Summary
In Corley, the plaintiff proposed several modifications to the MPO, and the court agreed to modify the provision requiring the pre-disclosure identification of experts.
Summary of this case from In re Google Assistant Privacy Litig.Opinion
Case No. 16-cv-00473-LHK (HRL)
06-22-2016
RYAN CORLEY, et al., Plaintiffs, v. GOOGLE, INC., Defendant.
ORDER DENYING DEFENDANT'S MOTION FOR PROTECTIVE ORDER AND GRANTING IN PART PLAINTIFFS' MOTION FOR PROTECTIVE ORDER
Re: Dkt. Nos. 89, 90
Plaintiffs allege Defendant Google, Inc. ("Google") violated the Electronic Communications Privacy Act "by intercepting and scanning" certain emails for commercial purposes and without consent. Dkt. No. 19 at 6-7. The parties tried and failed to agree upon a protective order which would govern discovery in this case, and the presiding district judge granted a joint request by the parties for leave to file cross-motions for a protective order. Dkt. No. 87.
Google moves the court to issue the protective order the parties developed together during their negotiations and to rule in Google's favor on the "only two substantive issues" the parties were unable to resolve: (1) the parties "should be required to disclose" the names of third parties who receive confidential information; and (2) protected information "produced in this case" may be used in "related" cases, but only in certain cases which currently exist or else are likely to soon exist, Dkt. No. 90 at 1-2. Plaintiffs respond that Google's non-standard proposed order has not actually been agreed upon by the parties, because Plaintiffs clearly communicated they would not enter any binding agreement unless a full protective order could be agreed upon.
Plaintiffs move the court to issue the model protective order that this court provides for use in highly sensitive cases because that model order is "presumptively reasonable[.]" Dkt. No. 89 at 2. Plaintiffs also seek four substantive modifications to that order: (1) a new definition for experts so that Plaintiffs may hire "past employees of Google or its competitors"; (2) broad permission to use any materials discovered in this case in "all future closely-related actions arising from the same acts"; (3) the inclusion of optional, court-authored provisions which govern the discovery of "HIGHLY CONFIDENTIAL - SOURCE CODE" materials; and (4) the inclusion of an optional, court-authored provision which would permit each party to keep the identity of any given non-testifying expert secret so long as the expert reviews no protected "source code" and the expert "is not an officer, director, or employee of a competitor of a [p]arty or expected to become one." Dkt. No. 89 at 2-3. Google responds that it would be unreasonable to throw out the terms the parties developed together during their negotiations; in the alternative, Google asks the court to issue the model order but raises several arguments for why the court should reject Plaintiffs' proposed amendments to that order.
Discussion
The court may, for good cause shown, "protect a party . . . from annoyance, embarrassment, oppression, or undue burden or expense" by issuing a protective order which "specif[ies] terms" for how discovery shall occur. See Fed. R. Civ. P. 26(c)(1). This court has "approved" the model protective orders it provides online, and the model order for use in highly sensitive cases contains "presumptively reasonable conditions" for managing the discovery of highly sensitive materials. See Barnes and Noble, Inc. v. LSI Corp., Case No. 11-cv-02709-LB, 2012 WL 601806, at (N.D. Cal. Feb. 23, 2012); CAND Model Order for Highly Sensitive Cases, available at http://www.cand.uscourts.gov/model-protective-orders. A party who unilaterally requests a deviation from the presumptively valid terms of that model order must show "specific harm or prejudice" will result if the court denies the request; otherwise, the party has failed to show good cause. See MasterObjects Inc. v. Google, Inc., Case No. 11-cv-01054-LB, 2012 WL 2958227, at *2-3 (N.D. Cal. Jul. 19, 2012).
The court denies Google's motion for lack of good cause. Google moves the court to issue the protective order developed by the parties in the course of their negotiations, but the parties reached neither a final agreement to adopt that protective order nor any binding subsidiary agreement to adopt a subset of its terms. Dkt. No. 94-1 at 6-8. Google has failed to show it will suffer any specific harm or prejudice if the court, instead, issues the model order. Indeed, Google asserts in its opposition brief that its interests will be sufficiently protected if the court adopts the model order "without modification." See Dkt. No. 93 at 3. The undersigned is therefore satisfied that the model order provides a reasonable starting point for setting the terms that will govern discovery in this case. The court shall issue the model order, but shall modify that order based on Plaintiffs' requested modifications and Google's arguments for why those requests should be denied.
The court agrees with Google that Plaintiffs have failed to justify their requested modification to the model-order definition of experts. The proposed modification would permit Plaintiffs to hire the former employees of Google and its competitors as experts. Plaintiffs argue they "need the assistance of" these former employees because Plaintiffs are not competitors to Google, and so they are less likely to misuse any confidential information they might obtain. See Dkt. No. 89 at 2. The undersigned is not persuaded. This court has repeatedly ruled that it is usually improper to hire an opponent's former employee as an expert because such an expert is substantially likely to inflict unfair prejudices which the former employer cannot realistically discover or guard against. See, e.g., Space Sys./Loral v. Martin Marietta Corp., Case No. 95-cv-20122-SW, 1995 WL 686369, at *2 (N.D. Cal. Nov. 15, 1995); Atari Corp. v. Sega of Am., 161 F.R.D. 417, 421 (N.D. Cal. 1994). Google also persuasively argues: (1) Plaintiffs can probably find "computer science professors" who would be capable experts in this case, Dkt. No. 93 at 3; and (2) it would create an unnecessary risk of competitive harm if the court permitted Plaintiffs to hire the former employees of Google's competitors as experts, Dkt. No. 93 at 4. The undersigned also notes that Plaintiffs' conclusion does not follow from their argument—even if Plaintiffs are less likely to misuse confidential information compared with Google's competitors, that fact does not show they have any heightened need to hire certain experts. The undersigned is therefore satisfied that Plaintiffs have failed to show good cause for their requested modification to the model-order definition of experts.
Google agrees in general with the second requested modification—materials discovered in this case should be usable in related cases—but Google argues the court should permit the use of discovery materials produced in this case only in a specific set of cases which exist, or are likely to soon exist, instead of broadly permitting the use of discovery materials in all future related cases. Dkt. No. 90 at 5. Plaintiffs emphasize that it would be proper to share discovery materials across the set of related cases which exist or are likely to exist, Dkt. No. 89 at 2-3, but they make no particular argument for why their broader proposed language is necessary here. The undersigned sees good cause to protect the parties from duplicative, undue expenses by permitting the parties to use materials discovered in this case in a specific set of related cases which exist or are likely to soon exist, see Dkt. No. 90-3 at 14; the undersigned does not see good cause to adopt the broader language proposed by Plaintiffs.
Plaintiffs propose the use of the optional court-authored provisions for a "HIGHLY CONFIDENTIAL - SOURCE CODE" designation "because source code will be at issue." Dkt. No. 89 at 3. Google makes no substantive argument for why these presumptively valid provisions would be improper here; instead, Google argues it might not "need[]" to produce source code in this case and that therefore there is no real dispute before the court, at this time, about how source code should be produced. See Dkt. No. 93 at 5. The undersigned is satisfied that source code is likely to be produced in this case, that the court faces a real dispute about how source-code discovery materials should be produced, and that Google's suggestion—to hold off on setting boundaries for source-code discovery until future disputes about source-code discovery have developed—seems likely to unduly delay this case and to unnecessarily increase litigation costs. The undersigned is therefore persuaded that the discovery order should include the optional court-authored source-code provisions in order to guard against undue burdens and expenses which would otherwise likely be incurred.
Plaintiffs also ask the court to modify the model order with optional, court-authored language which would permit them to withhold the identity of any non-testifying expert "so long as the information or item[s]" shown to the expert are not source code and "the [e]xpert is not an officer, director, or employee of a competitor of a [p]arty or expected to become one." Dkt. No. 89 at 2-3. Federal Rule of Civil Procedure ("FRCP") 26(b)(4)(D) ordinarily prohibits a litigant from discovering the "facts known . . . by an expert who has been retained or specially employed by another party . . . to prepare for trial and who is not expected to be called as a witness at trial." When a litigant attempts to discover the identities of any retained non-testifying experts who know certain facts, he is therefore indirectly seeking information that is ordinarily non-discoverable—is a certain "fact[] known" by any retained non-testifying expert? This court has also ruled that the identities of non-testifying experts should not be disclosed without a showing of "substantial need" because that information is "central to lawyering strategy" and is therefore protected work product. In re Pizza Time Theatre Securities Litigation, 113 F.R.D. 94, 98 (N.D. Cal. 1986) (interpreting Fed. R. Civ. P. 26(b)(3)).
Google argues non-testifying experts should not be permitted to remain anonymous because: (1) Google "is entitled to know the identit[ies] of" the people who possess its confidential and proprietary information, because otherwise Google cannot effectively guard against unjust harms those people might inflict, Dkt. No. 90 at 5; (2) Google faces a higher-than-usual risk that its confidential information will be misused, by some opponent's expert in some case, because Google is involved in more cases than the average litigant, Dkt. No. 90 at 5-6; and (3) Google will not gain any unfair insight into Plaintiffs' trial strategies if the names of the people who possess confidential information are not disclosed until this case ends, Dkt. No. 90 at 4. Plaintiffs respond: (1) the anonymity provision is presumptively valid because it was authored by the court; (2) it will be more difficult for Plaintiffs to find qualified experts in this case if they cannot remain anonymous, and "[s]everal" people have already refused to work on this case out of fear that Google will cut grant funding to their academic departments or will hire fewer of their students if Google learns they worked for Plaintiffs, Dkt. No. 94-1 at 12; and (3) Google relies on a vague argument which would apply "to virtually every" large company. Dkt. No. 94 at 3.
The undersigned is persuaded, for two independent reasons, that the protective order should permit the parties to withhold the identities of non-testifying experts who do not review source code and who are not officers, directors, or employees of a competitor of a party or expected to become one. First, the optional court-authored provision proposed by Plaintiffs is presumptively valid. Second, Plaintiffs have shown good cause to justify this provision—the undersigned is satisfied that it will be significantly more difficult for Plaintiffs to retain qualified non-testifying experts if those experts are not permitted to remain anonymous.
The undersigned also agrees with Plaintiffs' responses to Google's arguments here. Virtually any large company could assert in each of its pending civil cases, as Google asserts here, that non-testifying experts might misuse confidential information, that the company faces this risk in several cases, and that the company's agents should be permitted to track and manage these risks by obtaining the names of all non-testifying experts. Here, Google's concerns are speculative, the possible future harms are remote, and the underlying risks described by Google are unexceptional; indeed, the risk that confidential information might eventually be misused is inherent to the civil discovery of confidential information. Furthermore, the undersigned has already ruled in this order that former employees of Google and its competitors will not be permitted to work as experts in this case, and that ruling gives Google a reasonable degree of protection against the risk of experts improperly sharing confidential information with Google's competitors. Cf. Martin Marietta Corp., 1995 WL 686369, at *2; Atari Corp., 161 F.R.D. at 421. The undersigned is therefore satisfied that Google has failed to show a substantial need, in this particular case, for access to identifying information it is "[o]rdinarily" prohibited from discovering. Fed. R. Civ. P. 26(b)(4)(D); see In re Pizza Time Theatre Securities Litigation, 113 F.R.D. at 98 (interpreting Fed. R. Civ. P. 26(b)(3)).
Finally, the court addresses two arguments which appear in the footnotes of Google's opposition brief: (1) the proposed order unduly "gives Plaintiffs free reign" to decide who qualifies as a competitor of Google, Dkt. No. 93 at 5 n.5; and (2) the court "should not adopt Plaintiffs' proposed [s]ection 7.4(a)(2)" because "it is internally inconsistent"—it permits the anonymity of experts who review "HIGHLY CONFIDENTIAL - ATTORNEYS' EYES ONLY" discovery materials, but it also subsequently requires the disclosure of those experts, Dkt. No. 93 at 5 n.6. The court rejects the first argument; opposing lawyers must often rely on each other's integrity and good faith in the course of civil discovery, and the court sees no reason to suspect that Plaintiffs' lawyers will be unwilling or unable to determine in good faith whether a given person or entity is a competitor of Google. Furthermore, the protective order will leave the parties free to negotiate about who qualifies as a competitor of Google. The court accepts the second footnote argument, but only as a valid technical point which does not alter the court's substantive analysis on the propriety of expert anonymity; the court shall amend Plaintiffs' proposed section 7.4(a)(2) so that it will apply the substance of the pertinent court-authored provision without creating an internal inconsistency in the protective order.
Conclusion
The court will issue the model protective order which this court has written for use in highly sensitive cases. The model order, for good cause shown, will be amended: (1) to permit the parties to use materials produced in this case in a specific set of related cases which exist or are likely to soon exist; (2) to include optional court-authored provisions that govern the discovery of source code; and (3) to permit the parties to withhold the identities of non-testifying experts who have not reviewed protected source code. Plaintiffs' request for an amendment that would permit them to hire the former employees of Google or its competitors as experts is denied. The undersigned will also make several procedural amendments to facilitate compliance with his Standing Order re: Civil Discovery Disputes.
IT IS SO ORDERED. Dated: 6/22/16
/s/_________
HOWARD R. LLOYD
United States Magistrate Judge