Monday's filings were all about how to proceed on the copyright issue. That is, proposing what determinations the court needs to make with respect to copyright protection afforded Oracle before it can assess whether Google has infringed.
Back on October 19, at a case management conference, Judge Alsup invited each party to submit a précis describing how they thought the court should proceed on the copyright issues. Each party has done so:
Clearly, the parties have very different views on how to proceed. Oracle is focused "on the originality of the selection and arrangement of classes, subclasses, interfaces, fields, methods, error descriptions, and exceptions in the 37 API design specifications at issue." The Oracle approach favors ignoring the copyright protection that may be afforded (or not afforded) the actual elements within Java and to argue that copyright protection is afforded merely because of what was included and how it was organized. Think phone book, and ask how that turned out for for Rural Telephone in Feist.
Google, on the other hand, has focused on breaking down the copyright claims into their elements in order to show that, to the extent Google actually copied anything, what they copied was not afforded copyright protection. This approach is, among other cases, based on Apple Computer [Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435, 1443 (9th Cir. 1994)] and provides:
(1) The plaintiff must identify the source(s) of the alleged similarity between his work and the defendant’s work.
Google reaffirms this position, and takes odds with the Oracle approach in their response to Oracle's précis (582 [PDF; Text).
(2) Using analytic dissection, and, if necessary, expert testimony, the court must determine whether any of the allegedly similar features are protected by copyright. . . . [U]nprotectable ideas must be separated from potentially protectable expression; to that expression, the court must then apply the relevant limiting doctrines in the context of the particular medium involved, through the eyes of the ordinary consumer of that product.
(3) Having dissected the alleged similarities and considered the range of possible expression, the court must define the scope of the plaintiff's copyright— that is, decide whether the work is entitled to “broad” or “thin” protection. Depending on the degree of protection, the court must set the appropriate standard for a subjective comparison of the works to determine whether, as a whole, they are sufficiently similar to support a finding of illicit copying.
After all of this back and forth, Judge Alsup has decided to save all of the copyright issues of both fact and law until trial, but with one major exception. (584 [PDF; Text]) No later than this Friday, Google is invited to "submit a more developed précis limited to fifteen pages ... that more fully develops its request for an advance determination regarding the 37 API specifications and the issue of their selection, arrangement, and structure. Please use at least two APIs as examples so that the proposal will be concrete and not abstract." By the following Wednesday, Oracle is to respond. So it would appear Google has a shot at forcing Oracle into the details of what was allegedly copied in order to determine whether the underlying content is actually afforded copyright protection.
Winning on this point is not a slam dunk for Google, but, importantly, the burden for identifying exactly what copyright-protected material was infringed lies with Oracle. How the court proceeds on this issue after receiving the requested filings will be interesting.
In the only other matter of interest from Monday, the court clarified what could, or needed to be, filed under seal with respect to Oracle's motion to exclude portions of the Leonard and Cox reports (583 [PDF; Text]), and Oracle has now filed an unredacted version (585 [PDF; Text]) of its earlier motion (Document 558).
Skip To Comments
582 – Filed and Effective: 10/30/2011
Document Text: Letter from Robert A. Van Nest in response to Oracle's precis re originality. (Van Nest, Robert) (Filed on 10/30/2011) (Entered: 10/30/2011)
583 – Filed and Effective: 10/30/2011
Document Text: ORDER PARTIALLY GRANTING AND PARTIALLY DENYING MOTION TO FILE DOCUMENTS UNDER SEAL, Order by Hon. William Alsup granting in part and denying in part 557 Administrative Motion to File Under Seal.(whalc1, COURT STAFF) (Filed on 10/30/2011) (Entered: 10/30/2011)
584 – Filed and Effective: 10/31/2011
Document Text: ORDER REGARDING REQUESTS TO FILE MOTIONS re 562 Letter filed by Google Inc., 567 Letter filed by Oracle America, Inc., 566 Letter filed by Oracle America, Inc., 582 Letter filed by Google Inc.. Signed by Judge Alsup on October 31, 2011. (whalc1, COURT STAFF) (Filed on 10/31/2011) (Entered: 10/31/2011)
585 – Filed and Effective: 10/31/2011
Document Text: MOTION to Strike 583 Order on Administrative Motion to File Under Seal ORACLE AMERICA INC.S MOTION TO EXCLUDE PORTIONS OF THE EXPERT REPORTS OF GREGORY K. LEONARD AND ALAN J. COX filed by Oracle America, Inc.. Responses due by 10/28/2011. Replies due by 11/1/2011. (Attachments: # 1 Exhibit EXHIBITS 4-9 TO THE DECLARATION OF MEREDITH DEARBORN 559 )(Holtzman, Steven) (Filed on 10/31/2011) (Entered: 10/31/2011)
586 – Filed and Effective: 10/31/2011
Document Text: DOCUMENT E-FILED UNDER SEAL re 583 Order on Administrative Motion to File Under Seal EXHIBITS 1 & 2 TO THE DECLARATION OF MEREDITH DEARBORN IN SUPPORT OF ORACLE AMERICA, INC.S MOTION TO EXCLUDE PORTIONS OF THE EXPERT REPORTS OF GREGORY K. LEONARD AND ALAN J. COX 559 by Oracle America, Inc.. (Holtzman, Steven) (Filed on 10/31/2011) (Entered: 10/31/2011)
[Keker & Van Nest LLP letterhead]
October 30, 2011
The Honorable William Alsup
United States District Court, Northern District of California
450 Golden Gate Avenue
San Francisco, California 94102
Re: Oracle America, Inc. v. Google Inc., No. 3:10-CV-03561-WHA (N.D. Cal.)
Dear Judge Alsup:
Google writes in response to Oracle’s October 27 précis, in which Oracle requests leave
to file a summary judgment motion limited to the narrow issue of originality (Dkt. 567). Google
respectfully suggests that its own “copyright Markman” proposal (Dkt. 562) would be a better
use of the Court’s resources, and would do far more to clarify and narrow the issues for trial.
As Oracle points out, the Court’s order on Google’s copyright motion for summary
judgment left open the possibility that “the selection or arrangement of [the] names [in the APIs]
is subject to copyright protection.” (Dkt. 433 at 8.) Oracle proposes that the Court decide only
the narrow issue of originality (as Oracle puts it, deciding whether its selection, arrangement, and
structure is “more than a phone directory,” 10/19/11 Tr. at 48). That would not address whether
the selection, arrangement, and structure of the APIs are uncopyrightable for other reasons.
For example, “functional requirements for compatibility . . . are not protected by
copyright.” Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510, 1522 (9th Cir. 1992) (citing 17
U.S.C. § 102(b)). The selection, arrangement and structure of the APIs are all functionally
required in order to maintain compatibility with those APIs. As Dr. Astrachan has explained
in order for existing code in a language to be compatible and interoperable with new
software written in the same language, the API elements that constitute the language must
also be present, and named and organized identically.
Astrachan Opening Report (Dkt. 262-1) ¶ 131 (emphasis added). Thus, the selection,
arrangement and structure of the APIs are not copyrightable, which means that any similarities
between the Android and Oracle API specifications based on the selection, arrangement and
structure of the 37 API packages at issue are irrelevant to the issue of infringement, whether or
not original. See Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435, 1446 (9th Cir. 1994)
The Honorable William Alsup
October 30, 2011
(plaintiff cannot rely on similarity in expression due to unprotectable elements). For the same
reason, any such similarities are also irrelevant to the claim that the Android core libraries are a
derivative work. See Mirage Editions v. Albuquerque A.R.T. Co., 856 F.2d 1341, 1343 (9th Cir.
1988) (derivative work must incorporate material that could support an infringement claim).
Alternatively, if Oracle contends that Google copied aspects of the selection, structure
and arrangement of the APIs that were not functionally required for compatibility with the 37
API packages at issue, it should be required to identify those aspects, consistent with Apple
Computer. 35 F.3d at 1443 (requiring that the plaintiff “identify the source(s) of the alleged
similarity” (emphasis in original)). Moreover, in addition to the “functional requirements” issue,
the selection, structure and arrangement of the APIs are unprotectable for other reasons, such as
the doctrines of scenes a faire and merger. Addressing the entire issue of copyrightability would
do far more to narrow the issues for trial than would Oracle’s proposal to decide only originality,
which is merely one aspect of copyrightability and not determinative of the issue.
While conceding yet again that copyrightability is a question of law, Oracle nonetheless
asserts that there are “threshold factual determinations” for the jury. (See Dkt. 566 at 3.) But
nowhere in its proposed verdict form (Dkt. 531-1) does Oracle ask the jury to make any such
threshold factual determinations. Instead, Oracle states in its proposed jury instructions that
“upon the close of evidence, the Court should instruct the jury as to which works are protected
by copyright, and which standard of similarity and frame of reference to use when comparing the
allegedly infringing works to the copyrighted works.” (Dkt. 539 at 54.) That is, it proposes that
the Court decide copyrightability without first having the jury make any threshold factual
determinations. And, in fact, the Court should decide the copyrightability issue on its own. Lotus Dev. Corp. v. Borland Int’l, Inc., 788 F. Supp. 78, 96 (D. Mass. 1992) (“issues of
copyrightability, including any fact questions bearing upon them, must be determined by the
court, not the jury” (emphasis added)); see also Pivot Point, Int’l, Inc. v. Charlene Prods., Inc.,
The Honorable William Alsup
October 30, 2011
932 F. Supp. 220, 225 (N.D. Ill. 1996) (Easterbrook, J., sitting by designation).
The Court should also decide now what “works” must be compared as a whole when the
jury decides the issues of infringement. Newton v. Diamond, 388 F.3d 1189, 1195 (9th Cir.
2004); Apple Computer, 35 F.3d at 1443. In its jury instructions, Oracle argues that “[t]he ‘work
as a whole’ is not the entire Java platform, but rather each individual code file.” (Dkt. 539 at
70.) Yet in the very same instruction, it argues that the alleged literal copying must be
considered in conjunction with alleged copying outside those code files. Id. (“The de minimis
defense does not apply if you find that Google infringed Oracle’s copyrights in the 37 API
design specifications, because, collectively, the copying of the API design specifications and the
code files would be more than de minimis.”). The Court should decide what the appropriate
frame of reference is now—and to the extent that Oracle seeks to argue that the “work as a
whole” is something other the works it registered, it should be required to make a showing in
support of that position.
Resolving questions of patent claim construction before trial streamlines trial by allowing
the parties to tailor their factual presentations to the relevant law. The same principle applies to
the copyright claim. Resolving the copyrightability and related legal issues now will focus the
case, and allow for a more efficient trial. Leaving these issues unresolved until the close of
evidence will complicate questions of relevance, and in all likelihood will require the parties to
put in more evidence, to ensure an adequate record regardless of how the Court ultimately
decides the copyrightability and other issues. It also increases the risk of confusing the jury.
Google therefore requests that the Court decide all issues related to copyrightability at
this time, rather than adopting Oracle’s proposal to decide only the limited issue of originality—
which will require the Court’s time and resources but not meaningfully narrow the case.
Robert A. Van Nest
IN THE UNITED STATES DISTRICT COURT
ORACLE AMERICA, INC.,
FOR THE NORTHERN DISTRICT OF CALIFORNIA
No. C 10-03561 WHA
GRANTING AND PARTIALLY
DENYING MOTION TO FILE
DOCUMENTS UNDER SEAL
Plaintiff moves for permission to file under seal various documents supporting its motion
to exclude portions of the expert reports of Gergory K. Leonard and Alan J. Cox (Dkt. No. 557).
The motion is GRANTED IN PART AND DENIED IN PART as follows. As to the brief supporting the
motion and as to exhibits 4–9 to the Dearborn declaration, the motion is DENIED. Those
documents may be filed in the public record. As to exhibits 1 and 2 to the Dearborn declaration,
the motion is GRANTED. Those documents may be filed under seal. Please be aware this is
without prejudice to a future order that quotes from the sealed material and is public.
IT IS SO ORDERED.
Dated: October 30, 2011.
/s/ William Alsup
UNITED STATES DISTRICT JUDGE
IN THE UNITED STATES DISTRICT COURT
ORACLE AMERICA, INC.,
FOR THE NORTHERN DISTRICT OF CALIFORNIA
No. C 10-03561 WHA
At the case management conference on October 19, 2011, each party was invited to
submit one précis request describing a proposed motion. The parties have done so and have
responded to one another’s requests (Dkt. Nos. 562, 566, 567, 582). Having read all submissions,
both précis requests are DENIED except as stated in the next paragraph. The issues raised therein
are best left until after the judge has heard the evidence at trial. The judge will understand the
complications of the issues much better after the detailed trial evidence is laid out and cross
examined. It would be hard for the judge to master the material with the same grasp via a cold
written record. Moreover, the caseload and trial calendar of the Court do not allow time for going
through the witnesses twice. After the judge and jury hear the evidence, the judge then will give
proper law instructions and will rule on questions the court of appeals has designated as solely for
the judge. Counsel may wish to know as much as possible in advance of trial as to how the
rulings will develop, but counsel have no right to insist on such advance rulings.
By NOON ON NOVEMBER 4, 2011, Google Inc. may submit a more developed précis
limited to fifteen pages (double-spaced, twelve-point Times New Roman font, with no footnotes
and no attachments) that more fully develops its request for an advance determination regarding
the 37 API specifications and the issue of their selection, arrangement, and structure. Please use
at least two APIs as examples so that the proposal will be concrete and not abstract. Please stick
to this question and do not veer off into side or alternative issues. By NOON ON NOVEMBER 9,
Oracle America, Inc., may respond with up to fifteen pages. Please, no declarations or exhibits.
The fifteen pages should quote whatever is worth citing.
IT IS SO ORDERED.
Dated: October 31, 2011.
/s/ William Alsup
UNITED STATES DISTRICT JUDGE
585 (Note: This is document 558 with redacted sections re-inserted)
UNITED STATES DISTRICT COURT
ORACLE AMERICA, INC.
NORTHERN DISTRICT OF CALIFORNIA
SAN FRANCISCO DIVISION
Case No. CV 10-03561 WHA
ORACLE AMERICA INC.’S MOTION TO
EXCLUDE PORTIONS OF THE EXPERT
REPORTS OF GREGORY K. LEONARD
AND ALAN J. COX
Dept.: Courtroom 8, 19th Floor
Judge: Honorable William H. Alsup
TABLE OF CONTENTS
INTRODUCTION ...... 1
STANDARD OF REVIEW ...... 3
ARGUMENT ...... 4
I. Leonard’s and Cox’s Technical Opinions That Google Had “Good” And
CONCLUSION ...... 14
“Multiple Acceptable And Effective Non-Infringing Alternatives” Do Not
Meet The Standards Established For Expert Testimony And Should Be
Stricken ...... 4
A. Leonard and Cox Are Not Qualified To Provide Technical
II. Leonard And Cox Have No Experience, Skill, Or Qualifications That
Opinions Concerning Google’s Non-Infringing Alternatives And
Other Technical Matters ...... 4
B. Leonard’s and Cox’s Opinions Are Overwhelmingly Based On
Spoon-Fed Interview Statements By Google Employees ...... 6
C. The Interviews on Which Drs. Leonard and Cox Rely Are
Inconsistent With The Record Evidence ....... 8
D. Drs. Leonard and Cox Offer Opinions Unsupported By Any
Evidence At All ...... 9
Would Permit Them To Testify As To The Level Of Fragmentation Or
“Stagnation” In Java ...... 11
III. Leonard And Cox Offer Opinions Based On Incorrect Legal
Assumptions ...... 12
IV. Leonard Lacks The Knowledge And Expertise To Offer Any Opinions
Critical Of The Conjoint Survey ...... 13
TABLE OF AUTHORITIES
AT&T Wireless Services of California LLC v. City of Carlsbad,
308 F.Supp.2d 1148 (S.D. Cal. 2003) ...... 12
Cabrera v. Cordis Corp.,
134 F.3d 1418 (9th Cir. 1998) ...... 14
Carnegie Mellon Univ. v. Hoffmann-LaRoche, Inc.,
55 F.Supp.2d 1024 (N.D. Cal. 1999) ...... 14
Daubert v. Merrell Dow Pharmaceuticals, Inc.,
509 U.S. 578 (1993) ...... passim
DSU Medical Corp. v. JMS Co., Ltd.,
296 F.Supp.2d 1140 (N.D. Cal 2003) ....... 3
General Elec. Co. v. Joiner,
522 U.S. 136 (1997) ...... 10
Geo. M. Martin Co. v. Alliance Machine Systems Int’l, LLC,
2008 WL 2008638 (N.D. Cal. May 6, 2008) ...... 6
Integra Lifesciences, Ltd. v. Merck KGaA,
331 F.3d 860 (Fed. Cir. 2003) ...... 3
Kilgore v. Carson Pirie Holdings, Inc.,
2006 WL 3253490 (6th Cir.) ...... 11
Lust v. Merrell Dow Pharms., Inc.,
89 F.3d 594 (9th Cir. 1996) ...... 4
Matrix Motor Co. v. Toyota Jidosha Kabushiki Kaisha,
290 F. Supp. 2d 1083 (C.D. Cal. 2003) ...... 12
262 Fed. Appx. at 824-25 ...... 11
Polar Bear Prods, Inc. v. Timex Corp.,
384 F.3d 700 (9th Cir. 2004) ...... 13
Rambus Inc. v. Hynix Semiconductor Inc.,
254 F.R.D. 597 (N.D. Cal. 2008) ...... 6
298 F.3d at 1311 ...... 9
Robinson v. G.D. Searle & Co.,
286 F. Supp. 2d 1216 (N.D. Cal. 2003) ...... 9
Southland Sod Farms v. Stover Seed Co.,
108 F.3d 1134 (9th Cir. 1997) ...... 13
Sundance, Inc. v. DeMonte Fabricating Ltd.,
550 F.3d 1356 (Fed. Cir. 2008) ...... 5
Therasense, Inc. v. Becton, Dickinson and Co.,
2008 WL 2323856 (N.D. Cal. May 22, 2008) ...... 2, 6, 8, 9
Uniloc USA, Inc. v. Microsoft Corp.,
632 F.3d 1292 (Fed. Cir. 2011) ...... 3, 8
Wal-Mart Stores, Inc. v. Qore, Inc.
2009 WL 279096 (N.D. Miss. Feb. 5, 2009) ......12
17 U.S.C. § 504(b) ...... 13
FED. R. EVID. 702 ...... passim
FED. R. EVID. 703 ...... 1, 3, 6, 11
FED. R. EVID. 401 ...... 1
FED. R. EVID. 402 ...... 1
FED. R. EVID. 403 ...... 1
Plaintiff Oracle America, Inc. moves to strike significant portions of the expert reports of Dr.
Gregory K. Leonard and Dr. Alan J. Cox, damages experts for Defendant Google, Inc. Those Reports,
fail to meet the standards for expert opinion and testimony established by Daubert v. Merrell Dow
Pharmaceuticals, Inc., 509 U.S. 578 (1993) and as set forth in Federal Rules of Evidence 401, 402, 403,
702, and 703.
On October 3, 2011 Google served the damages expert reports of Professor Leonard (patent)
and Professor Cox (copyright). (Declaration of Meredith Dearborn (“Dearborn Decl.”) at Ex. 1
(Leonard Report); Ex. 2 (Cox Report).) Dr. Cox’s report incorporates Dr. Leonard’s report in its
entirety, and the two reports contain numerous passages that are identical. Dr. Leonard opines that
Oracle is entitled to a reasonable royalty of $27.8 million for Google’s infringement of its patents.
Using the same evidence, Dr. Cox opines that an appropriate award to Oracle for copyright
infringement would be $6.2 million under the theory of lost fair market value license or $21.1 million
based on purportedly comparable licenses that Sun had with Danger and handset manufacturers. Drs.
Leonard and Cox reach these conclusions by adopting essentially the same framework for calculation of
damages used by Oracle’s damages expert, Prof. Cockburn, but adjusting downward because, in their
opinions, Google had non-infringing alternatives to the patented and copyrighted elements, and any
injury to Sun and Java was actually lessened by Google’s infringement restoring order to a “highly
fragmented” Java mobile platform. Neither of these “opinions” is admissible.
First, Drs. Leonard and Cox, both economists with no technical expertise, offer extensive
opinions that Google had “multiple acceptable and effective non-infringing alternatives” to each of the
Java patents and the copyrights in suit. (Leonard Report, 10-31, 74-78, 80; Cox Report, 23-27, 61.)
These opinions purport to assess the technical and performance attributes of the asserted alternatives,
not their business implications. In large measure, as discussed below, they are based on Drs. Leonard’s
and Cox’s own say-so and last-minute interviews with Google employees. Even where they refer to
Google’s technical experts, Drs. Leonard and Cox do not merely adopt the opinions of those experts;
they go further to affirmatively assert that Oracle’s technical expert is wrong on technical matters.
Neither Dr. Cox nor Dr. Leonard is competent to offer such opinions. Neither should be permitted to
testify about purported non-infringing alternatives or base any economic analysis on their purported
assessment of the presence of such alternatives.
Second, Drs. Leonard and Cox fail to tie their damages opinions – including but not limited to
their core opinions regarding supposed non-infringing alternatives – to the contemporaneous evidence
or even the sworn testimony of witnesses. Instead, Drs. Leonard and Cox rely extensively on
interviews with Google employees arranged only after Prof. Cockburn had submitted his amended
damages report. Indeed, Dr. Leonard cites a grand total of 14 Google documents, compared to 66
citations to interviews. Dr. Cox cites only 18 Google documents (mostly the same ones Dr. Leonard
cites) but 44 interview references.
Analyses based on spoon-fed “facts” are inherently unreliable. See Therasense, Inc. v. Becton,
Dickinson and Co., 2008 WL 2323856, *1-3 (N.D. Cal. May 22, 2008) (Alsup, J.). But these opinions
are doubly unreliable because many of the supposed “facts” Drs. Leonard and Cox glean from the
interviews are contradicted by the contemporaneous record evidence. In many cases what the
interviewees supposedly said contradicts documents those very witnesses wrote. On still other factual
matters, Drs. Leonard and Cox cite to no evidence at all – they just say so. The failure to use reliable,
contemporaneous evidence, and instead substitute ex post interview statements and the experts’ own
invented facts render the reports unreliable and inconsistent with the standards of Daubert and the
Rules of Evidence. Each and every proposition in the reports that rests solely on interview statements
or ipse dixit should be stricken, and Drs. Leonard and Cox should be precluded from providing trial
testimony regarding or based on such points.
Third, Drs. Leonard and Cox opine – based, yet again, on interviews and web searches – that
Java was already “heavily fragmented” or “stagnant” before Google’s infringement began. Neither Dr.
Leonard nor Dr. Cox applies any standard or metric, scientific or otherwise, to measure the level of
supposed pre-existing fragmentation. Neither has any training, experience, or expertise that would
allow him to testify competently as to the level of fragmentation in a software standard or the extent to
which it is vital and dynamic, as opposed to “stagnant.” Consequently, their opinions regarding the
supposed fragmentation of Java, and the impact of such fragmentation on any hypothetical negotiation
between the parties or any Sun or Oracle lost profits, should be stricken and excluded at trial.
Fourth, Drs. Leonard and Cox both offer opinions that are contrary to law. Dr. Leonard argues
that Oracle’s patent damages should be reduced for failure to mitigate damages, despite the fact Google
pled no such defense, and it has sought no jury instruction for such a defense. Dr. Cox opines that
Google’s liability for infringer’s profits should be reduced if Google had non-infringing alternatives to
the Java copyrights, contrary to the very principle of disgorgement.
Fifth, Dr. Leonard ventures beyond his expertise to critique the scientific analysis – a conjoint
survey – conducted by Dr. Steven Shugan. Dr. Shugan is an expert in using surveys and sophisticated
statistical methods to assess consumer preferences. Dr. Leonard is not. As explained in Dr. Shugan’s
accompanying declaration, Dr. Leonard is unfamiliar with the relevant literature, cites articles for
propositions they do not contain or actually reject, advocates methodologies that are novel and untested
in place of the established and reliable techniques used by Dr. Shugan, and misapplies basic terms and
concepts that would be familiar to anyone with actual expertise in the field. The entirety of Dr.
Leonard’s report regarding conjoint analysis should be stricken. Dr. Leonard should be precluded from
offering testimony regarding Dr. Shugan’s work, Dr. Cockburn’s reliance on it, or conjoint analysis
STANDARD OF REVIEW
A witness who is qualified as an expert by knowledge, skill, experience, training or education
may testify in the form of an opinion or otherwise if the testimony is based on sufficient facts or data.
See FED. R. EVID. 702. Expert opinion based on methodologies that are scientifically unreliable and on
data upon which no reasonable expert would rely is inadmissible. See FED. R. EVID. 702, 703; see also
DSU Medical Corp. v. JMS Co., Ltd., 296 F.Supp.2d 1140, 1156 (N.D. Cal 2003) (“There is a threshold
issue of admissibility, however, requiring the Court to examine the connection between the opinion
proffered and the reconstructed market data.”). A damages expert, in particular, must sufficiently “tie
the expert testimony on damages to the facts of the case.” Uniloc USA, Inc. v. Microsoft Corp., 632
F.3d 1292, 1315 (Fed. Cir. 2011) (citing Daubert) (internal punctuation marks omitted); Integra
Lifesciences, Ltd. v. Merck KGaA, 331 F.3d 860, 870 (Fed. Cir. 2003) (“although an exercise in
approximation,” the hypothetical negotiation analysis “must be based on sound economic and factual
predicates”); vacated on other grounds, 545 U.S. 193 (2005).
Because expert testimony can “be both powerful and misleading, the Supreme Court
emphasized the “gatekeeping responsibility” federal judges have over the admission of expert
testimony. Daubert, 509 U.S. at 595, 597. As the proponent of the expert, Google has the burden of
proving admissibility. Lust v. Merrell Dow Pharms., Inc., 89 F.3d 594, 598 (9th Cir. 1996).
I. Leonard’s and Cox’s Technical Opinions That Google Had “Good” And “Multiple
Acceptable And Effective Non-Infringing Alternatives” Do Not Meet The Standards
Established For Expert Testimony And Should Be Stricken.
A core premise for both Dr. Leonard and Dr. Cox is that Google had good and acceptable noninfringing
alternatives to the Java patents and copyrights in suit. Drs. Leonard and Cox have neither the
qualifications nor the record support to proffer opinions on the technical subject of non-infringing
alternatives. They base those opinions on their own unqualified assertions and undocumented
interviews that occurred after Prof. Cockburn submitted his amended report.
A. Leonard and Cox Are Not Qualified To Provide Technical Opinions Concerning
“A witness qualified as an expert by knowledge, skill, experience, training or education, may
testify thereto in the form of an opinion or otherwise, if (1) the testimony is based upon sufficient facts
or data . . . .” FED. R. EVID. 702. Drs. Leonard’s and Cox’s opinions that Google had “good” and
“multiple acceptable and effective” non-infringing alternatives to the patents and copyrights in suit
(Leonard Report, 10-31, 74-78, 80; Cox Report, 23-27, 61), and that Oracle’s technical expert, Dr.
Mitchell, is wrong about alternatives (Leonard Report, 27-29), fail this test.
Google’s Non-Infringing Alternatives And Other Technical Matters.
Dr. Leonard holds a Bachelor of Science in Applied Mathematics-Economics and a Doctorate in
Economics. (Leonard Report, App’x A.) Dr. Cox holds a Masters degree in Economics and a
Doctorate in Business Administration. (Cox Report, App’x A.) Neither has any experience with or
training in computer science or other technical fields.
Nonetheless, Dr. Leonard offers extensive technical opinions, including that:
- “Google could have provided developers with a native compiler for the Java programming
language instead of using the Dalvik Virtual Machine in Android” Leonard Report, 11-13
(citing interviews and a blog, but no technical experts));
- “Google could have used a programming language other than Java” (id.,13-19 (primarily citing
“interviews” and only one Google-produced document));
- “Google could have retained the Dalvik Virtual Machine” without Oracle’s patents (id.,20-24
(citing interviews in 20 out of 30 footnotes, and no Google documents)); and
- “Dr. Mitchell’s [Oracle’s technical expert] claims are inconsistent with my discussions with
Google personnel and the opinions of Google’s technical experts regarding the effects of using
non-infringing alternatives in place of the functionalities allegedly covered by the patents.” (id.,28).
Where Drs. Leonard and Cox occasionally cite to Google’s technical experts concerning noninfringing
alternatives, their opinions extend beyond what they cite. For example, Dr. Leonard opines
that C++ “would have had certain advantages” over choosing Java. (Id.,14.) However, only one of the
purported advantages that Dr. Leonard lists is attributable to Google’s technical experts. (Id., 14 n.31.)
Similarly, while Dr. Leonard cites to Google’s technical expert in opining that the functionality
provided by the ’520 patent could have been removed (Id.,20 n.56, n.57, n.59), he has no citation to that
expert for his conclusion that “Google could have not included this functionality without any
performance loss for Android that was noticeable to end users” (id., 20). And in discussing the ’720
patent, Leonard cites to Google’s technical expert to state that “[i]t is trivial to remove this
functionality,” but fails to cite to any technical expert for his opinion that a non-infringing alternative
was the removal of the Zygote process that uses that functionality. (Id., 23.)
In all, his assessment of supposed non-infringing alternatives leads Dr. Leonard to conclude that
“Google would not have been willing to pay much of a royalty to obtain a license to the patents-insuit.”
(Id. 18.) Dr. Cox incorporates Dr. Leonard’s opinions into his report and also independently
opines that Google could have used a programming language other than Java and hence relied upon
different patents and copyrights. (Cox Report, 23.)
Neither Dr. Leonard nor Dr. Cox is “qualified as an expert” in computer science or
programming, and neither can “assist the trier of fact to understand the evidence or to determine a fact
in issue.” Sundance, Inc. v. DeMonte Fabricating Ltd., 550 F.3d 1356, 1364-65 (Fed. Cir. 2008)
(“Allowing a patent law expert without any technical expertise to testify on the issues of infringement
and validity amounts to nothing more than advocacy from the witness stand.”) Technical opinions
proffered by either expert are unreliable and cannot be presented to the jury. To qualify as an expert on
technical issues such as infringement or validity, a witness must be a person “of an ordinary skill in the
art.”) Id. at 1361; see also Rambus Inc. v. Hynix Semiconductor Inc., 254 F.R.D. 597, 603-05 (N.D.
Cal. 2008) (expert with 33 years experience in electrical engineering and semiconductor design
precluded from providing certain opinions about commercial success).
B. Leonard’s and Cox’s Opinions Are Overwhelmingly Based On Spoon-Fed Interview
Statements By Google Employees
Rule of Evidence 703 permits an expert to offer opinion testimony based on materials of a type
reasonably relied on by experts in a given field. In Therasense, this Court warned: “One of the worst
abuses in civil litigation is the attempted spoon-feeding of client-prepared and lawyer-orchestrated
‘facts’ to a hired expert who then ‘relies’ on the information to express an opinion.” 2008 WL
2323856, at *1; see also Geo. M. Martin Co. v. Alliance Machine Systems Int’l, LLC, 2008 WL
2008638, at *1 (N.D. Cal. May 6, 2008) (Alsup, J.) (“Important factual points should not be proven
through experts spoon-fed by wholly biased sources.”). Drs. Leonard and Cox are guilty of this abuse.
Unable to find adequate support for their opinions in the reports and testimony of Google’s technical
experts or in record evidence, they rely primarily on interviews with Google employees conducted the
week prior to service of the Reports and after the conclusion of fact depositions. In over 25 pages of
purported analysis regarding non-infringing alternatives (Leonard, 10-31, 74-78, 80), Dr. Leonard cites
interviews with Google employees 56 times; he cites precisely 2 Google documents. Similarly, in over
5 pages of discussion (Cox Report, 23-27, 61), Dr. Cox cites interviews 24 times, and zero
Such heavy reliance on interviews, to the near exclusion of the record evidence, eliminates the
reliability of Drs. Leonard’s or Cox’s technical opinions that Google supposedly had non-infringing
alternatives. Permitting such testimony would improperly allow Google to circumvent the hearsay rules
by filtering supposed facts through experts, while simultaneously keeping the facts secret from the other
side until after the close of discovery. Therasense, 2008 WL 2323856, at *1. That is precisely what
Google has done. Over and over again, Google’s damages experts rely only on interviews for critical
propositions. As just a few examples of many, Dr. Leonard:
- cites only an interview with Android engineer Dan Bornstein to opine that “[t]he out-of-pocket
cost and time to Google of developing a native compiler and associated tools would have been
approximately the same as the cost and time of developing the Dalvik virtual machine.” (Leonard, 12)
- cites only interviews with Google employees Tim Bray and Brian Swetland for the double
hearsay assertion that “Google has received complaints from certain developers regarding
Google’s choice of the Java programming language because it is slow.” (Id. at 14);
- cites only interviews with Google employees Andy Rubin, Bornstein, and Swetland to opine
that “[t]he 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.” (Id. at
- cites only interviews with Bray to opine that “the speed of an application often depends much
more crucially on the ‘Android applications framework’ provided by Android than on the way
the applications are coded.” (Id. at 26);
- cites only an interview with Bornstein to opine that Google had “myriad ways to improve
performance” at the time of the hypothetical negotiation that “would all be non-infringing
alternatives to the functionalities allegedly covered by the patents-in-suit.” (Id. at 27);
- cites only an interview with Rubin to opine that OEMs “would have accepted small reductions
in performance or somewhat more demanding handset specifications in order that they could
still be able to offer Android smartphones.” Id. at 30.)
Similarly, Dr. Cox relies almost exclusively on interviews with Google employees for his
opinion regarding non-infringing alternatives, citing to interviews 24 times, 14 of which serve as the
sole source for his stated opinion. For example, Dr. Cox
- cites only an interview with Bray to opine that the choice of the Java language has imposed “negative externalities on Android” because “some Java ME developers have done a poor job
porting their applications to Android”; “many Java programmers were not used to the sort of
memory and other limitations that come with the use of smaller, mobile processors”; “some Java
SE developers have had trouble adapting to the mobile environment;” and “programmers who
tend to create the most interesting, forward-thinking applications do not like or want to use the
Java programming language, because it is considered obsolete” (Cox Report, 22-23);
- cites only interviews with Bray and Swetland to opine that “[i]t would have been as easy, or
easier, for an applications developer who had written an iPhone application in Objective C to
port that application to Objective C or C++ for use in Android as it was to port it to or write it
from scratch in the Java programming language” (Id., 25);
- cites only interviews with Rubin, Bornstein, and Swetland to opine “[t]he incremental cost
and time to Google of using C++ as the applications programming language for Android would
not have been significant, and in light of this viable alternative Google would not have been
willing to pay a large royalty for a copyright license to the 37 API packages.” Id., 61.)
Opinions such as these – based on interview testimony without sufficient factual predicate – do
not meet the established standard for expert testimony, are unreliable, and must be stricken.
C. The Interviews on Which Drs. Leonard and Cox Rely Are Inconsistent With The
Compounding the problem inherent in its experts’ extensive reliance on interviews, Google has
refused to make available five of the seven interviewees even for limited two-hour depositions
regarding the interviews, all of which took place well after the close of fact discovery. (Dearborn Decl.
Ex. 3.)_ But even such depositions could not cure the defect in the Reports, because the spoon-fed
“facts” culled from these interviews – in many cases the only evidence cited for critical propositions –
contradict the record evidence, virtually all of which Drs. Leonard and Cox ignore. Consequently, Drs.
Cox and Leonard fail to “sufficiently tie the expert testimony on damages to the facts of the case.”
Uniloc, 632 F.3d at 1315.
The incurable problem with this practice occurs when the expert is permitted to regurgitate what
he has been fed as though it were, contrary to fact, the product of his independent analysis of an
objective record. As this Court has held, “no professional should reasonably rely on such a rigged and
biased source of information for any materially important fact to his or her opinion.” Therasense, 2008
WL 2323856, at *2. Drs. Leonard and Cox have gone one step further, relying on the “rigged and
biased source of information” to the exclusion of the record evidence. To list just a few examples:
- Drs. Leonard and Cox opine that C++ was a viable alternative to the Java programming
language, citing to interviews with Andy Rubin, Dan Bornstein, and Brian Swetland. (Leonard
Report at 13; Cox Report at 24), But they ignore that in October 2005, Mr. Rubin saw only
“two options” without a Java license: Microsoft’s C# or “Do Java anyway . . .making enemies
along the way,” (Dearborn Decl. Ex. 4 (GOOGLE-01-00019527)), not even mentioning C++ on
the list of the “sub-optimal” alternatives; that Mr. Swetland prepared an internal analysis in
January 2006 that exhaustively catalogued reasons that C++ was not a viable alternative (Dearborn Decl. Ex. 5 (GOOGLE-01-00019511)); and that in 2006, another Android engineer,
Andy McFadden, instructed members of the engineering team not to write code in C++, noting
schedule concerns. (Dearborn Decl. Ex. 6 (GOOGLE-01-00075935).
- Dr. Leonard states, citing to his interview of Andy Rubin, that in the late 2005 time frame,
Google was considering both the native compiler approach and the virtual machine approach,
and it was a close call for Google as to which direction to take (Leonard Report at 11), but he
ignores documents that – as early as August 2005 – made clear that “The JVM is going to be a
central piece of the system we’re building, not some little add-on on the side – so we can
provide some really good java application development and user experiences” Dearborn Decl.
Ex. 7 (GOOGLE-11-00000958, August 2005); and that Android should have “minimal native
code - write as much as possible in java - we are building a java based system: that decision is
final” (Dearborn Decl. Ex. 8 (GOOGLE-04-00055169).)
- Dr. Cox states that “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” (Cox Report at 20), but
completely ignores documents such as the October 2005 internal e-mail in which Mr. Rubin told
Google co-founder and current CEO Mr. Page that “We are making Java central to our solution
because a) Java, as a programming language, has some advantages because it’s the #1 choice for
mobile development . . ..” (Dearborn Decl. Ex. 4 (GOOGLE-01-00019527).)
The spoon-fed “facts” obtained during interviews conducted days prior to the service of the
Reports which are contradicted by contemporaneous record evidence are not “sufficient facts or data”
as required by Rule 702, nor do they form an adequate “factual predicate” as required by the Federal
Circuit. See Riles, 298 F.3d at 1311; Therasense, 2008 WL 2323856, at *1-2; Robinson v. G.D. Searle
& Co., 286 F. Supp. 2d 1216, 1221 (N.D. Cal. 2003) (expert’s testimony inadmissible when based on
factual premise directly contradicted by evidence on the record).
Moreover, the “facts” on which Drs. Leonard and Cox rely as to non-infringing alternatives are
materially important to their damages analyses. Dr. Leonard relies heavily on the alleged availability of
non-infringing alternatives to Google in concluding that Sun’s bargaining position was weak and
Google’s was strong (Leonard Report at 10, 44) and as a rationale for downward or neutral pressure
from Georgia-Pacific factors 6, 7, 8, 9,11, 13 (id. at 68-71) in his patent damages analysis. Dr. Cox
relies on alleged alternatives to conclude that copyright damages should be zero (Cox Report at 27, 61);
to assert that the “‘element of profit’ that is attributable to the allegedly infringed API claim contained
in the Android framework is very small or zero” (id. at 38); to explain why Google was unwilling to
enter into a license with Sun (id.); to determine that Sun could not have prevented any losses by
denying a license to Google (id. at 48); and as a significant factor in his hypothetical negotiation
analysis (id. at 60-61).
Each aspect of the reports relying on assessments of non-infringing alternatives should be
stricken and Drs. Leonard and Cox should be precluded from offering opinions at trial based on the
conclusion that Google had non-infringing alternatives to the copyrights and patents in suit.
D. Drs. Leonard and Cox Offer Opinions Unsupported By Any Evidence At All
In many cases, Drs. Leonard and Cox cite no evidence at all for key propositions. For example,
Dr. Leonard states that “Apple’s success using the native compiler approach demonstrate [sic] that it
would have been an acceptable and effective non-infringing substitute for Google” (Leonard Report at
12), that “even in the context of attracting developers to the Android platform, the advantage, if any, of
the Java programming language is marginal” (id. at 15), that “demand for a smartphone is a complex
function of all of its attributes, and a smartphone with slower application speed can still be highly
demanded by consumers” (id. at 25), and that smartphone manufacturers “would have accepted small
reductions in performance or somewhat more demanding handset specification in order that they could
still be able to offer Android smartphones.” (Id. at 30.) In each case, Dr. Leonard cites nothing in
support of the proposition.
Such reliance on ipse dixit is impermissible. See General Elec. Co. v. Joiner, 522 U.S. 136, 146
(1997) (“But nothing in either Daubert or the Federal Rules of Evidence requires a district court to
admit opinion evidence that is connected to existing data only by the ipse dixit of the expert.”);
Daubert, 509 U.S. at 590 (expert testimony must connote “more than subjective belief or unsupported
speculation”). This is particularly true where the ipse dixit is not within the expert’s area of expertise,
as with Dr. Leonard’s opinions about native compilers, attractiveness of programming languages, and
the importance of application speed.
Many of Dr. Cox’s opinions suffer from the same defect. Without citation, he speculates that “it
was not until handset manufacturers such as Motorola and Samsung decided to design and market
Android handsets that Android became successful” (Cox Report at 26) and thus “the success of Android
did not turn on the existence of the Java developers or applications written in the Java programming
language.” (Id.) Without citation, he opines that the APIs at issue contribute a small part of the
functionality of the Dalvik VM. Without citation, he states that “I understand Android phones use the
Dalvik virtual machine only about a third of the time.” (Id. at 28). Without citation, he declares that “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.” (Id.) Dr. Cox then uses this unsupported assertion to
support his claim “the contribution of the material covered by Oracle’s API claim provided little in
value compared to the elements contributed by Google” and thus a “low or zero damage for the alleged
copyright infringement is appropriate.” (Id. at 28) These unsupported, ipse dixit propositions render
Dr. Cox’s analysis of supposed non-infringing alternatives unreliable and inadmissible.
II. Leonard And Cox Have No Experience, Skill, Or Qualifications That Would Permit Them
To Testify As To The Level Of Fragmentation Or “Stagnation” In Java.
In near-verbatim sections of their respective reports, Drs. Leonard and Cox each opine that Sun
did not suffer significant losses from the deployment of Android because Java was already “stagnant”
and “highly fragmented” at the time of Android’s launch (Leonard Report at 31-37; Cox Report at 43-
48). Drs. Leonard and Cox have no expertise that would allow them to offer these opinions. Fed. R.
Evid. 702. Instead, they rely heavily on an interview with John Rizzo (Leonard Report at 35-36; Cox
Report at 47), internet blog posts and random comments on internet bulletin boards (Leonard Report at
35-36; Cox Report at 44, 46-48), and “various online sources” (Leonard Report at 35; Cox Report at
but ignore record evidence, including Google documents that extol the virtues of Java technology
based on its minimal fragmentation. For example, a document written by one of the interviewees
himself in April 2006 made clear:
“We need to provide an alternative to MSFT, and we need to do it in such a way as we
don’t fragment 3rd party developers. See the next slide in the deck for fragmentation:
Java has very little fragmentation, and it’s adoptable. If we play our cards right, we
can also leverage not only existing developers, but applications as well.” (Dearborn
Decl. Ex. 9 (GOOGLE-02-00111218).)
Similarly, Drs. Leonard and Cox completely ignore documents that show that Sun fought to ensure
compatibility, including in the license agreement that was being negotiated between Sun and Google.
(Dearborn Decl. Ex. 4 (GOOGLE-01-00019527 at 528) (“We’ll pay Sun for the license and the TCK.
Before we release our product to the open source community we’ll make sure our JVM passes all TCK
certification tests so that we don’t create fragmentation.”).)
In short, both Dr. Leonard and Dr. Cox fail to rely on facts or data customarily relied upon by
experts in the field. FED. R. EVID. 703; Nuveen, 262 Fed. Appx. at 824-25; Kilgore v. Carson Pirie
Holdings, Inc., 2006 WL 3253490, at *4 (6th Cir.) (internet article an unreliable basis for methodology
where expert did not know on what methodology the article was based and conducted no independent
research); Matrix Motor Co. v. Toyota Jidosha Kabushiki Kaisha, 290 F. Supp. 2d 1083, 1086 (C.D.
1 Both also repeatedly cite a YouTube video of James Gosling speaking on November 17, 2010 – three
months after this lawsuit was filed – and falsely refer to him as a “then Sun employee,” when he in fact
he had left Sun seven months earlier. (Leonard Report, 35-36; Cox Report, 47-48).
Cal. 2003) (expert reports “irrelevant” where they “merely recite hearsay statements, often verbatim,
culled from a variety of Internet websites”).
Moreover, Drs. Leonard and Cox neither attempt to define “stagnant” or “fragmented” nor apply
any methodology, scientific or otherwise, that would distinguish between “stagnant” and “dynamic”
software platforms, or between “highly fragmented” and “minimally fragmented” ones. Expert
opinions that neither employ nor can be evaluated against objective criteria do not meet the threshold
standard of admissibility. AT&T Wireless Servs. of Cali. LLC v. City of Carlsbad, 308 F.Supp.2d 1148,
1157 (S.D. Cal. 2003); see also Daubert, 509 U.S. at 593 (“Ordinarily, a key question to be answered in
determining whether a theory or technique is scientific knowledge that will assist the trier of fact will be
whether it can be (and has been) tested.”). As one court held:
Daubert’s gatekeeping function is specifically designed to keep standardless testimony
out of evidence. Majors opines on a variety of subjects, but he offers no guide as to
where his conclusions came from or how one might judge the value of those
conclusions. Operating in a standardless world allows Majors to get on the stand and
make any assertion he chooses without regard for the truth. There is no way to judge
those who operate without bounds. It is unfair to put a jury in the position of having to
Wal-Mart Stores, Inc. v. Qore, Inc., 2009 WL 279096, *2 (N.D. Miss. Feb. 5, 2009). Drs. Leonard’s
and Cox’s standardless opinions on “stagnation” and “fragmentation” fare no better.
III. Leonard And Cox Offer Opinions Based On Incorrect Legal Assumptions.
Prof. Cockburn made an upward adjustment to the starting point license to account for Sun’s
expectation of revenue from providing implementations for a compatible Android. Dr. Leonard opines
that this adjustment should be “zero” because Sun “could have mitigated” the loss of that opportunity
(Leonard Report at 83; see also id. at 84 n.278.) But Google pleaded no mitigation defense and it has
not offered any jury instruction on mitigation. Google should not be able to plead a defense it has never
once raised by offering an opinion through its expert.
Dr. Cox’s discussion of infringer’s profits also rests on an error of law. Dr. Cox argues that
infringer’s profits should be “very small or zero” because Google supposedly had non-infringing
alternatives to the APIs and code files that it infringed. (Cox Report, 38). That is wrong. A prevailing
plaintiff in a copyright case is entitled to recover the infringer’s profits attributable to the infringement,
to the extent not captured by actual damages. 17 U.S.C. § 504(b). The defendant is not permitted to
avoid disgorgement by arguing that it would have been able to earn those same profits without
infringing. To allow such a result would undermine the purpose of infringer’s profits, which requires
simple disgorgement in order “[t]o take away incentives for would-be infringers and to prevent the
infringer from unfairly benefitting from a wrongful act.” Polar Bear Prods, Inc. v. Timex Corp., 384
F.3d 700, 708 (9th Cir. 2004) (citations and quotations omitted).
IV. Leonard Lacks The Knowledge And Expertise To Offer Any Opinions Critical Of The
In analyzing apportionment, Oracle’s damages expert Prof. Cockburn worked with an expert in
conjoint marketing surveys, Dr. Steven M. Shugan, to conduct a survey and assess the importance to
consumers of the attributes provided by the patents and copyrights in suit. Dr. Leonard, offering
absolutely no affirmative apportionment analysis of his own, tries to critique the conjoint analysis and
claims that it cannot be used to calculate damages, but his attack shows only that he lacks the necessary
expertise. Dr. Shugan, who actually is an expert in survey-based research methods and statistical
analyses of their results, has submitted a declaration detailing the ways in which Dr. Leonard’s analysis
is inexpert, incorrect, and unreliable. (Declaration of Steven M. Shugan (“Shugan Decl.”).)
Dr. Leonard begins with an inappropriate bias against survey-based research that cannot be
reconciled with its proven and accepted use in courts, academia, and business. (Compare Leonard
Report at 109 (characterizing survey-based methods as “controversial”) with Southland Sod Farms v.
Stover Seed Co., 108 F.3d 1134, 1143 (9th Cir. 1997) (holding that “survey evidence should ordinarily
be found sufficiently reliable under Daubert”).) Given his hostility toward survey-based research, it is
perhaps unsurprising that he demonstrates little familiarity with its literature, methods, research, and
- Dr. Leonard asserts that there are “established results in the literature demonstrating
hypothetical bias in conjoint studies” (Leonard Report at 111 fn.340). In fact, there are not.
The literature establishes that hypothetical bias is irrelevant with established consumer products,
such as smartphones, and is a factor only in the context of abstract goods, such as environmental
clean-up. (Shugan Decl. ¶¶ 10, 12.)
- Dr. Leonard cites articles from the consumer behavior and psychological literature for
propositions that they do not contain. (Id. ¶ 13).
- Dr. Leonard cites articles from the consumer behavior and psychological literature for
propositions that they actually reject. (Id. ¶ 14.)
- Dr. Leonard ignores the fact that even the articles that he himself cites recognize that conjoint
analysis methods and related choice-based survey methods are routine, reliable, and based on
years of research. (Id. ¶ 15.)
- Dr. Leonard misinterprets the work of Prof. Min Ding on hypothetical bias as a critique of
conjoint analysis and as indicating that Dr. Shugan’s research is inaccurate. In fact, Prof. Ding
affirms the reliability of conjoint analysis. If Prof. Ding’s analyses were relevant, they would
establish that Dr. Shugan’s conclusions are conservative. (Id. ¶¶ 16-19.)
- Dr. Leonard’s challenge to the conjoint survey is based on his mistaken belief that the conjoint
survey is a “stated preference survey.” It is not. Anyone with expertise in the field would know
the difference between conjoint analysis and stated preference surveys, and would further know
that the problems with stated preference surveys cited by Dr. Leonard do not apply to conjoint
analysis. (Id. ¶¶ 20-21.)
- Dr. Leonard incorrectly assumes that consumers will always choose a lower price if other
attributes are held constant, despite extensive literature demonstrating that this is not so, and
explaining why. (Id. ¶ 22.)
- Dr. Leonard mischaracterizes standard robustness measures for conjoint analysis, such as U2 and
hit rate, as “tests” for “hypothetical bias.” (Id. ¶ 24.)
In short, Dr. Leonard’s discussion of the conjoint analysis (1) “fails to understand or apply the
basic principles that are accepted in the field of survey-based research” (Shugan Decl. ¶ 8); (2) “suffers
from sufficient ambiguities, errors and misrepresentations that any submitted article based on that
analysis would not survive a peer-review process of a scholarly journal” (id.); and (3) “reveal a
fundamental lack of understanding of choice-based-conjoint or statistical analyses” (id. ¶ 25). See
Daubert, 509 U.S. at 593-94 (emphasizing importance of peer review standards); Cabrera v. Cordis
Corp., 134 F.3d 1418 (9th Cir. 1998) (expert testimony unreliable because it was specifically prepared
for trial and had no support in medical literature); Carnegie Mellon Univ. v. Hoffmann-LaRoche, Inc.,
55 F.Supp.2d 1024, 1032-33 (N.D. Cal. 1999) (exclusion of expert testimony upon finding that studies
relied upon were not sufficient individually or in combination to support expert’s opinion).
For the aforementioned reasons, Oracle respectfully requests that the Court strike those portions of the
Leonard and Cox reports relating to, and preclude Drs. Cox or Leonard from offering opinions as to (1)
whether Google had acceptable non-infringing alternatives to the patents- and copyrights-in-suit; (2)
whether Java was fragmented at the time of the hypothetical negotiation; (3) whether Android causes
fragmentation of Java; (4) whether Prof. Shugan’s survey and conjoint analysis is unreliable; (5) the
impact of mitigation on damages; (6) the effect on copyright infringer’s profits damages of the presence
or absence of non-infringing alternatives; and (6) any matter that either Dr. Cox or Dr. Leonard learned
only through an interview. A proposed order consistent with the foregoing is submitted herewith.
Dated: October 21, 2011
BOIES, SCHILLER & FLEXNER LLP
By: /s/ Steven C. Holtzman
Steven C. Holtzman
Attorneys for Plaintiff
ORACLE AMERICA, INC.