Oracle Corporation et al v. SAP AG et al

Filing 831

Memorandum in Opposition to 769 Plaintiffs' Motion No. 4 to Exclude Expert Testimony of Donald Reifer filed by SAP AG, SAP America Inc, Tomorrownow Inc. (Attachments: # 1 Proposed Order)(Froyd, Jane) (Filed on 9/9/2010) Modified on 9/10/2010 (vlk, COURT STAFF).

Download PDF
Oracle Corporation et al v. SAP AG et al Doc. 831 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Robert A. Mittelstaedt (SBN 060359) Jason McDonell (SBN 115084) Elaine Wallace (SBN 197882) JONES DAY 555 California Street, 26th Floor San Francisco, CA 94104 Telephone: (415) 626-3939 Facsimile: (415) 875-5700 ramittelstaedt@jonesday.com jmcdonell@jonesday.com ewallace@jonesday.com Tharan Gregory Lanier (SBN 138784) Jane L. Froyd (SBN 220776) JONES DAY 1755 Embarcadero Road Palo Alto, CA 94303 Telephone: (650) 739-3939 Facsimile: (650) 739-3900 tglanier@jonesday.com jfroyd@jonesday.com Scott W. Cowan (Admitted Pro Hac Vice) Joshua L. Fuchs (Admitted Pro Hac Vice) JONES DAY 717 Texas, Suite 3300 Houston, TX 77002 Telephone: (832) 239-3939 Facsimile: (832) 239-3600 swcowan@jonesday.com jlfuchs@jonesday.com Attorneys for Defendants SAP AG, SAP AMERICA, INC., and TOMORROWNOW, INC. UNITED STATES DISTRICT COURT NORTHERN DISTRICT OF CALIFORNIA OAKLAND DIVISION ORACLE USA, INC., et al., Plaintiffs, v. SAP AG, et al., Defendants. Case No. 07-CV-1658 PJH (EDL) DEFENDANTS' OPPOSITION TO PLAINTIFFS' MOTION NO. 4 TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Date: Time: Courtroom: Judge: September 30, 2010 2:30 p.m. 3, 3rd Floor Hon. Phyllis J. Hamilton SVI-84676v1 DEFS.' OPP. TO PLS.' MOT. NO. 4: TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Case No. 07-CV-1658 PJH (EDL) Dockets.Justia.com 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 V. VI. VII. D. C. B. I. II. III. IV. TABLE OF CONTENTS Page INTRODUCTION .............................................................................................................. 1 LEGAL STANDARD......................................................................................................... 2 PLAINTIFFS' MOTION IS MOOT................................................................................... 3 REIFER'S OPINIONS ARE BASED ON RELIABLE DATA ......................................... 3 A. The Creation and Use of Counting Utilities............................................................ 4 1. 2. Pinto's Counting Utilities............................................................................ 4 Reifer's Replica Counting Utilities............................................................. 5 Reifer's Analysis of Pinto's Counters Is Reliable .................................................. 6 1. 2. 3. The Replicas Were Built Using Pinto's Disclosed Rules ........................... 6 Reifer Is Qualified to Develop Replica Counters ....................................... 6 Pinto's Use of FlightGear Was Necessary and Appropriate....................... 7 Any Alleged Flaws in Reifer's Data Do Not Render His Opinion Inadmissible ............................................................................................................ 7 Plaintiffs' Case Support Is Inapposite..................................................................... 8 REIFER IS QUALIFIED TO REBUT PINTO'S OPINION.............................................. 9 REIFER'S OPINIONs are HIGHLY RELEVANT TO THE JURY'S ANALYSIS........ 10 CONCLUSION ................................................................................................................. 10 -i- DEFS.' OPP. TO PLS.' MOT. NO. 4: TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Case No. 07-CV-1658 PJH (EDL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Cases TABLE OF AUTHORITIES Page(s) Andrews v. E.I. Du Pont De Nemours and Co., 447 F.3d 510 (7th Cir. 2006)........................................................................................................ 9 Clicks Billiards, Inc. v. Sixshooters, Inc., 251 F.3d 1252 (9th Cir. 2001)...................................................................................................... 8 Daubert v. Merrell Dow Pharms, Inc., 509 U.S. 579 (1993)................................................................................................................. 2, 3 Daubert v. Merrell Dow Pharms., Inc., 43 F.3d 1311 (9th Cir. 1995)........................................................................................................ 8 DSU Med. Corp. v. JMS Co., Ltd., 296 F. Supp. 2d 1140 (N.D. Cal. 2003) ....................................................................................... 9 Guidroz-Brault v. Missouri Pac. R.R. Co., 254 F.3d 825 (9th Cir. 2001)........................................................................................................ 9 In re Citric Acid Litig., 191 F.3d 1090 (9th Cir. 1999)...................................................................................................... 8 Kennedy v. Collagen Corp., 161 F.3d 1226 (9th Cir. 1998)...................................................................................................... 3 Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999)..................................................................................................................... 2 Microfinancial, Inc. v. Premier Holidays Int'l, Inc., 385 F.3d 72 (1st Cir. 2004) .......................................................................................................... 8 Primrose Operating Co. v. National Am. Ins. Co., 382 F.3d 546 (5th Cir. 2004)........................................................................................................ 7 QR Spex, Inc. v. Motorola, Inc., No. CV 03-6284-JFW (FMOx), 2004 WL 5642907 (C.D. Cal. Oct. 28, 2004).......................... 8 Redfoot v. B. F. Ascher & Co., No. C 05-2045 PJH, 2007 WL 1593239 (N.D. Cal. June 1, 2007)............................................ 10 Robinson v. G.D. Searle & Co., 286 F. Supp. 2d 1216 (N.D. Cal. 2003) ....................................................................................... 9 Salinas v. Amteck of Ky., Inc., 682 F. Supp. 2d 1022 (N.D. Cal. 2010) ..................................................................................... 10 Stilwell v. Smith & Nephew, Inc., 482 F.3d 1187 (9th Cir. 2007)...................................................................................................... 8 - ii - DEFS.' OPP. TO PLS.' MOT. NO. 4: TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Case No. 07-CV-1658 PJH (EDL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Rules TABLE OF AUTHORITIES (continued) Page(s) United States v. Chang, 207 F.3d 1169 (9th Cir. 2000)...................................................................................................... 9 United States v. Elkins, 885 F.2d 775 (11th Cir. 1989)...................................................................................................... 8 Fed R. Evid. 702 .............................................................................................................. 2, 3, 10, 11 - iii - DEFS.' OPP. TO PLS.' MOT. NO. 4: TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Case No. 07-CV-1658 PJH (EDL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 I. INTRODUCTION Plaintiffs designated Paul C. Pinto to testify solely on so-called "saved development costs." Defendants designated Donald Reifer and David P. Garmus solely to rebut Pinto. None of these experts are relevant to this case as a result of this Court's August 17, 2010 Order holding that Plaintiffs may not seek damages in the form of "saved development costs" for any cause of action. See D.I. 762 (8/17/10 Order) at 18-23; see also Declaration of Tharan Gregory Lanier in Support of Defs.' Opp. to Pls.' Mot. to Exclude Expert Testimony of Donald Reifer ("Lanier Decl.") ¶¶ 1-2, Ex. 1 (Defs.' Initial Expert Disclosures) at 5; Ex. 2. (Pls.' Supp. Expert Disclosures) at 3 (Pinto designated solely to "analyze, calculate, and testify to the costs associated with software product development"). Accordingly, Plaintiffs' Motion No. 4 is moot and should be denied as it seeks to exclude specific opinions of Reifer's that rebut Pinto's opinions on saved development costs. Additionally, were their motion not moot, Plaintiffs' attempt to exclude a discrete portion of Reifer's opinions should be denied. First, Plaintiffs argue that certain of Reifer's opinions are not supported by sufficient data when in fact they are supported by Plaintiffs' own disclosures. Specifically, in determining saved development costs, Pinto purports to use the Constructive Cost Model (COCOMO), a methodology used to estimate the size and cost of a software product. See Lanier Decl. ¶ 3, Ex. 3 (Pinto Report) at 34. As part of Pinto's COCOMO analysis, he uses counting utilities to count the number of source lines of code (SLOC) in two of the software suites at issue in this case. See Lanier Decl. ¶ 3, Ex. 3 (Pinto Report) at 16. Pinto created these counting utilities using parsing and counting rules, which rules he disclosed with his report. See Lanier Decl. ¶¶ 5-6, Ex. 5 (ORCLX-PIN-000066); Ex. 6 (ORCLX-PIN-000076). Reifer, an expert in the COCOMO model, examined Pinto's analysis. See Lanier Decl. ¶ 4, Ex. 4 (Reifer Report) at 7-8. Following the counting and parsing rules that Pinto provided, Reifer created replica utilities, which he then analyzed to determine if they accurately counted SLOC. See id. at 19-20. Thus, the portion of Reifer's opinion relevant to Plaintiffs' motion is based upon data disclosed by Plaintiffs' own expert. Plaintiffs have failed to show how Reifer's use of this data renders his opinion unreliable. SVI-84676v1 -1- DEFS.' OPP. TO PLS.' MOT. NO. 4: TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Case No. 07-CV-1658 PJH (EDL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Second, Plaintiffs' additional argument that Reifer is unqualified and that his opinion will not assist the trier of fact also fails. Reifer's extensive experience in software engineering and specifically in COCOMO more than qualifies him to critique Pinto's analysis. Reifer has worked in the software field for over forty years, possesses extensive expertise in the field of software cost estimating, and helped to calibrate the latest COCOMO model. See id. at 6, 7-8. Further, SLOC count is the primary input for a COCOMO analysis. See id. at 32. Because Reifer is highly familiar with methods for counting SLOC, his opinions will be of great use to the trier of fact in explaining Pinto's use of inaccurate SLOC counts. II. LEGAL STANDARD Rule 702 permits experts qualified by "knowledge, experience, skill, expertise, training, or education" to testify "in the form of an opinion or otherwise" based on "scientific, technical, or other specialized knowledge" if that knowledge will "assist the trier of fact to understand the evidence or to determine a fact in issue." Fed. R. Evid. 702. The Court serves as the "gatekeeper" in excluding expert testimony that fails to clear the threshold hurdles of relevance and reliability. Daubert v. Merrell Dow Pharms, Inc., 509 U.S. 579, 589-90, 597 (1993); see also Kumho Tire Co. v. Carmichael, 526 U.S. 137, 147-48 (1999) (holding that the gatekeeping function created by Daubert applies to evaluating technical experts). "This entails a preliminary assessment of whether the reasoning or methodology is scientifically valid and of whether that reasoning or methodology properly can be applied to the facts in issue." Daubert, 509 U.S. at 592-93. To make this determination, the Court must apply a three-part test: (1) Is the proffered expert qualified to testify in the area on which he is opining based on his knowledge, skill, experience, training, or education (qualification requirement)?; (2) Is the proffered expert testimony based on reliable scientific or specialized knowledge that is reliably applied to the facts of this case (reliability requirement)?; and (3) Will the proffered expert testimony assist the trier of fact in understanding the evidence or determining a fact in issue (relevancy requirement)? See Fed. R. Evid. 702; Daubert, 509 U.S. at 592-93. Rule 702 is applied consistent with "the `liberal thrust' of the Federal Rules and their SVI-84676v1 -2- DEFS.' OPP. TO PLS.' MOT. NO. 4: TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Case No. 07-CV-1658 PJH (EDL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 `general approach of relaxing the traditional barriers to `opinion testimony.'" Daubert, 509 U.S. at 588 (citations omitted); see also Fed. R. Evid. 702 advisory committee's notes (2000 Amendments) ¶ 6 (confirming that "rejection of expert testimony is the exception rather than the rule"). Opining on the flaws in another expert's methodology is a common, and admissible, form of expert testimony. See generally, e.g., Kennedy v. Collagen Corp., 161 F.3d 1226, 1230-31 (9th Cir. 1998) ("In arriving at a conclusion, the fact finder may be confronted with opposing experts, additional tests, experiments, and publications, all of which may increase or lessen the value of the expert's testimony. But their presence should not preclude the admission of the expert's testimony-they go to the weight, not the admissibility."). When the threshold for admissibility is met, differences in the experts' opinions simply go to the weight of the testimony and not the admissibility. See id. III. PLAINTIFFS' MOTION IS MOOT Plaintiffs' motion is moot and should be denied because it concerns testimony rebutting Pinto's calculation of now prohibited "saved development costs." Plaintiffs admit that Pinto measured "the amount that SAP would have spent to develop software of similar functionality to what it [allegedly] infringed here." D.I. 769 (Reifer Motion) at 1. Those are "saved development costs." This Court ordered that such damages are unavailable as a matter of law and barred Plaintiffs from seeking saved development costs for any cause of action in this case. See D.I. 762 (08/17/10 Order). As a result, any opinions regarding the costs that SAP allegedly would have spent to develop the software at issue--i.e., the entirety of Pinto's analysis and testimony--are irrelevant. Therefore, motion practice concerning Reifer's rebuttal of Pinto's now inadmissible saved development costs opinions is moot. IV. REIFER'S OPINIONS ARE BASED ON RELIABLE DATA Plaintiffs' challenge to one small portion of Reifer's rebuttal analysis, namely, Reifer's opinions on Pinto's estimates of SLOC counts, fails because (1) Reifer created accurate replicas based on the same parsing and counting rules employed by Pinto, and (2) any alleged flaws in Reifer's underlying data go to the weight of his opinion rather than its admissibility. SVI-84676v1 -3- DEFS.' OPP. TO PLS.' MOT. NO. 4: TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Case No. 07-CV-1658 PJH (EDL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 A. The Creation and Use of Counting Utilities. 1. Pinto's Counting Utilities. In performing his saved development costs analysis, Pinto purports to use COCOMO and Function Point Analysis--methodologies that, when properly employed, can determine the size and correlating development cost of software. See Lanier Decl. ¶ 3, Ex. 3 (Pinto Report) at 7-10. As part of his purported Function Point Analysis (Pinto's "Ten-Step Analysis"), Pinto asserts that he extracted source code and then counted the SLOC in the PeopleSoft and J.D. Edwards EnterpriseOne software suites. See id. at 15-17. The logical SLOC count was used by Pinto "as the basis for establishing the size of the code base." See id. at 16. This size was in turn used by Pinto to derive the "total cost of application development." See id. at 8. Pinto subsequently used his SLOC estimate as an input into his purported COCOMO analysis, again to estimate the size, and therefore cost, of the software suites at issue. See id. at 39-40 (using SLOC count in COCOMO analysis to derive estimated cost of developing J.D. Edwards World). The COCOMO model is an algorithmic model that estimates size of a software program based on the number of SLOC found in that program. Thus, Pinto's SLOC count was a major component in both of his estimating methodologies. In order to count SLOC, one generally uses software called "counting utilities" or "counters." Rather than using proven commercial counting utilities to derive his SLOC count, Pinto developed custom counting utilities to count the code in the Oracle software. See id. at 16. As an expert in the field, Reifer questions why Pinto would develop his own counting utilities when counters that are frequently used in the industry can be easily acquired. See Lanier Decl. ¶ 4, Ex. 4 (Reifer Report) at 18. Additionally, Pinto's counters fail to fully comply with the standards and conventions defined by the Software Engineering Institute (SEI). See id. at 19; Lanier Decl. ¶ 3, Ex. 3 (Pinto Report) at 16 (purporting to rely on SEI standards in the SLOC counting portion of his analysis). Because the source code that Pinto allegedly counted was destroyed by Pinto's team, and because Pinto used non-industry counters, Reifer determined that it was necessary to analyze the SVI-84676v1 -4- DEFS.' OPP. TO PLS.' MOT. NO. 4: TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Case No. 07-CV-1658 PJH (EDL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 accuracy of the Pinto counters themselves.1 See Lanier Decl. ¶ 7, Ex. 7 (Pinto Tr.) at 46:13-17; see also Lanier Decl. ¶ 4, Ex. 4 (Reifer Report) at 18. By analyzing the counters that Pinto used, Reifer could determine whether Pinto's SLOC count was accurate, and thereby whether Pinto's cost estimate of the relevant software was accurate. In other words, Reifer was able to use Pinto's counters to evaluate Pinto's final cost estimates. 2. Reifer's Replica Counting Utilities. When Pinto submitted his expert report, he did not provide his counting utilities--only the parsing and counting rules by which his utilities operate. In order to run and evaluate Pinto's counters, Reifer needed software files that he could execute (i.e., open and run). Reifer was not provided with executables of the Pinto counters until February 24, 2010, some three months after Pinto's report was first submitted and barely one month before Reifer's rebuttal report was due. See Lanier Decl. ¶ 8, Ex. 8 (2/24/10 e-mail from Amy Donnelly to Jeffrey Butler); D.I. 586 (12/22/09 Order) at 2. The files sent on February 24 could not be executed and thus were unusable for Reifer's rebuttal analysis; Reifer's requests for further information on these utilities were not effectively addressed. See Lanier Decl. ¶ 9, Ex. 9 (Reifer Tr.) at 127:12-16. Plaintiffs state that Reifer never sought the technical assistance of Oracle or Pinto with the counters. See D.I. 769 (Reifer Motion) at 5. However, the files provided by Pinto were inoperable. As explained by Reifer in his deposition testimony, if two software engineers--Reifer and his assistant--could not get the utilities to execute in a period of two weeks, it would have taken far longer than the two weeks that they had remaining for anyone else to make them work. See Lanier Decl. ¶ 9, Ex. 9 (Reifer Tr.) at 158:4-8. Pinto's late disclosure gave Reifer no choice but to build replicas. After creating the replica counters, Reifer first assessed how well the counters worked using an open source code program called FlightGear; this program served as an independent benchmark for the counting utilities. See Lanier Decl. ¶ 4, Ex. 4 (Reifer Report) at 14. The Plaintiffs' criticism that Reifer did not "count all of the lines of code in the software products at issue" ignores the fact that the extracted source code was impermissibly destroyed by Plaintiffs. In making this statement, Plaintiffs also appear to overlook Pinto's own failure to perform a SLOC count for each software product at issue in this case. See Lanier Decl. ¶ 3, Ex. 3 (Pinto Report) at 39-43 (extrapolating SLOC counts for J.D. Edwards World and Siebel). SVI-84676v1 1 -5- DEFS.' OPP. TO PLS.' MOT. NO. 4: TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Case No. 07-CV-1658 PJH (EDL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 FlightGear test revealed flaws in Pinto's SLOC counts and prompted Reifer to continue his evaluation of Pinto's counters. See id. at 19. Reifer tested a second set of counters for the Java J2EE programming language following the parsing and counting rules provided by Pinto. See Lanier Decl. ¶ 6, Ex. 6 (ORCLX-PIN000076). Reifer then located three C and two Java J2EE routines on the J.D. Edwards EnterpriseOne code library and ran them through the replica Pinto counters and the accepted industry counter. See Lanier Decl. ¶ 4, Ex. 4 (Reifer Report) at 19-20. The results showed a nine and one half percent error in Pinto's count. See id. Because there appeared to be a significant anomaly in Pinto's counters, Reifer then counted a larger sample of the C code in J.D. Edwards EnterpriseOne and found that the error for C code was 14.5%. See id. at 22. While "seemingly small," these errors are "significant when working with numbers of this magnitude." See id. Thus, Reifer's analysis serves to cast much doubt on Pinto's cost analysis. B. Reifer's Analysis of Pinto's Counters Is Reliable. 1. The Replicas Were Built Using Pinto's Disclosed Rules. Reifer replicated Pinto's counters using the exact counting and parsing rules disclosed by Pinto in the supporting materials to his report. See Lanier Decl. ¶ 9, Ex. 9 (Reifer Tr.) at 128: 911; see also Lanier ¶¶ 5-6, Ex. 5 (ORCLX-PIN-000066); Ex. 6 (ORCLX-PIN-000076). He did this with his assistant, Tom Tan, a Ph.D. student in computer science who creates code counters at the University of Southern California. See Lanier Decl. ¶ 4, Ex. 4 (Reifer Report) at 19-20. Pinto accurately, correctly, and fully reported the parsing and counting rules he used for his custom-made utilities, then the replicas thereof used by Reifer--which accurately followed all of those rules--were for all intents and purposes the same as those used by Pinto. See Lanier Decl. ¶ 9, Ex. 9 (Reifer Tr.) at 322:2-4. Thus, Reifer's analyses of these counters and of Pinto's correlating size and development cost estimates are reliable. 2. Reifer Is Qualified to Develop Replica Counters. If Plaintiffs acknowledge that the use of accurate replica counters is an appropriate method for performing this analysis. See D.I. 769 (Reifer Motion) at 9 (recognizing that "an experienced expert [may] design `replica' counters that [are] accurate"). However, unhappy with Reifer's SVI-84676v1 -6- DEFS.' OPP. TO PLS.' MOT. NO. 4: TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Case No. 07-CV-1658 PJH (EDL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 findings that Pinto's counters are highly inaccurate, Plaintiffs assert that the replica counters that Reifer used gave him faulty results because he did not engage "an experienced expert" to design the replicas. Id. Plaintiffs offer no evidence that Reifer is unqualified to build counters. To the contrary, the evidence shows that Reifer is immensely familiar with the calculation and use of SLOC counts and has previously led entire teams in developing code counters. See Lanier Decl. ¶ 9, Ex. 9 (Reifer Tr.) at 130:18-19, 22. Additionally, Tan, who assisted Reifer in creating the replicas, writes code counters as part of the code counting project at the University of Southern California. See id. at 138:24-139:1. Plaintiffs' additional complaint that Reifer did not develop the replica counters himself is curious considering that Pinto's SLOC counts were performed by a team in India and that Pinto's team modified his former employer's "in-house" counters for use in this case. See Lanier Decl. ¶ 7, Ex. 7 (Pinto Tr.) at 199:6-10; 40:20-42:4. 3. Pinto's Use of FlightGear Was Necessary and Appropriate. Although Plaintiffs attack Reifer's use of FlightGear by mischaracterizing his deposition testimony, Reifer has made clear that the use of this program was an important first step in testing the accuracy of Pinto's counters and was vital to the determination of whether to expend further time testing those counters. See Lanier Decl. ¶ 9, Ex. 9 (Reifer Tr.) at 150:3-24 (explaining that they used FlightGear as an initial experiment to compare Pinto's counters with the industry counters to determine if further investigation was necessary). Reifer's FlightGear test revealed flaws in Pinto's counters and suggested that Pinto's estimates regarding the software suites at issue were inaccurate. See Lanier Decl. ¶ 4, Ex. 4 (Reifer Report) at 19. That Plaintiffs do not like the results of this preliminary test is of no relevance to the reliability of Reifer's final opinion. Thus, any objection to Reifer's use of FlightGear as a test program is misplaced. C. Any Alleged Flaws in Reifer's Data Do Not Render His Opinion Inadmissible. Plaintiffs' complaints regarding Reifer's use of replica counters do not sufficiently call Reifer's opinions into question. "[A]s a general rule, questions relating to the bases and sources of an expert's opinion affect the weight to be assigned that opinion rather than its admissibility and should be left for the jury's consideration." Primrose Operating Co. v. National Am. Ins. Co., 382 F.3d 546, 562 (5th Cir. 2004). Even if the replica counters suffered from defects, this would SVI-84676v1 -7- DEFS.' OPP. TO PLS.' MOT. NO. 4: TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Case No. 07-CV-1658 PJH (EDL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 not provide sufficient basis to exclude this portion of Reifer's opinion. To the extent that the counters suffer from any defects, this goes to the weight, not admissibility, of Reifer's opinion on Pinto's SLOC counts. See Clicks Billiards, Inc. v. Sixshooters, Inc., 251 F.3d 1252, 1262-63 (9th Cir. 2001) (holding that faults in methodology and calculations, and critiques of conclusions, go to weight and not admissibility of expert opinions); United States v. Elkins, 885 F.2d 775, 786 (11th Cir. 1989) (holding weakness of basis for opinion by qualified expert goes to weight not to credibility); Microfinancial, Inc. v. Premier Holidays Int'l, Inc., 385 F.3d 72, 81 (1st Cir. 2004) (holding that objection regarding scope of expert's investigation went to weight, not admissibility, of his testimony). Moreover, Reifer's methodology is based on his extensive experience and expertise in the COCOMO model. Any alleged defects in the replica counters do not affect the overall relevance or reliability of Reifer's methodology or opinion. See Stilwell v. Smith & Nephew, Inc., 482 F.3d 1187, 1192 (9th Cir. 2007) ("The test for reliability, however, `is not the correctness of the expert's conclusions but the soundness of his methodology.'") (citing Daubert v. Merrell Dow Pharms., Inc., 43 F.3d 1311, 1318 (9th Cir. 1995)). Additionally, as Reifer explained, the only way that the replicas could be defective is if Pinto did not follow his own parsing and counting rules in creating his counters. See Lanier Decl. ¶ 9, Ex. 9 (Reifer Tr.) at 321:25-322:5. Thus, two possible scenarios exist: either the counters used by Reifer replicate Pinto's counters, meaning Pinto's final estimates are indeed incorrect, or Pinto's counters failed to follow his own counting and parsing rules, also indicating that Pinto's final estimates are questionable. Either way, Reifer's analysis is based on data provided by Pinto, and Pinto's SLOC counts are disputed by that analysis. D. Plaintiffs' Case Support Is Inapposite. To support their assertion that Reifer's opinion regarding SLOC count should be excluded for having "no basis in fact," Plaintiffs cite a myriad of cases that have no factual similarities to this case. In Citric Acid, a party sought to admit an undisclosed document by arguing that it was discussed in an expert report. See In re Citric Acid Litig., 191 F.3d 1090, 1102 (9th Cir. 1999). The court held that the expert's mere assertion that a company's market share had stayed constant was not enough to establish this fact for the record. See id. In QR Spex, Inc. v. Motorola, Inc., SVI-84676v1 -8- DEFS.' OPP. TO PLS.' MOT. NO. 4: TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Case No. 07-CV-1658 PJH (EDL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 No. CV 03-6284-JFW (FMOx), 2004 WL 5642907, at *9 (C.D. Cal. Oct. 28, 2004), the "expert" relied on a news article for underlying "facts," which were thereafter directly contradicted. Andrews addresses an expert who based calculations on the wrong highway ramp, and the expert in Robinson bases his opinion on a clear factual error--namely that the plaintiff had never experienced sleeping problems before taking the drug at issue. See Andrews v. E.I. Du Pont De Nemours and Co., 447 F.3d 510, 513 (7th Cir. 2006); Robinson v. G.D. Searle & Co., 286 F. Supp. 2d 1216, 1221 (N.D. Cal. 2003). The remaining cases Plaintiffs cite are equally distinguishable. See DSU Med. Corp. v. JMS Co., Ltd., 296 F. Supp. 2d 1140, 1157-58 (N.D. Cal. 2003) (expert left out crucial factors in estimating hypothetical market that would have existed but for patent infringing product); Guidroz-Brault v. Missouri Pac. R.R. Co., 254 F.3d 825, 83031 (9th Cir. 2001) (no physical evidence supported expert's assertion that if train engineers had kept proper lookout, they would have seen that track had been tampered with). These cases involve experts who attempted to offer opinions with no proper factual bases. Although Plaintiffs may allege that Reifer's replica counters were flawed, Reifer's opinion, based on in-depth analysis of the software examined by Pinto, is far from lacking factual basis. V. REIFER IS QUALIFIED TO REBUT PINTO'S OPINION Plaintiffs' assertion that Reifer is unqualified to "opine on the accuracy of Pinto's code counters" is without foundation. D.I. 769 (Reifer Motion) at 7-8. As discussed above, Reifer has over 40 years of experience in the field of software, has managed the development of softwareintensive systems, helped calibrate COCOMO (the most widely used software estimation model in the world), co-authored a book on the COCOMO method, and led teams that developed code counters for use in software estimation. See Lanier Decl. ¶¶ 4, 9, Ex. 4 (Reifer Report) at 8, 96; Ex. 9 (Reifer Tr.) at 130: 18-19, 22. Even Pinto cites to a white paper authored by Reifer discussing software cost, quality, and productivity benchmarks. See Lanier Decl. ¶ 3, Ex. 3 (Pinto Report) at 24. Given his training and experience, Reifer is more than qualified to analyze Pinto's SLOC counts and his overall use of the COCOMO model. Cases offered by Plaintiffs to support their assertion that Reifer is unqualified are inapposite. See United States v. Chang, 207 F.3d 1169, 1173 (9th Cir. 2000) (expert had practical SVI-84676v1 -9- DEFS.' OPP. TO PLS.' MOT. NO. 4: TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Case No. 07-CV-1658 PJH (EDL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 experience in international finance but no training or experience detecting counterfeit securities, the only issue on which he sought to offer an opinion); Salinas v. Amteck of Ky., Inc., 682 F. Supp. 2d 1022, 1029-30 (N.D. Cal. 2010), (forensics and workplace safety expert attempted to opine on safety labels, an area completely outside realm of expertise); Redfoot v. B. F. Ascher & Co., No. C 05-2045 PJH, 2007 WL 1593239, at *10-11 (N.D. Cal. June 1, 2007) (expert's "interest" in vaccines did not qualify him to opine on pediatrics, neurology, or mercury). Reifer is more than "interested" in software analysis and has extensive training in the specific area about which he seeks to offer testimony. Thus, Plaintiffs' arguments regarding Reifer's qualifications fail. VI. REIFER'S OPINIONS ARE HIGHLY RELEVANT TO THE JURY'S ANALYSIS Plaintiffs also assert that Reifer's opinion will not "assist the trier of fact to understand the evidence or to determine a fact in issue." Fed. R. Evid. 702. This contention does not support Plaintiffs' argument that Reifer's opinion should be excluded but rather relies on Reifer's opinion being excludable. As a preliminary matter, and as discussed above, Plaintiffs have failed to successfully challenge Reifer's qualifications and have similarly failed to explain how the replica counters are defective or unreliable given that they were created using Pinto's own parameters. See Lanier Decl. ¶ 4, Ex. 4 (Reifer Report) at 19-20. Moreover, Reifer spent a great deal of time and effort attempting to validate Pinto's logical SLOC counts because they represent the core basis of Pinto's entire estimating methodology. See id. at 33; ¶ 3, Ex. 3 (Pinto Report) at 16 (asserting that he "use[d] the logical Source Lines of Code count as the foundation for estimating software size and ultimately deriving the total cost of development") (emphasis added). Reifer's evaluation of these counts and his conclusions that they are seriously flawed are thus highly relevant. Reifer's opinion is also imperative given the jury's presumed lack of knowledge about software valuation and inability to independently evaluate Pinto's methods. Thus, Plaintiffs' argument regarding relevance, which is both factually inaccurate and fatally illogical, should be rejected. VII. CONCLUSION Plaintiffs' Motion No. 4 has been rendered moot by this Court's order excluding saved development costs as recoverable damages. In addition to being moot, Plaintiffs' motion offers SVI-84676v1 - 10 - DEFS.' OPP. TO PLS.' MOT. NO. 4: TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Case No. 07-CV-1658 PJH (EDL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 no basis in fact or law to exclude Reifer's opinion. Reifer is qualified to opine on Pinto's critical SLOC count and his opinions are sufficiently supported in accordance with Rule 702. For these reasons, the Court should deny Plaintiffs' motion to exclude Reifer's testimony regarding SLOC counts in Oracle software. Dated: September 9, 2010 JONES DAY By: /s/ Tharan Gregory Lanier Tharan Gregory Lanier Counsel for Defendants SAP AG, SAP AMERICA, INC., and TOMORROWNOW, INC. SVI-84676v1 - 11 - DEFS.' OPP. TO PLS.' MOT. NO. 4: TO EXCLUDE EXPERT TESTIMONY OF DONALD REIFER Case No. 07-CV-1658 PJH (EDL)

Disclaimer: Justia Dockets & Filings provides public litigation records from the federal appellate and district courts. These filings and docket sheets should not be considered findings of fact or liability, nor do they necessarily reflect the view of Justia.


Why Is My Information Online?