Free Declaration in Support - District Court of California - California


File Size: 573.4 kB
Pages: 35
Date: May 19, 2008
File Format: PDF
State: California
Category: District Court of California
Author: unknown
Word Count: 10,134 Words, 65,619 Characters
Page Size: Letter (8 1/2" x 11")
URL

https://www.findforms.com/pdf_files/cand/196709/153-9.pdf

Download Declaration in Support - District Court of California ( 573.4 kB)


Preview Declaration in Support - District Court of California
Case 5:07-cv-05248-JW

Document 153-9

Filed 05/19/2008

Page 1 of 35

N O R T H W E S T E R N JOURNAL OF TECHNOLOGY
AND

INTELLECTUAL PROPERTY

Identification of Trade Secret Claims in Litigation: Solutions for a Ubiquitous Dispute
Charles Tait Graves and Brian D. Range

Fall 2006
© 2006 by Northwestern University School of Law Northwestern Journal of Technology and Intellectual Property

VOL. 5, NO. 1

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW

Document 153-9

Filed 05/19/2008

Page 2 of 35
Volume 5, Number 1 (Fall 2006)

Copyright 2006 by Northwestern University School of Law Northwestern Journal of Technology and Intellectual Property

Identification of Trade Secret Claims in Litigation: Solutions for a Ubiquitous Dispute
By Charles Tait Graves and Brian D. Range *
I. INTRODUCTION ¶1 Trade secret cases pose special problems distinct from other intellectual property lawsuits because alleged trade secrets are rarely defined in advance of litigation. For every patent, registered copyright, or trademark, there is a document filed with the federal government identifying the details and boundaries of the asserted intellectual property. But in trade secret cases, the plaintiff's attorneys typically develop the trade secret claims at the same time that they file a misappropriation suit against a former employee or business partner ­ or afterwards. Trade secret plaintiffs rarely provide a precise and complete identification of the alleged trade secrets at issue without a court order requiring them to do so. This is a strategy, not an accident. The tactical advantages a plaintiff gains from non-identification are too tempting for a plaintiff to voluntarily provide such identification. In a typical case, the plaintiff has filed suit against a recently departed employee who joined or started a competing business, and then provide little more than a list of high-level, generic categories in which its alleged secrets are said to reside. It is also common for a trade secret plaintiff to alter its list of trade secret claims as the case proceeds ­ sometimes dramatically, by replacing entire categories of information or technology, or by re-combining slippery, multi-element "combination trade secret" claims into new subsets. In a 2006 New Jersey case, for example, the plaintiff first identified four alleged secrets in an initial interrogatory response, then added six, later added thirty-nine in an expert report, then claimed ninety-two in a supplemental interrogatory response, and had its expert give what the court called "moving target" testimony about the alleged secrets. 1 We have seen many similar alterations during the course of cases in several jurisdictions, and defendants sometimes

¶2

¶3

Charles Tait Graves and Brian D. Range are associates at Wilson Sonsini Goodrich & Rosati in San Francisco, California and Austin, Texas. This article is the fourth in a series that seeks to clarify frequently litigated but obscure areas of trade secret law, in the interest of protecting employee mobility and the right to use information in the public domain. 1 See, e.g., Civil Action No. 01-4677 (SRC) 2006 WL 1344084, at *1-3 (D.N.J. May 16, 2006) (acknowledging plaintiff's "moving target" allegations but holding that defense had sufficient opportunity to take discovery on shifting claims); Thomas & Betts Corp. v. Richards Mfg. Co., No. Civ. 01-4677, 2006 WL 902148, at *8-9 (D.N.J. Apr. 4, 2006) (barring some expert testimony).

*

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW Vol. 5:1]

Document 153-9

Filed 05/19/2008

Page 3 of 35

Charles Tait Graves and Brian D. Range

¶4

¶5

¶6

spend months disproving one set of allegations only to face a new, replacement set as the close of discovery nears. Precise identification of the alleged trade secrets is a crucial component of trade secret litigation. This is especially true when complex technology is at issue, but also in cases featuring simpler claims. Without a precise identification of software code elements, hardware architecture components, or even the names of allegedly secret customers, the defendant cannot compare the claims against public domain information to challenge the alleged secrecy of such information. It is difficult to conduct a public domain analysis if the claims are identified at only a generic level such as "source code," "ASIC design elements," or "customer lists." Stated differently, the parties and the court cannot accurately decide the question of whether a trade secret exists without first understanding what precisely is asserted as a secret. It is not surprising, then, that disputes over identification of the plaintiff's alleged trade secrets are common. However, it is notable that courts nationwide have not yet articulated a set of guidelines and standards to assist judges who must rule on identification questions. Indeed, courts in various jurisdictions have issued rulings on the same identification issues with almost opposite results ­ a majority requiring more identification, but a minority allowing the plaintiff to proceed with only general, conclusory descriptions of the alleged trade secrets. The published rulings indicate that courts subjectively reach their own conclusions, often with little or no reference to other published decisions. The problem is further exacerbated because courts rarely, if ever, quote precise descriptions of potentially valuable trade secrets in publicly available opinions, and thus most published opinions do not clarify the degree of specificity required. 2 As a result, the scattered case law in this area does not always serve as precedent, and it is difficult for parties to predict the result in any given case. The lack of a uniform set of guidelines invites trade secret plaintiffs to avoid identifying their alleged secrets in detail, because there is always a fair chance that the reviewing court will not require them to do so. Despite the frequency of disputes over the identification of alleged trade secrets, the issue has received little scholarly attention. 3 Perhaps this is because the case law is scattered across multiple jurisdictions, in state and federal courts, without any unifying doctrine apart from citations to a few well-known cases. Most treatises refer to the problem in only general terms, 4 and a handful of practitioner-oriented articles note particular cases without a nationwide survey or detailed proposals. 5
See Savor, Inc. v. FMR Corp., No. Civ. A. 00C-10-149JRS, 2004 WL 1965869, at *6-7 (Del. Super`. Ct. Aug. 16, 2004) (redacting alleged trade secrets). 3 The single full-length article we have found devoted to the identification question is Kevin R. Casey, Identification of Trade Secrets During Discovery: Timing and Specificity, 24 AIPLA Q.J. 191, 281 (1996) (summarizing case law and noting different stages of the case, but apparently recommending that "precise" detail need not be disclosed until trial; no specific focus on identification of different types of information). We disagree that identification should be withheld until trial, for reasons discussed below. The majority of the case law since 1996 supports our position. 4 See, e.g., RUDOLF CALLMANN, CALLMANN ON UNFAIR COMPETITION, TRADEMARKS AND MONOPOLIES § 14:30 (4th ed. Supp. 2006) (briefly noting issue with citations); MELVIN A. JAGER, TRADE SECRETS LAW §§ 5:32, 9:8 (West 2006) (1985) (general description of some identification cases); ROGER M. MILGRIM, MILGRIM ON TRADE SECRETS §§ 15.01[1][d][i], 16.01[5][b] (Matthew Bender 2006) (1967) (general summary of issue and its importance with a few citations); ROBERT I. WEIL & IRA A. BROWN, JR., CALIFORNIA PRACTICE GUIDE: CIVIL PROCEDURE BEFORE TRIAL § 8:58 (2006) (general description of
2

69

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW

Document 153-9

Filed 05/19/2008

Page 4 of 35 [2006

NORTHWESTERN JOURNAL OF TECHNOLOGY AND INTELLECTUAL PROPERTY

Given that most of the identification disputes on record have occurred in the past decade, the time is right for a detailed analysis of the identification problem. We believe that courts nationwide facing trade secret disputes should apply a recognized set of guidelines and standards to resolve disputes over the identification of alleged trade secrets in litigation. This would reduce unpredictable and inconsistent rulings, deter plaintiffs from strategically withholding information by gambling that any given court will not require a precise identification, and deter defendants from making the wrong demands as to identification at the wrong stage of the case. We believe that a proper standard would take into account the type of information being alleged as secret, the stage of the case, and the defendant's own diligence in raising the issue. ¶8 To that end, we have reviewed substantially all of the trade secret identification rulings available, including unpublished rulings and slip opinions, in order to paint a picture of how courts are reacting to similar arguments and disputes across the country. We will describe both outlier and mainstream opinions to illustrate why a set of common standards is needed. 6 ¶9 Yet merely summarizing holdings does not provide sufficient information to propose guidelines. Few courts have attempted to state rules of general applicability beyond the dispute at issue. Outlining the policy grounds supporting detailed identification of trade secret claims is therefore necessary. We will organize the discussion by the stage of the case, the type of information at issue, and the skill with which the defense has properly raised the issue. We will also offer solutions that courts can apply for the specific types of alleged secrets at issue in a given case. ¶10 In short, we offer five proposals as a standard for identification of alleged secrets in litigation:

¶7

California's Code Civ. Proc. sec. 2019.210). The most detailed existing commentary on the identification question is JAMES POOLEY, TRADE SECRETS § 11.02 (Supp. 2005). However, we respectfully disagree with the view expressed there that courts should not always require precise identification, especially during the discovery stage, because without precision the plaintiff has too easy an opportunity to continually alter its claims, and because plaintiffs who file multi-million dollar technology lawsuits are fully capable of providing such precision. Moreover, we have seen many cases where highly technical details were located in the public domain, and such research would have been impossible absent precision. 5 See, e.g., JOHN F. HORNICK & MARGARET A. ESQUENET, TRADE SECRET IDENTIFICATION OF TIMING IN DISCOVERY (2005), http://www.finnegan.com/publications/news-popup.cfm?id=1248&type=article (focusing on jurisdictions with pre-discovery requirements); Brent Caslin, Secret Weapon: Understanding What Constitutes "Reasonable Particularity" Can Be the Decisive Element in Trade Secret Litigation, L.A. LAW., Apr. 2004, at 44 (noting leading cases and advising defendants about various motions to file regarding an insufficient identification); York M. Faulkner & Margeret A Esquenet, Putting an End to the Trade Secret Plaintiff's Game of Bait and Switch, IP LITIGATOR, Sept. 2001, at 24-25 (noting problem and describing a few leading cases); Randy Kay, Identifying Trade Secrets in Litigation - Managing the Process, 28 NEW MATTER 1/2, Sept. 2003, 19-22 (noting problem and providing advice for plaintiffs and defendants); James R. Mckown, Discovery of Trade Secrets, 10 SANTA CLARA COMPUTER & HIGH TECH. L.J. 35, 47 (1994) (noting leading cases on identification disputes as of 1994 in article about procedural aspects of disclosing trade secrets in various litigation contexts); David R. Sugden, C.C.P. Section 2019.210: A Sword, Not a Shield, ORANGE COUNTY LAW., Mar. 2006, at 51-55 (advising defense counsel to take advantage of California's pre-discovery identification statute and use a plaintiff's failure to identify to move for summary judgment later in the case). 6 Of course, rulings on most identification disputes take place on discovery calendars in trial courts and thus are never published, so a complete nationwide survey is not possible.

70

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW Vol. 5:1]

Document 153-9

Filed 05/19/2008

Page 5 of 35

Charles Tait Graves and Brian D. Range

(1) Courts should first separate the question of whether the plaintiff has identified an alleged secret from the question of whether the information is, in fact, a trade secret; (2) Courts should require a reasonably detailed identification of the alleged secrets before a plaintiff can proceed with discovery ­ a rule already required by statute in California, and by common law in several other states; (3) Perhaps most importantly, if a defendant properly requests a precise identification during discovery, the court should require a precise, complete identification of whatever type of information is at issue ­ including an identification of alleged "combination trade secrets"; (4) No trade secret plaintiff should obtain an injunction or avoid summary judgment without a precise identification of alleged secrets; (5) Trade secret plaintiffs should not be permitted to alter or amend an identification of trade secret claims without a showing of good cause. These standards would remove the uncertainty that prevails in identification disputes. A trade secret plaintiff with a valid claim would have incentive to identify the alleged secrets early on, and a defendant who has misappropriated trade secrets could more easily be shown to have done so. A trade secret plaintiff with a meritless claim, however, would more easily be exposed. These standards would make it more difficult to sue former employees for harassment purposes and to spend them into the ground with protracted litigation, and more difficult to claim public domain information as trade secrets. ¶12 We believe that these proposals are equally applicable in jurisdictions which have enacted the Uniform Trade Secrets Act and those which follow the Restatement (Second) of Torts. Under both, a plaintiff must establish that the information at issue does, in fact, constitute trade secrets. 7 Requiring identification is a necessary case management tool to ensure that the parties and the court accurately understand what information is being asserted as secret in order to proceed to the secrecy question. 8 ¶13 This Article is primarily for courts ruling on identification questions, who must research the issue when parties file pleadings that cite only decisions that favor their positions, and who may need a single source for all of the nationwide case law. We believe that courts do not always realize the degree to which the identification of trade
See, e.g., CAL. CIV. CODE § 3426.1(d)(1) (West 2006) (California Uniform Trade Secrets Act provision defining trade secrecy); RESTATEMENT (SECOND) OF TORTS § 757 cmt. b (1979) (Restatement definition of trade secrecy). 8 We note, however, that identification may not always matter in jurisdictions where a former employer can use contract law to bind former employees with non-competition covenants that bar use of non-secret material for some period after resignation. See, e.g., Lason Serv., Inc. v. Rathe, No. Civ.A.3:02CV2110-D, 2003 WL 21728184, at *6 (N.D. Tex. Mar. 14, 2003) (rejecting defendant's argument that plaintiff failed to identify allegedly secret customers because non-solicitation covenant barring solicitation of customers was valid in jurisdiction and it was rrelevant whether customers were secret or non-secret); Paper Mfg. Co. v. Weiss, No. 1040, 1972 WL 15994, at *3 (Pa. Ct. Com. Pl. Mar. 13, 1972) (denying motion to require identification of alleged trade secrets given that plaintiff's claim was premised on restrictive noncompetition contract valid in that jurisdiction, and thus plaintiff need not prove trade secrets to prevail).
7

¶11

71

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW

Document 153-9

Filed 05/19/2008

Page 6 of 35 [2006

NORTHWESTERN JOURNAL OF TECHNOLOGY AND INTELLECTUAL PROPERTY

secret claims is a lawyers' game, where one side holds back information and hopes that the opposing party will fail to raise the issue, or hopes that that a court will accept the argument that generic references to complex technology are sufficient. ¶14 This Article should also be of use to litigants, both to plaintiffs who have valid claims and seek to identify them in good faith and to defendants who lack the time or money to canvass the vast jurisprudence on identification, or who want to craft the best possible discovery requests to pin down a trade secret plaintiff and force a precise identification. ¶15 In what follows, we will describe the common mistakes that courts make when ruling on identification disputes, the problems that arise in the absence of a precise identification of trade secret claims, and how courts nationwide have ruled on identification disputes. We will then offer specific proposals for resolving identification questions. II. THE PROBLEM ¶16 A trade secret case usually begins shortly after a former employee has resigned and either joined a competitor or formed a new, competing business. The plaintiff files a lawsuit that accuses the defendant of misappropriating trade secrets, but the complaint contains only generalities about the nature of the alleged secrets. Once the parties have entered into a protective order that allows the confidential sharing of information between the parties and their attorneys, 9 the plaintiff sometimes issues a written statement that purports to identify the trade secret claims at issue, and sometimes provides such a statement as a discovery response. Because the statement rarely if ever identifies the alleged secrets with precision, a dispute over the specificity of the identification often arises within the first three months of the litigation. In other cases, the identification dispute arises later, when the plaintiff alters its claims or raises a new set of claims. ¶17 In the typical dispute over identification of alleged trade secrets, the defense argues that the plaintiff's list of categories and concepts is not specific enough, and then cites a scattering of cases that may not have been decided at the same stage of the case, that may not have addressed the same type of information, or that provided no real guidance on how much precision was required. The plaintiff, in turn, often responds by arguing that the defendant already knows what the alleged trade secrets are because the defendant knows what it stole, and thus no identification is necessary. The plaintiff will typically argue that a list of high-level concepts is sufficient and make selective citations to cases so holding, and it may also argue that because it alleges "combination" trade

A protective order in a trade secret case usually creates a three-tier classification for information produced in discovery: material that is designated as non-confidential, material that is designated such that only the parties can view it and not the public, and material that only the attorneys and expert witnesses can view. The parties usually agree on terms allowing the individuals actually accused of misappropriating trade secrets to see the list of items they are accused of having stolen. In most cases, employees of the parties rarely see documents and other information produced by the other party, so there is little if any risk that one side will misuse information produced by the other.

9

72

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW Vol. 5:1]

Document 153-9

Filed 05/19/2008

Page 7 of 35

Charles Tait Graves and Brian D. Range

secrets rather than individual trade secrets, no identification should be required. 10 Also, if the case is at an early stage, the plaintiff may argue that it cannot yet know every trade secret that was stolen and therefore cannot identify all trade secrets claims it may assert in the future. ¶18 Assessing an identification dispute can be difficult given the lack of uniformity in the published cases and the absence of a set of standards to which courts can look, especially in jurisdictions where trade secret lawsuits are less common. A. Twelve Policy Considerations Relating to Vague Trade Secret Allegations ¶19 The best place to begin is to step back from the facts of the case at issue and outline the policy reasons why requiring a detailed identification of the trade secret claims makes sense. We believe that there are at least twelve such considerations. Courts have cited some of these considerations when requiring identification, and we have come to recognize the others in the course of litigating scores of trade secret cases. ¶20 The first consideration is that the identification of an alleged trade secret can only come from the plaintiff, because there is no other pre-existing source one can look to for the details of the alleged intellectual property. Trade secrets are not listed claim-byclaim like patents or set forth in their entirety like registered copyrights or trademarks. There is no public source to which the defendant can look in order to determine what potential trade secrets the plaintiff possesses. The plaintiff's decision on what to claim as its trade secrets in litigation is a subjective one. Even if a former employee defendant could guess what trade secret claims the former employer might make, the specifics of the actual claim could differ from that speculation. On this point, it is important to note that the plaintiff's identification of its own alleged intellectual property does not depend on the defendant's documents, because the preliminary question is whether the plaintiff has a trade secret, which must be answered before determining whether the defendant used the secret in its products or services. ¶21 The second consideration is that requiring an early and precise identification of alleged trade secrets will encourage compliance with Federal Rule of Civil Procedure 11 and similar state rules governing frivolous pleadings. Such statutes and the Uniform Trade Secret Act all require that a trade secret plaintiff make allegations of misappropriation in good faith ­ either directly or through general rules against meritless accusations. 11 Any plaintiff's pre-lawsuit review of a potential trade secret matter should therefore consider whether the basic elements of the tort claim can be met, such as (1) the existence, in fact, of a trade secret; (2) the plaintiff's use of reasonable measures to maintain that secrecy; and (3) the defendant's misappropriation of one or more such secrets. 12 A plaintiff cannot adequately investigate these elements without internally identifying the trade secret claims under consideration. A vague identification of alleged trade secrets once the litigation has begun suggests that the plaintiff did not undertake a

For a discussion of courts which have accepted this "combination trade secret" argument, see infra Section III.A. 11 For example, California's version of the Uniform Trade Secrets Act provides remedies for bad faith trade secret claims, and the state also has a general statute governing frivolous pleadings. See CAL. CIV. CODE § 3426.4 (West 2006); CAL. CIV. PROC. CODE § 128.7(b) (West 2006). 12 See, e.g., CAL. CIV. CODE § 3426.1 (West 2006) (definitions of elements of a trade secret claim).

10

73

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW

Document 153-9

Filed 05/19/2008

Page 8 of 35 [2006

NORTHWESTERN JOURNAL OF TECHNOLOGY AND INTELLECTUAL PROPERTY

pre-lawsuit review. If parties know that identification will be required early in litigation, they have greater incentive to do a careful review before filing suit. ¶22 Third, and similarly, a lax identification standard could allow a party to avoid the statute of limitations. As one court noted on this point: These facts of cutting edge technological life, to say nothing of the limitations of language generally, have the effect of giving the person asserting trade secret claims in these kinds of cases, very wide definitional latitude, and, with that, considerable power to manipulate the articulation of claims for tactical purposes, e.g., to escape what might otherwise be a perfectly sensible application of the statute of limitations. 13 The fourth consideration is that an exact identification is necessary to set the proper bounds and scope of discovery ­ the areas in which discovery will be permitted. Litigating parties often make and sell products outside the scope of the alleged trade secrets, and requiring a precise identification prevents burdensome and expensive discovery into irrelevant areas. Also, without a detailed identification, a plaintiff may later define its alleged secrets by first looking at the defendant's documents and then crafting claims that match those documents, but that do not match the plaintiff's own information. ¶24 The fifth consideration is that requiring a detailed identification early in a lawsuit may reduce future disputes over identification and thereby reduce the length of the discovery process. By contrast, a plaintiff who identifies the claims only later may cause new rounds of depositions and other discovery as the parties scramble to assess the newly-provided information. Knowing the claims early allows for a focused and efficient discovery plan from the start, and thus serves as a useful case management tool. ¶25 The sixth consideration is that without a precise identification, the defense lacks the information it needs to conduct public domain research and challenge the alleged secrecy of the information at issue. Seventh, and for the same reasons, the reviewing court cannot rule on a secrecy claim with accuracy and confidence if the exact parameters of the information remain unclear. Without details, who can be sure that the alleged secret has not been published elsewhere, by a market competitor, an engineering publication, or in academic work? ¶26 The eighth consideration is that requiring a precise identification prevents a plaintiff from making overbroad secrecy claims encompassing vast categories of information. Plaintiffs sometimes use the trade secret claim as a sort of non-competition covenant by listing wide areas of technology that encompass the former employee's entire set of skills, knowledge, and experience. Ninth, and on a related point, a defendant cannot formulate its arguments for differentiating its technology from the alleged trade secrets at issue without knowing precisely what those trade secrets are. For example, a former employee may well have worked in good faith to ensure that software written for a subsequent employer would not copy or otherwise make use of the former employer's ¶23

Intermedics, Inc. v. Ventritex, Inc., 822 F. Supp. 634, 656 (N.D. Cal. 1993) (regarding trade secret statute of limitations under California's UTSA).

13

74

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW Vol. 5:1]

Document 153-9

Filed 05/19/2008

Page 9 of 35

Charles Tait Graves and Brian D. Range

¶27

¶28

¶29

¶30

non-public information, but cannot make that distinction if the plaintiff asserts only general functionalities common to that general type of software. The tenth consideration is that expert reports from both sides will be more useful and focused if the exact nature of the alleged secrets is known before they begin their work. Without an exact identification, the plaintiff's expert may simply declare that broad areas of technology are trade secrets, without actually conducting specific public domain research that could support a valid scientific opinion. Meanwhile, the defense experts may be forced to spend a great deal of effort researching and opining on wider areas of technology than are at issue, without being able to focus on the exact technology the plaintiff claims as intellectual property. The eleventh consideration is that the trade secret plaintiff should not be able to unilaterally alter and amend its claims without judicial review as the lawsuit proceeds. Too many trade secret cases feature a kaleidoscope of amorphous, frequently-altered claims that prevent the defense from pinning down and defeating any solid set of claims. Requiring an early and precise identification of alleged secrets can make attempts at unsupervised alteration apparent. By contrast, a broad and non-specific list of concepts and categories allows a plaintiff to alter the details of the claims in a manner more likely to escape judicial review. Finally, and perhaps most broadly, requiring a detailed identification protects employees' interests in moving to the jobs of their choice and using public domain information, as they have a right to do. When a plaintiff makes vague trade secret claims against a former employee without narrowly defining the actual items at issue, it is essentially requesting a court-created non-competition agreement against the defendant. 14 Such claims cut into the former employee's general right to take non-secret knowledge, skills, and experience from job to job. 15 But when discovery focuses only on detailed clams susceptible to examination and study, there is less risk that these important public policy interests will be encroached upon. At the same time, we do not agree with all of the policy grounds that courts have given for requiring a detailed identification of alleged trade secrets. Some courts and commentators have asserted that identification is needed to prevent a malicious plaintiff from filing suit for the purpose of getting into the defendant's own information and taking the defendant's trade secrets. 16 While we have learned from experience that
14 It should be noted that an employee seeking to act in good faith may have no affirmative ability to obtain a specific list of potential trade secrets in order to avoid using them. In Lane v. Baxter Healthcare Corp., 905 S.W.2d 39, 40 (Tex. Ct. App. 1995), a former employee sought a declaratory judgment that would force his former employer to list any trade secrets that it contended the former employee could not utilize in his new job. The court held the issue was not justiciable because the former employee had "not invented, and may never invent, anything that is of concern to Baxter." Id. at 42. Thus, the former employee's attempt to use a declaratory judgment action to determine how he could fairly compete with his former employer failed. 15 The concept that former employees are free to use general skills, knowledge, and experience is well established. See, e.g., Winston Research Corp. v. Minn. Mining & Mfg. Co., 350 F.2d 134, 143-144 (9th Cir. 1965) ("[plaintiff's] former employees cannot be denied the right to use their general skill, knowledge and experience, even though acquired in part during their employment by [plaintiff]); Eaton Corp. v. Appliance Valves Corp., 526 F. Supp. 1172, 1180 (N.D. Ind. 1981) ("the general knowledge and skill obtained by an employee during his employment belongs to that employee and can be utilized by him after terminating his employment"), aff'd, 688 F.2d 842 (7th Cir. 1982). 16 See, e.g., Computer Econ., Inc. v. Gartner Group, Inc., 50 F. Supp. 2d 980, 985 (S.D. Cal. 1999)

75

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW

Document 153-9

Filed 05/19/2008

Page 10 of 35 [2006

NORTHWESTERN JOURNAL OF TECHNOLOGY AND INTELLECTUAL PROPERTY

attorneys are willing to assist their clients by altering claims and by not defining claims, we believe that it is less likely that a law firm would assist a client's scheme to take information from a competitor by filing a lawsuit against the competitor and turning over information produced in discovery and designated as confidential under a protective order. We believe that courts should avoid relying on this purported basis for requiring an identification, and focus instead on the real problems caused by non-identification. B. Mistakes Courts Make When Ruling on Identification Questions Reflecting on these policy considerations should assist courts in making a ruling appropriate to the case at issue, and in avoiding the mistakes courts sometimes make when deciding identification questions. ¶32 Several such mistakes merit special attention. To begin with, a tiny minority of courts have nullified the identification requirement altogether, refusing to require a trade secret plaintiff to identify the trade secret claims even in the face of the defendant's discovery requests or controlling law in the jurisdiction. As one example, a California trial court sanctioned a trade secret defendant in a 2006 case for not providing discovery after the plaintiff failed to serve a trade secret identification statement, and it took an appeal to reverse the sanctions ­ even though California has a statute requiring that a trade secret plaintiff provide a reasonably particular identification of alleged secrets before pursuing discovery. 17 In another case, a Massachusetts court repeatedly rejected the defendant's efforts to learn the details of the alleged secrets. 18 ¶33 Rulings like these are outliers. But three other mistakes that courts make are more common, and perhaps less obvious. One is not recognizing that the identification of the trade secret claims is a separate issue distinct from whether the information at issue is secret, because that identification is a necessary preliminary step leading to the secrecy analysis. In such cases, courts conflate identification questions with the secrecy question, confusing the discussion. 19 ¶31
(identification "prevents plaintiffs from using the discovery process as a means to obtain the defendant's trade secrets."); Data General Corp. v. SCI Sys., Inc., Civ. A. No. 5662, 1978 WL 22033, at *1 (Del. Ch. Nov. 27, 1978) (identification serves to prevent the plaintiff from going through the defendant's information to gain "an unfair business advantage."); WEIL & BROWN, supra note 4, at § 8:58 (claiming that purpose of California's identification statute is "to prevent plaintiff from conducting `fishing expeditions' into competitors' business files by unfounded claims of trade secret misappropriation."). 17 See Roup v. Super. Ct., No. B188652, L.A.S.C. No. BC328972, 2006 WL 710891, at *1 (Cal. Ct. App. Mar. 22, 2006) (unpublished). 18 See, e.g., Storage Tech. Corp. v. Custom Hardware Eng'g & Consulting, Inc., No. Civ.A. 02-12102RWZ, 2005 WL 283200, at *1 (D. Mass Feb. 7, 2005) (referring to unspecified "rules" when refusing again to require detailed identification: "CHE objects to the Master's denial of yet another attempt to gain detailed identification of Storage Tech's copyright and trade secret claims. Much of what the Master rejected, the court had previously ruled out of order, or Storage Tech had in the interim supplied. The rules do not sanction the degree of detail CHE seeks to extract."); Storage Tech. Corp. v. Custom Hardware Eng'g & Consulting, Inc., No. Civ.A. 02-12102-RWZ, 2003 WL 22462494, at *1 (D. Mass Oct. 30, 2003) (on pre-discovery order in software case, rejecting defendant's attempt to find out exactly what source code was claimed as secret: "Defendant proposed that plaintiff provide a detailed list of the trade secret allegedly misappropriated to which plaintiff objects. This appears to be another request for the identification of each line of the maintenance source code implicated in these proceedings. I remain unpersuaded that defendant is either entitled to discover that level of detail or that such is necessary to mount an adequate defense."). 19 See, e.g., Excelligence Learning Corp. v. Oriental Trading Co., No. 5:03-CV-4947 JF (RS), 2004 WL

76

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW Vol. 5:1] ¶34

Document 153-9

Filed 05/19/2008

Page 11 of 35

Charles Tait Graves and Brian D. Range

The second mistake is assuming that descriptions of technical categories and concepts such as "source code" or "placement of capacitors in the schematic for the ABC project" are sufficient to identify alleged secrets ­ perhaps because courts do not always recognize how much public domain information exists in such broadly defined technical categories. We will describe several such rulings below. ¶35 The third common mistake, and perhaps the worst, is accepting a plaintiff's argument that because "combination trade secrets" are claimed, the plaintiff need not make a precise identification of such alleged secrets. A "combination trade secret" is a multi-element claim that, when valid, ties non-secret items of information together in a unique manner to form a trade secret. The concept is dangerous, however, because trade secret plaintiffs often string together non-secret items that have no functional interrelationship to avoid a defendant's showing that the individual items are non-secret. Attorneys also use the "combination" concept to alter claims, and to create more sets of alleged intellectual property than the plaintiff ever actually thought about or linked together. But combination claims can fail for a number of reasons, just as individual trade secret claims can, and thus they should not provide a shield against identification. 20 ¶36 Trade secret plaintiffs use the "combination" argument during identification disputes to confuse courts and escape a precise identification of the alleged secrets. If a defendant argues for a precise identification, the plaintiff will often respond that it claims "combination" secrets in an "entire process" rather than individual secrets, and that identifying individual elements should therefore be unnecessary. Or, to the same effect, the plaintiff argues that its secret is its entire base of software source code, its entire chip design, and so forth, without regard to identifying what exactly about those broad categories are asserted as an allegedly misappropriated secret. ¶37 The "combination" concept can create a major loophole for a plaintiff seeking to avoid identification. Perhaps most dangerous, a plaintiff can use the "combination" theory to alter and revise its claims in different mixes of subsets, to gerrymander a claim so that the defense cannot focus its research efforts on defeating the final version. If we take, for example, seven software algorithms and assume that five are in the public domain, the plaintiff might alter the claim several times to create subsets of the seven where at least one of the included algorithms is secret, in order to claim the non-secret algorithms as secret as well. With a simple set of seven, there is a multiplicity of mathematically possible variations of subsets of two or more elements. Of course, the
2452834, at *3-4 (N.D. Cal. June 14, 2004) (reviewing whether plaintiff's Cal. Civ. Proc. Code. sec. 2019.210 identification statement was sufficient and asking whether the information was secret instead of whether identification was sufficiently specific); Momswin, L.L.C. v. Lutes, No. Civ.A. 02-2195-KHV, 2003 WL 21554944, at *7 (D. Kan. July 8, 2003) (denying plaintiff's motion for summary judgment for failure to establish conclusively that information was secret, but seemingly conflating identification requirement with secrecy requirement in discussion); Lovell Farms, Inc. v. Levy, 641 So. 2d 103, 105 (Fla. Ct. App. 1994) (affirming denial of request for temporary restraining order and requiring that plaintiff first show existence of trade secret; seemingly blending an identification requirement and proof of secrecy through an in-camera review procedure); Gabriel Int'l, Inc. v. M&D Indus. of Lou., Inc., 719 F. Supp. 522, 524-25 (W.D. La. 1989) (in confusing ruling, court required that a trade secret plaintiff prove the existence of a secret early in the lawsuit, apparently before seeking third party discovery, but in so doing appears to have conflated the need to identify the alleged secrets with the ultimate proof of secrecy or non-secrecy). 20 For a detailed description of "combination trade secret" claims and citations to numerous cases where such claims have failed, see Tait Graves & Alexander Macgillivray, Combination Trade Secrets and the Logic of Intellectual Property, 20 SANTA CLARA COMPUTER & HIGH TECH. L.J. 261 (2004).

77

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW

Document 153-9

Filed 05/19/2008

Page 12 of 35 [2006

NORTHWESTERN JOURNAL OF TECHNOLOGY AND INTELLECTUAL PROPERTY

algorithms may not in reality interoperate together, and perhaps should properly be defined as individual trade secret claims rather than "combination" claims. But without requiring a precise identification of the "combination" trade secret at issue, the defense may be unable to raise that argument. Courts need to be aware of this common litigation tactic and nail down exactly what permutations and subsets are actually claimed as secret. ¶38 But all too often, courts have confused the question of (1) whether the plaintiff claims "combination" secrets instead of individual secrets with (2) whether the plaintiff should be required to precisely identify its alleged trade secrets regardless of whether they are claimed in combinations or individually. ¶39 There are several published rulings where courts have failed to require a precise identification of the alleged secrets as a result of conflating these two issues. 21 In one, a plaintiff vaguely claimed secrets somewhere within 500 pages of materials, and the defendant argued on appeal after a jury verdict that the plaintiff did not specify what the secrets were. The court merely noted that a plaintiff can have a combination trade secret in material that includes elements that are individually non-secret, and did not ask whether any such combination had been properly identified. 22 And in one of the most questionable trade secret rulings on record, a Pennsylvania district court rejected a defendant's identification argument by holding that the plaintiff's set of generalized business concepts for running a tanning salon was an "entire methodology" that added up to form a trade secret, without any specific identification. By avoiding the identification analysis and stringing together vague business concepts into a combination trade secret, the decision essentially created a non-competition agreement and held the defendant liable for violating it.23 Of course, a plaintiff certainly is capable of properly identifying an alleged combination trade secret. 24 Indeed, many courts have required precise identification of "combination trade secrets." We will discuss those cases below when proposing solutions. ¶40 With these policy considerations and common mistakes in mind, we will survey the nationwide case law, and then offer specific proposals for addressing identification disputes.
21 Courts sometimes sidestep the identification question by accepting a plaintiff's argument that its entire process or software was at issue. See, e.g., Trandes Corp. v. Guy F. Atkinson Co., 996 F.2d 655, 662 (4th Cir. 1993) (refusing to overturn jury verdict despite plaintiff's failure to identify precise secrets, and affirming on questionable ground that entire object code software was secret without addressing whether any secrets were identified in the functionality or other aspects of the executable software; even with finalized object code, we believe that a plaintiff should identify what about the software is not generally known to the trade); Touchpoint Solutions, Inc. v. Eastman Kodak Co., 345 F. Supp. 2d 23, 28-29 (D. Mass. 2004) (granting in part plaintiff's motion for preliminary injunction despite defendant's argument that plaintiff had not specifically identified alleged secrets; court accepted plaintiff's characterization of "source code, implementation, overall design and `distributed computing model'" as a design that was "unique, complex, and apparently valuable" without focusing on whether specific details had been identified). 22 See 3M Co. v. Pribyl, 259 F.3d 587, 595-96 (7th Cir. 2001). 23 See Tan-Line Studios, Inc. v. Bradley, Civ.A. No. 84-5925, 1986 WL 3764, at *7 (E.D. Pa. Mar. 25, 1986) (ruling following bench trial; referring to general definition of a combination trade secret). 24 See Rohm & Haas Co. v. Adco Chem. Co., 689 F.2d 424, 432 n.7 (3d Cir. 1982) (plaintiff claimed combination trade secret in chemical process and "did put in evidence detailed descriptions of the Process," though specificity unclear from published decision).

78

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW Vol. 5:1]

Document 153-9

Filed 05/19/2008

Page 13 of 35

Charles Tait Graves and Brian D. Range

III. HOW COURTS NATIONWIDE HAVE ADDRESSED IDENTIFICATION DISPUTES ¶41 There are several stages of a case where courts face disputes over the identification of alleged trade secrets: the pleading stage, on a request for a temporary restraining order or preliminary injunction, just before discovery begins, during discovery, on a motion for summary judgment, during trial, and after trial. ¶42 For the most part, the majority trend is to require a detailed identification of alleged secrets ­ particularly on a request for an injunction, during discovery, and on a motion for summary judgment. Rulings are inconsistent at the pleading stage, probably because of liberal notice pleading requirements. In turn, rulings after trial often go against the defendant, likely because the defense waited too long to raise the identification issue and, more importantly, because courts are loath to overturn jury verdicts. A. The Pleading Stage ¶43 In some cases, trade secret defendants have moved to dismiss the complaint, or to obtain a more definite statement, when the plaintiff is vague regarding the alleged secrets. Some courts have required more information at the pleading stage, 25 though plaintiffs obviously need not disclose material believed to be secret in a public complaint, and liberal notice pleading standards often mean little is required. 26 As one court explained the general rule, "courts are in general agreement that trade secrets need not be disclosed in detail in a complaint alleging misappropriation . . . . The query is whether
See, e.g., Metis Int'l, L.L.C. v. Ace INA Holdings, Inc., No.Civ.A.SA.04CA-1033-XR, 2005 WL 1072587, at *5 (W.D. Tex. May 6, 2005) (granting Rule 12(e) motion and requiring plaintiff to plead more about the categories of alleged secrets being claimed); Cambridge Internet Solutions, Inc. v. Avicon Group, No. 99-1841, 1999 WL 959673, at *2 (Mass. Ct. App. Sept. 21, 1999) (granting in part motion for a more definite statement where plaintiff alleged that some alleged secrets where in specific document, but did not do so for alleged "customer material" secrets); Diodes, Inc. v. Franzen, 260 Cal. App. 2d 244, 253 (Cal. Ct. App. 1968) (sustaining demurrer to third amended complaint where plaintiff failed to provide more than highly generalized description of alleged secrets; "the complainant should describe the subject matter of the trade secret with sufficient particularity to separate it from matters of general knowledge in the trade or of special knowledge of those persons who are skilled in the trade, and to permit the defendant to ascertain at least the boundaries within which the secret lies."); AMP, Inc. v. McCaughey, 38 Pa. D. & C.2d 109, 118 (Pa. Ct. Cm. Pl. 1966) ("In requiring a more specific pleading, we are not indicating that plaintiff must plead trade secret processes, trade secrets, or the like. However, it is apparent that plaintiff can provide defendants with at least an identification and nature of the products, persons, time and places involved, so as to permit them to know that against which they are to prepare a defense."). 26 See, e.g., Automed Tech., Inc. v. Eller, 160 F. Supp. 2d 915, 921 (N.D. Ill. 2001) (general references to categories of alleged secrets sufficient to withstand motion to dismiss, but "plaintiff will ultimately need to identify which specific designs, software or research defendant allegedly misappropriated."); Kosower v. Gutowitz, No. 00 Civ. 9011(JGK), 2001 WL 1488440, at *8 (S.D.N.Y. Nov. 21, 2001) (general references to claimed secrets permissible because facts as alleged must be treated as true; motion to dismiss denied); Labor Ready, Inc. v. Williams Staffing, L.L.C, 149 F. Supp. 2d 398, 411 (N.D. Ill. 2001) (denying Rule 12(e) motion because general identification of alleged secrets met notice pleading standards; noting that more would be required at the summary judgment stage;); Cinebase Software v. Media Guaranty Trust, Inc., No. C98-1100 FMS, 1998 WL 661465, at *7-8 (N.D. Cal. Sept. 22, 1998) (on motion to dismiss, identification "in general terms" sufficed, but "plaintiff will have to identify its alleged trade secrets with much greater particularity in order to prevail on this claim"); Savor, Inc. v. FMR Corp., No. CIV.A. 00C10-249JRS, 2002 WL 393056, at *4 (Sup. Ct. Del. Mar. 14, 2002) (dismissing third amended complaint with prejudice where plaintiff did not provide any detail for allegedly misappropriated trade secrets, even though complaint filed partially under seal and plaintiff pointed to attached documents, claiming "whole process" as secret), rev'd, 812 A.2d 894, 897 (Del. 2002) (reversing based on liberal pleading standards).
25

79

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW

Document 153-9

Filed 05/19/2008

Page 14 of 35 [2006

NORTHWESTERN JOURNAL OF TECHNOLOGY AND INTELLECTUAL PROPERTY

the allegations provide the defendant with notice as to the substance of the claims." 27 In some cases where courts have required more detail in a complaint, it is unclear whether the amended complaint would be filed under seal, though it would seem that a confidential or redacted filing would be appropriate. 28 B. Requests for Injunctive Relief ¶44 Many trade secret cases begin with a rushed request for a temporary restraining order or preliminary injunction. At that stage, trade secret plaintiffs sometimes make only vague references to the alleged secrets, without providing details. As a result, courts have repeatedly used the identification requirement as a ground to deny the request. 29 Given the courts' general reluctance to enjoin a defendant's
See Dick Corp. v. SNC-Lavalin Constr., Inc., No. 04 C 1043, 2004 WL 2967556 (N.D. Ill. Nov. 24, 2004) (on motion to dismiss and for a more definite statement, general reference to "Joint Venture Data" sufficed). 28 See Combined Metals of Chi. Ltd. Part. v. Airtek, Inc., 985 F. Supp. 827, 832 (N.D. Ill. 1997) (to prevent cross-plaintiff from modifying claims later, and to provide fair notice, court ordered party to identify "specific, concrete secrets" in amended cross-complaint). 29 See, e.g., MAI Sys. Corp. v. Peak Computer, Inc., 991 F.2d 511, 522-23 (9th Cir. 1993) (reversing preliminary injunction in part where plaintiff failed to "specifically identify" alleged secrets claimed to reside within a "diagnostic software and operating system"); Litton Sys., Inc. v. Sundstrand Corp., 750 F.2d 952, 956 (Fed. Cir. 1984) (applying apparent federal common law to affirm denial of preliminary injunction where, among other things, "Litton's Achilles Heel on this record is its insistence on postponing identification or description of such a broad universe of thousands of unidentified trade secrets"); Cataphote Corp. v. Hudson, 422 F.2d 1290, 1296 (5th Cir. 1970) (noting that trial court had correctly denied earlier permanent injunction request because plaintiff had no first "specifically identified and proved" its alleged secrets); Internet Inc. v. Tensar Polytech., Inc., No. Civ. 05-317RHKAJB, 2005 WL 2453170, at *6 n.8 (D. Minn. Oct. 3, 2005) (denying request for preliminary injunction when, inter alia, plaintiff did not "identif[y] specific products that have been successfully modified with the use of information from Brand, much less with the use of specific information alleged to be Internet's trade secrets"); Diamond Power Int'l, Inc. v. Clyde Bergemann, Inc., 370 F. Supp. 2d 1339, 1346 (N.D. Ga. 2005) (denying motion for preliminary injunction when plaintiff claimed secrets in dry hub carriage for sootblower cleaning system but could not identify specific items claimed as trade secrets); Hypred S.A. v. Pochard, No. Civ.04-2773(JNE/JGL), 2004 WL 1386149, at * 5 (D. Minn. June 18, 2004) (rejecting trade secret portion of requested preliminary injunction because plaintiff failed to identify specific alleged secrets and claimed only generic categories of "product formation and manufacturing secrets"); Vital State Can., Ltd. v. Dreampak, LLC, 303 F. Supp. 2d 516, 521-23 (D.N.J. 2003) (denying preliminary injunction where plaintiff identified "overbroad and vague" technology categories, and repeatedly "shifted" its attempted identifications); Compuware Corp. v. IBM, No. 02-CV-70906, 2003 WL 23212863, at *6-7 (E.D. Mich. Dec. 19, 2003) (rejecting motion for preliminary injunction where plaintiff claimed secrets in "unique combination of functions and capabilities" in certain software, "customer preferences," product documentation, and source code; court rejected claims, noting "As for its source code, Compuware has for whatever reason failed to identify any specific lines of source code that have been taken by IBM."); Motorola, Inc. v. Dbtel Inc., No. 02 C 3336, 2002 WL 1610982, at *16-17 (N.D. Ill. July 22, 2002) (plaintiff offered somewhat detailed testimony making general claims about several aspects of its technology, but court found that evidence "confusing, inconsistent, and lacks specificity;" court was unable to tell what exactly plaintiff wanted to enjoin, particularly where "many aspects of Motorola's process for creating a cellular phone involve publicly available equipment and standards," and refused to issue injunction); FSI Int'l v. Shumway, No. CIV.02-402RHKSRN, 2002 WL 334409, at *9 (D. Minn. Feb. 26, 2002) (same where plaintiff sought temporary restraining order but offered only generalized categories instead of identifying alleged secrets); Newleaf Designs, L.L.C v. Bestbins Corp., 168 F. Supp. 2d 1039, 1043-44 (D. Minn. 2001) (same where plaintiff failed to identify several alleged secrets); Am. Sci. & Eng'g Inc. v. Kelly, 69 F. Supp. 2d 227, 238 (D. Mass. 1999) (preliminary injunction denied where plaintiff failed to identify alleged secrets in collimator device); Visionair, Inc. v. James, 606 S.E.2d 359, 364 (N.C. Ct. App. 2004) (affirming denial of request for preliminary injunction where plaintiff "has failed to identify
27

80

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW Vol. 5:1]

Document 153-9

Filed 05/19/2008

Page 15 of 35

Charles Tait Graves and Brian D. Range

employment or business on an expedited proceeding without a careful review of the evidence, it is not surprising that plaintiffs' failures to identify the information at issue provides a ready reason for a denial. ¶45 Of course, some trade secret plaintiffs do provide a sufficient identification of their claims when seeking injunctive relief. 30 Indeed, one would expect that in egregious cases ­ where, say, a former employee downloaded key non-public files the day before resigning ­ a plaintiff should easily be able to make a precise identification of the alleged secrets to support its request. ¶46 At least one court has required little identification on a request for injunctive relief. That court held that "source code" was a sufficient identification on a request for a preliminary injunction, and attacked the defendant for considering its own software to be confidential while also seeking a better identification of the plaintiff's claim. 31 The court may have conflated the separate considerations of whether or not alleged trade secret is truly secret and whether or not the plaintiff adequately identified the alleged trade secret,

with any specificity the trade secrets allegedly misappropriated, mentioning only broad product and technology categories."); IBM Corp. v. Seagate Tech., Inc., 941 F. Supp. 98, 100 (D. Minn. 1992) (same where plaintiff admitted that its descriptions of alleged disk drive secrets "do not purport to define precisely the boundaries of each trade secret."); CVD, Inc. v. Raytheon Co., Civil Action No. 81-2216-S, 1981 WL 2162 (D. Mass. Dec. 3, 1981) (request denied where plaintiff failed to identify alleged secrets); Lee Pharms. v. Den-Mart, Inc., 1976 WL 21026 (C.D. Cal. 1976) (somewhat confusing early case where court delayed hearing on requested preliminary injunction in part because plaintiff had refused to identify alleged secrets); Analog Devices, Inc. v. Michalski, 579 S.E.2d 449, 453-54 (N.C. Ct. App. 2003) (affirming denial of motion for preliminary injunction where plaintiff claimed combination trade secret in chip designs, submitted documents and schematics, and listed general concepts such as "the overall design and implementation of Analog's 94xx products"; Court stated "Analog invites this Court to acknowledge the existence of trade secrets in the submitted information without bearing the burden of identifying those trade secrets. We will not read into Analog's claims specific identification of devices worthy of trade secret protection when it is Analog's burden to come forward with evidence of such devices."); Grow Co., Inc. v. Chokshi, 2006 WL 551367, at *4 (N.J. Super. Ct. Mar. 3, 2006) (unpublished) (motion to reconsider denial of preliminary injunction denied where plaintiff "fails to specifically identify a trade secret" in "processes and formulations"); Iron Age Corp. v. Dvorak, 880 A.2d 657, 665-66 (Pa. Ct. App. 2005) (denying preliminary injunction based on "inevitable disclosure"-type theory denied, in part because plaintiff failed to identify allegedly confidential information at issue); Southwest Research Inst. v. Keraplast Tech., Ltd., 103 S.W.3d 478, 483 (Tex. Ct. App. 2003) (vacating temporary restraining order where plaintiff "failed to identify any specific trade secret that should be protected," and instead broadly claimed that everything defendant learned from plaintiff was secret). 30 See e.g., Whyte v. Schlage Lock Co., 101 Cal. App. 4th 1443, 1453 (2002) (holding that plaintiff sufficiently identified alleged secrets during request for temporary restraining order though combination of generic categories and "descriptions made pursuant to" California Code of Civil Procedure section 2019.210 and other pleadings; published opinion did not quote from the latter documents, no doubt to preserve the claimed secrecy, thus leaving unclear what level of precision was supplied); Lowell Anderson, Litigation Issues, in TRADE SECRETS PRACTICE IN CALIFORNIA § 11.21 (2d ed. Supp. 2005) ("Unfortunately for practitioners seeking guidance on §2019(d), the court based its ruling in part on information contained in the plaintiff's §2019(d) statement, which was not disclosed in the published opinion."). At least two subsequent federal cases have incorrectly assumed that Whyte stood for the proposition that California's section 2019.210 allows identification by generic categories ­ though the subsequent Advanced Modular Sputtering case, discussed below, mooted the question by setting forth the California standard. See, e.g., Pixion, Inc. v. Placeware, Inc., 421 F. Supp. 2d 1233, 1242 (N.D. Cal. 2005); Excelligence Learning Corp v. Oriental Trading Co., Inc., No. 5:03-CV-4947 JF (RS), 2004 WL 2452834, at *3-4 (N.D. Cal. June 14, 2004). 31 See Cisco Sys., Inc v. Huawei Tech., Co., 266 F. Supp. 2d 551, 555-56 (E.D. Tex. 2003) (granting preliminary injunction; no sign in published opinion that plaintiff provided specific identification of alleged software code trade secrets).

81

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW

Document 153-9

Filed 05/19/2008

Page 16 of 35 [2006

NORTHWESTERN JOURNAL OF TECHNOLOGY AND INTELLECTUAL PROPERTY

because it did not provide a separate analysis for identification and then secrecy in sequential order. ¶47 An issue related to injunctive relief is whether the court itself properly identifies the alleged secrets in the order so that the defendant knows exactly what it is forbidden from using. Some appellate courts have reversed orders for injunctive relief on this basis. 32 It stands to reason that if a trade secret plaintiff fails to provide a precise identification of its alleged secrets, a court will not be able to issue an appropriately specific order against the defendant. C. The Pre-Discovery Stage If a defendant does not challenge the specificity of the complaint, and if the plaintiff does not seek immediate injunctive relief, the next phase of the case where identification disputes arise is the moment when the plaintiff seeks to commence discovery on its trade secret accusations. Several jurisdictions have required that trade secret plaintiffs make at least a "reasonably particular" identification of their alleged secrets as a condition precedent to seeking discovery from the defendant. In those jurisdictions, pre-discovery identification disputes are common. ¶49 Courts in several states (or federal courts applying state law) have required a pre-discovery identification procedure by common law: Delaware, 33 Illinois, 34 Massachusetts, 35 Minnesota ­ which requires the most detail 36 ­ and possibly Florida. 37 ¶48
See, e.g., Corning Inc. v. Picvue Elec., Ltd., 365 F.3d 156 (2d Cir. 2004) (remanding where trial court's preliminary injunction was too vague under FRCP Rule 65; impossible to tell from order what information was subject of the injunction); Roton Barrier, Inc. v. Stanley Works, 79 F.3d 1112, 1121-22 (Fed. Cir. 1996) (two post-judgment inunctions overbroad under Rule 65(d) for failure to specify exactly what information was barred from use); American Can Co. v. Mansukhani, 742 F.2d 314, 333 (7th Cir. 1984) (reversing preliminary injunction in part because order was too vague under Rule 65(d); order barred used of "compositionally similar" inks which could be "attributed principally" to the plaintiff's trade secrets without identifying precisely what was barred); E.W. Bliss Co. v. Struthers-Dunn, Inc., 408 F.2d 1108, 1114-15 (8th Cir. 1969) (setting aside preliminary injunction where order was "excessively broad" under FRCP 65(d) and listed only high-level technical categories instead of specifying exactly what information was subject of injunction); Microstrategy, Inc. v. Business Objects, S.A., 331 F. Supp. 2d 396, 431 (E.D. Va. 2004) (in post-trial, permanent injunction following finding of misappropriation, plaintiff's proposed injunction "too broad" in scope; limiting injunction to "specific documents identified as trade secrets"). 33 See, e.g., Leucadia, Inc. v. Applied Extrusion Tech., Inc., 755 F. Supp. 635, 637 (D. Del. 1991) (denying discovery against defendant because plaintiff's statement too high-level); Miles Inc. v. Cookson Am., Inc., Civ. A. No. 12,310, 1992 WL 136381, at *1 (Del. Ch. June 16, 1992) (unpublished) (statement generally outlining alleged secrets insufficient to take discovery against defendant); Magnox v. Turner, Civ. A. No. 11951, 1991 WL 182450, at *1 (Del. Ch. Sept. 10, 1991) (unpublished) (same where plaintiff served discovery without first making identification of alleged secrets with reasonable particularity); Engelhard Corp. v. Savin Corp., 505 A.2d 30, 33 (Del. Ch. 1986) (same where plaintiff failed to make required statement with reasonable particularity); Data Gen. Corp. v. SCI Sys., Inc., Civ. A. No. 5662, 1978 WL 22033, at *2 (Del. Ch. Nov. 27, 1978) (unpublished) (early case allowing defendant to file papers with court stating which alleged secrets in plaintiff's identification it was "unable to comprehend" in order to rule on motion for protective order). 34 See Automed Tech., Inc. v. Eller, 160 F. Supp. 2d 915, 925-26 (N.D. Ill. 2001) (citing Delaware law to hold that plaintiff must identify alleged secrets with "reasonable particularity" before taking discovery "so that we can evaluate the relevance of plaintiff's discovery and address any objections."). 35 See L-3 Comm. Corp. v. Reveal Imaging Tech., Inc., 2004 WL 2915743, at *13 (Mass. Sup. Ct. Dec. 2, 2004) (holding that plaintiff must specifically identify alleged secrets before commencing discovery). This decision is only an unpublished trial court order, however, so the status of pre-discovery identification remains unclear in Massachusetts.
32

82

KUWAYTI EXHIBIT H

Case 5:07-cv-05248-JW Vol. 5:1]

Document 153-9

Filed 05/19/2008

Page 17 of 35

Charles Tait Graves and Brian D. Range

In addition, a federal court in New York has addressed a defendant's argument that the plaintiff should not be able to start discovery on the trade secret claim without first making an identification of the alleged secrets by sending the dispute to a discovery master, but it did not attempt to craft a rule of general application to other cases. 38 ¶50 California is the only state to codify a pre-discovery identification rule by 39 statute. California's legislature enacted what is now Code of Civil Procedure section 2019.210 in 1985 as part of the state's Uniform Trade Secrets Act. The impetus for the statute came from a state bar memorandum to the state legislature proposing the text of the statute and noting the discovery abuses that trade secret plaintiffs often engage in when they fail to identify the alleged secrets: One area not addressed by the Uniform Act is the area of plaintiff's abuse in initiating trade secret lawsuits for the purpose of hara