Waymo LLC v. Uber Technologies, Inc. et al

Filing 2219


Download PDF
1 2 3 4 5 IN THE UNITED STATES DISTRICT COURT 6 FOR THE NORTHERN DISTRICT OF CALIFORNIA 7 8 9 WAYMO LLC, 11 For the Northern District of California United States District Court 10 12 13 14 15 16 17 No. C 17-00939 WHA Plaintiff, v. ORDER RE MOTIONS FOR PARTIAL SUMMARY JUDGMENT AND MOTION TO STRIKE ASSERTED TRADE SECRET NUMBER 96 UBER TECHNOLOGIES, INC.; OTTOMOTTO LLC; and OTTO TRUCKING LLC, Defendants. / (UNDER SEAL) INTRODUCTION In this action for trade secret misappropriation, all parties move for partial summary 18 judgment on select issues. Plaintiff’s motion is GRANTED IN PART and DENIED IN PART. 19 Defendants’ motion is also GRANTED IN PART and DENIED IN PART. Defendants’ motions to 20 strike and for summary judgment as to asserted trade secret number 96 are GRANTED. 21 STATEMENT 22 The factual background of this action has been detailed in a prior order. In brief, 23 plaintiff Waymo LLC sued defendants Uber Technologies, Inc., Ottomotto LLC (collectively, 24 “Uber”), and Otto Trucking LLC for alleged misappropriation of trade secrets concerning Light 25 Detection and Ranging (LiDAR) technology that helps self-driving cars “see” their 26 surroundings. The centerpiece of Waymo’s case is its evidence that its former star engineer, 27 non-party Anthony Levandowski, downloaded over 14,000 confidential files from Waymo 28 immediately before leaving his employment there to start Ottomotto and Otto Trucking 1 (collectively, “Otto”). Uber then acquired Ottomotto and hired Levandowski as the head of its 2 self-driving car efforts (Dkt. No. 426).1 3 By agreement of both sides, this action was set for a jury trial on October 10. After the 4 belated production of non-party Stroz Friedberg’s due diligence report on the Otto acquisition, 5 however, Waymo successfully moved to continue the trial date to December 4 (Dkt. No. 1954). 6 Waymo moves for partial summary judgment on certain affirmative defenses (Dkt. No. 7 1418). Uber moves for partial summary judgment that Waymo’s asserted trade secret number 8 nine is not a trade secret, and Otto Trucking moves for summary judgment that it has not 9 misappropriated any alleged Waymo trade secret (Dkt. No. 1423). All defendants also move for partial summary judgment on Waymo’s asserted trade secret number 96 on multiple grounds, 11 For the Northern District of California United States District Court 10 including that it was inadequately disclosed (Dkt. Nos. 1514, 1518). The latter motion also 12 represents the final piece of defendants’ prior motion to strike certain trade secret claims (Dkt. 13 Nos. 1108, 1129), which has already been denied as to all challenged claims except asserted 14 trade secret number 96 (see Dkt. No. 1260 at 108:5–15, 110:13–111:3, 133:24–134:13).2 15 This order follows full briefing and oral argument. 16 ANALYSIS 17 1. 18 Summary judgment is appropriate when there is no genuine dispute of material fact and 19 the moving party is entitled to judgment as a matter of law. A genuine dispute of material fact 20 is one that “might affect the outcome of the suit under the governing law.” Anderson v. Liberty 21 Lobby, Inc., 477 U.S. 242, 247–48 (1986). In deciding a motion for summary judgment, the 22 court must believe the non-movant’s evidence and draw all justifiable inferences in their favor. 23 Id. at 255. “The mere existence of a scintilla of evidence” or “some metaphysical doubt as to 24 the material facts” in the non-movant’s favor, however, will not suffice. Id. at 252, 260–61 25 (quoting Matsushita Elec. Indus. Co., Ltd. v. Zenith Radio Corp., 475 U.S. 574, 586 (1986)). LEGAL STANDARD. 26 27 28 1 For the benefit of the court of appeals and with apologies to the public, most record citations herein are to the unredacted versions of the documents cited. 2 Defendants also moved for summary judgment of noninfringement of Waymo’s United States Patent No. 9,368,936, but Waymo’s dismissal of its patent claim moots that part of the motion (Dkt. No. 1593). 2 1 “Where the record taken as a whole could not lead a rational trier of fact to find for the non- 2 moving party, there is no genuine issue for trial.” Matsushita, 475 U.S. at 587 (quotation and 3 citation omitted). 4 2. 5 Waymo initially moved for summary judgment on eleven of defendants’ affirmative 6 defenses (Dkt. No. 1418). In their opposition brief, defendants waived all but two (Dkt. No. 7 1524-3 at 2). Only defendants’ failure-to-mitigate defense and Otto Trucking’s unclean hands 8 defense remain in dispute on this motion. 9 A. Failure to Mitigate. Waymo contends defendants have no evidence that it failed to mitigate its damages 11 For the Northern District of California United States District Court 10 WAYMO’S MOTION RE AFFIRMATIVE DEFENSES. because (1) Waymo sued “within weeks of obtaining evidence of . . . trade secret 12 misappropriation,” i.e., the email dated December 13, 2016, from LiDAR component vendor 13 Gorilla Circuits (see Dkt. No. 426 at 5); (2) any “delay” in bringing suit was de minimis; and (3) 14 “[t]he length of the head start created by Uber’s trade secret misappropriation does not depend 15 on the timing of Waymo’s suit” (Dkt. No. 1418 at 8–9). 16 Defendants note that Waymo “began investigating a compromise of its alleged trade 17 secrets” months before receiving the Gorilla email, involved outside counsel in said 18 investigation by July and August 2016, “accelerated its investigation after learning about Uber’s 19 imminent acquisition of Ottomotto,” and internally discussed when to reach out to Uber about 20 its “potential concerns” (Dkt. No. 1524-3 at 16). Far from showing failure to mitigate, these 21 facts actually indicate that Waymo acted diligently by actively investigating its “potential 22 concerns” and “accelerat[ing] its investigation after learning about Uber’s imminent acquisition 23 of Ottomotto.” These facts do nothing to rebut Waymo’s point that it filed suit promptly after 24 receiving the Gorilla email that appeared to contain first evidence of actual trade secret 25 misappropriation by defendants. 26 As noted in the prior order granting provisional relief, defendants essentially “contend 27 that Waymo lacks sufficient evidence of use but simultaneously criticize Waymo for not suing 28 before it had any evidence of use” (Dkt. No. 426 at 19–20). This dissonance remains on full 3 1 display in defendants’ opposition to the instant motion. For example, defendants specifically 2 argue that “Waymo has failed to meet its duty to mitigate by delaying in seeking injunctive 3 relief as to Otto Trucking” and in the very next paragraph claim “Waymo has no evidence that 4 Otto Trucking is involved in the development of LiDAR or that it has used the purported trade 5 secrets” (Dkt. No. 1524-3 at 15). Were this really true, it would be a mystery how or why 6 Waymo could or would have sought earlier relief. 7 Insinuations about Waymo’s ulterior motives, including that “Waymo did not care about 8 Mr. Levandowski or his companies” and “made the strategic choice to delay in suing Otto 9 Trucking until it could cobble together a way to sue Uber,” permeate the defense brief (see id. at 16). Such facts, if true, may bear on the merits of Waymo’s trade secret misappropriation 11 For the Northern District of California United States District Court 10 claims insofar as they may speak to the value of the asserted trade secrets. But they do not 12 further indicate, as defendants suggest, that Waymo had some prior opportunity to mitigate its 13 damages and squandered it by filing suit later rather than sooner. This order agrees with 14 Waymo that, taken as a whole, the record reflects only de minimis, if any, delay by Waymo in 15 bringing this action. 16 Defendants have failed to show a genuine dispute of material fact on their failure-to- 17 mitigate defense because this record, taken as a whole, could not lead a rational trier of fact to 18 find for defendants on said defense as to any claim to be tried by a jury. Waymo’s motion for 19 summary judgment on this defense is GRANTED without prejudice to any future argument that, 20 as an equitable matter, Waymo’s timeliness (or lack thereof) in suing may somehow diminish 21 its entitlement to equitable relief. 22 B. Unclean Hands. 23 Unclean hands “closes the doors of a court of equity to one tainted with inequitableness 24 or bad faith relative to the matter in which he seeks relief.” Jarrow Formulas, Inc. v. Nutrition 25 Now, Inc., 304 F.3d 829, 841 (9th Cir. 2002) (quoting Precision Instrument Mfg. Co. v. Auto. 26 Maint. Mach. Co., 324 U.S. 806, 814 (1945)). As our court of appeals has said, “troubling” 27 actions and “hands [that] are not as ‘clean as snow’” do not necessarily rise to the level of 28 unclean hands. Id. at 842. Rather, “the defense of illegality or unclean hands is recognized 4 1 only rarely, when the plaintiff’s transgression is of serious proportions and relates directly to the 2 subject matter of the . . . action” — for example, when a plaintiff “misused the process of the 3 courts by falsifying a court order or evidence.” Dream Games of Ariz., Inc. v. PC Onsite, 561 4 F.3d 983, 990–91 (9th Cir. 2009) (quotation omitted). “The application or rejection of the clean 5 hands doctrine in a given case is equitable in nature and within the discretion of the trial court.” 6 Wash. Capitols Basketball Club, Inc. v. Barry, 419 F.2d 472, 478 (9th Cir. 1969). 7 Here, Waymo contends Otto Trucking has no evidence that “Waymo engaged in 8 misconduct directly related to the claims that Waymo is asserting” (Dkt. No. 1418 at 9). In 9 response, Otto Trucking claims “Waymo has ‘sought at every turn to create the false impression’ that its forensic investigation proved that Mr. Levandowski had stolen valuable 11 For the Northern District of California United States District Court 10 trade secrets from Waymo by downloading 14,000 files — the entire repository — from 12 Waymo’s SVN server,” citing evidence tending to show weaknesses in Waymo’s case (Dkt. No. 13 1524-3 at 13–14). For example, Otto Trucking points to Sasha Zbrozek, a Waymo engineer, 14 who told Waymo’s lawyers the downloaded files were “low-value” and added that “it was 15 regular practice for engineers” to download the entire SVN repository (ibid.), revelations that 16 had to be pried out of Waymo by Court order. Given this stonewalling, Otto Trucking argues, 17 Waymo’s “presentation of its case contrary to facts” constitutes unclean hands. 18 Since this equitable defense will have meaning only if Waymo ultimately prevails on the 19 merits, there is no need to reach it now. Indeed, it would be premature to decide the issue at 20 this stage, since litigation remains ongoing and further evidence of Waymo’s misconduct may 21 yet come to light. It is conceivable that Waymo might win at trial and then seek some equitable 22 relief, whereupon its own conduct in this litigation might loom large as an inequity and fairness 23 may (or may not) require moderation or even denial of any relief. Moreover, inasmuch as 24 Waymo’s motion relates specifically to a defense asserted by Otto Trucking, it is mooted by this 25 order, which grants summary judgment of no liability in favor of Otto Trucking. Thus, both 26 because it is premature and because it is moot, Waymo’s motion for summary judgment on the 27 defense of unclean hands is DENIED. 28 5 1 3. 2 Both the California Uniform Trade Secrets Act and the federal Defend Trade Secrets 3 UBER’S MOTION RE ASSERTED TRADE SECRET NUMBER NINE. Act define “trade secret” as information that: 4 (1) Derives independent economic value, actual or potential, from not being generally known to the public or to other persons who can obtain economic value from its disclosure or use; and 5 6 (2) Is the subject of efforts that are reasonable under the circumstances to maintain its secrecy. 7 CAL. CIV. CODE § 3426.1; see also 18 U.S.C. 1839. Here, Waymo’s asserted trade secret 8 number nine claims (Dkt. No. 25-7 at 8): 9 [A] LiDAR system comprising a transmit block comprising a plurality of laser diodes, wherein said transmit block is comprised of a plurality of PCBs [printed circuit boards], and wherein each laser diode is configured to emit beams and wherein a cylindrical lens is positioned in front of each laser diode, said cylindrical lens being configured to 11 For the Northern District of California United States District Court 10 12 13 14 . Defendants’ motion relies on a compilation of evidence from various sources to argue 15 that various subparts of Waymo’s asserted trade secret number nine are generally known in the 16 field. Defendants point out that Waymo’s own United States Patent No. 8,836,922 (Dkt. No. 17 23-1) disclosed at all relevant times “a LiDAR system comprising a transmit block comprising a 18 plurality of laser diodes, wherein said transmit block is comprised of a plurality of PCBs” 19 (see Dkt. No. 1419-4 at 10). They also quote the deposition testimony of Waymo’s engineers 20 for the proposition that using FAC lenses to 21 in the field, as is the general notion that the lens in front of a light source can be is a well-known concept (see id. at 10–12). Finally, they point to Velodyne’s 22 23 commercially-available LiDAR system as an example of another LiDAR system that uses FAC 24 lenses to 25 (see id. at 13). None of the foregoing evidence conclusively shows that Waymo’s entire asserted trade 26 secret number nine was generally known to the public or in the field. The ’922 patent did not 27 disclose a “cylindrical lens being configured to .” 28 6 1 The same Waymo engineers quoted by defendants also testified to the effect that the strategy of 2 using FAC lenses to 3 known in the field (see Dkt. Nos. 1526-10 at 230:16–231:19; 1526-25 at 217:16–218:20). And 4 while Velodyne has used its FAC lenses to 5 Dkt. No. 1526-28 at 68:7–69:6), Waymo contends this in a LiDAR system is not generally in its LiDAR systems (see 6 7 8 was used to , not to in the first instance (see Dkt. No. 1526-4 at 16–17). For the latter purpose of , it seems, Velodyne relies on the strategic 19). Additionally, Velodyne itself insists — albeit in general terms — that the relevant details 11 For the Northern District of California (see, e.g., Dkt. No. 1526-27 at 10 United States District Court 9 of its LiDAR system constitute trade secret information (Dkt. No. 1455). 12 In their reply brief, defendants seize upon Waymo’s explanation that the design claimed 13 in asserted trade secret number nine “stems in part from the fact that Waymo, unlike Velodyne, 14 positions multiple laser diodes on a single PCB, with multiple PCBs stacked in parallel” (see 15 Dkt. Nos. 1526-4 at 10, 1636-4 at 2 (emphasis added)). (Velodyne uses 16 .) 17 Defendants accuse Waymo of “acknowledg[ing] that it cannot bar its former engineers from 18 using multiple diodes on multiple PCBs” because of its ’922 patent, yet “seek[ing] to bar them 19 from including in such a design the basic optical concept of 20 ” (Dkt. No. 1636-4 at 2). But this reasoning merely begs the 21 question, i.e., whether Waymo’s use of FAC lenses to 22 its patented LiDAR system is actually a “basic optical concept.” 23 24 within True, substantial parts of asserted trade secret number nine are encompassed by the ’922 patent. But it remains a question of fact whether or not Waymo’s use of FAC lenses to 25 was generally known to the public or in the field. For 26 example, both sides agree that Velodyne’s LiDAR system meets the definition of a LiDAR 27 system “comprising a transmit block comprising a plurality of laser diodes, wherein said 28 transmit block is comprised of a plurality of PCBs” (see Dkt. Nos. 1526-4 at 7, 1636-4 at 4 n.1). 7 1 Yet, as explained above, Velodyne’s LiDAR system is significantly different from Waymo’s, 2 including in how each system uses its FAC lenses to 3 . In short, it remains possible that a LiDAR design might use that part of 4 asserted trade secret nine that is encompassed by the ’922 patent but nevertheless not arrive at 5 Waymo’s particular solution of using FAC lenses to 6 7 8 9 . Defendants also brush off the question of “whether Velodyne’s FAC design is public or confidential,” urging instead that “the point is that the general concept of using an FAC to is known in the LiDAR field, and follows fundamental principles of optics” (Dkt. No. 1636-4 at 4). This argument dismisses the evidence without responding to it. The point is that whether or not Velodyne considers its design to be trade secret information may bear on 11 For the Northern District of California United States District Court 10 whether or not Waymo’s asserted trade secret number nine is generally known to the public or 12 in the field, as defendants contend. 13 In summary, defendants’ arguments as to asserted trade secret number nine — including 14 whether or not Waymo’s asserted trade secret number nine encompasses no more than ordinary 15 concepts and skills inherent in the head of every LiDAR engineer, the degree to which 16 Waymo’s ’922 patent and Velodyne’s commercially-available LiDAR system disclose asserted 17 trade secret number nine, and whether or not Velodyne’s designs constitute sufficient evidence 18 that asserted trade secret number nine is generally known in the field — all present genuine 19 disputes of material fact. They are accordingly reserved for the jury, and defendants’ motion 20 for summary judgment that asserted trade secret number nine is not a trade secret is DENIED. 21 22 4. OTTO TRUCKING’S MOTION RE MISAPPROPRIATION. Otto Trucking has never been involved in the development or use of LiDAR systems, 23 and exists simply as a holding company to own trucks equipped with Velodyne’s LiDAR 24 system (Dkt. No. 1419-4 at 16–20). Unlike Ottomotto, it has not been acquired by Uber and 25 remains a separate company. Waymo nevertheless seeks to preserve its trade secret 26 misappropriation claims against Otto Trucking by lumping it together with Uber and 27 Levandowski under various unpersuasive theories, which this order now discusses. 28 8 1 A recurring problem in Waymo’s briefing must be addressed at the outset. Waymo 2 repeatedly suggests that Otto Trucking should be deemed automatically liable for trade secret 3 misappropriation by Levandowski. To give just a few non-exhaustive examples, Waymo 4 purports to identify “Otto Trucking’s role in the misappropriation” but goes on to recite only 5 that “a central figure in the misappropriation of Waymo’s trade secrets — Anthony Levandowski 6 — is Otto Trucking’s Executive Chairman, co-Managing Member, and largest stockholder” 7 (Dkt. No. 1526-4 at 19 (emphasis added)). Waymo also contends its claims “encompass Otto 8 Trucking’s acquisition — by way of Mr. Levandowski’s theft — of the asserted trade secrets 9 (id. at 21). And even Waymo’s supplemental offer of proof as to Otto Trucking relies primarily on evidence of Levandowski’s misappropriation (see Dkt. No. 2055-3 at 2–7). In short, 11 For the Northern District of California United States District Court 10 Waymo’s case against Otto Trucking rests largely on — in Waymo’s words — its “very 12 existence as Mr. Levandowski’s company” (Dkt. No. 1526-4 at 23). 13 To be very clear, whether or not Otto Trucking and Levandowski are, as a practical 14 matter, similarly situated or even interchangeable in some sense, they are not interchangeable 15 for purposes of this particular litigation. In a special way, this is a key limitation of Waymo’s 16 own making. To avoid having to arbitrate this case, Waymo has repeatedly represented to both 17 this Court and to the Federal Circuit that it has not sued Levandowski herein. By doing so, 18 Waymo has successfully avoided arbitration and kept its claims in a public court. It must keep 19 its word. Having made and benefitted from its strategic choice to not name Levandowski as a 20 defendant, Waymo may not renege and suggest that Otto Trucking — or any other defendant — 21 is somehow a stand-in for Levandowski, or that misappropriation by Levandowski is somehow 22 automatically transmogrified into misappropriation by Otto Trucking — or any other defendant 23 — such that Waymo need not separately prove the latter. This order evaluates Waymo’s 24 arguments on the instant motion through the lens of this foundational principle. 25 In its original opposition brief, Waymo contends Otto Trucking is vicariously liable for 26 Levandowski’s misappropriation because it failed to take adverse action against him after he 27 formed it. Moreover, Waymo claims, Levandowski’s possession of misappropriated trade 28 secrets will increase Otto Trucking’s value as an acquisition target for Uber. Thus, Waymo 9 1 reasons, Otto Trucking “accepted and retained” the benefit of Levandowski’s misappropriation, 2 thereby ratifying his continued possession of Waymo’s trade secrets (see Dkt. No. 1526-4 at 3 20–25). But it is undisputed that Otto Trucking has no practical ability to take adverse action 4 against Levandowski. Waymo’s argument, if accepted, would mean that Otto Trucking must 5 always be deemed to have “ratified” whatever Levandowski has done, and it would be of no 6 moment for vicarious liability purposes that Otto Trucking has no meaningful choice in the 7 matter. For all intents and purposes, Waymo’s proposed theory of vicarious liability is 8 indistinguishable from automatic equivalence of Levandowski and Otto Trucking. Under either 9 approach, Waymo simply wants to be able to hold Otto Trucking liable by proving only that Levandowski misappropriated its asserted trade secrets. As explained, this will not be allowed. 11 For the Northern District of California United States District Court 10 Otto Trucking’s potential future with Uber does not compel a different result. Insofar as 12 Waymo is relying on Otto Trucking’s possible future actions, those actions are not properly the 13 subject of the operative complaint in this litigation (although Waymo remains free to file 14 another lawsuit in the future if defendants’ future actions give rise to additional claims). With 15 respect to the trade secret misappropriation claims in this litigation, the plain text of CUTSA 16 and DTSA provide for liability only through acquisition, use, or disclosure of trade secret 17 information. Waymo cites no authority for its suggestion that Otto Trucking can be liable now 18 merely because it might benefit in the future from the misappropriation of other individuals and 19 entities. Indeed, this suggestion puts the cart before the horse. The question of whether a 20 defendant benefitted from trade secret misappropriation would typically be one of damages, 21 reached only after the misappropriation itself has been established in the first instance. 22 Counting the mere possibility of such benefit as proof of misappropriation would turn this 23 sequence on its head and drastically expand the scope of liability under CUTSA and DTSA. 24 Waymo correctly notes that any effective injunctive relief against Uber may have to 25 include Otto Trucking as well, but incorrectly asserts that this practical fact bars summary 26 judgment as to liability (see Dkt. No. 1526-4 at 23). The effectiveness of injunctive relief is a 27 matter of remedies, not of liability. 28 10 1 Waymo’s supplemental opposition brief similarly fails to raise a genuine issue of substantially overlaps with Waymo’s prior arguments and again relies heavily on evidence of 4 Levandowski’s misappropriation to state a case against Otto Trucking (see Dkt. No. 2055-3 at 5 2–9). It also summarizes evidence that Don Burnette, another co-founder and shareholder of 6 Otto Trucking, misappropriated trade secrets in motion planning software that Waymo had 7 sought to add as new claims late in the case (see id. at 9–10, 16). Those proposed new claims, 8 however, will not be added to our upcoming trial this year, so they do not justify preserving 9 Waymo’s current claims against Otto Trucking (although they may justify adding Otto 10 Trucking back in for a second trial later on if Waymo is eventually allowed to add in its 11 For the Northern District of California material fact as to Otto Trucking’s liability for the claims at issue in this litigation. It 3 United States District Court 2 proposed new software misappropriation claims) (see Dkt. No. 2129). 12 In its supplemental opposition brief, Waymo reiterates in one short paragraph that 13 “Levandowski possessed and retained accessed [sic] to Waymo trade secret information while 14 Mr. Levandowski was working at Otto Trucking,” “that trade secret information was used in 15 LiDAR development efforts by Mr. Levandowski and other individuals who were 16 simultaneously employed by Uber and Otto Trucking,” and “that work was done for the benefit 17 of accelerating development of autonomous driving in both cars and trucks” (Dkt. No. 2055-3 at 18 10–11). Waymo styles this as a showing of direct liability but provides no explanation or 19 authority for why this evidence against other individuals and entities adds up to a case against 20 Otto Trucking on Waymo’s current claims. This evidence fails to raise a genuine dispute of 21 material fact as to Otto Trucking’s liability. 22 In its supplemental opposition brief, Waymo also expands its vicarious liability and 23 ratification theories to include both Levandowski and Lior Ron, another co-founder and 24 shareholder of Otto Trucking. As to Levandowski, these theories are rejected for the same 25 reasons already stated. As to Ron, Waymo’s supplemental offer of proof cites the due diligence 26 report as evidence that Ron retained possession of some “confidential Waymo data.” 27 Significantly, although Waymo describes this data in some detail — and states in later argument 28 that Ron possessed its trade secrets — it never points to specific evidence that the data retained 11 1 by Ron actually included any of the asserted trade secrets at issue in this case (see id. at 4). 2 Waymo also summarizes evidence that both Levandowski and Ron attempted to cover their 3 tracks and otherwise acted suspiciously during the due diligence process, but proffers no 4 evidence that any of this shady behavior related to or indicated trade secret misappropriation on 5 behalf of Otto Trucking (see id. at 4–5). In light of these deficiencies, Otto Trucking’s 6 “refus[al] to take any remedial actions” against Ron is of little significance (see id. at 13–14). 7 In short, Waymo’s supplemental evidence regarding Ron’s actions fails to improve on its 8 vicarious liability and ratification theories (see id. at 11–16). 9 Finally, in another short paragraph of its supplemental opposition brief Waymo argues that all defendants are jointly and severally liable “[r]egardless of the specific theory of liability 11 For the Northern District of California United States District Court 10 the jury may use to find Otto Trucking liable” because many of the same individuals involved 12 in Uber’s LiDAR-development efforts are also Otto Trucking employees, and Otto Trucking 13 stands to benefit in the future from Uber’s development efforts. Again, Waymo provides no 14 explanation or authority for why these assertions add up to a case against Otto Trucking on 15 Waymo’s current claims. Its blip about joint and several liability therefore fails to raise a 16 genuine dispute of material fact as to Otto Trucking’s liability. 17 In short, the record taken as a whole could not lead a rational jury to find Otto Trucking 18 liable on Waymo’s current trade secret misappropriation claims. Waymo’s strongest evidence 19 on misappropriation is about Levandowski, not about Otto Trucking, and as a result of its own 20 litigation strategy, Waymo cannot treat the two as fungible targets. Nor can it circumvent this 21 limitation of its own making by restyling its theory as one of vicarious liability, ratification, or 22 joint and several liability. With this foundational principle in mind, the remainder of Waymo’s 23 evidence against Otto Trucking fails to raise a genuine dispute of material fact that Otto 24 Trucking misappropriated any asserted trade secrets. Summary judgment is therefore proper as 25 to Waymo’s current claims for trade secret misappropriation against Otto Trucking. 26 That said, it remains a practical concern that, if Waymo prevails against Uber at trial, 27 any effective injunctive relief may have to include Otto Trucking as well. The Court will 28 therefore retain jurisdiction over Otto Trucking for the limited purpose of possibly facilitating 12 1 such relief. To the foregoing extent, Otto Trucking’s motion for summary judgment is 2 GRANTED. Otto Trucking will not be a part of the upcoming trial this year. DEFENDANTS’ MOTIONS RE ASSERTED TRADE SECRET NUMBER 96. 3 5. 4 Pursuant to its case management responsibility as well as Section 2019.210 of the 5 California Code of Civil Procedure, which requires trade secrets to be identified “with 6 reasonable particularity” prior to discovery thereon, the Court required Waymo to disclose up 7 front its asserted trade secrets. (Otherwise, trade-secret plaintiffs are liable to get into the 8 accused files and conveniently assert that contents therein were trade secrets stolen from them.) 9 Waymo’s asserted trade secret number 96 claimed “the G[B]r3 PCB Transmit Board C design schematics and layouts contained in folder ‘ 11 For the Northern District of California United States District Court 10 Waymo’s disclosure added that, “[f]or example, these details [in the schematics] include unique 12 and unknown design characteristics such as the 13 14 15 .’” The rest of , the selection and layout of individual electrical components, and the required manufacturing tolerances” (Dkt. Nos. 25-7 at 55). Prior to the instant motion for summary judgment, defendants had moved to strike 16 asserted trade secret number 96 as overbroad because “Waymo fails to identify with 17 particularity what specific aspects of the hundreds of components disclosed in the ‘detailed 18 engineering schematics’ it contends to be a trade secret” (Dkt. Nos. 1107-4 at 4, 1129). In 19 response, Waymo stated, “It is not clear how a trade secret disclosure that identifies one specific 20 PCB in one specific LIDAR design could be defined with any more ‘reasonable particularity’” 21 (Dkt. No. 1159-4 at 7). In other words, Waymo attempted to avoid the overbreadth problem by 22 insisting that its asserted trade secret claimed only the specific design of one specific PCB (see 23 also Dkt. No. 1491 at 32:3–4). 24 The judge personally examined the PCB schematic in question during a closed 25 demonstration. It became readily apparent that the schematic in question included hundreds of 26 components and specifications with no clue as to what part thereof might be considered trade 27 secret information. Worse, Waymo argued that its asserted trade secret number 96 28 encompassed not only the information contained in the schematic itself but also strategies or 13 1 concepts “reflected” in the schematic (see, e.g., id. at 54:18–55:6). It also comprehended, 2 Waymo said, “unique and unknown design characteristics such as the 3 , the selection and layout of individual electrical components, and the 4 required manufacturing tolerances” (Dkt. Nos. 25-7 at 55, 1260 at 113:10–21 (emphasis 5 added)). Such wide-ranging expanders compounded the misery in trying to place arms around 6 the breadth of asserted trade secret number 96. Modular Sputtering, Inc. v. Superior Court, 132 Cal. App. 4th 826, 836 (2005) (“The degree of 9 ‘particularity’ that is ‘reasonable’ will differ, depending on the alleged trade secrets at issue in 10 each case.”). The vastness of Waymo’s disclosure was most unfair, for it placed defendants in 11 For the Northern District of California This disclosure came nowhere near the required reasonable particularity. See Advanced 8 United States District Court 7 the untenable position of having to prove that hundreds of subparts of the design schematic — 12 and innumerable other strategies and concepts arguably “reflected” therein, as Waymo put it — 13 either did not qualify as trade secret information or were not used in defendants’ technology. 14 See ibid. (“Where, as here, the alleged trade secrets consist of incremental variations on, or 15 advances in the state of the art in a highly specialized technical field, a more exacting level of 16 particularity may be required to distinguish the alleged trade secrets from matters already 17 known to persons skilled in that field.”) 18 Waymo’s broad claim might have been forgiven had it accused defendants of 19 misappropriating the entire schematic lock, stock, and barrel. Its eventual accusation, however, 20 was exactly the opposite — seizing upon a seemingly-random subpart of a subpart and claiming 21 it had been misappropriated. When Waymo submitted its offer of proof as to defendants’ 22 alleged misappropriation of trade secrets, it focused on one tiny portion of the schematic and 23 claimed that defendants copied it — but even that collapsed upon examination. Out of 24 PCBs in Waymo’s LiDAR system, only one was even claimed to have been copied in any 25 respect into Uber’s LiDAR system. The only aspect of the PCB design supposedly copied was 26 the 27 28 , nothing else. It soon turned out that even the supposed equivalence of the was untrue, as will be explained in a moment. But the overbreadth of the disclosure itself was further 14 1 amplified when read in the context of asserted trade secret number one, so the reader will please 2 bear with us as we revisit that history. 3 Waymo’s narrowed offer of proof as to asserted trade secret number 96 in large part 4 turned out to be merely a more detailed rendition of the evidence it had offered with respect to 5 asserted trade secret number one — which claimed the concept of 6 printed circuit board in a LiDAR system — at the provisional relief stage (compare Dkt. No. 7 1371-4 at 17–25 with Dkt. Nos. 25-61 at 8–11, 245-3 at 3–4). Significantly, the provisional 8 relief order had already specifically found that Waymo’s asserted trade secret number one was 9 “nothing more than Optics 101,” deleting that supposed “trade secret” from the case long before the parties began to argue over asserted trade secret number 96 (see Dkt No. 426 at 16–17). 11 For the Northern District of California United States District Court 10 on a Defendants filed a supplemental brief pointing out that Waymo was essentially 12 attempting to circle back to claiming a “ 13 confirmed that it was attempting to reframe its claim over an entire printed circuit board 14 schematic to a claim over “its implementation of its 15 16 ” concept. Waymo’s response and the resulting .” Waymo attempted to distinguish number 96 from number one by suggesting that the former claimed its specific 17 configuration rather than the general concept of (see, e.g., Dkt. No. 1449-4 at 3–4). Moreover, Waymo argued that its 18 disclosure was adequate as to asserted trade secret number 96 because it “identified this exact 19 PCB schematic, and identified for Defendants that one unique aspect of the schematic reflected 20 ‘the 21 ’” (id. at 4 (emphasis added)). This raised problems. First, as explained, it effectively resurrected Waymo’s asserted 22 trade secret number one, previously rejected as nothing more than “Optics 101.” Second, it cast 23 into sharp relief the unreasonableness of Waymo’s initial disclosure. Although Waymo did list 24 the 25 asserted trade secret, it would have been wholly unreasonable to expect defendants to discern, 26 from the complex schematic and from Waymo’s nonexclusive list of examples — not to 27 mention its expansion of the asserted trade secret to include “reflected” strategies and concepts 28 — that Waymo’s true focus in asserted trade secret number 96 all along was only on its as one example of many “characteristics” encompassed by its 15 1 or strategy. Third, and in a similar vein, Waymo’s retreat back to its 2 asserted trade secret number one actually exacerbated the unreasonableness of its disclosure. 3 Even assuming for the sake of argument that defendants were obliged to guess what specific 4 component of the claimed schematic would emerge as the issue, the very last thing they could 5 be expected to guess would be something already deleted from the case. 6 To repeat, Waymo’s disclosure might still perhaps have been reasonably adequate had 7 Waymo presented evidence that defendants misappropriated its specific implementation (see 8 Dkt. No. 1159-4 at 6). But the PCB in question was not copied in its entirety, not even close. 9 copied. The latter proved to be conclusively different, and no jury could reasonably find otherwise. 11 For the Northern District of California United States District Court 10 Nor, as it turns out, was the For starters, Waymo’s PCB while Uber’s (see, 12 e.g., Dkt. No. 1357-3 at 24). The Court nevertheless asked both sides to provide detailed 13 comparisons of their 14 PCBs in both LiDAR systems) (see Dkt. Nos. 1408, 1416). In response, Waymo submitted a 15 declaration by its hired expert Lambertus Hesselink (Dkt. No. 1456-3). In a series of parabolic 16 graphs, Dr. Hesselink compared the Petzval surfaces of Waymo and Uber’s respective transmit 17 lenses, even scaling for the discrepancy between their focal lengths — 120 mm versus 150 mm 18 — to maximize the overlap between the resulting parabolic graphs (id. ¶¶ 13–23).3 19 strategies for both PCBs in question (and for all Two identical lenses, however, will always have identical Petzval surfaces — always. 20 An overlay of Petzval surfaces proves nothing in this case. Waymo insists that “Prof. 21 Hesselink’s analysis shows that the lens material and lens shape do not meaningfully contribute 22 to the shape of the Petzval surface in the case of the two transmit boards under comparison,” so 23 “his analysis shows that the transmit lenses [are] substantially similar in terms of their overall 24 optical properties” (see, e.g., Dkt. No. 1634-4 at 18–19). But asserted trade secret number 96, 25 as broad as it is, has nothing to do with the “optical properties” of any transmit lens. Rather, 26 what supposedly matters is the along the Petzval surfaces, 27 28 3 In an optics system, the image corresponding to a linear object focused by a lens will lie on a curved surface also called the Petzval surface. 16 1 which . Even with Dr. 2 Hesselink’s overlay of the scaled Petzval surfaces, however, it is apparent from his own graphs 3 that the 4 do not overlap at all (see Dkt. No. 1456-3 ¶ 23).4 5 where Waymo and Uber’s fall on the overlaid Petzval surface Nevertheless, Dr. Hesslink concluded (id. ¶ 24 (emphasis added)): 6 This comparison allows evaluation of whether the from the Uber system fall on the Waymo parabola. They do, as reflected by the fact that the curves are quite similar. The are similar as well, but not identical, reflecting that Uber took a similar approach to Waymo in designing their . 7 8 9 In other words, Dr. Hesselink manufactured a visually-similar match between the Petzval 10 visually-similar match in Petzval surfaces somehow translated to similar For the Northern District of California United States District Court surfaces of Waymo and Uber’s respective transmit lenses, and then tried to suggest that this 11 . 12 And despite acknowledging that the are “not identical” even on the scaled and 13 overlaid curves, he parroted, with no additional analysis, Waymo’s unsupported litigation 14 position that Uber “took a similar approach to Waymo in designing their .” 15 To repeat, two identical lenses will always have identical Petzval surfaces — always. 16 That proves nothing for Waymo. What matters with respect to asserted trade secret number 96 17 is where on the Petzval surfaces the in Waymo and Uber’s respective LiDAR systems 18 are placed. Of course, all must be placed somewhere on the Petzval surface — 19 otherwise they will be out of focus. But, as stated, Uber’s accused PCB while 20 Waymo’s claimed PCB (see, e.g., Dkt. No. 1357-3 at 24). Moreover, the 21 of Uber’s did not overlap with any on Waymo’s claimed PCB, and the 22 for each side produced dramatically different . 23 Dr. Hesselink’s emphasis on the similarity of the Petzval surfaces is a trick — smoke 24 and mirrors. His inexplicable leap from that similarity to the conclusion that Uber “took a 25 similar approach to Waymo in designing their ” — despite all the aforementioned 26 differences between the two sides’ approaches — renders his opinion unreliable, 27 28 4 similar This is consistent with other responses by both sides showing that their LiDAR systems share neither nor a similar strategy (see, e.g., Dkt. Nos. 1458-5, 1461-5). 17 1 and thus inadmissible, under Federal Rule of Evidence 702 and Daubert v. Merrell Dow 2 Pharmaceuticals, Inc., 509 U.S. 579, 597 (1993). See Gen. Elec. Co. v. Joiner, 522 U.S. 136, 3 146 (1997) (a district court may exclude an expert opinion where “there is simply too great an 4 analytical gap between the data and the opinion proffered”). Moreover, his suggestion that the 5 overlap of the Petzval surfaces — which, of course, are visually similar — somehow constitutes 6 evidence that Uber misappropriated Waymo’s 7 misleading to the jury, more prejudicial than probative, and thus inadmissible under FRE 403 as 8 well. Dr. Hesselink’s opinion as to asserted trade secret number 96 is therefore STRICKEN. 9 strategy would be grossly In short, summary judgment is appropriate as to asserted trade secret number 96 because (1) it was inadequately disclosed at the outset of this litigation, (2) Waymo’s evidence of 11 For the Northern District of California United States District Court 10 misappropriation by defendants relied upon an unreliable expert opinion that would be more 12 prejudicial than probative to the jury, (3) insofar as Waymo reformed its asserted trade secret to 13 resurrect asserted trade secret number one, it is rejected again as “Optics 101,” and (4) the 14 evidentiary record shows that even under Waymo’s reformed version of asserted trade secret 15 number 96, there is no genuine dispute of material fact that defendants use neither Waymo’s 16 17 nor its and for summary judgment as to asserted trade secret number 96 are therefore GRANTED. 18 19 strategy. Defendants’ motions to strike CONCLUSION For the foregoing reasons, plaintiff’s motion is GRANTED IN PART and DENIED IN PART. 20 Defendants’ motion is also GRANTED IN PART and DENIED IN PART. Defendants’ motions to 21 strike and for summary judgment as to asserted trade secret number 96 are GRANTED. 22 23 IT IS SO ORDERED. 24 25 Dated: October 30, 2017. WILLIAM ALSUP UNITED STATES DISTRICT JUDGE 26 27 28 18

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?