TV Interactive Data Corporation v. Sony Corporation et al

Filing 686

ORDER by Judge Joseph C. Spero denying 546 Motion in Limine; granting in part and denying in part 548 Motion in Limine (jcslc2, COURT STAFF) (Filed on 3/1/2013)

1 2 3 4 UNITED STATES DISTRICT COURT 5 NORTHERN DISTRICT OF CALIFORNIA 6 7 TV INTERACTIVE DATA CORPORATION Plaintiff 8 v. 9 10 SONY CORPORATION, et al. ORDER RE DAUBERT MOTIONS. Dkt. Nos. 546, 548. Defendants 11 United States District Court Northern District of California Case No.:3:10-cv-00475-JCS 12 13 I. INTRODUCTION Before the Court are Motions in Limine re Daubert filed by Plaintiff TV Interactive Data 14 15 Corporation (“TVI”), and Defendants Sony Corporation, Sony Corporation of America, Sony 16 Electronics, Inc., Sony Computer Entertainment, Inc., and Sony Computer Entertainment America 17 LLC (collectively “Sony”). TVI and Sony both contend that certain testimony from the opposing 18 party’s experts must be excluded because the information is unreliable and thus inadmissible 19 under Rule 702 and Daubert. The Motions came on for hearing February 22, 2013. For the 20 reasons explained below, TVI’s Partial Daubert Motion is GRANTED in part and DENIED in 21 part, and Sony’s Daubert Motion is DENIED.1 22 II. 23 LEGAL STANDARD Rule 702 of the Federal Rules of Evidence permits expert testimony if it “will help the trier 24 of fact to understand the evidence or to determine a fact in issue.” Fed.R.Evid. 702. The expert 25 may provide testimony if “(1) the testimony is based on sufficient facts or data; (2) the testimony 26 27 28 1 The parties have consented to the jurisdiction of the undersigned magistrate judge pursuant to 28 U.S.C. § 636(c). 1 is the product of reliable principles and methods; and (3) the expert has reliably applied the 2 principles and methods to the facts of the case.” Id. In Daubert v. Merrell Dow Pharmaceuticals, 3 the Supreme Court held that the Rule 702 analysis “entails a preliminary assessment of whether 4 the reasoning or methodology underlying the testimony is scientifically valid and of whether that 5 reasoning or methodology can be applied to the facts in issue.” 509 U.S. 579, 593-94 (1993). 6 Thus, district courts “are charged with a ‘gatekeeping role,’ the objective of which is to ensure that 7 expert testimony admitted into evidence is both reliable and relevant.” Sundance, Inc. v. DeMonte 8 Fabricating Ltd., 550 F.3d 1356, 1360 (Fed. Cir. 2008). In Daubert, the Court suggested four non-exhaustive guiding factors to make this 10 determination: (1) whether the expert’s methodology has been tested or is capable of being tested; 11 United States District Court Northern District of California 9 (2) whether the technique has been subjected to peer review and publication; (3) the known and 12 potential error rate of the methodology; and (4) whether the technique has been generally accepted 13 in the proper scientific community. Daubert, 509 U.S. at 593-95. The Court also emphasized that 14 the focus of this inquiry, “of course, must be solely on principles and methodology, not on the 15 conclusions that they generate.” Id. at 595. When an expert’s principles and methodology are 16 sound, and the evidence relied upon sufficiently related to the case at hand, disputes about the 17 degree of relevancy or accuracy may go to the testimony’s weight, not its admissibility. Primiano 18 v. Cook, 598 F.3d 558, 564-65 (9th Cir. 2010). 19 III. TVI’S DAUBERT MOTION 20 A. 21 TVI contends the Court should exclude the opinion of Sony expert Mr. John Byrd, who has Mr. Byrd 22 proposed what Sony considers to be the next-best noninfringing alternatives to TVI’s patented 23 technology. See Declaration of Patrick M. Arenz in Support of TVI’s Partial Daubert Motion 24 (“Arenz Decl.”), Ex.1 (Responsive Expert Report of John Byrd on Noninfringement). In the event 25 TVI prevails on its infringement claims at trial, Mr. Byrd’s proposed alternatives will be relevant 26 to the assessment of a reasonably royalty. See Grain Processing Corp. v. American Maize- 27 Product Co., 185 F.3d 1341 (Fed. Cir. 1989). TVI argues, however, that both Mr. Byrd’s 28 noninfringing alternatives must be excluded because they are based on speculation and are thus 2 1 unreliable. TVI also contends that one of Mr. Byrd’s noninfringing alternatives−the “press any 2 button” alternative−must also be excluded because it still infringes TVI’s patents. 3 4 1. Mr. Byrd explains the “press any button” alternative in his expert report: 5 In the Option 1 redesign method, once a compatible disc insertion has been detected, the user would see (and optionally hear) a prompt on the screen requesting them to press a button to play the movie or other content. The device would then pause to wait for the user to press any button before the content on the disc starts playing. 6 7 8 9 The “Press Any Button” Alternative2 Arenz Decl. Ex. 1 ¶ 167. TVI contends the “press any button” alternative infringes TVI’s patents because it is contrary to the established claim construction of the term “automatically.” TVI notes that Judge 11 United States District Court Northern District of California 10 White construed this claim in TVI’s previous case against Microsoft−which involved the same 12 patents as in this case−when he wrote that the term ‘automatically’: 13 [D]oes not eliminate all user input, it only eliminates user input of a file name. Therefore, there is still a possibility of user input reserved by the patentee during prosecution. The prosecution history of the patent serves to clarify what the patentee meant by “automatically” and “without any additional user input.” The Court adopts the definition of automatically used in the specification as clarified by the prosecution history as: without user input of a file name. 14 15 16 17 18 Arenz Decl. Ex. 7 at 21 (Judge White’s Claim Construction Order) (emphasis in original). TVI 19 also notes that in this case, Sony stipulated to Judge White’s construction of this term in a joint 20 submission before Judge Fogel. See Arenz Decl. Ex. 8 at 2-3 (On November 9, 2010, the parties 21 stipulated that the construction of the term “automatically” was “without user input of a file 22 name”). Additionally, TVI contends this construction became part of the prosecution history of 23 the patents when Microsoft represented to the United States Patent and Trademark Office−when 24 seeking re-examination of the patents after Judge White’s construction−that “[u]nder the Court’s 25 construction, a substantial new question of patentability is created[.]” Declaration of Sarah E. 26 27 28 2 This issue of whether Mr. Byrd’s “press any button” alternative still infringes upon TVI’s patents is also discussed in TVI’s Motion in Limine No. 4 to Exclude Evidence Based on Improper Claim Construction. See Dkt. No. 599. 3 1 Huddleston in Support of TVI’s Reply Memorandum in Support of Its Partial Daubert Motion, Ex. 2 A (applying Judge White’s construction that “‘automatically’ means ‘without user input of a file 3 name’”). 4 Sony does not dispute that it stipulated to Judge White’s construction of the term 5 “automatically.” Nor does Sony dispute that the “press any button” alternative does not require 6 the user to input a file name. See Arenz Decl. Ex. 2 (Deposition of Mr. Byrd) at 410 (“In my 7 redesign the user would not be required to input a file name in this specific redesign”).3 8 Nevertheless, Sony contends the “press any button” alternative does not infringe TVI’s patents 9 because “TVI has clearly and unequivocally disclaimed coverage of a system with a button to activate the start of a disc.” Sony Memorandum in Opposition to TVI’s Daubert Motions (“Sony 11 United States District Court Northern District of California 10 Opp.”) at 7:1-2. Sony argues that during the Markman hearing in this case, TVI’s counsel was 12 very clear that use of a system in which the play of the disc was activated by pressing a button is 13 not an infringing product. 14 Sony points to representations from TVI’s counsel during the Markman hearing to show 15 how TVI purportedly disclaimed Judge White’s construction of the term “automatically” to mean 16 “without user input of a file name.” Excerpted in the proper context, TVI’s counsel made the 17 following representations while explaining the prior art: 18 The patent explains that, for example, to use a CD-ROM based multimedia book, the user must do the following on an IBM PC host device, you have to start the Windows operating environment. Second, insert the CD-ROM into the drive. And then third, find the appropriate icon and double click on the icon. If you’ve used a Mac, that’s a very common approach where it pops in a dialog box with the contents of the disk. But you have to navigate what’s on there and decide which item you’re going to click on. Sometimes people label the files, click my first install, but there’s some complexity there. 19 20 21 22 23 24 Declaration of Fahd K. Majiduddin In Support of Defendant’s Opposition to TVI’s Partial 25 Daubert Motion (“Majiduddin Decl.”), Ex. 2 (Transcript of Markman Proceedings before Judge 26 27 28 3 Sony’s counsel argued for the first time at the motion hearing that Mr. Byrd’s “press any button” alternative in fact does require the user to input the filename. Aside from the fact Sony’s counsel was unable to cite any supporting evidence for this argument, it is rejected because it was not included in Sony’s opposition brief. 4 1 Fogel) at 13:6-18. TVI’s counsel also explained that TVI’s patents entailed the following 2 improvements: 3 4 5 6 And so here we are told that automatic startup of an application on insertion of a storage media allows even preschool children to use applications encoded on a storage media without adult supervision, therefore using interactive media in accordance with this invention is made as simple as playing a video cassette recorder or a VCR tape, and even preschool children can read interactive media without adult supervision. 7 Id. at 22:14-22 (emphasis in Sony’s Opposition). Sony also argues that TVI’s PowerPoint from 8 the Markman hearing makes the disclaimer most clearly because Slide 11 states that the invention 9 allows applications “to start up automatically without any additional user input.” Majiduddin Decl. Ex. 3 at 4. Sony contends the foregoing representations show that TVI’s counsel “clearly 11 United States District Court Northern District of California 10 and unequivocally disclaimed coverage of a system with a button to activate the start of a disc.” 12 Sony Opp. at 7:1-2. 13 The Court disagrees. TVI’s counsel did not make any disclaimer regarding the 14 construction of the term “automatically,” much less a “clear and unequivocal” disclaimer as Sony 15 contends here. On November 9, 2010, the parties stipulated to use Judge White’s construction of 16 the term “automatically.” See Arenz Decl. Ex. 8 at 2-3. On February 22, 2011, when TVI’s 17 counsel explained the prior art and patented technology before Judge Fogel during the Markman 18 hearing, construction of the term “automatically” was simply not in dispute. Moreover, TVI’s 19 counsel did not say what Sony would like the Court to believe. While the excerpts above suggest 20 the patented technology enables users to start up automatically without any additional user input, 21 these statements do not say that the patented technology eliminates all user input such that 22 pressing any button will avoid infringement. Such a construction was expressly rejected by Judge 23 White. Arenz Decl. Ex. 7 at 21 (Judge White stated that the term ‘automatically’ “does not 24 eliminate all user input, it only eliminates user input of a file name.”). In light of the parties’ 25 stipulation to Judge White’s construction, there should be no further dispute on this matter. All 26 testimony regarding the “press any button” alternative is therefore excluded. 27 28 5 1 2 2. The “Operating System Reload” Alternative TVI sets forth several other reasons why the Court should exclude Mr. Byrd’s testimony regarding both proposed noninfringing alternatives. Because the Court will exclude the first 4 alternative, the analysis in this section is limited to Mr. Byrd’s second “operating system reload” 5 alternative. See Arenz Decl. Ex. 1 ¶¶ 187-218. There is no dispute that Mr. Byrd’s second 6 proposed noninfringing alternative was not actually in existence at the relevant time period in 7 1997−the period in which a hypothetical negotiation between TVI and Sony for a licensing 8 agreement would have taken place. See Majiduddin Decl. Ex. 5 at 14. Thus, “the burden shifts to 9 [Sony] to show that a substitute nonetheless was ‘available’ during this period….” Conceptus, 10 Inc. v. Hologic, Inc., 771 F.Supp.2d 1164, 1179 (N.D. Cal. 2010) (citing DePuy Spine, Inc. v. 11 United States District Court Northern District of California 3 Medtronic Sofamor Danek, Inc., 567 F.3d 1314, 1331 (Fed. Cir. 2009)). 12 Whether a purported noninfringing alternative was “available” requires an assessment of 13 whether Sony had the “necessary materials, equipment, know-how, and experience to make an 14 alternative product during the relevant time frame.” Conceptus, 771 F.Supp.2d at 1179 (citing 15 Grain Processing, 185 F.3d at 1353). “Mere speculation or conclusory assertions will not suffice 16 to overcome the inference” that the noninfringing alternative was not available at that time. Grain 17 Processing, 185 F.3d at 1353. “Thus, the trial court must proceed with caution in assessing proof 18 of the availability of substitutes not actually sold during the period of infringement.” Id. 19 “Acceptable substitutes that the infringer proves were available during the accounting period can 20 preclude or limit lost profits; substitutes only theoretically possible will not.” Id. (citing Minco 21 Inc. v. Combustion Eng'g, Inc., 95 F.3d 1109, 1119 (Fed. Cir. 1996)). 22 TVI contends Sony has not shown the noninfringing alternative would have been 23 “available” because Mr. Byrd’s expert analysis is laden with fundamental flaws and speculation. 24 TVI puts forth several specific arguments pertaining to the inadequacy of Mr. Byrd’s analysis. 25 For the following reasons, the Court finds that each of TVI’s arguments go to the weight of Mr. 26 Byrd’s testimony, not its admissibility. 27 28 TVI first argues Mr. Byrd’s cost estimates are unreliable because Mr. Byrd relied on his own expertise rather than obtaining information from Sony regarding engineer salary rates or 6 1 overhead costs. TVI also notes that Mr. Byrd failed to apply an exchange rate despite the fact 2 Sony’s Japanese engineers are generally paid in Japanese Yen. However, the fact Mr. Byrd relied 3 on his own expertise to assess the cost and time required to implement the alternative, as opposed 4 to retrieving data from Sony, does not render Mr. Byrd’s methodology unreliable. TVI does not 5 challenge Mr. Byrd’s expertise, and TVI may cross-examine Mr. Byrd on his source of his data at 6 trial. This goes to the weight of Mr. Byrd’s testimony. 7 Next, TVI argues the redesigns could not have been “available” because they would have 8 required modification of source code to which Sony did not have access. At the time of Mr. 9 Byrd’s deposition, Mr. Byrd was not aware that Sony did not have access to the source code of various accused products because of restrictions imposed by third parties. Mr. Byrd stated in his 11 United States District Court Northern District of California 10 deposition, however, that the two alternatives “don’t require access to all the source code at all 12 times in order to implement them,” and that “Sony had enough information in its possession at the 13 time these devices were released certainly to do these redesigns.” Majiduddin Decl. Ex. 1 14 (Deposition of Mr. Byrd) at 320:11-15, 324:3-6. TVI challenges these statements as conclusory 15 and devoid of any support, and may TVI may make such challenges on cross-examination. This 16 goes to the jury’s assessment Mr. Byrd’s credibility on whether the alternative was “available” 17 when the source code was not. The lack of source code, however, does not render Mr. Byrd’s 18 methodology unreliable so as to exclude his opinion on a Daubert motion. 19 TVI also contends Mr. Byrd’s cost and time estimations are unreliable, as Mr. Byrd only 20 did an analysis for a single model in each product category (DVD-S700, BDP-S1, and the 2006 21 PlayStation3). In response, Sony contends that TVI misconstrues the scope of an acceptable 22 noninfringing alternative because after implementing the initial design for one product in a class, 23 Sony would not have to redesign the later versions since they would have been designed with the 24 alternative from the start. Again, the Court finds this matter appropriate for consideration by the 25 jury. Both parties are entitled to present their opinions on how the alternatives would have been 26 implemented, and the jury will accord each side’s testimony the weight it is due. 27 28 Finally, TVI challenges Mr. Byrd’s failure to analyze the acceptability of the noninfringing alternative to consumers. Mr. Byrd stated at his deposition that he had not analyzed whether the 7 1 proposed alternatives would be acceptable with respect to consumers or the marketplace. Arenz 2 Decl. Ex. 2 at 417. Although this is an important element pertaining to whether the alternatives 3 were “available,” this is not a basis to exclude Mr. Byrd’s alternatives. Sony states that it will 4 introduce other evidence on this issue. TVI has not cited any case in which an expert’s opinion 5 was excluded for this reason alone. The jury can hear evidence on whether these proposed 6 alternatives would have been acceptable to consumers, and the jury can resolve this dispute. 7 B. 8 Sony retained Mr. Creighton Hoffman to opine on damages and to rebut TVI’s damages 9 Mr. Hoffman expert’s opinion. In the event the jury finds TVI’s patents are valid and infringed, TVI will be awarded damages “adequate to compensate for the infringement, but in no event less than a 11 United States District Court Northern District of California 10 reasonable royalty for the use made of the invention by the infringer.” 35 U.S.C. § 284. To 12 calculate a reasonably royalty rate, Mr. Hoffman analyzed a hypothetical negotiation based on 13 fifteen Georgia-Pacific factors. See Arenz Decl. Ex. 12 (Expert Report of Creighton G. 14 Hoffman); see also Lucent Technologies Inc. v. Gateway, Inc., 580 F.3d 1301, 1324-25 (Fed. Cir. 15 2009). TVI challenges several portions of Mr. Hoffman’s hypothetical negotiation analysis as 16 incomplete and unreliable under Rule 702 and Daubert. The Court will assess each of TVI’s 17 challenges separately. As noted below, several portions of Mr. Hoffman’s opinion must be 18 excluded from evidence. 19 20 1. Reliance on Mr. Byrd’s Non-Infringing Alternatives TVI first challenges Mr. Hoffman’s reliance on Mr. Byrd’s noninfringing alternatives 21 which TVI contends should be excluded as unreliable. Mr. Byrd is precluded from presenting an 22 expert opinion on his first “press any button” alternative, but his opinion pertaining to the 23 “operating system reload” alternative is admissible. Thus, TVI’s motion to exclude this evidence 24 is only granted to the extent Mr. Hoffman relies on Mr. Byrd’s “press any button” alternative. 25 It is not exactly clear where Mr. Hoffman relies on Mr. Byrd’s “press any button” 26 alternative in his expert report. During Mr. Hoffman’s Georgia-Pacific analysis, he states that 27 “[t]he cost of implementing one of these non-infringing alternatives (disabling autoplay) has been 28 estimated to be approximately $25,000 to $419,000. This amounts to less than $0.01 per accused 8 1 unit sold by Sony from March 1997 through the life of the patents-in-suit.” Arenz Decl. Ex. 12 at 2 29. It is not clear whether this statement refers to just one of Mr. Byrd’s proposed noninfringing 3 alternatives, or to both. In order to be admissible, this estimate may not be based, in whole or in 4 part, on the “press any button” alternative. If Mr. Hoffman’s estimate is not based on the excluded 5 evidence, it will be admitted. The witness must lay a proper foundation in this regard. 6 TVI also contends that Mr. Hoffman relies on Mr. Byrd’s noninfringing alternatives when he critiques Professor Srinivasan’s assessment of reasonable damages because, according to Mr. 8 Hoffman, the survey fails to compare TVI’s patented technology to the best noninfringing 9 alternatives. However, a review of Mr. Hoffman’s critique on this matter does not reveal a single 10 reference to the “press any button” alternative, which the Court would exclude. See Arenz Decl. 11 United States District Court Northern District of California 7 Ex. 12 at 33. Nor does Mr. Hoffman’s critique expressly mention the “operating system reload” 12 alternative, which the Court would not exclude. See id. Rather, Mr. Hoffman merely states that 13 Professor Srinivasan failed to consider “[m]any other more simple and user friendly noninfringing 14 alternatives [which] were available.” Arenz Decl. Ex. 12 at 33. There will be at least one such 15 alternative in evidence. Therefore, this testimony will not be excluded under Daubert. 16 17 2. Sony’s Licenses TVI next contends that Mr. Hoffman’s reference to the Sony licenses must be excluded 18 from testimony because by Mr. Hoffman’s own admission, the Sony licenses are not comparable. 19 The Federal Circuit requires that “use of past patent licenses under [Georgia-Pacific] factors 1 and 20 2 … account for differences in the technologies and economic circumstances of the contracting 21 parties.” Finjan, Inc. v. Secure Computing Corp., 626 F.3d 1197, 1211 (Fed. Cir. 2010) (citing 22 Wordtech Sys. v. Integrated Networks Solutions, Inc., 609 F.3d 1308, 1319–20 (Fed. Cir. 2010) 23 (“[C]omparisons of past licenses to the infringement must account for ‘the technological and 24 economic differences’ between them”) (quoting ResQNet.com, Inc. v. Lansa, Inc., 594 F.3d 860, 25 870–73 (Fed. Cir. 2010) (per curiam) (reversing the trial court’s assessment of damages because 26 similar to the district court in Lucent Technologies, it “made no effort to link certain licenses to the 27 infringed patent”))). Mr. Hoffman references the Sony licenses in connection with his analysis of 28 Georgia-Pacific Factor 2, which considers “[t]he rates paid by the licensee for the use of other 9 1 patents comparable to the patent in suit.” Lucent Technologies, 580 F.3d at 1325 (quoting 2 Georgia-Pacific Corp. v. U.S. Plywood Corp., 318 F.Supp. 1116, 1120 (S.D.N.Y. 1970)). 3 4 5 6 7 8 Mr. Hoffman wrote the following regarding Georgia-Pacific Factor 2 in his expert report: Georgia-Pacific Factor 2 − The rates paid by the licensee for the use of other patents comparable to the patents-in-suit. I have reviewed the patent and technology license agreement produced by Sony in this matter. These licenses are summarized on Exhibit G. Although these agreements are not for technologies directly comparable to those described in the patents-in-suit, they indicate that Sony has agreed to pay running royalties ranging from $0.01 to $0.25 per unit. Majiduddin Decl. Ex. 5 at 18 (emphasis added). Thus, Mr. Hoffman admitted that the Sony 10 licenses are for technologies that are not comparable to the technology of the patents-in-suit. 11 United States District Court Northern District of California 9 Therefore, the Court will exclude Mr. Hoffman from referring to the royalty rates associated with 12 the Sony licenses. 13 Sony argues that Mr. Hoffman should be permitted to opine on the Sony licenses because 14 Mr. Hoffman only uses the Sony licenses for the limited purpose of noting that they “support” his 15 royalty rate and do not support TVI’s expert Mr. Wagner’s royalty rate. This argument is without 16 merit. The purpose of the Georgia-Pacific factors is to “support” an expert’s proposed royalty 17 rate, and the Federal Circuit has clearly prohibited the use of noncomparable licenses to do so. 18 See id. Moreover, Mr. Hoffman used the Sony licenses in the context of Factor 2, which expressly 19 requires the “use of other patents comparable to the patent in suit.” Lucent Technologies, 580 20 F.3d at 1325 (emphasis added). 21 The Court is mindful that the Federal Circuit recently stated that “[t]he degree of 22 comparability of [other] license agreements …[is a] … factual issue[] best addressed by cross- 23 examination and not by exclusion.” ActiveVideo Networks, Inc. v. Verizon Communications, Inc., 24 694 F.3d 1312, 1333 (Fed. Cir. 2012). This statement, however, did not overrule the Federal 25 Circuit’s well-established precedent requiring some degree of comparability between licenses used 26 in the Georgia-Pacific analysis for the jury to weigh. Here, Mr. Hoffman expressly stated that the 27 Sony licenses “are not for technologies directly comparable” and offered no further testimony on 28 the matter. Majiduddin Decl. Ex. 5 at 18. Moreover, six days after the Federal Circuit issued its 10 1 decision in ActiveVideo, the Federal Circuit criticized a district court’s denial of a Daubert motion 2 because that “ruling erroneously permitted continued reliance on this evidence where 3 comparability between it and a hypothetical license to the [infringed patent] was absent.” 4 LaserDynamics, Inc. v. Quanta Computer, Inc., 694 F.3d 51, 80 (Fed. Cir. 2012).4 5 6 3. Industry Patent Pools In his expert report, Mr. Hoffman bases his royalty rate of $0.07 per unit in part on 7 “[i]ndustry ‘patent pools’ for the hundreds of patents necessary to implement the DVD and BD 8 standards [which] include royalty rates ranging from $2.50 to $9.00 per player.” Majiduddin Ex. 5 9 at 29. Before Mr. Hoffman notes that the royalty rates from these patent pools support his 10 calculation of a $0.07 reasonable royalty rate, he explains: United States District Court Northern District of California 11 There are several industry patent pools and licensing organizations that provide manufacturers of DVD and BD players rights to the patents necessary to manufacture DVD and BD players and discs. The following discussion of relevant patent pools provides a general overview of the industry royalty rates in the DVD and BD disc player markets. 12 13 14 15 Id. at 24. Mr. Hoffman continues to give an overview of the patent pools, including leaders and 16 licensors in the pools, the general technology involved, and the royalty rates associated with each 17 patent pool. See id. at 23-27. This overview is included in Mr. Hoffman’s analysis of Georgia- 18 Pacific Factor 12, which considers “[t]he portion of the profit or of the selling price that may be 19 customary in the particular business or in comparable businesses to allow for the use of the 20 invention or analogous inventions.” Georgia-Pacific, 318 F. Supp. at 1120. 21 22 TVI contends that Mr. Hoffman’s testimony on the patent pools must be excluded for the similar reason as the Sony licenses: because Mr. Hoffman fails to explain the technological or 23 24 25 26 27 28 4 Sony also argues Mr. Hoffman should be able to reference the Sony licenses because TVI expert Mr. Wagner will also reference the Sony licenses in his Georgia-Pacific Factor 2 analysis. In the same regard, and in the context of the next argument pertaining to industry patent pools, Sony argues that because Mr. Wagner also reviews the royalty rates associated with these patent pools, Mr. Hoffman should be permitted to opine on the same matter. However, the Court finds TVI’s distinction on these points persuasive. Mr. Wagner found the Sony licenses and the patent pools were not comparable and therefore considered Factors 2 and 12 to be “neutral.” Unlike Mr. Wagner, Mr. Hoffman expressly referred to the royalty rates associated with the Sony licenses and patent pools to support his calculation of the reasonable royalty rate. Such testimony will not be permitted. 11 economic comparability between any one of the patents in the patent pools and license that would 2 have been reached had the hypothetical negotiation taken place. TVI shows that other courts have 3 excluded evidence pertaining to patent pools when the expert fails to explain how they are 4 technologically or economically comparable to the hypothetical license. TVI Motion at 17-18 5 (citing Dataquill Ltd. v. High Tech Computer Corp., No. 08-0543, 2010 U.S. Dis. LEXIS 53164, 6 2012 WL 1284381, *24-25 (S.D. Cal. April 16, 2012) (excluding an expert’s opinion regarding a 7 patent pool and noting that it “is a worldwide license for hundreds of patents covering an entire 8 standard, and, therefore, is much larger in scope than the license that would have been reached at 9 the hypothetical negotiation.”); LaserDynamics, Inc. v. Quanta Computer, Inc., No. 06-0348, 2011 10 U.S. Dist. LEXIS 42590, 2011 WL 7563818, *11 (E.D. Texas Jan. 7, 2011) (“Particularly where a 11 United States District Court Northern District of California 1 license covers a portfolio of patents or includes other intellectual property or services, Plaintiff 12 must present evidence sufficient to allow the jury to weigh the economic value of the patented 13 feature against the economic value of the features and services covered by the license 14 agreement.”)). The Court finds these well-reasoned principles persuasive, and will exclude Mr. 15 Hoffman’s testimony on the royalty rates associated with the patent pools as supporting the 16 calculation of royalty rates in this case. 17 Sony contends Mr. Hoffman’s information on the patent pools should not be excluded 18 because Mr. Hoffman did not “derive” his proposed $0.07 reasonably royalty rate from the range 19 of royalties associated with the patent pools, but merely used the patent pool rates to insure that 20 his royalty rate was reasonable in comparison. This argument is without merit. Mr. Hoffman 21 expressly “based” his conclusion that TVI and Sony would have negotiated a $0.07 per unit 22 royalty rate in part on industry patent pools with “running royalties ranging from $2.50 to $9.00 23 per player.” Majiduddin Decl. Ex. 5 at 29. Moreover, Mr. Hoffman uses this information in his 24 analysis of Georgia-Pacific Factor 12, which also expressly requires some degree of comparability 25 between the businesses. See Georgia-Pacific, 318 F. Supp. at 1120 (Factor 12 considers “[t]he 26 portion of the profit or of the selling price that may be customary in the particular business or in 27 comparable businesses to allow for the use of the invention or analogous inventions.”). Mr. 28 Hoffman failed to show any degree of comparability. Therefore, he may not refer to the royalty 12 1 2 rates for the patent pools. Nevertheless, Mr. Hoffman’s consideration of the patent pools in Georgia-Pacific Factor 13 does not suffer from the same fatal flaws. Georgia-Pacific Factor 13 considers the “portion of 4 the realizable profit that should be credited to the invention as distinguished from non-patented 5 elements, the manufacturing process, business risks, or significant features or improvements added 6 by the infringer.” Georgia-Pacific, 318 F.Supp. at 1120. Within his Factor 13 analysis, Mr. 7 Hoffman describes Sony’s “substantial intellectual property holdings” including Sony’s own 8 patents and the patent pools to which it is a licensor. Majiduddin Decl. Ex. 5 at 27. Mr. Hoffman 9 notes that “[t]hese patented and proprietary technologies have contributed significantly to the 10 commercial success and profitability of Sony’s accused products.” Id. This opinion does not 11 United States District Court Northern District of California 3 improperly refer to royalty rates associated with noncomparable patent pools. There is no mention 12 of the royalty rates at all. Therefore, so long as the royalty rates are not admitted into evidence, 13 the Court finds Mr. Hoffman’s opinion proper in this regard. 14 15 4. The Microsoft Royalty Agreement TVI also argues that Mr. Hoffman should not be permitted to testify about an effective 16 $0.02 royalty rate he calculated from the lump-sum settlement and licensing agreement between 17 TVI and Microsoft. In his expert report, Mr. Hoffman notes that Microsoft and TVI entered into 18 an agreement wherein Microsoft agreed to pay TVI $60 million for its alleged infringing sales of 19 1,134,170,852 PC units between January 29, 1997 and June 30, 2005. Arenz Dec. Ex. 12 at 16. 20 Mr. Hoffman notes this payment included a license for past sales, as well as future sales made 21 during the nine additional years of life remaining on TVI’s patents. Id. Mr. Hoffman does not 22 state in his expert report that this lump-sum agreement translates to a $0.02 effective rate. See id. 23 At his deposition, however, Mr. Hoffman was asked: “At the hypothetical negotiations, 24 were the parties aware that Microsoft would later license the TVI’s patents?” Arenz Decl. Ex. 13 25 at 59:14-16. Mr. Hoffman first responded that this question was “wholly irrelevant,” but 26 eventually stated that “Microsoft settled out something like two cents a copy was the effective 27 rate.” Arenz Decl. Ex. 13 at 58:18-19, 59:23-25. When asked how he reached that conclusion, 28 Mr. Hoffman explained: 13 1 What happened was Mr. Wagner has told us how many units were involved and he’s got that in his report and we’ve got the dollar value, and that was as of a certain date. And the patent still had another nine years to go, and you divide these units into the dollar, you get about five cents, and Microsoft sales, everyone would have expected−truth is, they have, you know, increased since then. So I think a reasonable number through the end of time would be somewhere in the two cent range. 2 3 4 5 6 7 8 9 10 United States District Court Northern District of California 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Id. at 60:12-23. In Sony’s Opposition to TVI’s Daubert motion, Sony states that Mr. Hoffman took Microsoft’s sales, doubled them to account for the fact that the paid up license also would cover Microsoft’s sales for an additional nine years, and divided the $60 million payment by these sales. Sony Opp. at 30. TVI contends that Mr. Hoffman’s testimony on the $0.02 effective rate which he calculated from the lump sum agreement between TVI and Microsoft should be excluded for two reasons: first, because Mr. Hoffman failed to disclose this effective rate in his expert report, and second, because the opinion is legally deficient for Mr. Hoffman’s failure to account for “significant” and “fundamental” differences between lump-sum licenses and running royalty licenses. See Lucent Technologies, 580 F.3d at 1325-36, 1329-30. In response, Sony asserts that all evidence pertaining to the settlement between TVI and Microsoft should be excluded in its entirety.5 Sony contends that Mr. Hoffman’s calculation of an effective royalty rate was merely in rebuttal to TVI’s expert Mr. Wagner’s reference to the $60 million lump-sum settlement agreement without any effort to convert that lump-sum into a per unit payment. TVI’s prior settlement agreements are not admissible for the purpose of calculating a reasonable royalty rate. Mr. Hoffman is therefore excluded from opining on any effective royalty rate calculated from the lump-sum agreement between TVI and Microsoft. In addition, as the $0.02 effective rate was not disclosed in his expert report, Mr. Hoffman is precluded from testifying to this matter at trial. 5. Critique of Dr. Srinivasan’s Conjoint Analysis TVI’s final challenge is to the portion of Mr. Hoffman’s opinion which critiques Professor 5 In Sony’s Motion in Limine No. 1, Sony moves to exclude all evidence and arguments pertaining to TVI’s settlement agreements in which other companies agreed to license TVI’s patents. The Court has granted this motion. 14 1 Srinivasan’s conjoint analysis. As described in below in the discussion of Sony’s Daubert motion, 2 Professor Srinivasan is an expert in the area of marketing research known as “conjoint analysis,” 3 and has written twenty-four peer-reviewed research papers on the topic. According to Professor 4 Srinivasan, conjoint analysis quantifies customer preferences for certain attributes in a product, 5 and has enabled Professor Srinivasan to estimate the “market’s willingness to pay” for TVI’s 6 patented technology as an incremental benefit in Sony’s accused products. See Declaration of 7 Charles P. Kennedy in Support of Sony’s Daubert Motion (“Kennedy Decl.”), Ex. 8A at 4. 8 A section of Mr. Hoffman’s expert report (Opinion III) critiques Professor Srinivasan’s conjoint analysis. See Arenz Decl. Ex. 12 at 38-41. In particular, Mr. Hoffman offers his expert 10 opinion that (a) Professor Srinivasan’s methodology was inappropriate given the large number of 11 United States District Court Northern District of California 9 features of the products which were not tested; (b) the results were nonsensical in the manner that 12 the respondents interpreted the price discounts used in the surveys; (c) the reboot times were 13 vastly overstated in comparison to reboot times determined by TVI’s own expert; and (d) the price 14 discount levels were not appropriately chosen. See id. 15 TVI argues that Mr. Hoffman should not be allowed to opine on Professor Srinivasan’s 16 conjoint survey analysis because Mr. Hoffman admitted at his deposition that he is not an expert in 17 21 conjoint analysis: Q: − are you an expert in conjoint analysis A: No, no, not at all, never held myself out to that Q: You’re only − you’re not going to tell the ladies and gentlemen of the jury, for example, that you’re an expert in conjoint analysis? A: No, no. I don’t think I’ve got that in my report anyway. Q: You’re not going to tell the court you’re an expert in conjoint analysis? A: No, no…. 22 Arenz Decl. Ex. 13 at 109:15-25, 110:1-5. TVI contends that because Rule 702 only permits 23 expert testimony from “[a] witness who is qualified as an expert by knowledge, skill, experience, 24 training, or education,” Mr. Hoffman’s opinion on the conjoint analysis should be excluded. 25 Fed.R.Evid. 702. In response, Sony argues that Mr. Hoffman offers proper testimony in his area 26 of expertise because he testifies to such matters as whether Professor Srinivasan’s surveys are 27 appropriate for determining a reasonable rate. 18 19 20 28 15 1 The Court finds that Mr. Hoffman’s opinion is proper in this regard. Mr. Hoffman is an 2 expert in the area of damages determined through the Georgia-Pacific analysis. His critiques of 3 Professor Srinvasan’s surveys are in the context of his expertise. Therefore, the Court will not 4 exclude his opinion under Daubert. 5 IV. SONY’S DAUBERT MOTION 6 A. 7 Sony first argues that the entirety of TVI expert Professor Srinivasan’s conjoint surveys Professor Srinivasan should be excluded from evidence as they are unreliable in the face of Daubert. Professor 9 Srinivasan has served as a tenured and emeritus professor at the Stanford Graduate School of 10 Business for the last thirty-nine years. Declaration of Professor V. Seenu Srinivasan (“Srin. 11 United States District Court Northern District of California 8 Decl.”) ¶ 2. He has authored over one-hundred articles in peer-reviewed journals, including 12 twenty-three on conjoint analysis. Id. In 1978, Professor Srinivasan coined the term “conjoint 13 analysis” in a peer-reviewed article that he co-authored with his professor. Id. 14 Professor Srinivasan describes conjoint analysis as a type of survey or market research, 15 which, at the most general level, conceptualizes products as bundles of attributes, treating price as 16 an attribute. Id. ¶ 3. Conjoint analysis uses customer surveys to determine “values” for each 17 attribute. By choosing among multiple bundles of attributes, survey participants make implicit 18 tradeoffs one would make in real-world purchasing decisions. Id. “For example, conjoint analysis 19 offers respondents hypothetical products in several combinations, some of which might contain 20 feature 1 (but not feature 2), and some of which might contain feature 2 (but not feature 1).” Id. 21 Professor Srinivasan explains that by comparing respondents’ choices when presented with 22 different features, one can estimate the quantitative values of specific features. Id. According to 23 Professor Srinivasan, studies have validated that this implicit tradeoff is more reliable than asking 24 consumers directly what they would pay for a specific feature. Id. 25 In his particular study for this case, Professor Srinivasan estimated the “market’s 26 willingness to pay” (“MWTP”) for TVI’s patented technology as an incremental benefit in Sony’s 27 accused products. Srin. Decl. Exs. 1-3 at 3. Professor Srinivasan conducted three surveys to 28 measure the MWTP for the benefit of using the autoplay feature (which he termed “next disc 16 1 playback”) in a DVD player, Blu-ray player, and a PlayStation 3 console. Id. at 3-4. The surveys 2 incorporate two noninfringing alternatives to autoplay identified by Dr. Wolfe in his expert report 3 for TVI: (1) a reboot option (measured by the increased time it takes to restart and play the next 4 disc), and (2) a filename option (the device displays a virtual keyboard on the screen and instructs 5 the user to type in a file name to play the next disc). Id. at 3; see also Arenz Decl. Ex. 7 (Opening 6 Expert Report of Dr. Andrew Wolf) ¶ 97. The MWTP is the amount by which Sony would be 7 able to increase the price of its accused products yet suffer no reduction in the number of units 8 sold if Sony were to use autoplay rather one of Professor Srinivasan’s two proposed noninfringing 9 alternatives. Srin. Decl. Exs. 1-3 at 4. 10 To conduct his conjoint analysis, Professor Srinivasan used a mathematical formula to United States District Court Northern District of California 11 determine the MWTP for the autoplay feature in each of the three accused products. Id. at 5. He 12 did this by using a particular type of conjoint analysis called Choice-Based Conjoint Analysis, in 13 which survey respondents chose among various options of combinations of the product attributes. 14 Professor Srinivasan recognized that a challenge to his study was that each of the accused products 15 had various attributes (18 in Blu-ray, 20 in DVD, and 18 in PS3). Id. The accepted methodology 16 for conjoint analysis is to limit the number of attributes to six or fewer. If survey respondents 17 were asked to evaluate 18 − 20 attributes, Professor Srinivasan states there would been an 18 information overload on the respondent, leading to poor quality data. Id. To remedy this problem, 19 Professor Srinivasan conducted his conjoint analysis in two phases. 20 In Phase 1 of his study, Professor Srinivasan asked respondents to prioritize 18 attributes 21 of each accused product to come up with a list of six attributes that have similar values as the 22 autoplay feature. Id. at 6. Professor Srinivasan asserts that Phase 1 also allowed him to determine 23 a price discount that approximately has the same or smaller value as the autoplay, which led to a 24 conservative bias of the MWTP. Id. 25 In Phase 2 of his study, Professor Srinivasan determined the estimate values which the 26 survey respondents attached to autoplay as compared to the two noninfringing alternatives. Id. 27 The six comparable attributes chosen from Phase 1 were broken into set A and set B, each of three 28 attributes. The survey respondents were then presented with hypothetical products with five total 17 1 attributes (the attribute of interest (autoplay); three other attributes of similar value; and a price 2 discount of $0, $3, $5, $10 or $20), and were asked to assume that all other attributes were the 3 same across the four options in each choice set. Id. at 5. Professor Srinivasan tested autoplay 4 against his two noninfringing alternatives. Id. at 6. Each survey respondent saw fifteen 5 randomized choice sets. Id. at 6, 14. Professor Srinivasan then analyzed the data using industry-standard software according to 6 7 peer-reviewed research and presented his conclusions in a “Dashboard” exhibit. Id. at 6. The 8 Dashboard is dynamic in that it shows the MWTP for autoplay over reboot on a second-by-second 9 basis. For example, considering the values of 20 seconds for autoplay and 75 seconds for reboot (a 55 second difference), the MWTP for autoplay is $9.86 for Blu-ray players. Id. at 6. If the 11 United States District Court Northern District of California 10 difference between autoplay and reboot is faster or slower, then the MWTP adjusts accordingly. 12 Professor Srinivasan’s results are shown below considering the MWTP over the two noninfringing 13 alternatives, for each of the three products (Blu-ray player, DVD and PS3). The estimated MWTP 14 results below account for the margin of error due to sample size at a 95% confidence interval: 15 AUTOPLAY v. REBOOT 16 Product Type MWTP 17 Blu-Ray $7.61 18 DVD $2.33 19 PS3 $5.73 20 AUTOPLAY v. FILENAME 21 Product Type MWTP 22 Blu-ray $10 23 DVD $5 24 PS3 $10 25 26 Srin. Decl. Exs. 1-3 at 6, 16-17. The estimated MWTP was used as a baseline by TVI’s other 27 expert Mr. Wagner in his calculation of a reasonable royalty rate. See Declaration of Michael J. 28 Wagner Ex. 1 at 37. 18 1 Sony moves to exclude the entirety of Professor Srinivasan’s conjoint surveys on the basis that they are fundamentally flawed and unreliable, and therefore should be excluded under 3 Daubert. The Court notes from the outset, however, that Sony’s criticism of the survey designs is 4 more appropriate for consideration by a jury, rather than the Court on a Daubert motion. The 5 Ninth Circuit has stated that “[u]nlike novel scientific theories, a jury should be able to determine 6 whether asserted technical deficiencies undermine a survey’s probative value.” Southland Sod 7 Farms v. Stover Seed Co., 108 F.3d 1134, 1143 n.8 (9th Cir. 1997). See also Click Billards, Inc. v. 8 Sixshooters Inc., 251 F.3d 1252, 1263 (9th Cir. 2001) (“[I]ssues of methodology, survey design, 9 reliability, the experience and reputation of the expert, critique of conclusions, and the like go to 10 the weight of the survey rather than its admissibility.”). Such reasoning has also been applied to 11 United States District Court Northern District of California 2 deny a Daubert motion seeking to exclude a conjoint analysis from evidence. See Microsoft Corp. 12 v. Motorola Mobility, Inc., No. 10-1823, 2012 U.S. Dist. LEXIS 152244, at *32 (W.D. Wash. Oct. 13 22, 2012) (criticisms of a conjoint analysis went “to issues of methodology, survey design, 14 reliability, and critique of conclusions, and therefore go to the weight of the survey rather than 15 admissibility”) (internal quotes omitted)). 16 1. Price Discounts 17 Sony first criticizes Professor Srinivasan’s choice to use price discounts—as opposed to 18 price itself—as one of the attribute variables in Phase 2 of his study. Sony argues that Professor 19 Srinivasan’s “price discount methodology” has not been accepted in the relevant scientific 20 community, a consideration which weighs heavily in the Daubert analysis. In response, TVI 21 argues that this argument underscores Sony’s fundamental misunderstanding of the conjoint 22 analysis. 23 The Court agrees with TVI. Professor Srinivasan did not create a novel “price discount 24 method” solely for this litigation. Rather, price discount was just one of the variables in the 25 conjoint analysis, which is accepted by the relevant community.6 Professor Srinivasan explains: 26 27 28 6 TVI points to a handful of cases to demonstrate that conjoint analysis is increasingly used in litigation. See, e.g., Arenz Decl. Ex. 8 (transcript of proceedings for Apple v. Samsung, Case No. 11-1846, Aug. 10, 2012). 19 In conjoint studies, some aspect of price is almost always included as an attribute of the product. However, the particular way price is operationalized varies across studies depending upon the objective of the study. A very common use of conjoint analysis is in deciding a new product’s feature specifications and in setting the price. In these situations, conjoint researchers include the product price itself as an attribute to predict customer choice under alternative product specifications and prices. But that is not to say that product price is the only way of incorporating price in conjoint analysis. 1 2 3 4 5 6 7 Srin. Decl. ¶ 4. Professor Srinivasan provides a handful of examples in which price was 8 incorporated in ways other than product price, such as a discount card or as a percentage of the 9 price of the base product. Id. ¶ 5. 10 Professor Srinivasan also explains why using price discounts was better than using price in United States District Court Northern District of California 11 this particular case. Id. ¶ 8. Here, the objective was to measure the MWTP of a single feature of 12 a product, such as Blu-ray which has prices ranging from about $140 to $210. If Professor 13 Srinivasan had used price, there would have been three levels of prices for the Blu-ray ($140, $170 14 and $210), and survey respondents would have been comparing a single feature (autoplay v. 15 reboot) against price differentials of $30, $40 or $70. Professor Srinivasan states that the 16 measuring interval would have been much greater than the likely value for the feature of interest 17 and could have created a non-conservative bias. Id. 18 Sony also argues that the choice to use price discounts made the results of his survey 19 illogical, and demonstrate that Professor Srinivasan’s survey method was unreliable. Sony expert 20 Mr. Klein re-analyzed the data from Professor Srinivasan’s surveys using the same survey 21 software. Unlike Professor Srinivasan, however, Mr. Klein did not constrain the utility of the 22 price variable. “Constrained estimates for price, by definition, constrains the results so that a 23 lower discount has no greater value than a higher discount.” Srin Decl. ¶ 17. According to 24 Professor Srinivasan, literature on conjoint analysis shows that constraining estimations of 25 desirable features produces better predictive results.7 Id. Mr. Klein agrees that when there are 26 27 28 7 Professor Srinivasan notes that Mr. Klein’s company’s cofounder, Dr. John Hauser of M.I.T., used constrained estimations in his expert report for Apple v. Samsung. Srin. Decl. ¶ 17. 20 1 only a few variables, constraining the price variable may improve estimates, but believes it was 2 not useful in this case where the surveys entailed several variables. Klein Decl. ¶ 11. 3 By re-analyzing Professor Srinivasan’s data and not constraining the price variable, Mr. 4 Klein infers that approximately one-half of the survey respondents violated the rational 5 assumption that consumers prefer a larger discount to a smaller discount: 6 Blu-ray: 24% prefer $0 to $10 discount 25% prefer $0 to $5 discount 36% prefer $5 to $10 discount 50% with some price preference inconsistency DVD: 20% prefer $0 to $5 discount 25% prefer $0 to $3 discount 38% prefer $3 to $5 discount 52% with some price inconsistency PS3: 23% prefer $0 to $20 25% prefer $0 to $10 31% prefer $10 to $20. 46% with some price preference inconsistency 7 8 9 10 United States District Court Northern District of California 11 12 13 14 15 Klein Decl. ¶ 7. Mr. Klein attributes this illogic to the possibility that survey respondents 16 misunderstood the price discount for a price surcharge, or did not follow Professor Srinivasan’s 17 instructions to hold all the non-tested attributes constant. Id. ¶ 12. 18 In response, Professor Srinivasan criticizes Mr. Klein’s failure to prove that any of his 19 results are statistically significant. Srin. Decl. ¶¶ 10-16. Professor Srinivasan notes that testing for 20 statistically significancy is an “elementary” principle in survey analysis, and without having 21 conducted such a test, Mr. Klein cannot know whether the perceived inconsistencies are true.8 Id. 22 ¶ 8. To rebut Mr. Klein’s findings, Professor Srinivasan oversaw a test for statistical significance 23 at a 95% confidence level, and arrived at the following results: 24 Blu-ray: 1.7% prefer $0 to $5 discount (compare Mr. Klein’s 25%) 25 26 27 28 8 Professor Srinivasan also notes that Mr. Klein did perform a statistical significance test to see whether the respondents preference for higher discounts are significantly different from their preference for no discount at all. Therefore, Professor Srinivasan believes that Mr. Klein either chose not to perform the same statistical significance test for his primary numbers, or he obtained the same results as Professor Srinivasan that the inconsistencies are immaterial to the validity of the surveys. Id. ¶ 16. 21 2.7% prefer $0 to $10 discount 2.7% prefer $5 to $10 discount 9.5% with some inconsistency (compare Mr. Klein’s 24%) (compare Mr. Klein’s 36%) (compare Mr. Klein’s 50%) DVD: 3.0% prefer $0 to $3 discount 4.0% prefer $0 to $5 discount 2.0% prefer $3 to $5 discount 9.5% with some inconsistency (compare Mr. Klein’s 25%) (compare Mr. Klein’s 20%) (compare Mr. Klein’s 38%) (compare Mr. Klein’s 52%) PS3: 6.8% prefer $0 to $20 5.1% prefer $0 to $10 3.7% prefer $10 to $20 13.9% with some inconsistency (compare Mr. Klein’s 23%) (compare Mr. Klein’s 25%) (compare Mr. Klein’s 31%) (compare Mr. Klein’s 46%) 1 2 3 4 5 6 7 8 9 Id. ¶ 12. These results demonstrate that a statistically insignificant number of respondents valued lower discounts over greater discounts. As Professor Srinivasan explained, the 95% confidence 11 United States District Court Northern District of California 10 level means that even if every one of the respondents’ true preferences are consistent with 12 economic rationality, we would incorrectly identify approximately five percent of the respondents 13 as violating economic rationality. Id. ¶ 14. Consequently, in interpreting the results above, the 14 amount in excess of 5% is a more accurate indication of the prevalence of true economic 15 irrationality. An examination of the table above therefore indicates that the prevalence of true 16 economic irrationality is less than 5% in 11 out of the 12 cells in the table above. Because only in 17 the PS3 case did approximately 9% of respondents violate the rational economic assumption 18 19 20 21 22 23 24 25 26 27 28 (13.9% - 5% = 8.9%), Professor Srinivasan concludes that there is very little prevalence of true economic irrationality. Id. Based on Professor Srinivasan’s findings, there is certainly not enough evidence to render the method unreliable under Daubert. Mr. Klein also claims in his expert report that he determined the extent to which respondents who were presented with two choices (out of a total of four) in which all the features were identical except the price discount illogically selected the alternative with the smaller discount. Klein Decl. ¶ 10. Mr. Klein’s results purportedly show that 26% chose the lower discount for DVDs, 28% chose the lower discount for Blu-ray, and 29% chose the lower discount for PS3. Id. and Exs. 4-6 thereto. In response, Professor Srinivasan points out that this data is taken from 82 choice sets out of a total of 14,535, which is one-half of one percent of the total number of choice. Srin Decl. ¶ 20. Again, this small sample of illogical choices is not significant, 22 1 and cannot bar introduction of the surveys under Daubert. Even considering this small portion of 2 the survey results to be material, Professor Srinivasan notes that Mr. Klein ignored the very 3 rational possibility that these survey respondents may have simply not paid attention to the price 4 discount, considering it to be a much less important factor than the other tested product attributes. 5 Id. ¶ 19. Professor Srinivasan also accuses Mr. Klein and Sony of merely speculating that any price 6 inconsistencies arise from his choice to use the price discount, as opposed to product price. Id. ¶ 8 18. Professor Srinivasan states that such price inconsistencies often arise in conjoint analysis, and 9 this is true regardless of whether the product price or a price discount is used. Id. Moreover, 10 Professor Srinivasan conducted what he calls a “robustness check” in response to Mr. Klein’s 11 United States District Court Northern District of California 7 critique of his report. See id. ¶ 21. According to Professor Srinivasan, the Dashboard results 12 provide an opportunity exclude lower quality data and tested at a 95% level, thereby excluding 13 respondents who exhibit great error in their responses distort the MWTP results. When examining 14 only the higher-quality data, the MTWP numbers for the reboot option decrease on average by two 15 cents, demonstrating that his original MTWP calculations were sound. Id. In the face of Professor Srinivasan’s convincing rebuttal evidence, Sony cannot genuinely 16 17 argue that his use of price discounts rendered the surveys unreliable. Mr. Klein failed to test 18 whether the results of his own re-analysis of Professor Srinivasan’s survey were statistically 19 significant. Without the help of Mr. Klein’s re-analysis of the survey data, Sony has no basis to 20 assert that survey respondents were confused by Professor Srinivasan’s use of a price discount as 21 opposed to price. Sony’s motion to exclude the Srinivasan surveys on this basis is therefore 22 denied. 23 24 2. Testing Few Product Features Sony also argues that Professor Srinivasan’s surveys are unreliable because he only tested 25 a portion of all the attributes of each product. As noted above, Professor Srinivasan asked 26 respondents to rank 18 – 20 product attributes in Phase 1 of his study. In Phase 2, he only used the 27 five attributes respondents chose in Phase 1 to have the closest value to autoplay. Sony contends 28 that the exclusion of the majority of product attributes from Phase 2 led respondents to overvalue 23 1 2 each of the product attributes that were tested. In response, TVI claims this argument disregards long-standing peer-reviewed literature 3 that using six or fewer variables leads to better predictive results because survey respondents are 4 not overwhelmed by too much data. See Srin. Decl. Ex. 1-3 at 5-6. TVI quotes Mr. Klein’s 5 partner, Professor Hauser, in a declaration for his work in the Apple v. Samsung case in which he 6 wrote that “from a scientific perspective, it is unnecessary to include all features of a product when 7 designing a conjoint study. Such a requirement is not consistent with the academic literature or 8 commercial use of conjoint studies.” Arenz Decl. Ex. 13 ¶ 7. Sony does not respond to this point. 9 Once again, Sony uses Mr. Klein’s re-analysis of Professor Srinivasan’s survey data to attempt to discredit Professor Srinivasan’s findings by showing how testing only a portion of the 11 United States District Court Northern District of California 10 product features encouraged respondents to overvalue the product features which were tested. 12 Using the data from Phase 2 of the survey and employing the same software used by Professor 13 Srinivasan, Mr. Klein measured the MWTP for the six other features tested in Phase 2. Klein Dec. 14 ¶ 14 and Ex. 7 thereto. Mr. Klein shows the MWTP for each of the six other tested features for 15 the Blu-ray player: 16 Feature Ability to play video from computer Adjustable picture settings Camera memory slot Video noise reduction Surround sound Instant skip/replay Total value of six features: 17 18 19 20 MWTP $26.79 $18.24 $16.89 $20.59 $31.22 $19.40 $133.11 21 Id. ¶ 15. Mr. Klein opines that such MWTP results are “absurd” because $133.11 is the total value 22 for only six of the 18 possible features chosen for inclusion in the study, and a fully-featured Blu- 23 ray player can cost as little as $150.9 Id. Mr. Klein states that the MWTP as measured by the 24 Srinivasan studies therefore bears no relationship to the market realities of what consumers 25 actually pay as represented by retail prices. Id. 26 27 28 9 In this same regard, Mr. Klein noted that survey respondents valued the six features of the DVD player at $55.16, which is 80% of Sony’s $68.66 retail price, and valued the six features of the PS3 player at $272.83, which is 91% of Sony’s $299 retail price. Klein Decl. ¶ 16. 24 1 In response to this analysis, Professor Srinivasan accuses Mr. Klein of violating “a very 2 basic principle in conjoint analysis that the analyst should not ‘extrapolate.’” Srin. Decl. ¶ 22 3 (“Any reasonable data analyst, and certainly a conjoint expert should know that extrapolation is to 4 be avoided.”). Professor Srinivasan explains: The price discounts varied from $0 to $5 in the case of DVD players, $0 to $10 in the case of Blu-ray player, and $0 to $20 in the case of PS3 player. Consequently, the respondent never compared two choice options that varied more than $5, $10, and $20 in the cases of DVD, Blu-ray, and PS3 players, respectively. 5 6 7 8 Id. Professor Srinivasan gives an example of the results for autoplay v. filename for Blu-ray 9 players, where the MTWP value exceeded $10 but Professor Srinivasan nevertheless entered $10 for the MWTP. Id. ¶ 23. Professor Srinivasan states that “[m]aking projections beyond these 11 United States District Court Northern District of California 10 values is considered “extrapolation,” and that “Mr. Klein’s refusal to apply this basic tenet of 12 conjoint analysis makes his opinions regarding values of individual features that exceed the dollar 13 ranges above for the respective product categories ... completely wrong.” Id. In response to this, 14 Sony notes that even avoiding such extrapolation, six features of the PS3 would still amount to 15 $120, which is more than a reasonable portion of the retail price. 16 The foregoing dispute is a classic example of the “battle of the experts” for the jury to 17 decide. While the Court is skeptical of Mr. Klein’s violation of what Professor Srinivasan calls a 18 basic principle in conjoint analysis to avoid extrapolation, the Court recognizes the oddity in 19 finding such a high valuation of each individual product feature. Nevertheless, the literature on 20 conjoint analysis condones testing six or fewer variables to produce results with a better predictive 21 value, and the Court will not exclude the Srinvasan surveys for failing to depart from this accepted 22 methodology. 23 The Court is aware that Judge Alsup recently excluded a conjoint analysis on a Daubert 24 motion for its illogical results, noting that testing only a portion of the total number of features 25 may have caused the problem. See Oracle America, Inc. v. Google, Inc., No. 10-3561-WHA, 26 2012 WL 850705, *9-13 (N.D. Cal. Mar. 13, 2010). However, in that study, the expert tested 27 seven features in the analysis, three of which were covered by the patented technology, and four of 28 which were not. Id. at *10. Judge Alsup remarked that the expert offered no principle basis for 25 1 selecting the four non-patented features for inclusion in the analysis. In contrast, Professor 2 Srinivasan conducted his conjoint analysis in two phases, and the objective of Phase 1 entailed a 3 principled basis for choosing the five features which would be tested alongside autoplay in Phase 4 2—their similar values relative to the other features. 5 6 3. Failure to Test the Closest Noninfringing Alternatives Finally, Sony contends that Professor Srinivasan’s surveys should be excluded for failure 7 to compare the value of the autoplay feature next to what Sony considers to be the best two 8 noninfringing alternatives. As discussed above, Sony’s “press any button” alternative is excluded 9 from evidence because it still infringes TVI’s patents. Moreover, because Sony’s other proposed noninfringing alternative—the operating system reload alternative—was not in existence at the 11 United States District Court Northern District of California 10 time the hypothetical negotiation would have taken place, Sony bears the burden of proving before 12 the jury that this alternative was “available.” Conceptus, 771 F.Supp.2d at 1179 (citing Grain 13 Processing, 185 F.3d at 1353) (whether a purported noninfringing alternative was “available” 14 requires an assessment of whether Sony had the “necessary materials, equipment, know-how, and 15 experience to make an alternative product during the relevant time frame.”). Sony has not yet met 16 its burden of proving “availability” because this is a fact issue for the jury to decide. The Court 17 will not exclude an expert report based on Sony’s assumption which may be disproven at trial. 18 B. 19 TVI hired Mr. Wagner to offer an expert opinion on a reasonable royalty rate owed to TVI 20 in the event TVI prevails on its infringement claims. Sony moves to exclude Mr. Wagner’s expert 21 opinion on five grounds−the first and last of which the Court quickly rejects. The Court will not 22 exclude Professor Srinivasan’s conjoint analysis, thus Mr. Wagner’s use of Professor Srinivasan’s 23 calculation of the MWTP as a baseline for his royalty rates is admissible. The Court will also not 24 exclude Mr. Wagner’s opinion because he considered qualitative factors−the essence of Sony’s 25 final argument. Mr. Wagner based his assessment of a reasonable royalty on several underlying 26 considerations and performed the analysis required by Georgia-Pacific. See Declaration of 27 Michael J. Wagner (“Wagner Decl.”) Ex. 1. The Court will now address Sony’s more interesting 28 arguments. Mr. Wagner 26 1 2 1. Price Elasticity Sony contends Mr. Wagner’s opinion should be excluded from evidence because of his 3 purported failure to account for price elasticity−the amount by which Sony’s sales would decrease 4 as a result of the increase in price associated with Mr. Wagner’s reasonable royalties. Sony cites 5 Crystal Semiconductor Corp. v. Tritech Microelectronics International, Inc., where a district court 6 set aside a jury award for lost profits based on a price erosion theory because it lacked adequate 7 record support, and the Federal Circuit affirmed this determination. 246 F.3d 1336, 1361 (Fed. 8 Cir. 2001). The Federal Circuit noted several deficiencies of the patent-holder’s expert’s report, 9 including the fact the expert “did not present any evidence of how a hypothetical increase in price would have affected Crystal’s profits due to lost sales.” Id. at 1360. The court thought this 11 United States District Court Northern District of California 10 problematic because “[l]ost sales and price erosion damages are inextricably linked.” Id. 12 TVI persuasively distinguishes Crystal Semiconductor from the instant case. That case 13 arose in the context of price erosion damages, a theory which is not at issue in this case and which 14 differs from an expert’s calculation of a reasonable royalty. To be entitled to damages on a price 15 erosion theory, the patent holder must “show that ‘but for’ infringement, it would have sold its 16 product at a higher price.” Id. at 1357. The Federal Circuit has noted that “in a credible economic 17 analysis, the patentee cannot show entitlement to a higher price divorced from the effect of that 18 higher price on the demand for the product.” Id. Therefore, “the patentee’s price erosion theory 19 must account for the nature, or definition, of the market, … and the effect of the hypothetically 20 increased price on the likely number of sales at that price in that market.” Id. Thus, Crystal 21 Seminconductor requires a price elasticity analysis when a patent holder seeks to collect damages 22 on a price erosion theory. 23 Mr. Wagner’s rebuttal evidence persuades the Court that the reasoning in Crystal 24 Semiconductor cannot simply extend to calculations of a reasonably royalty. To collect price 25 erosion damages, a patent-holder must show that the infringing conduct affected retail prices. See 26 Crystal Semiconductor, 246 F.3d at 1357 (to recover price erosion damages, a patent-holder must 27 show that “‘but-for infringement, it would have sold its product at higher prices”). One cannot 28 assume, however, that a royalty will always affect retail prices because a supplier of goods does 27 1 not always burden its consumers by passing additional costs downstream. Without a showing that 2 the hypothetical retail prices would have affected Sony’s retail prices, the reasoning in Crystal 3 Semiconductors is wholly inapplicable to this case. 4 According to Mr. Wagner, his proposed royalty for the Playstation3 would have resulted in 5 a total added cost of $24.5 million to Sony based on the sales information that Sony has provided. 6 Wagner Decl. ¶ 5. Sony’s total cost of goods sold for Playstation3 is $10,592,355,058 (not 7 including any sales, general, or administrative costs). Id. The $24.5 million royalty expense 8 would only be an increase of two-tenths of one percent of the total costs. Id. According to Mr. 9 Wagner, “Sony has already agreed not to charge over $3.1 billion dollars of costs of goods sold to their customers.” Id. In that case, the $24.5 million would be absorbed as a portion of the costs 11 United States District Court Northern District of California 10 Sony already chooses not to pass onto its customers. See id. 12 13 2. Retail vs. Wholesale Prices Sony next argues that Mr. Wagner’s opinion must be excluded for his calculation of a 14 reasonably royalty using retail prices instead of whole sale prices. Upon a finding of infringement 15 this Court must award damages “adequate to compensate for the infringement, but in no event less 16 than a reasonably royalty for use made of the invention by the infringer ….” 35 U.S.C. § 284 17 (emphasis added). Sony contends that as a wholesaler, Sony’s “use made of the invention” must 18 be measured by wholesale prices as opposed to retail prices, which Mr. Wagner recognizes is 19 usually double in relation to the wholesale price. 20 Sony does not cite any case in support of this argument. TVI contends that nothing in 21 § 284 or Federal Circuit precedent requires the use of wholesale prices to calculate a reasonable 22 royalty to be paid by a wholesaler. TVI notes that Mr. Wagner is required to “carefully tie proof 23 of damages to the claimed invention’s footprint in the market place,” ResQNet.com, Inc. v. Lansa, 24 Inc., 594 F.3d 860, 869 (Fed. Cir. 2010), and contends Mr. Wagner has done just that by 25 considering the value of TVI’s patented technology to consumers in the context of the Georgia- 26 Pacific factors. TVI also argues that Sony’s argument makes no sense in a real-world application. 27 TVI asserts that because it may recover on the theory of indirect infringement through its retailers’ 28 or end-consumers’ infringing use, Mr. Wagner’s consideration of the value to end consumers is 28 1 appropriate. In the same regard, TVI argues that under the theory of patent exhaustion, 2 downstream infringers (retailers and end-consumers) effectively receive a license to use TVI’s 3 patented technology and thus consideration of retail prices is appropriate. 4 The Court will not exclude Mr. Wagner’s opinion from evidence on this basis. While 5 Sony puts forth an interesting argument, Sony lacks any precedential support for the argument that 6 a hypothetically-calculated reasonable royalty must be tied to wholesale prices as opposed to retail 7 prices. Sony has failed to prove that Mr. Wagner’s use of retail prices renders his opinion 8 unreliable under Daubert. If Mr. Wagner’s use of retail prices rendered his calculation of a 9 reasonable royalty inaccurate, that goes to the weight of his testimony rather than admissibility. 10 3. Royalty Values from 2011/2012 United States District Court Northern District of California 11 Sony’s final argument is that Mr. Wagner improperly used Professor Srinivasan’s MWTP 12 as a baseline for his royalty rates because Professor Srinivasan’s surveys were taken in 2011 and 13 2012, fifteen years after the date of the hypothetical negotiation in 1997. Sony cites Laser 14 Dynamics, Inc. v. Quanta Computer, Inc., 694 F.3d 51 (Fed. Cir. 2012), where the Federal Circuit 15 stated that “[a] reasonably royalty determination for purposes of making a damages evaluation 16 must relate to the time infringement occurred, and not be an after-the-fact assessment.” Id. at 75 17 (quoting Riles v. Shell Exploration & Prod. Co., 298 F.3d 1302, 1313 (Fed. Cir. 2002)). In Laser 18 Dynamics, the Federal Circuit reversed the district court for requiring the parties to set their 19 hypothetical negotiation at the time the lawsuit was filed as opposed to when infringement began. 20 Id. Here, Mr. Wagner calculated a reasonable royalty using a hypothetical negotiation date in 21 1998, which is the date Sony began selling the accused products. Wagner Decl. Ex. 1 at 31. 22 Sony’s challenge to Mr. Wagner’s opinion is therefore more subtle than the issue presented 23 in Laser Dynamics. Sony contends that even though Mr. Wagner agrees a reasonable royalty rate 24 should correspond to the hypothetical negotiation in 1997-1998, Mr. Wagner’s consideration of 25 surveys from 2011-2012 render his calculations unreliable. More specifically, Sony argues that 26 consumers value autoplay in relation to reboot more than they did fifteen years ago. In response, 27 TVI argues that because DVD prices have decreased since they were first introduced, survey 28 respondents likely placed less importance on price than they would have done in 1997. 29 1 Sony may be correct in its assessment that consumers value autoplay more today than they 2 did fifteen years ago, but this is an issue for the jury to decide. It simply does not follow that Mr. 3 Wagner’s consideration of the Srinivasan surveys from 2011-2011 render his calculations 4 unreliable in the face of Daubert. Rather, the lapse in time goes to the weight of Mr. Wagner’s 5 testimony. In the same regard, Mr. Wagner’s failure to consider the 1997 licensing agreement 6 negotiated between TVI and Intel for the same technology bears on the weight of his testimony, 7 but does not merit total exclusion. Accordingly, the Court will not exclude Mr. Wagner’s opinion 8 on this basis. 9 C. Dr. Wolfe Dr. Wolfe was hired by TVI as an expert on the best noninfringing alternatives. Arenz 11 United States District Court Northern District of California 10 Decl. Ex. 7 (Opening Expert Report of Dr. Andrew Wolfe). As mentioned in the discussion of 12 Professor Srinivasan’s surveys, Dr. Wolfe’s expert report suggests that the two next-best 13 noninfringing alternatives are the filename and the reboot alternatives. Id. ¶ 97. In filename 14 alternative, a user inputs a filename in order to begin playback of content from a disc. In the 15 reboot alternative, a user reboots between disc insertions. According to Dr. Wolfe, “[t]his would 16 cause the user interface, both on the video display and the device front panel, to disappear for at 17 least some period of time. It would also cause an additional delay in playing the new disc.” Id. 18 Dr. Wolfe conducted a series of experiences to quantify the impact of the delay incurred as 19 a result of using the reboot noninfringing alternative. Id. ¶ 98. Dr. Wolfe took two time-estimates 20 to document the time it would take for content from a disc to be displayed on the screen using both 21 autoplay and the reboot noninfringing alternative. Dr. Wolfe captured the playback of each disc 22 on HD video and then extracted the playback time by reviewing the time codes for each recording 23 to the nearest 1/30th of a second. Id. ¶ 100. Dr. Wolfe displayed his time measurements on two 24 charts in of his opening expert report. See id. at 47, 49. 25 Sony challenges Dr. Wolfe’s time measurements on grounds that his “tests were not 26 derived from scientific methods.” Sony Daubert Motion at 23. Sony asserts that “Dr. Wolfe has 27 not pointed to any peer review or publication to support that his tests rest on scientifically valid 28 principles.” Id. Sony criticizes Dr. Wolfe’s failure at his deposition to explain the details about 30 1 how he conduc his tests and failure to include t tape-reco w cted s, e the ordings of D Wolfe’s ti Dr. ime 2 me easurements in his expert report as re t equired by R Rule 26 of th Federal Ru of Civil Procedure. he ules l 3 Id. 4 Sony’s contentions are meritles Dr. Wolf more than adequately explained each step of ss. fe n 5 the process in his expert report. Sony makes much of the fact that such tim e h m h me-measurem ments have 6 not been “peert -reviewed” and tested fo accuracy. However, w respect to measurem of time a or with ment 7 by tape-recordi ings, no such peer-review is necessa h w ary. The Cou will there urt efore not exc clude Dr. 8 olfe’s expert opinion. t Wo Further rmore, the fa that Dr. Wolfe tape-re act W ecorded his tests does no mean he v ot violated the 10 Fed deral Rules of Civil Proc o cedure by no including these record ot dings in his e expert report Pursuant t. 11 United States District Court Northern District of California 9 to Rule 26, Dr. Wolfe was required to disclose in h opening r R d his report “(i) a complete sta atement of 12 all opinions the witness will express an the basis a reasons for them; [a e nd and and] (ii) the f facts or data 13 con nsidered by the witness in forming th t i hem.” Fed.R R.Civ.P. 26(a a)(2)(B). He did so. Son learned e ny 14 of the videotap at deposition but neit t pes ther sought t them, moved to compel, nor moved t d to 15 sup pplement its expert opini ions. This motion came too late. m e 16 V. 17 18 19 20 CONCLUSION For the reasons stat above, So ted ony’s Daube Motion is DENIED a TVI’s Partial ert and aubert Motio is GRANT on TED in part and DENIE in part as set forth abo ED ove. Da IT IS SO ORDERE ED. Da ated: March 1, 2013 1 21 __ ___________ __________ ____ Jos seph C. Sper ro Un nited States M Magistrate J Judge 22 23 24 25 26 27 28 31