To read comments to this article, go here
|Oracle v. Google - A Slow Day
Wednesday, November 09 2011 @ 11:35 AM EST
Although we had expected to see Oracle's response to Google's 15-page copyright précis, the only filing of significance yesterday was Google's opposition to Oracle's motion to exclude portions of the Leonard and Cox reports. (604 [PDF; Text]) This is the same document as 581, (Opposing Positions (to Motions) except that, as ordered by the court (602 [PDF]; Text]) the quoted paragraph on page 15 is no longer redacted.
Other than that, it was a pretty quiet day.
602 – Filed and Effective: 11/07/2011
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 580 Administrative Motion to File Under Seal.(whalc1, COURT STAFF) (Filed on 11/7/2011) (Entered: 11/07/2011)
603 – Filed and Effective: 11/08/2011
Document Text: ORDER REGARDING CONSTRUCTION OF CLAIMS TO BE TRIED. Signed by Judge Alsup on November 8, 2011. (whalc1, COURT STAFF) (Filed on 11/8/2011) (Entered: 11/08/2011)
604 – Filed and Effective: 11/08/2011
Document Text: RESPONSE (re 558 MOTION to Strike ORACLE AMERICA INC.S MOTION TO EXCLUDE PORTIONS OF THE EXPERT REPORTS OF GREGORY K. LEONARD AND ALAN J. COX - REDACTED ) (modified per 11/7/2011 Order (Dkt. 602)) filed by Google Inc. (Van Nest, Robert) (Filed on 11/8/2011) (Entered: 11/08/2011)
605 – Filed and Effective: 11/08/2011
Document Text: EXHIBITS re 581 Opposition/Response to Motion, Exhibit K to Declaration of David Zimmer In Support Of Google's Opposition to Oracle's Motion to Strike Expert Reports of Gregory K. Leonard and Alan J. Cox filed byGoogle Inc.. (Related document(s) 581 ) (Van Nest, Robert) (Filed on 11/8/2011) (Entered: 11/08/2011)
606 – Filed and Effective: 11/08/2011
Document Text: DOCUMENT E-FILED UNDER SEAL re 602 Order on Administrative Motion to File Under Seal Google's Opposition to Motion to Exclude Portions of the Expert Reports of Gregory K. Leonard and Alan J. Cox (Unredacted) by Google Inc.. (Van Nest, Robert) (Filed on 11/8/2011) (Entered: 11/08/2011)
607 – Filed and Effective: 11/08/2011
Document Text: DOCUMENT E-FILED UNDER SEAL re 602 Order on Administrative Motion to File Under Seal Exhibits D, E, F, G, and J to the Declaration of David Zimmer in Support of Google Inc.'s Opposition to Oracle America, Inc.'s Motion to Exclude Portions of the Expert Reports of Gregory K. Leonard and Alan J. Cox (Dkt. No. 581) by Google Inc. (Van Nest, Robert) (Filed on 11/8/2011) (Entered: 11/08/2011)
608 – Filed and Effective: 11/08/2011
Document Text: Declaration of David Zimmer in Support of 592 Administrative Motion to File Under Seal ORACLE AMERICA, INC.S ADMINISTRATIVE MOTION TO FILE UNDER SEAL PORTIONS OF ITS REPLY TO GOOGLE INC.S OPPOSITION TO MOTION TO EXCLUDE PORTIONS OF THE EXPERT REPORTS OF GREGORY K. LEONARD AND ALAN J. COX filed by Google Inc. (Attachments: # 1 Proposed Order)(Related document(s) 592 ) (Van Nest, Robert) (Filed on 11/8/2011) (Entered: 11/08/2011)
609 - Filed and Effective:
ORDER PARTIALLY GRANTING AND PARTIALLY DENYING REQUEST TO FILE DOCUMENTS UNDER
SEAL, Order by Hon. William Alsup granting in part and denying in part
592 Administrative Motion to File Under Seal.(whalc1, COURT STAFF)
(Filed on 11/8/2011) (Entered: 11/08/2011)
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
Defendant moves to file under seal portions of its brief opposing plaintiff’s motion to exclude portions of the Leonard and Cox expert reports, as well as certain materials supporting its
opposition (Dkt. No. 580). The motion is GRANTED IN PART AND DENIED IN PART as follows. As to lines 12:14–17 and 12:22–13:2 of the opposition brief, and as to exhibits D, E, F, G, and J to the Zimmer declaration, the motion is GRANTED. Those materials may be filed under seal.
Please be aware that this is without prejudice to a future order that quotes from the sealed material and is public. As to lines 15:8–10 of the opposition brief and exhibit K to the Zimmer
declaration, the motion is DENIED. Those materials may be filed in the public record.
Additionally, defendant must file a copy of its opposition brief in the public record with only
lines 12:14–17 and 12:22–13:2 redacted.
IT IS SO ORDERED.
Dated: November 7, 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
CLAIMS TO BE TRIED
With respect to the 26 claims to be tried, the Court expects to advise the jury that all terms
have their ordinary meaning, unless previously construed. If any party contends otherwise, then it must specify the claim, the term therein, and its proposed construction by NOON ON DECEMBER 2, 2011. Drafts must be exchanged seven calendar days before and an in-person conference must be held to work out differences within the seven-day period. Please try to stipulate to any meaning other than the ordinary meaning. By NOON ON DECEMBER 9, 2011, each side shall respond stating with particularity its objection, if any, to the proposed constructions. This pleading will be used by the judge for case and trial management purposes. A party may not deviate from any earlier claim construction contentions unless it is to accept the ordinary
meaning, to stipulate to an agreed meaning, or to accept the other side’s earlier contention.
IT IS SO ORDERED.
Dated: November 8, 2011.
/s/ William Alsup
UNITED STATES DISTRICT JUDGE
UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF CALIFORNIA
SAN FRANCISCO DIVISION
ORACLE AMERICA, INC.,
Case No. 3:10-cv-03561-WHA
GOOGLE'S OPPOSITION TO MOTION
TO EXCLUDE PORTIONS OF THE
EXPERT REPORTS OF GREGORY K.
LEONARD AND ALAN J. COX
Dept. Courtroom 8, 19th Floor
Judge: Hon. William Alsup
TABLE OF CONTENTS
I. INTRODUCTION ...... 1
II. ARGUMENT ...... 1
A. Dr. Leonard and Dr. Cox do not offer any "technical" expert opinions
of their own ...... 1
B. Dr. Leonard and Dr. Cox appropriately rely on interviews with
witnesses as part of the basis of their opinions ...... 3
1. Experts may rely on witness interviews, and Google has
C. The fact that Oracle may be able to offer evidence to dispute Dr.
acknowledged that it must (and will) present the underlying
facts through testimony of percipient witnesses before its
experts may rely on those facts ...... 3
2. Google did not hide the ball from Oracle in discovery and has
proposed reasonably limited additional discovery from
witnesses who have not yet been deposed ...... 5
Leonard's and Dr. Cox's opinions does not make those opinions
inadmissible ...... 8
D. Google's damages experts do not offer any ipse dixit statements ...... 9
E. Google's damages experts properly discuss the implications of the fact
that the Java platform was stagnant and fragmented before Google's
alleged infringement ...... 10
F. Google's damages experts offer opinions consistent with settled law ...... 13
G. Dr. Leonard is fully qualified to explain why Dr. Shugan's marketing
analysis is unreliable as a matter of basic economics ...... 14
III. CONCLUSION ...... 17
TABLE OF AUTHORITIES
Children's Broad. Corp. v. Walt Disney Co.
357 F.3d 860 (8th Cir. 2004) ...... 9
Daubert v. Merrell Dow Pharmaceuticals, Inc.
509 U.S. 579 (1993) ...... 1, 4, 8
Geo. M. Martin Co. v. Alliance Machine Sys. Int'l, LLC
No. C 07-00692 WHA, 2008 WL 2008638 (N.D. Cal. May 6, 2008) ...... 3, 4
Hangarter v. Provident Life and Acc. Ins. Co.
373 F.3d 998 (9th Cir. 2004) ...... 8
Inline Connection Corp. v. AOL Time Warner Inc.
470 F. Supp. 2d 435 (D. Del. 2007) ...... 4
Int'l Adhesive Coating Co., Inc. v. Bolton Emerson Int'l Inc.
851 F.2d 540 (1st Cir. 1988) ...... 4
Mackie v. Rieser
296 F.3d 909 (9th Cir. 2002) ...... 14
McLean v. 988011 Ontario, Ltd.
224 F.3d 797 (6th Cir. 2000) ...... 9
Polar Bear Productions, Inc. v. Timex Corp.
384 F.3d 700 (9th Cir. 2004) ...... 13, 14
Stecyk v. Bell Helicopter Textron, Inc.
295 F.3d408 (3d Cir. 2002) ...... 9
Therasense, Inc. v. Becton, Dickinson & Co.
No. C 04-02123 WHA, 2008 WL 2323856 (N.D. Cal. May 22, 2008) ...... 3, 4, 5, 9
Federal Rule of Civil Procedure 30(b)(6) ...... 5, 6, 7
Federal Rule of Evidence 703 ...... 4, 9
Federal Rule of Evidence 705 ...... 9, 10
Facing a second motion to strike critical parts of its own damages expert's revised—but still arbitrary and result-oriented—expert report, Oracle has filed a tit-for-tat motion to strike portions of the reports of Google's damages experts Dr. Gregory Leonard and Dr. Alan Cox. But none of Oracle's transparent complaints offer any reason to exclude expert testimony under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), or its progeny.
Oracle offers seven arguments in moving to strike, but each lacks merit. Each argument is either factually or legally mistaken, or both.
A. Dr. Leonard and Dr. Cox do not offer any "technical" expert opinions of their own.
First, Oracle is factually wrong to argue that either Dr. Leonard or Dr. Cox has offered or will offer at trial any "technical" expert opinions on any issue. They do not and will not. Oracle mentions two "technical" issues in its motion: (1) the technical viability of Google's noninfringing alternatives to the allegedly patented functionality and copyrighted materials at issue; and (2) the extent to which Sun's Java mobile platform was "fragmented" prior to Google's introduction of the Android platform. As to both of these issues, both Dr. Leonard and Dr. Cox rely for technical background on the opinions of Google's technical experts, interviews with personally knowledgeable Google and third-party witnesses, and documentary evidence about the mobile-platform market. Dr. Leonard and Dr. Cox then offer their own economic expert opinions based on those underlying facts and technical expert opinions, which they are entitled to do as economists. This is entirely proper, and no different from what Oracle's own expert Dr. Iain Cockburn claims to have done—offer economic opinions based on a technical understanding shaped at least in part by Oracle's technical experts, employees, and documentary evidence.
In fact, if one reads Oracle's motion carefully, it is obvious that Oracle lacks any basis for accusing Dr. Leonard and Dr. Cox of offering "technical" opinions. For example, one of the purported "technical opinions" that Oracle insists must be stricken from Dr. Leonard's testimony is his statement that the opinions of Oracle's technical expert Dr. John Mitchell "are inconsistent
with my discussions with Google personnel and the opinions of Google's technical experts."
Oracle Mot. at 5:3-4 (citing Dearborn Decl. Ex. 1 (Leonard Report) at 28). That is to say, Oracle is demanding that the Court strike a harmless, foundational sentence where Dr. Leonard identifies the existence of a factual dispute between the parties and explains the basis of his own economic opinions. Oracle cites no authority to support its apparent belief that Dr. Leonard cannot even tell the jury where the parties disagree or identify the basis of his opinions.
Oracle's other complaints about "technical opinions" are similarly baseless. The quotes from Oracle's own brief make clear that each of the "technical opinions" Oracle ascribes to Dr. Leonard and Dr. Cox are actually either the opinions of Google's technical experts or factual statements by witnesses (including at deposition). In each case, Dr. Leonard and Dr. Cox are plainly relying on this other testimony for foundational technical statements about Google's noninfringing alternatives to the technology at issue. They then do what all experts do, offering their own (economic) opinions based on those foundational statements. Specifically, Oracle objects to Dr. Leonard's statements that:
None of this is an economist improperly offering opinions on technical computer-science issues,
as Oracle claims. These are statements of the foundations of Dr. Leonard's economic opinions.
- Google could have used a Java-language native compiler instead of using the allegedly infringing Dalvik virtual machine. Oracle Mot. at 4 (citing Leonard Report at 11-13). But this is not based on Dr. Leonard drawing any "technical conclusions" of his own. The statement is partly based on the success of Apple's iPhone, which uses a native compiler and not a virtual machine, and thus proves the viability of the native compiler approach generally. The statement is also based on Dr. Leonard's interviews with and deposition testimony from percipient Google witnesses Andy Rubin and Dan Bornstein—engineers who are familiar both with native compilers generally and Google's consideration and the viability of that approach for Android.
- Google could have used a programming language other than Java. Oracle Mot. at 4 (citing Leonard Report at 13-19). Here, Dr. Leonard relies on four separate reports from Google's technical experts for the basic proposition that Google could have used other programming languages, then relies on the technical experts and Google witnesses to explain how Google could have made a non-Java-language approach work.
- Google could have used a virtual machine, but designed around Oracle's patents. Oracle Mot. at 5 (citing Leonard Report at 20-24). Here, Dr. Leonard cites to percipient witness interviews explaining that, in the original releases of Android, the Dalvik virtual machine did not contain most of the allegedly infringing functionality. Dr. Leonard further cites to Google's technical experts to describe how specific non-infringing alternatives would have worked; he does not offer any opinions on that technical issue himself.
B. Dr. Leonard and Dr. Cox appropriately rely on interviews with witnesses as part of
the basis of their opinions.
1. Experts may rely on witness interviews, and Google has acknowledged that it
Second, Oracle objects generally to Dr. Leonard's and Dr. Cox's reliance on information obtained in interviews with Google employees and third-party witnesses. Oracle argues that permitting Dr. Leonard and Dr. Cox to repeat information allegedly provided by Google employees would be an end run around the hearsay rule. But Google has already explained that it will, and accepts that it must, offer the underlying factual testimony from the percipient witnesses first, before its experts may testify based on those facts. Oracle will have a full and fair opportunity to cross-examine each of the underlying percipient witnesses before Dr. Leonard or Dr. Cox takes the stand.
must (and will) present the underlying facts through testimony of percipient
witnesses before its experts may rely on those facts.
In support of this part of its motion, Oracle cites this Court's opinions in Therasense, Inc.
v. Becton, Dickinson & Co., No. C 04-02123 WHA, 2008 WL 2323856 (N.D. Cal. May 22,
2008), and Geo. M. Martin Co. v. Alliance Machine Sys. Int'l, LLC, No. C 07-00692 WHA, 2008
WL 2008638 (N.D. Cal. May 6, 2008). But those cases endorse the exact procedure outlined
above, which procedure Google will follow at trial. As this Court wrote in Therasense:
The traditional and correct way to proceed is for a foundational witness to testify firsthand at trial to the foundational fact or test and to be cross-examined. Then the expert can offer his or her opinion on the assumption that the foundational fact is accepted by the jury. The expert can even testify before the foundation is laid so long as counsel represents in good faith that the foundational fact will be laid before counsel rests. When the foundational fact is tested during fact discovery, as by a deposition, for example, it is often true that opposing counsel forego any objection and allows the expert to summarize the foundation.
Therasense, 2008 WL 2323856, at *2. What Therasense and Alliance Machine condemn is what Google has no intention of doing (and the Court would not permit it to do anyhow): having its experts testify as to "facts" entirely on the say-so of Google witnesses, without allowing Oracle a fair chance to cross-examine those Google witnesses. Oracle's hearsay objection is a straw man; Google has never proposed to offer hearsay at trial.
With that out of the way, Oracle has no valid objection to Dr. Leonard or Dr. Cox using statements by Google employees as part of the factual basis of their opinions. Neither
Therasense nor Alliance Machine suggests, much less holds, that party employee interviews are an inherently unreliable basis for an expert opinion. And Oracle offers no other case so holding, likely because courts routinely approve of experts using such interviews under Federal Rule of Evidence 703. See Int'l Adhesive Coating Co., Inc. v. Bolton Emerson Int'l Inc., 851 F.2d 540, 545 (1st Cir. 1988) (approving accounting expert's reliance on plaintiffs "business and financial records" and "interviews with company personnel" because "these are sources of information normally and reasonably relied upon by accountants"); Inline Connection Corp. v. AOL Time Warner Inc., 470 F. Supp. 2d 435, 441-42 (D. Del. 2007) (rejecting challenge to expert's reliance on party employees where plaintiff "was provided the opportunity to depose AOL and Earthlink employees and acquire the information that it needed to prepare for trial").
In fact, the subject matter of the witness interviews here make clear why those interviews constitute facts and data "of a type reasonably relied by experts in the particular field." Fed. R. Evid. 703. In order to offer an opinion on the economic significance of Google's non-infringing alternatives, Dr. Leonard and Dr. Cox had to investigate (1) what alternatives existed; (2) how viable each alternative was; (3) what alternatives Google considered; and (4) the advantages and disadvantages each alternative offered to Google. The best place to go for first-hand, percipient answers to each of these questions is Google's own engineering staff. To the extent Oracle would argue that Dr. Leonard and Dr. Cox could not rely on Google employees, but instead had to rely only on Google's outside technical experts, that would make no sense. It would introduce more hearsay to the proceedings, since the technical experts would have to get the underlying information from the Google employees before relaying it to the economists. Not only was this approach consistent with the law, it was the sensible and efficient one. Oracle is free to cross-examine them about the interviews, but this is not a Daubert objection.
Finally, Oracle undoubtedly will argue in reply that the Court should strike Dr. Leonard's
and Dr. Cox's opinions because Dr. Leonard and Dr. Cox accepted as true the statements of
Google technical experts and employees that Google had good non-infringing alternatives.
According to Oracle, Dr. Leonard and Dr. Cox should have tested the accuracy of those
statements by modifying Android phones to replace the allegedly infringing functionality with an
alternative, then measuring the performance impact. But Oracle is the plaintiff here, not Google. It is fundamentally the job of Oracle and its experts to quantify the impact (if any) of the allegedly infringing functionality on handset performance. Oracle's argument that Dr. Leonard and Dr. Cox should be disqualified for failing to conduct tests to impeach Google's witnesses amounts to an attempt to shift the burden of proof. Oracle is certainly free to point out on cross-examination that Dr. Leonard and Dr. Cox didn't conduct tests, but that should be the extent of it. An expert is entitled to rely on the accuracy of statements of party witnesses regarding matters within those witnesses' personal knowledge, then render opinions based on the assumption that the jury will believe those witnesses—just as this Court held in Therasense.
2. Google did not hide the ball from Oracle in discovery and has proposed
Oracle's motion to strike goes so far as to make veiled accusations of discovery abuse, by implying that Google has "ke[pt] the facts secret from the other side until after the close of discovery." Oracle Mot. at 6. Of course, Oracle does not identify a single fact Google "kept secret," or any subject covered by the witness interviews on which Oracle did not actually take discovery. The primary subjects of the witness interviews were Google's non-infringing alternatives to the patents and copyrights at issue and the evolution of the Android software over time. Oracle took ample discovery on both subjects. Those were two of Oracle's 15 allotted Rule 30(b)(6) topics. Declaration of David Zimmer in Support of Google's Opposition ("Zimmer Decl.") Ex. A (June 21, 2011 Rule 30(b)(6) notice). As one would expect given the size of this case and scope of discovery, Oracle also propounded document requests, id. Ex. B (Oracle's Corrected Second Set of Requests for Production) at RFPs 102 & 126, and an interrogatory, id. Ex. C (Oracle's Fourth Set of Interrogatories) atRogs 20-21, about Google's non-infringing alternatives and how much additional cost they would have imposed. Moreover, as discussed more below, Oracle deposed the very same employees Dr. Leonard and Dr. Cox interviewed on these same subjects.
reasonably limited additional discovery from witnesses who have not yet
Likewise, Oracle puts great weight on the fact that some of the interviews took place after
the close of discovery. But this is Oracle's own fault, because the expert discovery schedule
dictated the timing of the witness interviews. Oracle served its initial damages report on May 21, 2011. Had that report not been so riddled with legal and factual errors that the Court struck it and ordered Oracle back to the drawing board, Google's responsive report would have been due before the July 29, 2011 discovery cutoff, or at worst very shortly thereafter. Any witness interviews would have been conducted earlier. But as things turned out, Oracle did not serve its revised damages report until September 12, 2011—six weeks after fact discovery ended.
Finally, Oracle complained in its précis letter, and again at the recent case management conference, that Dr. Leonard and Dr. Cox had interviewed two witnesses that Google had allegedly failed to disclose—Google employee Tim Bray and third-party witness John Rizzo.1 Oracle mentions this issue only briefly in its motion, Oracle Mot. at 8:4, but Google has taken steps to resolve this issue, as the Court suggested at the case management conference. Because neither Bray nor Rizzo has yet been deposed, Google has proposed that Oracle depose each of them for two hours, as the Court recommended. Rizzo is not under Google's control, but Google has reached out to him and his employer Aplix Corporation of America regarding potential deposition dates, and has agreed to facilitate Oracle's service of a subpoena. Zimmer Decl. ¶ 5. Google has proposed deposition dates to Oracle for both Bray and Rizzo. Id.
This is not enough for Oracle, which now insists on depositions of each of the other five
witnesses who spoke to Dr. Leonard and/or Dr. Cox, even though all of those witnesses were
well known to Oracle since this case began, and four of the five have previously been deposed in
this case, often for multiple days. Taking these witnesses one at a time:
- Andy Rubin. Oracle has already deposed Rubin for three days in this case— for a full day in his personal capacity on April 5, 2011, for a full day in his personal capacity and on three Rule 30(b)(6) topics on July 27, 2011, and then a further examination on the Rule 30(b)(6) topics on August 18, 2011. Zimmer Decl. 6. Oracle exhaustively covered the subject matter of Rubin's interview with Google's experts—the existence of non-infringing alternatives with Rubin. E.g., id. Ex. D (April 5, 2011 Rubin Dep.) at 37:1-38:23, 62:5-63:4.
- Dan Bornstein. Oracle deposed Bornstein for two days in this case—for a full day in his personal capacity on May 16, 2011, and then for a further full day as a Rule 30(b)(6) designee on July 22, 2011. Zimmer Decl. | 8. As with
1 As Google explained in opposing Oracle's précis, Oracle was wrong—Google had disclosed Bray and Rizzo. Oracle subsequently corrected its précis accusing Google of nondisclosure.
Accordingly, Oracle has already examined Rubin, Bornstein, Swetland, and Agarwal on the subject matter of their interviews with Dr. Cox and Dr. Leonard. As to Yellin, because Dr. Leonard relies on him only for one issue that is not in dispute, there would be no benefit to Oracle from taking his deposition. And Oracle will have a chance to depose Bray and Rizzo prior to trial. Given this, there is no danger of any prejudice to Oracle from Dr. Leonard's and Dr. Cox's entirely appropriate reliance on those witnesses. Further, Oracle has examined both Dr. Cox and Dr. Leonard for a full day apiece. It had a complete opportunity during those
Rubin, Oracle examined Bornstein comprehensively about the subject matter of Bornstein's interview with Google's experts—Google's non-infringing alternatives. E.g., id. Ex. F (May 16, 2011 Bornstein Dep.) at 48:14-52:19, 114:3-116:6; id. Ex. E (July 22, 2011 Bornstein Dep.) at 149:5-157:8, 164:2-165:9, 171:13-174:11. In fact, Bornstein was Google's corporate designee on that topic. Id. Ex. E (July 22, 2011 Bornstein Dep.) at 168:25-169:12.
- Brian Swetland. Oracle deposed Swetland for a full day on July 7, 2011. Zimmer Decl. If 11. Oracle examined Swetland thoroughly about the subject matter of Swetland's interview with Google's experts—Google's noninfringing alternatives and the addition of the allegedly infringing functionality to Android. E.g., id. Ex. G (Swetland Dep.) at 43:24-47:2, 92:1-94:21, 150:3-152:16.
- Aditya Agarwal. Oracle deposed Agarwal for a half day on April 8, 2011. Zimmer Decl. ¶ 13. Agarwal is a financial analyst for Android and was Google's Rule 30(b)(6) designee on Android finances. Id. Ex. H (Agarwal Dep.) at 7:14-21, 10:12-11:3. Oracle examined him for three hours on that topic. Agarwal is not cited at all in the Leonard report and is cited only once in the Cox report, for the single factual proposition that "[t]he Engineering category on Google's Android Profit and Loss statements include costs that are entirely and directly attributable to the Android platform." Dearborn Decl. Ex. 2 (Cox Report) at 32n.l17.
- Frank Yellin. Yellin is the only one of the five witnesses who has not yet been deposed, but that is only because Oracle made the decision not to depose him. Yellin is a former Sun employee and the lead named inventor on the '520 patent-in-suit, so Oracle has always known that Yellin had potentially relevant information. Further, Yellin is not cited at all in the Cox report and just twice in the Leonard report. Both times, he is cited as one of several sources to support the same proposition—that Sun never incorporated the functionality of the '520 patent into its Java ME platform. Dearborn Decl. Ex. 1 (Leonard Report) at 21 n.61 & 77 n.262. As far as Google knows, the issue is not in dispute. Oracle has never offered any evidence that Sun incorporated the '520 functionality into Java ME. In fact, on August 3, 2011, in response to Google's request for Rule 30(b)(6) testimony on the issue, Oracle's counsel informed Google that "[ajfter diligent inquiry, Oracle cannot locate an Oracle witness to testify based on personal knowledge as to 'actual use' of the" '520 patent. Zimmer Decl. Ex. I (August 3, 2011 Muino e-mail to Francis). At his deposition earlier today, Dr. Leonard testified that he spoke to Yellin for only about five or 10 minutes.
depositions to collect any discoverable information about all the interviews—including how they
were arranged, who was present, and what was said.
C. The fact that Oracle may be able to offer evidence to dispute Dr. Leonard's and Dr.
Cox's opinions does not make those opinions inadmissible.
Third, Oracle essentially contends that Dr. Leonard and Dr. Cox should not be allowed to rely on evidence or offer opinions inconsistent with Oracle's case theory and preferred evidence. Oracle chastises Dr. Leonard and Dr. Cox for citing interviews, testimony, documents, and market facts that support Google's position in the case, instead of the material Oracle plans to use for cross-examination. But the existence of factual disputes is not a Daubert issue. If it were, no expert would ever get to testify at any trial (since trials exist to resolve factual disputes). Oracle will get to cross-examine the Google percipient witnesses before Google's experts testify. If Oracle thinks the witnesses have failed to lay a foundation for Google's experts, Oracle can object to the expert testimony. If the Court permits Google's experts to testify, Oracle still will have an opportunity to cross-examine those experts with any evidence it deems meaningful.
As the Supreme Court explained in Daubert itself, this is the way to deal with competing expert testimony at trial—by letting the trier of fact decide whether the expert's testimony is persuasive, not through a motion to strike. The Court specifically stated that "[vjigorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evidence." Daubert, 509 U.S. at 596 (emphasis added). Obviously, given that the Court identified the "presentation of contrary evidence" as the proper means of discrediting an opposing expert, an expert's mere reliance on disputed facts or evidence cannot be a basis for disqualification.
To the contrary, experts may rely on disputed evidence without risking disqualification. The Ninth Circuit and other circuit courts have followed the Supreme Court's lead and made clear that contrary evidence should be used on cross-examination, not as the basis of a motion to strike. See Hangarter v. Provident Life and Acc. Ins. Co., 373 F.3d 998, 1017 n.14 (9th Cir. 2004) ("The factual basis of an expert opinion goes to the credibility of the testimony, not the admissibility, and it is up to the opposing party to examine the factual basis for the opinion in
cross-examination.") (quoting Children's Broadcasting Corp. v. Walt Disney Co., 357 F.3d 860, 865 (8th Cir. 2004)); McLean v. 988011 Ontario, Ltd., 224 F.3d 797, 800-01 (6th Cir. 2000) (an "expert's conclusions ... must have a basis in established fact and cannot be premised on mere suppositions," but "mere weaknesses in the factual basis of an expert witness' opinion . . . bear on the weight of the evidence rather than on its admissibility") (citation omitted); Stecyk v. Bell Helicopter Textron, Inc., 295 F.3d 408, 414 (3d Cir.,2002) ("Rule 705, together with Rule 703, places the burden of exploring the facts and assumptions underlying the testimony of an expert witness on opposing counsel during cross-examination").
Oracle undoubtedly will argue in its reply that the Court should strike Dr. Leonard's and Dr. Cox's opinions because Dr. Leonard and Dr. Cox accepted as true the statements of Google technical experts and employees that Google had good non-infringing alternatives. According to Oracle, Dr. Leonard and Dr. Cox should have tested the accuracy of those statements by modifying Android phones to replace the allegedly infringing functionality with an alternative, then conducting the performance tests and measuring the results. But Oracle is the plaintiff here, not Google. It is fundamentally the job of Oracle and its experts to quantify the impact (if any) of the allegedly infringing functionality on handset performance. Oracle's argument that Dr. Leonard and Dr. Cox should be disqualified for failing to conduct tests to impeach Google's witnesses amounts to an attempt to shift the burden of proof. Oracle is certainly free to point out on cross-examination that Dr. Leonard and Dr. Cox didn't conduct tests, but that should be the extent of it. An expert is entitled to rely on the accuracy of statements of party witnesses regarding matters within those witnesses' personal knowledge, and render opinions based on the assumption that the jury will believe those witnesses—just as this Court held in Therasense.
D. Google's damages experts do not offer any ipse dixit statements.
Fourth, Oracle gins up the accusation that Dr. Leonard and Dr. Cox engage in ipse dixit
reasoning, stating conclusions unsupported by evidence. Oracle concocts this criticism by
yanking from context individual sentences that happen to lack citations. But each of the
supposedly citation-free sentences Oracle identifies is just one part of a broader discussion, in
the middle of paragraph or sections with ample citation and support.
Oracle identifies four instances of alleged ipse dixit by Dr. Leonard. In one—a statement on page 30 of Dr. Leonard's report—Dr. Leonard does in fact cite to evidence. Dearborn Decl. Ex. 1 (Leonard Report) at 30 & n.l 10. In another case, Oracle has created the illusion of ipse dixit by truncating Dr. Leonard's report to remove the introductory phrase "As discussed above," which makes clear that Dr. Leonard is incorporating statements and citations appearing earlier on that same page. Id. at 25 & nn.85-87. In the other two cases, the statements identified by Oracle are summary sentences in the middle or at the end of a fuller discussion with numerous cited sources. Id. at 11-12 & nn. 18-21; id. at 15-16 & nn.38-40.
The same is true for Oracle's criticism of Dr. Cox. Oracle focuses nearly all its criticisms
on a summary section at the end of Dr. Cox's damages analysis discussing the relative
contribution to the total Android functionality of the 37 API packages at issue. Dearborn Decl.
Ex. 2 (Cox Report) at 27-28. But Oracle ignores that this summary refers to and incorporates the
lengthy analysis earlier in the report of the full functionality of Android and the limited role of
the API packages therein. Id. at 6-8 & nn.5-10. Similarly, Oracle objects to Dr. Cox's
conclusion that the success of Android did not turn on Google's decision to use the Java
language, but ignores that Dr. Cox supports that conclusion with a raft of evidence regarding
(and citation to) the massive success of Apple's iPhone, which uses a proprietary (not open)
architecture and the previously obscure Objective C language (not the well known and widely
used Java). Id. at 25-26 & nn.97-101. None of this is ipse dixit; it all has record support.
E. Google's damages experts properly discuss the implications of the fact that the Java
platform was stagnant and fragmented before Google's alleged infringement.
Fifth, Oracle objects to Dr. Leonard and Dr. Cox discussing the extent to which the Java mobile platform was stagnant and "fragmented" before Google ever launched Android, and for reasons having nothing to do with Google. This objection is also misplaced.
With respect to stagnation, Oracle cannot even object that Dr. Leonard and Dr. Cox rely on interviews with Google employees. Instead, Dr. Leonard and Dr. Cox base their opinions about stagnation almost entirely on admissions by Sun's and Oracle's own employees. Both Dr. Leonard and Dr. Cox cite the deposition testimony of former Sun CEO Jonathan Schwartz,
former Sun employees Jeet Kaul and Craig Gering, and current Oracle employee Hasan Rizvi, in addition to four separate Oracle documents and third-party commentary. Dearborn Decl. Ex. 1 (Leonard Report) at 31-33 & nn.l 13-121 (Leonard Report); id. Ex. 2 (Cox Report) at 44-45 & rm.137-146. All of the Sun and Oracle employees and documents confirmed that Sun had failed to update the Java platform and was unable to secure consensus in the Java Community Process, which in turn led to dissatisfaction among carriers, device manufacturers, and developers. All those things created opportunities for other, competing platforms to displace Java. Id. Neither Dr. Leonard nor Dr. Cox cites a single Google employee or document in this discussion. Nor is this discussion "technical" in any sense; it involves strictly economic business issues, on which Dr. Leonard and Dr. Cox are plainly qualified to opine.
With respect to so-called "fragmentation," Oracle once again criticizes Dr. Leonard and Dr. Cox for relying on evidence that Oracle doesn't like. Again, this is a point for cross-examination, not a proper basis of a motion to strike. But when one looks closely at the evidence Dr. Leonard and Dr. Cox did consider, Oracle's complaint is shown to be especially baseless. In particular, Oracle objects to Dr. Leonard's and Dr. Cox's reliance on statements by knowledgeable, percipient, third-party witnesses like: (1) James Gosling, the former Sun employee recognized as "the father of Java," who bemoaned Sun's carelessness in fragmenting Java; and (2) John Rizzo of Aplix, who discussed Sun's habit of creating incompatible fragments of its Java platform for wireless carrier partners, and the fact that Sun's purported compatibility test tested only 70% of Java functionality, leaving the possibility of incompatibility in the remaining 30%. Dearborn Decl. Ex. 1 (Leonard Report) at 34-37 & nn.122-135; id. Ex. 2 (Cox Report) at 45-48 & nn. 147-160. Dr. Leonard and Dr. Cox also relied on the deposition testimony of Sun witnesses Schwartz and Vineet Gupta and Sun documents. Id. All these sources are reliable bases for an economic analysis. Neither Dr. Leonard nor Dr. Cox cited any Google employee interviews or documents to support their opinions on fragmentation. Id.
Oracle further objects that Dr. Leonard and Dr. Cox apply no technical methodology that could define what constitutes "fragmentation." This complaint is especially bizarre given that Oracle is the one that has injected "fragmentation" into the case, without ever explaining what
that term means or how any alleged fragmentation has supposedly harmed Oracle. Oracle never argues in its motion that Dr. Leonard or Dr. Cox ignored any technical, industry-standard definition of "fragmentation," or even that such a definition exists. Oracle certainly never offers evidence from any of its experts supporting such a definition. Just as Oracle's experts have done, Dr. Leonard and Dr. Cox use "fragmentation" to mean real-world results contrary to Sun's "write once, run anywhere" Java marketing slogan—where applications or operations designed to run in the Java mobile space do not work on a given device or platform. Dearborn Decl. Ex. 1 (Leonard Report) at 34-37 & nn.122-135; id. Ex. 2 (Cox Report) at 45-48 & nn. 147-160. Even as Oracle objects to Dr. Leonard's and Dr. Cox's opinions, it never suggests Google's experts are using "fragmentation" incorrectly, much less explains why. Its objection is makeweight.
Indeed, as Google pointed out in its supplemental brief in support of its motion to strike Oracle's damages expert's report, Oracle's "fragmentation" claim is essentially vapor. At his deposition, Dr. Cockburn could not explain how one could determine whether Oracle was being harmed by gragmentation. First he suggested that [REDACTED] Zimmer Decl. Ex. J (Cockburn Dep.) at 139:5-11 (emphasis added). His only factual support for the proposition that fragmentation had harmed Oracle was not Oracle financial data—it was a calculation showing that Java developer message boards have recently been less active, in terms of new threads started, than Android developer message boards. Id. at 143:6-16. In the end, he could not cite a single piece of evidence demonstrating harm from fragmentation:
Id. at 156:18-157-10. Dr. Leonard's and Dr. Cox's opinions about fragmentation simply respond to Oracle's own expert economist's opinions on that same subject. Any uncertainty about what constitutes "fragmentation" reflects deeper problems with that aspect of Oracle's case.
F. Google's damages experts offer opinions consistent with settled law.
Sixth, Oracle asserts that Google's experts have offered opinions that misapply governing law. Oracle is wrong.
Oracle first argues that Dr. Leonard may not opine that Oracle's alleged damages should be reduced because Sun and Oracle failed to take reasonable steps to mitigate those damages by developing a commercial product. Oracle Mot. at 12. Oracle criticizes this opinion because "Google pleaded no mitigation defense and it has not offered a jury instruction on mitigation." Id. Oracle either misunderstands or mischaracterizes Dr. Leonard's point (or does both). Dr. Leonard is responding to Dr. Cockburn's upward adjustment of Oracle's patent damages from under $30 million to $176 million, based on the belief that Google's release of the supposedly incompatible Android platform prevented Sun from making millions off a compatible product. Dr. Leonard's point is that nothing about the alleged infringement prevented Sun or Oracle from developing a version of the Java platform that would be compatible with Android. In fact, Sun tried to develop such a product but failed. Dr. Leonard's opinion is that Sun's failure suggests either that there was never much value in a Sun implementation of Android in the first place, or that Sun never could have developed such its own product because of its own lack of engineering skill or executive will. This isn't a contract-style "failure to mitigate" defense; it is a refutation of the factual basis supporting Dr. Cockburn's damages adjustment.
Oracle is also wrong to argue that Polar Bear Productions, Inc. v. Timex Corp., 384 F.3d 700, 708 (9th Cir. 2004), holds that the existence of non-infringing alternatives is irrelevant to how much of a copyright infringer's profits are attributable to the infringement. Polar Bear never addresses that question. In fact, Polar Bear affirms the basic principles that "a causal link between the infringement and the monetary remedy sought is a predicate to recovery of both
actual damages and profits," that "a plaintiff in a [copyright infringement] action must establish
this causal connection, and that this requirement is akin to tort principles of causation and
damages." Id; see also Mackie v. Rieser, 296 F.3d 909, 915-16 (9th Cir. 2002). The existence
of non-infringing alternatives tends to disprove the causal link between infringement and profits,
by suggesting that the profits are actually attributable to other, non-infringing, creative features
added by the alleged infringer. If anything, the Polar Bear principles support the argument that
non-infringing alternatives are relevant to a copyright damages analysis.
G. Dr. Leonard is fully qualified to explain why Dr. Shugan's marketing analysis is
unreliable as a matter of basic economics.
Seventh, Oracle argues that Dr. Leonard lacks adequate knowledge and expertise to reply to the opinions of Oracle marketing expert Dr. Steven Shugan. Along with this brief, Google is filing a detailed declaration from Dr. Leonard rebutting each of Dr. Shugan's points. Dr. Leonard's declaration, and the primary sources he cites and quotes, demonstrate that it is Dr. Shugan, not Dr. Leonard, who has mischaracterized the relevant literature and misunderstood the fundamental economic limitations of his marketing analysis.
To begin with, Dr. Shugan wrote a report regarding a "conjoint analysis," which purports to measure the relative importance to consumers of particular features of a product. But his analysis was made-for-litigation in every possible sense. Dr. Shugan did not even participate in the initial one-on-one interviews and focus group that led to the selection of the smartphone features tested in the conjoint analysis; he outsourced that work to Dr. Cockburn's assistants at Analysis Group. Zimmer Decl. Ex. K (Shugan Dep.) at 42:6-50:24. He left it up to Analysis Group to determine which product features mentioned in the initial one-on-one interviews would subsequently be discussed by the focus group. Id. at 46:17-47:7.
Neither Dr. Shugan nor Analysis Group conducted any analysis of the focus group data to identify the particular features that the small focus-group sample identified as most important. Zimmer Decl. Ex. K (Shugan Dep.) at 50:5-24. Instead, Dr. Shugan cherry-picked seven smartphone features—e.g., application multitasking, application start time, number of applications, operating system brand, price, screen size, and voice command capabilities—for
the conjoint analysis, out of dozens of potential features that might be relevant to a consumer in
the market for a smartphone. Dr. Shugan was not even the person who made the decision to
include five of the seven features. He made the decision to include only the price and operating
system brand variables. Dr. Cockburn instructed Dr. Shugan to include the number of
applications, while Dr. Cockburn's assistants at Analysis Group instructed Dr. Shugan to include
four other product features. Id. at 29:7-30:11. Dr. Shugan admitted at deposition that the
decisions as to which features to test were driven by Oracle's litigation needs:
In the end, I had to decide which ones to include based on the ability of the conjoint analysis to deal with the feature and the requirements of the case. So there was a specific objective to the case regarding supplying information to Cockburn, and that had to be the primary determinant of how the study was conducted to give him the most accurate possible information.
Id. at 31:18-25. Dr. Shugan did nothing to account for the possibility that other features—overall
design and styling, handset manufacturer brand, wireless carrier, or keyboard type—might be
even more important to consumers than the ones he tested. Then, after using the conjoint
analysis to test the relative importance of the cherry-picked features, Dr. Shugan in essence
concludes that, if 25% of the respondents value application start time most highly of the tested
features, a delay in application start time would result in a roughly 25% drop in market share. Id.
Oracle does not cite a single case where any court accepted the use of a conjoint analysis for any purpose in litigation, let alone as the basis of a damages calculation. It certainly does not identify any case where any court approved what Dr. Shugan has done here—calculate relative consumer preferences for an arbitrary number of randomly selected features, then translate the results into a purported drop in market share, then use that purported decrease in market share to justify a nine- or ten-figure damages award.
Given the shaky ground beneath Dr. Shugan's feet, it is no surprise Oracle is trying to bar
Dr. Leonard from telling the jury exactly why Dr. Shugan's analysis—whatever value it might
have in helping companies with the guesswork of product design or marketing—is unreliable, as
a matter of economics, to estimate damages in litigation. That is the essence of Dr. Leonard's
criticism, and it is one Dr. Shugan is not competent to rebut. Dr. Shugan is not an economist; Dr.
Leonard is. But even as to the marketing aspects of Dr. Shugan's study, Dr. Leonard explains in detail why his criticisms are valid and Dr. Shugan's retorts miss the mark.
- Dr. Shugan's study was a "stated preference survey," despite Dr. Shugan's rejection of that label. As Dr. Leonard explains, any study that requires consumers to value products or choose between alternatives hypothetically, as Dr. Shugan's did, is a "stated preference survey." This is in contrast to a "revealed preference survey," where consumer preferences are revealed by consumers' actual market actions. Declaration of Gregory K. Leonard in Support of Google's Opposition ("Leonard Decl.") ¶¶ 17-19.
- Contrary to Dr. Shugan's claim that he is "biased" against survey-based research, Dr. Leonard has both conducted and analyzed hypothetical choice surveys, and published papers about such surveys. He was even hired recently by Oracle's counsel at Boies Schiller to critique a hypothetical choice survey in another litigation. Leonard Decl. ¶¶ 10, 20-24. Dr. Leonard has no bias against surveys; he is just critiquing a badly-designed survey used for an inappropriate purpose.
- Dr. Shugan is simply wrong when he asserts that the literature on hypothetical choice surveys expresses no concern about hypothetical bias, or that the potential for such bias exists only when evaluating consumer preferences for abstract goods, instead of consumer products like smartphones. In his declaration alone, Dr. Leonard cites and quotes a dozen primary sources directly refuting Dr. Shugan on both of these points. Leonard Decl. ¶¶ 25-30.
- Dr. Leonard has not cited any articles for propositions they do not contain or that they reject, as Dr. Shugan claims. Dr. Leonard explains why he has accurately characterized the articles he cites. Leonard Decl. ¶¶ 38-46.
- Dr. Leonard has never denied that conjoint analyses are commonly done and based on years of research. But that is irrelevant. As Dr. Leonard explains, he is simply pointing out that conjoint analysis is not suited for the use Dr. Shugan has made of it in this case, because of the hypothetical bias problem and other reasons. Leonard Decl. ¶¶ 47-52.
- Dr. Leonard has properly relied on Professor Min Ding's critique of the sort of conjoint analysis performed by Dr. Shugan. That critique has been repeatedly peer reviewed. While Dr. Shugan calls Professor Ding's critique "controversial," he does not identify any scholarly analysis actually criticizing Professor Ding's approach. Moreover, after decrying Professor Ding's work as "controversial," Dr. Shugan plays the other side, by wrongly arguing that Professor Ding's work supports his analysis. In fact, Professor Ding describes the point Dr. Shugan identifies as supportive as "conjecture" requiring "[fjurther study." Leonard Decl. ¶¶ 53-58.
- Dr. Shugan defends his survey even though its results indicate that consumers actually prefer higher-priced smartphones to lower-priced ones, all other things being equal. Even the software package Dr. Shugan used describes such results as "nonsense." Dr. Leonard explains that this counter-intuitive finding shows Dr. Shugan's conjoint analysis has failed, because the survey participants are not evaluating certain product attributes—in this case, price— while holding all other product attributes constant. Instead, they are injecting other, undefined attributes into their decision-making process. In other words,
Oracle has not offered any basis to strike Dr. Leonard's critique of Dr. Shugan; at most, it has identified a dispute between experts. Again, factual disputes or disagreements between experts are not grounds for striking expert testimony. They are the reason we have trials, rather than disposing of cases on summary judgment. For all the reasons stated in Dr. Leonard's report and declaration, Google will show at trial that Dr. Shugan's analysis is litigation-driven, ignores fundamental economic principles, and is unworthy of reliance by the jury.
they are actually valuing attributes other than the ones being tested, making Dr. Shugan's conclusions valueless. Leonard Decl. ¶¶ 62-65.
- Dr. Shugan is simply wrong that Dr. Leonard referred to standard robustness tests as tests for hypothetical bias. Dr. Leonard said the opposite—that such tests had "low power to detect hypothetical bias." Leonard Decl. ¶¶ 66-69.
For all the above reasons, the Court should deny Oracle's motion to strike in its entirety.
Dated: October 28, 2011
KEKER & VAN NEST LLP
By: s/ Robert A. Van Nest
ROBERT A. VAN NEST
Attorneys for Defendant
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 to file under seal portions of its reply brief supporting its motion to exclude portions of the Leonard and Cox expert reports, as well as certain materials supporting its reply (Dkt. No. 592). The motion is GRANTED IN PART AND DENIED IN PART as follows. As to exhibits A and B to the Norton declaration, the motion is GRANTED. Those materials may be filed under seal. Please be aware that this is without prejudice to a future order that quotes from the sealed material and is public. As to exhibit C to the Norton declaration and as to the reply brief itself, the motion is DENIED. Those materials may be filed in the public record.
IT IS SO ORDERED.
Dated: November 8, 2011.
/s/ William Alsup
UNITED STATES DISTRICT JUDGE