172148)
[e-mail]
GREENBERG TRAURIG, LLP
1900 University Avenue
East Palo Alto, CA 94303
Telephone: [phone]
Facsimile: [phone]

SCOTT T. WEINGAERTNER (Pro Hac Vice)
[e-mail]
ROBERT F. PERRY
[e-mail]
BRUCE W. BABER (Pro Hac Vice)
[e-mail]
KING & SPALDING LLP
1185 Avenue of the Americas
New York, NY 10036-4003
Telephone: [phone]
Facsimile: [phone]

ROBERT A. VAN NEST (SBN 84065)
[e-mail]
CHRISTA M. ANDERSON (SBN 184325)
[e-mail]
KEKER & VAN NEST LLP
710 Sansome Street
San Francisco, CA 94111-1704
Telephone: [phone]
Facsimile: [phone]

7
8
9
10
11
12
13
14
15

Attorneys for Defendant
GOOGLE INC.

16

UNITED STATES DISTRICT COURT

17

NORTHERN DISTRICT OF CALIFORNIA

18

SAN FRANCISCO DIVISION

19
20

Case No. 3:10-cv-03561-WHA
ORACLE AMERICA, INC.
Honorable Judge William Alsup

21

Plaintiff,

22

Hearing Date: July 21, 2011
Hearing Time: 2:00 p.m.

v.
23
GOOGLE INC.
24
25
26
27
28

Defendant.

GOOGLE INC.’S BRIEF IN SUPPORT OF
DAUBERT MOTION
HIGHLY CONFIDENTIAL ATTORNEYS’ EYES ONLY

1
2

TABLE OF CONTENTS
I.

STATEMENT OF FACTS

3

3

A.

Patents-in-Suit

3

4

B.

Accused Functionality And Alternatives

4

5

C.

Valuation Of The Asserted Patents And Java Generally

5

6

D.

Fragmentation Of Java

6

7 II.

LEGAL STANDARDS

7

8 III.

ARGUMENT

8

9

A.

10

Cockburn’s reasonable royalty calculation rests entirely on improper
categories not recoverable as patent damages

11

1.

12
2.

13
14
15

3.

16
17

B.

18
19
20

C.

27

11

Cockburn’s report makes no effort to base his
royalty rate on the value of the patented technology

12
12
16

16

Cockburn improperly sets the date of the hypothetical
negotiation in 2008, when Android phones first launched,
rather than when the Android code was first written

17

Cockburn Ignores Record Facts And Undisputed Market Realities

18

3.

D.

Cockburn improperly includes damages for socalled “fragmentation” in his royalty base

Cockburn improperly includes Google’s revenues
from international sales in his damages calculation

24

26

10

16

2.

25

Cockburn improperly smuggles Oracle’s lost
profits in his royalty calculation, with no
attempt to satisfy the stringent Panduit test

Cockburn inflates his damages estimate by double
counting, including Oracle’s alleged future
damages despite Oracle’s request for an injunction

22
23

9

Cockburn’s methodology contains numerous other legal errors
1.

21

Cockburn improperly includes all of Google’s
worldwide advertising revenue from Android
phones in his royalty calculation

Cockburn also calculates an unreasonable, inflated royalty rate by
refusing to value the patented technology at issue, among other errors
1.

9

28
i
GOOGLE INC’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

1.

2
3

2.

4
5

Cockburn ignores the actual market value of Java, as proven
by numerous licenses granted by Sun, Google’s negotiations
with Sun, and the value assigned to Java by Oracle

18

Cockburn assumes, without any factual basis, that Sun and
Google would have negotiated an expensive, Java-incompatible
license rather than a cheap, Java-compatible license

19

i.

6
ii.

7
8
9

iii.

10
3.

11
12
13

E.
IV.

Far from prohibiting fragmentation of Java, Sun promoted
such fragmentation through its licensing program

20

Even if Cockburn’s assumptions about fragmentation
are correct, it is improper to base a damages award
on compatibility with nonpatented features

21

Cockburn relies on fragmentation as a reason to ignore
relevant evidence, including Java-compatible licenses

22

Cockburn relies on inapposite licenses and ignores
relevant ones to inflate the damages calculation

Cockburn Offers No Meaningful Analysis Regarding Copyright Damages

CONCLUSION

23
23
25

14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
ii
BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

TABLE OF AUTHORITIES

2

Cases

3

Cornell Univ. v. Hewlett-Packard Co., 2008 WL 2222189 (N.D.N.Y. 2008) (Rader, J.)

7

4 Cornell Univ. v. Hewlett-Packard Co., 609 F. Supp. 2d 279 (N.D.N.Y. 2009) (Rader, J.)
5 Fonar Corp. v. General Electric Co., 107 F.3d 1543 (Fed.Cir.1997)

15
9

6 Garretson v. Clark, 111 U.S. 120 (1884)

12

7 Georgia-Pacific Corp. v. U.S. Plywood Corp., 318 F. Supp. 1116 (S.D.N.Y. 1970)

12, 13

8 Interactive Pictures Corp. v. Infinite Pictures, Inc., 274 F.3d 1371 (Fed. Cir. 2001)

17

9 Kalman v. Berlyn Corp., 914 F.2d 1473 (Fed.Cir.1992)

9

10 Kori Corp. v. Wilco Marsh Buggies & Draglines, Inc., 761 F.2d 649 (Fed. Cir. 1985)

10

11 Lucent Tech. Inc. v. Gateway Inc., 580 F.3d 1301, 1325-35 (Fed. Cir. 2009) 12, 13, 15, 16, 23
12

Minks v. Polaris Indus., Inc., 546 F.3d 1364 (Fed. Cir. 2008)

12

13 Paice LLC v. Toyota Motor Corp., 504 F.3d 1293 (Fed. Cir. 2007)

16

14 Panduit Corp. v Stahlin Bros. Fibre Works, Inc. 575 F.2d 1152 (6th Cir. 1978)

11

15 Paper Converting Machine Co. v. Magna-Graphics Corp., 745 F.2d 11 (Fed.Cir.1984)
16 Radio Steel & Mfg. Co. v. MTD Prods., Inc., 788 F.2d 1554 (Fed. Cir. 1986)
17 ResQNet.com, Inc. v. Lansa, Inc., 594 F.3d 860 (Fed. Cir. 2010)
18 Riles v. Shell Exploration & Production Co., 298 F.3d 1302 (Fed. Cir. 2002)
19 Uniloc USA, Inc. v. Microsoft Corp., 632 F.3d 1292 (Fed. Cir. 2011)

9
10

11, 14, 23
8, 16
8, 9, 15

20
21
22

Statutes
35 U.S.C. § 284

14

23
24
25
26
27
28
iii
GOOGLE INC’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

PLEASE TAKE NOTICE that on July 21, 2011 at 9:00 a.m., or as soon thereafter as

2

counsel may be heard, Defendant Google Inc. (“Google”) will, and hereby does, respectfully

3

move to exclude the opinions and testimony of Oracle America, Inc.’s (“Oracle”) damages

4

expert Dr. Iain M. Cockburn. This Motion is based on the following memorandum of points and

5

authorities in support, the Declaration of Scott T. Weingaertner (“Weingaertner Decl.”) and

6

accompanying exhibits, the Declaration of Gregory K. Leonard, Ph.D. (“Leonard Decl.”) and

7

accompanying exhibits, the entire record in this matter, and on such evidence as may be

8

presented at the hearing of this Motion.

9

This Court should exclude the unreliable and results-oriented opinions and testimony of

10

Oracle’s damages expert Dr. Cockburn. Ignoring the settled legal framework for evaluating

11

patent damages and many years of evidence establishing the modest market value of Java

12

technology, Cockburn opines that Google would owe Oracle an amount in the breathtakingly

13

wide range of

14

patents at issue in this case. That range is orders of magnitude beyond any reasonable valuation

15

of the intellectual property at issue.

if Google is held to infringe any claim of any of the seven

16
17
18
19
20

Cockburn’s legal errors are fundamental and disqualifying. Although Cockburn purports

21

to calculate a reasonable royalty, he ignores the guidelines established by the Supreme Court and

22

Federal Circuit. He fails to separate out and value the patented technology at issue, instead using

23

the purported value of the entire Java platform as the basis for his damages claim. He also fails

24

to calculate the incremental value of the alleged infringing technology to the Android platform.

25

Instead, he improperly adds together as the basis for calculating a royalty rate: (1) Google’s

26

incremental profit from advertising on all Android devices world-wide, and (2) the purported

27

harm to Sun’s and Oracle’s business from lost profits and alleged “fragmentation” of the Java

28

platform.
1
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

None of these are proper components of a royalty calculation for patent damages:

2



Google’s advertising business is not the accused product. The accused product is the
Android platform software, which Google does not sell and from which it earns no
direct revenue. The value of the Android software and of Google’s ads are entirely
separate: the software provides for certain phone functionality, whether or not the
user is viewing ads, and Google’s ads are viewable on any software and are not
uniquely enabled by Android.



Oracle’s lost profits cannot be used to inflate the royalty calculation. Lost profits are
a separate category of patent damages, authorized only if the patentee meets a distinct
and demanding test that Cockburn does not even attempt to satisfy.



Oracle’s purported damages from so-called “fragmentation” of Java relate to
theoretical downstream harm to a wholly different Oracle product that is not
recoverable as damages in this action.

3
4
5
6
7
8
9

After inflating his royalty calculation as described above, Cockburn opines that Oracle should be
10
awarded more than 50% of the net benefit he claims Google would receive from licensing the
11
patents and copyrights at issue from Oracle, resulting in a greater than 20% royalty rate. This
12
“methodology” is not authorized by the law or consistent with any real-world negotiations
13
regarding any aspect of the Java technology. Cockburn treats the patents and copyrights at issue
14
as a single, indivisible unit, casually dismissing critical differences among the patents, including
15
expiration dates over a decade apart.
16
17
18
Cockburn’s factual errors and omissions are just as fundamental. He disregards market
19
evidence about the value of the Java platform in favor of speculation. He dismisses Sun’s
20
history of licensing Java for a modest royalty and the fact that, during its acquisition of Sun,
21
Oracle assigned a value of

to the entire Java platform—which includes far

22
more than the Java mobile platform, only part of which is at issue here. He ignores Google’s
23
negotiation history with Sun regarding a Java license for the mobile space, which would have
24
included far more than the patents-in-suit and during which Google rejected a proposal to pay
25
Sun orders of magnitude less than what Cockburn now says Google would have paid. Even
26
more glaring, he mischaracterizes Sun’s and Microsoft’s 2004 settlement of litigation that
27
involved far more than claims of fragmentation, asserting that Sun demanded and received $900
28
2
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

million to cover the risk of fragmentation to Java, even though Sun and Microsoft settled

2

litigation relating to fragmentation in 2001 and Microsoft paid Sun just $20 million.

3

Finally, as to copyright damages, Cockburn provides no separate analysis at all.

4

All of these errors are foundational, and any one of them would be enough to render

5

Cockburn’s report unreliable, misleading, and inappropriate for presentation to the jury. Taken

6

together, they require exclusion of his opinions.

7

I.

STATEMENT OF FACTS
Oracle alleges infringement of seven patents 1 and two copyrights.

8

(See Amended

9 Complaint for Patent and Copyright Infringement (“Amended Complaint”) (Dkt. 36) at ¶ 10
10 (Exs. A-G) and ¶ 11 (Ex. H).) On May 21, 2011, Oracle served the Expert Report of Iain M.
11 Cockburn (“Cockburn Report” or “Cockburn Rep.”) regarding its purported damages on these
12 claims. (See Weingaertner Decl. at ¶ 1 (Ex. A), Cockburn Rep.)
13

A.

14

The patents-in-suit fall into two technology categories: the ‘447 and ‘476 patents relate

15

to a security framework, and the five remaining patents relate to optimizations for use with a

16

virtual machine. During claim construction, Oracle’s counsel confirmed that the patents-in-suit

17

involve only these technology categories. (See, e.g., Weingaertner Decl. at ¶ 2 (Ex. B), Oracle’s

18

Technology Tutorial Supplement at 1; April 6, 2011 Transcript of Proceedings (Dkt. 110) at

19

15:6-8.)

20

Patents-in-Suit

The patents-in-suit, either individually or collectively, do not encompass Java as a whole

21 or even a significant part of it. The ‘720 patent provides a concrete example. The purported
22 novelty of the ‘720 patent lies in optimizations useful for multitasking. (‘720 Patent (Dkt. 36,
23 Ex. D) at 1:27-33.) Oracle has asserted that it uses this optimization in only two minor products,
24 not in the Java platform generally.

(Weingaertner Decl. at ¶ 3 (Ex. C), Oracle’s Second

25 Supplemental Patent Local Rule 3-1 Disclosure Of Asserted Claims And Infringement
26
27
28

1

U.S. Patent Nos. 6,125,447 (“the ‘447 patent”); 6,192,476 (“the ‘476 patent”); 5,966,702
(“the ‘702 patent”); 7,426,720 (“the ‘720 patent”); RE38,104 (“the ‘104 patent”); 6,910,205
(“the ‘205 patent”); and 6,061,520 (“the ‘520 patent”).
3
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

Contentions (“Oracle’s Infringement Contentions”) at 11.) Each of the other patents is also

2

directed to a specific, discernible feature. The patents also differ substantially in their remaining

3

life. The ‘104 patent will expire in 2012, and five of the other six patents will expire in just over

4

six years. Only one patent—the ‘720 multitasking patent discussed above—has a longer life.

5

B.

6

According to Oracle, the accused products break down into two groups: a) devices

7

running or containing the Android operating system (“Android Devices”); and b) computers

8

running or containing the Android software development kit (“SDK”). The Android operating

9

system and the SDK are separate sets of code with distinct functionalities, and Oracle has

10

Accused Functionality And Alternatives

accused the two of infringing different patents.

11

Because Oracle’s patents relate to either a security framework or virtual machine

12 optimizations, its Infringement Contentions are similarly limited.

Oracle alleges that only

13 portions of Android code referring to security infringe the security patents (see, e.g.,
14 Weingaertner Decl. at ¶ 4 (Ex. D), Oracle’s Infringement Contentions, Exhibit D at 2-21 (citing
15 code referring to SecurityManager class)) and that only portions of code relating to the Dalvik
16 Virtual Machine infringe the optimization patents (see, e.g., id. at ¶ 5 (Ex. E), Oracle’s
17

Infringement Contentions, Exhibit G at 1-20 (citing Android functionality within Dalvik)).

18

The accused features of Android frame any damages analysis, yet they comprise only a

19

fraction of Android as a whole. Some of the accused features have never been used and are

20

therefore hardly essential to Android’s functioning or success.

21

SecurityManager, and is not aware of any third party that uses or has ever used

22

SecurityManager”;

23

functionality for the Gingerbread version of Android (version 2.3)”; and “[i]n the current version

24

of Android, as of December 6, 2010, users and application programs are prevented from even

25

installing an instance of SecurityManager.” (See Weingaertner Decl. at ¶ 6 (Ex. F), Defendant

26

Google Inc.’s Fourth Supplemental Responses to Plaintiff’s Interrogatories, Set One, No. 3 at

27

28.)

28

programming language at all, instead allowing Android developers to implement applications
4

Google

“implemented

changes

to

disable

“Google does not use

SecurityManager-related

In addition, the Android Native Development Kit, (“NDK”) does not use the Java

GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

using native-code languages like C and C++.

2

http://developer.android.com/sdk/ndk/index.html.)

(See Weingaertner Decl. at ¶ 7 (Ex. G),

3

C.

4

The Java platform and its components are not new. As a result, the record contains

5

evidence of numerous instances when market participants have quantified the value of both Java

6

as a whole and the fraction of that value attributable to the asserted patents, including Sun’s

7

historical Java licensing practices. Sun’s practice was always to license Java when asked to do

8

so.

Valuation Of The Asserted Patents And Java Generally

9
10
11
12
13
14
15
16
17
18
19

Oracle also made statements to the SEC regarding the value of the entire Java system in
connection with its acquisition of Sun.

20
21
22
23
24
25
26
27
28
5
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1
2
3

This case does not involve the entire Java platform. It involves only certain features of

4

one small portion of the Java Platform. Sun’s and Oracle’s overall investment in Java was

5

centered in the larger desktop, server, and enterprise markets, not in the mobile space.

6
7
8
9
10

Even the features of the patents—which are not specific to or required by the Java
platform—are capable of valuation based on the documentary record.

11
12
13
14
15
16
17
18

Finally, there is no need to speculate about the terms Google and Sun might have reached

19

in a hypothetical negotiation because Google and Sun actually engaged in negotiations regarding

20

a license to the entirety of Java.

21
22
23
24

D.

25

A substantial portion of Cockburn’s damages calculation is based on so-called

Fragmentation Of Java

26

“fragmentation” of the Java platform into numerous incompatible sub-standards.

27

“fragmentation”—which is the purported downstream harm to a separate Oracle product—is not

28

a recognized category of damages for patent infringement. Cockburn not only includes it, he
6
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

Such

1

vastly inflates the value Sun assigned to preventing that alleged harm. In 2002, Sun sued

2

Microsoft for various alleged unfair business practices and trademark infringement, including

3

Microsoft’s attempt to create its own version of Java, which might have led to fragmentation of

4

Java standards and practices. That case eventually settled in 2004. Cockburn focuses on this

5

2004 agreement, which was a broad business arrangement involving more than settlement of

6

litigation, including covenants not to sue under the parties’ entire patent portfolios. (Ex. A,

7

Cockburn Rep. ¶ 312.) Cockburn completely ignores a 2001 settlement agreement in which

8

Microsoft paid Sun $20 million to settle litigation involving claims related to fragmentation.

9

(Weingaertner Decl. at ¶ 14 (Ex. .N), Microsoft Reaches Agreement to Settle Contract Dispute

10

With Sun Microsystems at 1; Weingaertner Decl. at ¶ 15 (Ex. O), Settlement Agreement and

11

Mutual Limited Release at 2.)

12
13

Even prior to the Microsoft litigations, Sun tolerated and even promoted fragmentation of
Java through its own licensing program.

14
15
16
17
18
19
20
21

II.

LEGAL STANDARDS

22

Trial judges serve as gatekeepers for both the relevance and reliability of expert

23

testimony. Federal Rule of Evidence 702 permits a trial court to admit only expert “testimony

24

that is (1) based upon sufficient facts or data, (2) the product of reliable principles and methods,

25

and (3) delivered by a witness who has applied the principles and methods reliably to the facts of

26

the case.” Cornell Univ. v. Hewlett-Packard Co., 2008 WL 2222189, *1 (N.D.N.Y. 2008)

27

(Rader, J.) (citing Fed. R. Evid. 702). The Federal Circuit and district courts have excluded

28
7
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

expert testimony on damages that fails to meet these criteria. See Uniloc USA, Inc. v. Microsoft

2

Corp., 632 F.3d 1292 (Fed. Cir. 2011); Cornell, 2008 WL 2222189, at *4.

3

Oracle bears the burden of proving damages. Here, Cockburn calculated damages by

4

employing the commonly used “hypothetical negotiation” approach to determining a reasonable

5

royalty. That approach attempts to determine what a willing licensor and a willing licensee

6

would have agreed “just before infringement began.” Id. The hypothetical negotiation must be

7

based in “sound economic and factual predicates.” Riles v. Shell Exploration & Production Co.,

8

298 F.3d 1302, 1311 (Fed. Cir. 2002).

9

III.

ARGUMENT

10

Cockburn’s opinion is unreliable and should be excluded because he (1) ignored settled

11

legal standards; (2) relied on assumptions that are uniformly contradicted by the record; and (3)

12

offered no analysis whatsoever with respect to copyright damages. The predictable result is a

13

hyper-inflated

14
15

Cockburn’s errors are several:


16
17

damages calculation.

he fails to tie his royalty calculation to sales of the allegedly infringing product
(the Android platform software);



18
19
20



he inflates his royalty rate to 20 percent, ignoring the time-honored GeorgiaPacific analysis;



he does not even try to value the technology embodied in the asserted claims of
the patents-in-suit, the relative contribution of that technology to the overall
Android platform, or the value of individual patents, or to apportion the value of
patented and non-patented features; and



he ignores all of Sun’s actual licenses, which establish a relatively modest value
for Java as a whole and even more minuscule values for the Java components
embodied by the patents-in-suit, crediting instead a handful of licenses for wholly
unrelated technologies between other companies.

21
22
23
24
25
26

The
from advertising on all Android Devices. No one has ever paid anywhere near

27 this amount for even the entire Java platform, much less the minor components of that platform
28 system that relate to the patents-in-suit.
8
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

A.

Cockburn’s Reasonable Royalty Calculation Rests Entirely On Improper
Categories Not Recoverable As Patent Damages.

2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27

Cockburn’s first task in his reasonable royalty analysis was to identify the allegedly
infringing product and determine how much revenue Google, as the alleged infringer, earned
from sales of that product. Cockburn fails even this simple initial step, stuffing his royalty
calculation with numerous unreliable, inflated, and improper components.
1.

Cockburn improperly includes all of Google’s worldwide advertising
revenue from Android phones in his royalty calculation.

First, Cockburn has no basis for including in his royalty base Google’s worldwide
revenue from advertising displayed on Android phones.

The accused product here is the

Android software, which Google does not sell, and from which Google receives no direct
revenue. Google receives no payment, fee, royalty, or other remuneration for its contributions to
the Android platform. Only in specific circumstances is it proper to include revenues derived
from downstream sales of non-infringing products, like advertising on Android phones, in a
royalty base. The royalty base may include non-infringing components when those infringing
components are the basis for customer demand for the entire machine, Fonar Corp. v. General
Electric Co., 107 F.3d 1543, 1552 (Fed.Cir.1997); when physically separate products are sold
together as a functional unit or parts of a complete machine, Paper Converting Machine Co. v.
Magna-Graphics Corp., 745 F.2d 11, 23 (Fed.Cir.1984); and when separate components are
analogous to a single functioning unit, Kalman v. Berlyn Corp., 914 F.2d 1473, 1485
(Fed.Cir.1992). Cockburn’s analysis ignores all these limitations.
Because Google’s advertising is not sold with the Android software and does not consist
of a single functioning unit along with that software, Google advertising revenue could be
included in the royalty base only if Oracle could satisfy the requirements of the entire market
value rule—i.e., that the patented features of the Android software drive demand for Google’s
advertising or that Google’s advertising create the value of the software. See Uniloc, 632 F.3d at
1318-20. Cockburn neither analyzes these questions nor answers them. In fact, the value drivers
of the Android software and of Google’s ads are entirely separate: the software provides for

28
9
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

certain phone functionality, whether or not the user is viewing ads; and Google’s ads are

2

viewable on any software and are not uniquely enabled by Android.

3

Because he fails to address the entire market value rule, Cockburn increases the

4

reasonable royalty by recovery of Oracle’s purported lost profits even though “[t]he

5

determination of a reasonable royalty . . . is based not on the infringer’s profit, but on the royalty

6

to which a willing licensor and a willing licensee would have agreed at the time the infringement

7

began.” Radio Steel & Mfg. Co. v. MTD Prods., Inc., 788 F.2d 1554, 1557 (Fed. Cir. 1986). At

8

most, an infringer’s profits from the infringing product are a factor to be considered in the

9

royalty analysis. See Kori Corp. v. Wilco Marsh Buggies & Draglines, Inc., 761 F.2d 649, 655

10

(Fed. Cir. 1985). Here, Cockburn’s analysis is even farther afield of what the law allows,

11

because he bases his damages estimate on Google’s profits from non-infringing products.

12

Cockburn also increases damages by laying out, at considerable effort, speculative

13

theories as to direct and indirect revenue that Google purportedly attained or could have

14

anticipated with respect to Android. That shows what Oracle is really after—any revenues that

15

could be associated with advertising on Android Devices.

16

Google’s advertising revenues associated with Android Devices would have been fair game to

17

Sun and that Google would have given Sun “an even split” of its incremental profits from

18

advertisements, plus compensated Sun for half of what Sun lost. (See Ex. A, Cockburn Rep. ¶¶

19

283-284.) Out of the myriad actual licenses Sun granted to Java and its components, Cockburn

20

fails to point to a single one that adopted a structure whereby a licensee paid Sun any percentage

21

of its profits on downstream products associated with the use of the product that included Java

22

technology.

23
24

2.

Cockburn assumes that all of

Cockburn improperly smuggles Oracle’s lost profits in his royalty
calculation, with no attempt to satisfy the stringent Panduit test.

25

Second, Cockburn improperly includes a lost profits recovery in his royalty calculation.

26

(See Leonard Decl. at ¶¶ 14-15.) He assumes that “Sun would have required a relatively high

27

royalty rate to compensate it for the immediate losses” allegedly related to Android. (Ex. A,

28

Cockburn Rep. ¶ 303.) He thereby conflates a reasonable royalty recovery with a lost-profits
10
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

recovery, and bypasses the separate and well-developed legal framework for computing lost

2

profits in patent cases. See, e.g., Panduit Corp. v Stahlin Bros. Fibre Works, Inc. 575 F.2d 1152

3

(6th Cir. 1978). If Cockburn’s technique were valid, it would render the Panduit doctrine

4

pointless. Patentees would have no need to try to prove lost profits in a “but-for” world or to

5

meet the high standards of Panduit when they could, instead, simply assert a reasonable royalty

6

would compensate for purported lost profits. It would also open the door, as it has here, to an

7

increased recovery that would punish infringers instead of compensating patentees.

8
9

3.

Cockburn improperly includes damages for so-called “fragmentation”
in his royalty base.

10

Third, Cockburn includes as a component of his royalty calculation money Sun

11

purportedly would have demanded as compensation for so-called “fragmentation” of the Java

12

platform into multiple, competing sub-standards.

13

allegedly infringing product (or any product, for that matter) sold by Google, or any product sold

14

by Sun or Oracle embodying the patents-in-suit. They relate only to theoretical harm to a

15

separate product that is not even at issue in this case. This is neither a reasonable patent royalty

16

nor lost profits, and is not a patent damages theory at all; it is a general tort damages theory.

17

Oracle has not asserted any tort claims in this case. Cockburn does not justify his inclusion of

18

fragmentation in his royalty calculation by reference to any case law. In fact, he admits that the

19

“fragment or fork [of] Java generally” is “an effect going beyond the use of the specific

20

[purportedly] infringed intellectual property itself.” (E.g., Ex. A, Cockburn Rep. ¶ 157; see also

21

id. at Ex. C-2 (citing same for Georgia-Pacific Factor 3 regarding “scope of the license”).) The

22

only conclusion is that he explicitly seeks royalties for compatibility with something other than

23

the intellectual property at issue.

These payments are unconnected to any

24

Nothing in the Patent Act or case law supports an award for such attenuated “damages.”

25

To the contrary, an award unrelated to the patented technology “punishes beyond the reach of the

26

statute.” ResQNet.com, Inc. v. Lansa, Inc., 594 F.3d 860, 869 (Fed. Cir. 2010). As will be

27

discussed further below, there is also no factual basis for Cockburn’s approach, as Sun regularly

28

approved of and encouraged the sort of fragmentation it now claims would be fatal to Java.
11
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

B.

Cockburn Also Calculates An Unreasonable, Inflated Royalty Rate By
Refusing To Value The Patented Technology At Issue, Among Other Errors.

2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

“A determination of the royalty stemming from a hypothetical negotiation is often made
by assessing factors such as those set forth in Georgia-Pacific Corp. v. U.S. Plywood Corp., 318
F. Supp. 1116, 1120 (S.D.N.Y. 1970).” Minks v. Polaris Indus., Inc., 546 F.3d 1364, 1372 (Fed.
Cir. 2008). Not all of the 15 Georgia-Pacific factors are relevant in every case, and the Federal
Circuit has recently weighed some factors more heavily than others. See Lucent Tech. Inc. v.
Gateway Inc., 580 F.3d 1301, 1325-35 (Fed. Cir. 2009) (discussing factors 2, 10, 11, and 13).
Cockburn has not analyzed these factors in any meaningful way. He buries a cursory analysis of
the factors in an appendix, but essentially reduces the analysis to the catch-all factor 15—the
hypothetical negotiation between Sun and Google—thereby giving himself maximum freedom to
maximize damages.
1.

Cockburn’s report makes no effort to base his royalty rate on the
value of the patented technology.

The most glaring of Cockburn’s departures from the Georgia-Pacific standard is his
express refusal to tie his calculation to the value of the patented technology at issue, both to Sun
and Oracle as part of the overall Java platform and to Google in the context of the broader
Android platform. (See Leonard Decl. at ¶¶ 8-11.)
The Supreme Court held nearly 130 years ago that

19
20
21
22
23

[t]he patentee . . . must in every case give evidence tending to separate or
apportion . . . the patentee’s damages between the patented feature and the
unpatented features, and such evidence must be reliable and tangible, and
not conjectural or speculative; or he must show, by equally reliable and
satisfactory evidence, that the profits and damages are to be calculated on
the whole machine, for the reason that the entire value of the whole
machine, as a marketable article, is properly and legally attributable to the
patented feature.

24

Garretson v. Clark, 111 U.S. 120, 121 (1884). Just two years ago, the Federal Circuit drove

25

home that point by reversing a $357 million damages award because of “the glaring imbalance

26

between infringing and non-infringing features” in the accused product. Lucent, 580 F.3d at

27
28
12
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

1333; see also id. at 1337 (quoting Garretson). Cockburn has not even attempted to distinguish

2 between the relevant claims of the patents-in-suit and the rest of either Android or Java.
3

The principle of Garretson and Lucent is embodied in two Georgia-Pacific factors and

4 has figured prominently in recent Federal Circuit cases. Georgia-Pacific factor 10 looks to
5 “[t]he nature of the patented invention; the character of the commercial embodiment of it as
6 owned and produced by the licensor; and the benefits to those who have used the invention.”
7 Georgia-Pacific, 318 F. Supp. at 1120. Factor 13 looks to “[t[he portion of the realizable profit
8 that should be credited to the invention as distinguished from nonpatented elements” or other
9 improvements. Id. The point of these factors is “to elucidate how the parties would have valued
10 the patented feature during the hypothetical negotiation.” Lucent, 580 F.3d at 1332 (emphasis
11 added). Yet Cockburn essentially ignores the patented features entirely, both by failing to
12 consider the patents’ claimed technology on a claim by claim basis and in view of the expiration
13 dates, and by assuming all Java and Android revenues are attributable to the patented features.
14

Cockburn treats the patents-in-suit—and the copyrights, too—as a single, indivisible unit,

15

casually dismissing critical differences in the patents by deeming them all “essential.” (See, e.g.,

16

Ex. A, Cockburn Rep. ¶¶ 130-132.) Continuing to build this house of cards, he then equates the

17 value of the patents-in-suit with the value of Java as a whole. He never states a basis for that
18 conclusion, points to any evidence in the record, or explains how his aggregate analysis could
19

possibly satisfy Garretson and Lucent. The closest he comes to analyzing any patented feature is

20

to state that “five of the seven patents-in-suit relate to improved performance, including efficient

21

memory utilization and optimized execution speed, of the virtual machines.” (Id. at ¶ 19.) He

22

jumps from that simple, descriptive statement to the conclusion that “the patents-in-suit are

23

wholesale enablers of the Java architecture on a smartphone,” without explaining why. (Id. at

24 ¶ 24).
25
26
27
28

Accordingly, just as was the case
13
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

in Lucent, the ‘720 patent is “but a tiny feature of one part of a much larger software program.”

2

Lucent, 580 F.3d at 1332. As a matter of law, Cockburn’s conclusion that the ‘720 patent

3

accounts for all of Java’s value is unreasonable. 2 His silence regarding the other patents means

4 there is no basis to conclude that those patents are essential to enabling Java architecture on a
5 smartphone either.
6

On the other side of the ledger, Cockburn also assumes the asserted patents are “essential

7 to the device,” which is an apparent reference to Android Devices (see, e.g., Ex. A, Cockburn
8

Rep. ¶ 130), but does not explain how they could possibly account for the entire value of

9

Android. Even though the Federal Circuit has declared Georgia-Pacific factor 13—which looks

10

at the value of the patented features to the infringing product—as one of the most important

11

factors for purpose of calculating a royalty rate, Cockburn refuses to apply that factor at all.

12

Buried in an Appendix to his report, he states that there is “no clear economic basis at this time

13

for apportioning the total value of Android into value attributable to the patents and copyrights in

14

suit and any additional value added by Google.” (Ex. A, Cockburn Rep., Appx. C, at C-5 - C-6.)

15

That conclusion alone renders Cockburn’s report unreliable as a matter of law because “the trial

16

court must carefully tie proof of damages to the claimed invention’s footprint in the market

17

place.” ResQNet.com, 594 F.3d at 869 (emphasis added). Instead, Cockburn has admitted to

18

endorsing a damages calculation that includes royalties for features beyond the scope of the

19

claimed invention and, therefore, inherently beyond what is “adequate to compensate for the

20

infringement.” 35 U.S.C. § 284.

21

Finally, Cockburn’s refusal to separate out each patented feature, either from the broader

22

Java system or from each other, leads to another obviously wrong conclusion: that Oracle should

23

recover the same damages whether it proves complete infringement of all seven patents and both

24
25
2

26
27
28

Cockburn briefly analogizes to patents required to practice an industry standard. (Ex. A,
Cockburn Rep. ¶ 131.) With a standards essential patent, however, demonstrating essentiality is
simple: one can compare the claim with the published standard and, if it matches, the patent is
essential to practice the standard. Cockburn cites no published reference standard, and instead
vaguely asserts the patents are essential to “Java on a smartphone.” (Id. at ¶ 24.)
14
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

copyrights or only infringement of a single claim of a single patent. (Ex. A, Cockburn Rep.

2

¶ 347 (“[A]lthough an adjustment could be made to my calculations to the extent Oracle

3

succeeds on some but not all of its claims, it is not necessary to do so.”).) Even apart from the

4

inherent differences in value in the patented features, the patents have widely disparate

5

expiration dates. The ‘720 patent expires nearly eight years after every other patent-in-suit. If

6

Google does not infringe that patent, there is no basis to extend damages liability to 2025.

7

Similarly, if Google is found to infringe only the ‘104 patent, there is no basis to extend liability

8

beyond 2012. Yet Cockburn concludes that damages run throughout the life of the ‘720 patent

9

no matter what. (See, e.g., id. at ¶ 215, n.284.) He attempts to justify that approach by asserting

10

“that Oracle generally offered portfolio licenses” for Java (Id. at ¶ 347), but the question in a

11

hypothetical negotiation is what deal a willing licensor and licensee would strike for the patent-

12

in-suit—not whether the licensor generally tried to license a larger portfolio.

13

Valuing each patented feature individually, as the law requires, would burst Cockburn’s

14

overinflated damages estimate. After all, Oracle has admitted the incremental value of the ‘720

15

patent is $0. See p. 6, supra. If Google is found to infringe only that patent, the damages award

16

may well be only a nominal figure. Yet Cockburn incorporates

17

reasonably royalty calculation for the time period between 2018 and 2025, when the ‘720 patent

18

would be the sole remaining patent. (See Ex. A, Cockburn Rep., Appx. C, at C-4 (citing 2025 as

19

cutoff date in support of Georgia-Pacific Factor 7).) There is no plausible way that a feature

20

admittedly without any commercial value could “constitute[ ] a substantial portion of the value”

21

of either Java or Android. Lucent, 580 F.3d at 1332.

into his

22

Cockburn’s use of the Nash Bargaining Solution is not appropriate because he has not

23

tied its use to the particular value of the patented features. It makes no difference that the Nash

24

Bargaining approach has been widely used. (Ex. A, Cockburn Rep. ¶¶ 281-82.) It is not

25

appropriate to apply the Nash Bargaining Solution in a way that does not reflect the reality the

26

parties would have confronted in an actual real-world negotiation. The real question is “whether

27

[Cockburn] has justified the application of a general theory to the facts of the case.” Uniloc, 632

28

F.3d at 1316. He has not.
15
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

There is no “credible and sufficient economic proof that the patented invention drove

2 demand for” Android, and Cockburn offers none. Cornell Univ. v. Hewlett-Packard Co., 609 F.
3 Supp. 2d 279, (N.D.N.Y. 2009) (Rader, J.). (See also Leonard Decl. at ¶¶ 18-19, 21.) The
4 accused SecurityManager has never been used by Google and almost certainly never will be. See
5 pp. 5, supra. There are indisputably a host of “significant features or improvements added” to
6 Android. (See, e.g., Weingaertner Decl. at ¶¶ 22-24 (Exs. T-V).) These other features of
7 Android “account for the overwhelming majority of the consumer demand” of Android
8 smartphones. Lucent, 580 F.3d at 1333. Cockburn’s “model [does not support the award
9 because it] does not associate [the] proposed royalty with the value of the patented method at all,
10 but with the unrelated cost of the entire” Android platform. Riles, 298 F.3d at 1312.
11

C.

12

Cockburn’s Methodology Contains Numerous Other Legal Errors.
1.

13

Cockburn inflates his damages estimate by double counting, including
Oracle’s alleged future damages despite Oracle’s request for an
injunction.

14
Cockburn boosts his damages number by straight-up double counting. He includes
15
“future damages from May 20, 2011 through the different terminal dates I use in my hypothetical
16
license calculations: October 2018 (ten years from the date the infringement began), December
17
2021 (six years following the date of the longest projections for which I currently have data, and
18
based on the assumption that sales would decline to zero over an approximately six-year period
19
after that point), and December 2025 (following the expiration date of the last-to-expire patent).”
20
(Ex. A, Cockburn Rep. ¶ 320.) That calculation is error because future damages amount to
21
double counting in view of Oracle’s request for an injunction. See, e.g., Paice LLC v. Toyota
22
Motor Corp., 504 F.3d 1293, 1314 (Fed. Cir. 2007) (ongoing royalty may be appropriate “in lieu
23
of an injunction”). (See also Leonard Decl. at ¶¶ 28-29; Amended Complaint (Dkt. 36) at ¶ 46,
24
¶ B.)
25
26
27
28
16
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

2.

Cockburn improperly includes Google’s revenues from international
sales in his damages calculation.

2
3
4
5
6

Cockburn makes no attempt to exclude extraterritorial—and therefore noninfringing—
activities. (See Leonard Decl. at ¶¶ 22-23.)

In discussing Java’s purported value, Cockburn

references the total number of Java-enabled phones, operators and developers without regard to
where they are located. (Ex. A, Cockburn Rep. ¶ 26.)

7
8
9
10
11
12
13
14
15

3.

Cockburn improperly sets the date of the hypothetical negotiation in
2008, when Android phones first launched, rather than when the
Android code was first written.

16

Cockburn also inflates his calculation by arbitrarily setting the date of the hypothetical

17

Google-Sun negotiation very late in time, after Android was allegedly commercially viable and

18

Google was purportedly locked in to using certain technology. (See, e.g., Ex. A, Cockburn Rep.

19

¶ 233 (“with the window of opportunity closing”); ¶ 263 (“[Google] could not get to market in

20

time with a product to justify the expense of the undertaking”).) This is another error. The

21

relevant date is much earlier, when there was no guarantee Android would ever gain a foothold

22

in the market and before its development was locked in to any particular path.

23

The appropriate date of a hypothetical negotiation is “the time infringement began.”

24

Interactive Pictures Corp. v. Infinite Pictures, Inc., 274 F.3d 1371, 1385 (Fed. Cir. 2001)

25

(emphasis added). Without explanation, Cockburn “chose” October 22, 2008—the date that the

26

“first Google phone shipped”—as the date of the negotiation. (Ex. A, Cockburn Rep. ¶ 119.)

27

But Oracle’s infringement contentions allege that Google’s infringement began long before that

28

date. Oracle accuses the Android SDK, which, according to the same web page cited by
17
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

Cockburn, (see Weingaertner Decl. at ¶ 19 (Ex. S), http://www.android.com/timeline.html), was

2 released on November 12, 2007, almost a year before the G1 phone first shipped. Cockburn’s
3 date of hypothetical negotiation is, at a minimum, eleven months later than it should be for the
4 SDK. Cockburn completely ignored his client’s own allegations in setting an artificially late
5 date.
6

Cockburn magnifies the impact of this error by attributing so much significance to the

7 2008 date and the purported “window closing” or need to come to market so quickly that
8 allegedly existed as of that date. Most obviously, Cockburn uses the advanced date to justify his
9 inflated damages number by asserting that Google knew Android would be successful and thus
10 would pay anything to protect it. Cockburn also uses the advanced date to posit a hypothetical
11 negotiation between Google and Oracle, an unwilling negotiator who is seeking
12

rent in litigation, rather than between Google and Sun, which always licensed Java and its

13 components on reasonable terms. (Ex. A, Cockburn Rep. ¶ 134.) This is another critical error,
14 because Cockburn claims that the value of Oracle’s Java-based products supports a higher
15 royalty rate. (Id. at ¶¶ 143-156.) But Google and its proper negotiating counterparty Sun never
16 would have negotiated license terms based on the value of Oracle’s, or any third party’s,
17 products, in 2008 or at any other time.
18
19
20

D.

Cockburn Ignores Record Facts And Undisputed Market Realities.
1.

Cockburn ignores the actual market value of Java, as proven by
numerous licenses granted by Sun, Google’s negotiations with Sun,
and the value assigned to Java by Oracle.

21
22
23
24
25
26
27
28
18
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1
2
3
4

Not coincidentally, the real world values Java
much less dearly than Oracle would prefer. (See Leonard Decl. at ¶¶ 12-13.)
Cockburn acknowledges that Oracle valued Sun’s assets prior to acquiring Sun, but he
ignores the implications of that valuation.

5
6
7
8
9

Instead,

10

Cockburn argues that these figures should not place a “cap” on the damages award. (Ex. A,

11

Cockburn Rep. ¶ 327). Of course they should. Hypothetically, Google would be negotiating for

12

U.S. rights to a portion of the JavaME subset of the Java platform. Essentially, Cockburn is

13

telling the Court that it is reasonable to value a single thread in a carpet at three times the price of

14 the entire carpet. That does not pass any test, and certainly not Daubert.
15

Moving on to Java specifically, Oracle’s own public filings disclose concrete numbers

16 valuing the entire Java platform at a fraction of the figure Cockburn’s assigns to the asserted
17 claims of the patents-in-suit.
18
19
20
21
22
23
24
25
26
27
28
19
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1
2
3
4

Cockburn assumes, without any factual basis, that Sun and Google
would have negotiated an expensive, Java-incompatible license rather
than a cheap, Java-compatible license.

5

As noted above, one of the fundamental premises of Cockburn’s analysis is that Sun

6

desperately feared, and wanted to avoid, any fragmentation of the Java standard. According to

7

Cockburn, avoiding fragmentation was so critical to Sun that it would have insisted on an

8

effective 50% royalty rate for the right to fragment Java.

9
10
11
12

2.

i.

Far from prohibiting fragmentation of Java, Sun promoted such
fragmentation through its licensing program.

Cockburn’s major premise, that Sun vigorously fought Java fragmentation, is false. As
discussed below, Sun promoted fragmentation through its own licensing program.

13
14
15

and paid a modest

16 TCK fee to Sun. In particular, JavaME—the subset of Java at issue here—was fragmented for
17 several years before Android had even been conceived. (See, e.g., Weingaertner Decl. at ¶¶ 2418 29 (Exs. X-CC).) In other words, Sun was willing to live with some amount of fragmentation as
19 long as it was compensated at fairly modest levels.
20

Cockburn also wrongly assumes that an “incompatible Android implementation . . .

21 fragments and undermines not only Oracle’s Java licensing business but also the value of Java as
22 a whole.” (E.g., Ex. A, Cockburn Rep. ¶ 336.) Treating “Java” as one overarching technology
23 for purposes of fragmentation is contrary to reality and would make no sense. JavaME is just
24 one of several branches of the overall Java platform.

Most of Java encompasses

25 implementations for desktop, servers, or enterprise systems.

Even if Android fragmented

26 JavaME, it would have no effect on desktop- or server-based implementations of Java.
27

Cockburn makes one other effort to argue that fragmentation would have been the central

28 issue driving Sun in a hypothetical negotiation. As discussed above, he analogizes to Sun’s
20
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

settlement with Microsoft in a legal dispute over Java in the early 2000s. (See p. 11, supra.) As

2

Cockburn notes, the case itself concerned trademark infringement, false advertising, breach of

3

contract, and similar unfair competition claims—none of which are at issue here. (Ex. A,

4

Cockburn Rep. ¶ 178 n.245.) Cockburn focuses on the $900 million settlement of the case in

5

2004 even though the case involved far more than fragmentation and the agreement was a broad

6

business arrangement that involved more than settlement of litigation (i.e., included covenants

7

not to sue under the parties’ entire patent portfolios). At the same time, Cockburn completely

8

ignores a 2001 settlement in which Microsoft paid Sun just $20 million to settle litigation

9

involving claims related to fragmentation. (Exs. N & O.) In any event, the Sun–Microsoft

10

settlements relate to all of Java, not JavaME. Cockburn gives no reason why the recovery here,

11

related only to JavaME, should be several times that amount.

12

ii.

13

Even if Cockburn’s assumptions about fragmentation are correct,
it is improper to base a damages award on compatibility with
nonpatented features.

14

Cockburn opines that Sun would have charged Google a premium for “an incompatible

15

implementation of the intellectual property” because of its great fear of fragmentation. (Ex. A,

16

Cockburn Rep. ¶ 157.)

17

fragmentation for a modest licensing fee. But in any event, Cockburn has no basis for assuming

18

that Sun and Google hypothetically would have negotiated an “incompatible” license that would

19

have caused purported problems for JavaME. By definition, a license to the patents-in-suit by a

20

willing licensor to a willing licensee would be “compatible” with those patents. In any event,

21

Cockburn never explains what exactly is “incompatible” about the hypothetically negotiated

22

product.

As discussed above, Sun was more than willing to live with

23

Remarkably, Cockburn does not even consider a “compatible” product as a potential

24

“next-best alternative” in his analysis. (See, e.g., Ex. A, Cockburn Rep. ¶¶ 230-263.) In other

25

words, Cockburn ignores that this is supposed to be a hypothetical negotiation between willing

26

parties, instead assumeing that Sun would have been unwilling to license a compatible product,

27

notwithstanding its longstanding TCK program. (See Leonard Decl. at ¶ 20.)

28
21
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1
2
3

Cockburn’s emphasis on fragmentation, and his conclusion that fear of fragmentation

4

would have caused Sun not to license the patents-in-suit, contradicts his comparison of the

5

patents-in-suit to essential standards patents. If indeed Oracle’s patents are analogous to such

6

patents, then Oracle has an obligation to license them on Fair Reasonable And Non-

7

Discriminatory (FRAND) terms.

8

acknowledges that FRAND obligations arise because of “the significant bargaining power their

9

[essential] patents may provide,” but concludes “[o]f course, in the context of a hypothetical

10

license for an implementation of a set of technologies that is by definition incompatible, those

11

considerations do not apply.” (Ex. A, Cockburn Rep. ¶ 130 n.196.) Cockburn wants to have it

12

both ways—he gives Oracle all of the benefits of essential standards patents while refusing the

13

obligations inherent in such patents. No sound economic principle supports that approach.

14
15

iii.

Cockburn dismisses that obligation in a footnote.

He

Cockburn relies on fragmentation as a reason to ignore relevant
evidence, including Java-compatible licenses.

16
17
18
19
20
21
22

Cockburn offers contradictory statements that illustrate the unreliability of his entire

23

analysis on this issue. He states that “Oracle’s practice is not to license particular patents, but

24

rather to license Java.” (Ex. A, Cockburn Rep. ¶ 23.) That suggests that prior Java licenses were

25

for compatible products, and contradicts his refusal to consider the possibility that Sun and

26

Google would have negotiated a Java-compatible license. Sun’s earlier Java licenses, which

27

were for the entire Java platform, not just specific patents, would be highly instructive to a

28

reasonable royalty. Again, Cockburn tries to have it both ways. On the one hand, he refuses to
22
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

value the patented technology at issue here, instead valuing Java as a whole.

2
3
4
5
6
7
8
9

3.

10
11

Cockburn relies on inapposite licenses and ignores relevant ones to
inflate the damages calculation.

After discarding Java licenses actually entered into by Sun, Cockburn trumpets various

12 “Other Mobile IP-Related Agreements.” (Ex. A, Cockburn Rep. ¶ 342.) Those agreements
13 include a license between Qualcomm Inc. and Nokia Corp. for what Cockburn vaguely
14 characterizes as “various wireless technologies.” (Id. at ¶ 342.) Using this contract as an
15 exemplar while ignoring Sun’s agreement with Nokia for Java technology is methodologically
16 unsound, especially given the Federal Circuit’s “vigilance when considering past licenses to
17 technologies other than the patent in suit.” ResQNet, 594 F.3d at 869 (emphasis in original)
18 (citation omitted). Cockburn offers no reason that mere wireless technology “kinship imparts
19 enough comparability to support” his damages estimate. Lucent, 580 F.3d at 1328. Cockburn’s
20 purpose in citing Qualcomm’s agreement with Nokia while ignoring Sun’s agreement with
21 Nokia for Java technology serves only “to inflate the reasonable royalty analysis with
22 conveniently selected licenses without an economic or other link to the technology in question.”
23 ResQNet, 594 F.3d at 872.
24
25

Cockburn also purports to look to “Other Mobile Operating Systems.” (Ex. A, Cockburn

26 Rep. ¶ 341.)
27
28
23
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1
2
3
4

E.

5

Cockburn’s analysis with respect to the asserted copyrights is even more cursory and

Cockburn Offers No Meaningful Analysis Regarding Copyright Damages.

6

unreliable.

7

copyright-protected expression may be “essential or important to a product’s success” and that

8

such success may allow the copyright owner to “charge accordingly.”

(See Leonard Decl. at ¶ 30.)

Cockburn begins with the general premise that

(See, e.g., Ex. A,

9 Cockburn Rep. ¶ 133.) Yet he does not explain how the asserted copyrights are essential to
10 Java’s so-called success, leaving that critical issue to be filled in by “other witnesses.” (Id. at ¶¶
11 21-22.)
12

Nor does Cockburn explain how any copyrighted features of the Java platform have

13 benefited Google. He merely describes Google’s decision to “focus on a single programming
14 language: Java” to leverage the “large Java developer community” to allow Android to quickly
15 and seamlessly “enter the existing mobile ecosystem.” (See, e.g., id. at ¶ 124.) How the asserted
16 copyrights allowed Google to leverage this community goes unexplained.

To the extent

17 Cockburn relies on developers’ ability to write applications in the Java programming language,
18 Oracle has renounced any rights in the language itself. (See February 9, 2011 Transcript of
19 Proceedings (Dkt. 87) at 8:16-18 (“[T]he Java Programming Language, we’re not asserting that
20 we own that programming language for purposes of this case.”).) The rest of the report simply
21 conflates copyright and patent damages. (See, e.g., Ex. A, Cockburn Rep. ¶ 9.2 (“a reasonable
22

royalty for the infringement in this case would reflect the key enabling virtues of the patents and

23

copyrights at issue”); see also id. at ¶ 9.3; ¶ 112; ¶ 113; ¶ 131; ¶ 232; ¶ 340.)

24

Cockburn offers no other copyright damages analysis. He summarizes Oracle’s entire

25

basis for copyright damages in a single, conclusory sentence at the very end of his report: “I

26

understand that, because Android infringes the expression embodied in Oracle’s copyrighted

27

class libraries or APIs, Oracle may also be entitled to receive Google’s profits attributable to the

28
24
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

infringement to the extent that they are not taken into account in computing actual damages.”

2

(Id. at ¶ 349.)

3

IV.

4

CONCLUSION
For all the above reasons, the Court should exclude Cockburn’s report under Daubert.

5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
25
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA

1

DATED: June 14, 2011

KING & SPALDING LLP

2

By: /s/ Scott T. Weingaertner

3

SCOTT T. WEINGAERTNER (Pro Hac Vice)
[e-mail]
ROBERT F. PERRY
[e-mail]
BRUCE W. BABER (Pro Hac Vice)
[e-mail]
1185 Avenue of the Americas
New York, NY 10036-4003
Telephone: [phone]
Facsimile: [phone]

4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

ROBERT A. VAN NEST - #84065
[e-mail]
CHRISTA M. ANDERSON - #184325
[e-mail]
KEKER & VAN NEST LLP
710 Sansome Street
San Francisco, CA 94111-1704
Telephone: [phone]
Facsimile: [phone]
DONALD F. ZIMMER, JR. (SBN 112279)
[e-mail]
CHERYL A. SABNIS (SBN 224323)
[e-mail]
KING & SPALDING LLP
101 Second Street – Suite 2300
San Francisco, CA 94105
Telephone: [phone]
Facsimile: [phone]
IAN C. BALLON (SBN 141819)
[e-mail]
HEATHER MEEKER (SBN 172148)
[e-mail]
GREENBERG TRAURIG, LLP
1900 University Avenue
East Palo Alto, CA 94303
Telephone: [phone]
Facsimile: [phone]
ATTORNEYS FOR DEFENDANT
GOOGLE INC.
26
GOOGLE INC.’S BRIEF IN SUPPORT OF DAUBERT MOTION
HIGHLY CONFIDENTIAL - ATTORNEYS’ EYES ONLY
CIVIL ACTION NO. CV 10-03561-WHA