The Attorney General has written an amicus brief in search of a lawsuit. He must continue searching. What he has written certainly has no application here. He defends a fictitious injunction that restricts only the distribution of computer code and only over the Internet. He defends a fictitious trade secret that he imagines was subject only to “an appearance” on the Internet. He seeks to import into this case the record of a different case, to which Bunner was not a party, in order to prove, among other things, Bunner's knowledge and motivations. And like the other amici curiae supporting DVDCCA, he seeks to use this case as a vehicle for amending the California Uniform Trade Secrets Act, conceding that in its present form the statute cannot justify the preliminary injunction.

    Once the thicket of factual errors and legal confusions is cleared away, the brief reveals itself to be of no assistance to this Court in resolving the very narrow issues presented here.




        1.     The Preliminary Injunction Is Not Merely a Bar to the Republication of Computer Code on the Internet

The preliminary injunction prevents Bunner from “[p]osting or otherwise disclosing or distributing, on their websites or elsewhere, the DeCSS program, the master keys or algorithms of the Content Scrambling System ('CSS'), or any other information derived from this proprietary information.” [AA712 (emphasis added)] By its own terms, therefore, the ban applies not only to computer code, but also to any other information derived from the purported trade secrets regardless of the language or form of the republication or the “functional capability” of the republished information. The ban applies not only to republication on the Internet, but elsewhere, thus encompassing any medium. Thus, the Attorney General's characterization of the preliminary injunction as “prohibiting only the dissemination of computer code,” [AG at 20] its assertion that “the injunction restrained nothing more than Bunner's dissemination of DeCSS and CSS on the Internet,” [AG at 21] and its claim that “[t]he only positive duty created by the injunction is the requirement that DeCSS itself (or the CSS encryption algorithm and master keys if they are separated from DeCSS) not be posted on a web site,” [AG at 36] are all in error. These errors are material, indeed central, to the Attorney General's position as amicus. The Attorney General relies solely on the purported “functionality” of computer code posted on the Internet, the only expression he believes is at issue, to justify its proposed constitutional test.

        2.     The Preliminary Injunction Was Not Intended to Protect Against Piracy of Copyrighted Material

    The Attorney General argues that the preliminary injunction furthers a governmental interest in combating piracy [AG at 34-36] and mischaracterizes the constitutional issues as whether a “free speech claim can[] be used to facilitate the wholesale piracy of DVDs via the Internet.” [AG at 2] But DVDCCA itself admits that the preliminary injunction is not aimed at fighting copyright infringement: “[N]either the trade secrets statute nor the preliminary injunction are focused directly at preventing copyright infringement. Rather, both are directed at protecting and preserving DVD CCA's trade secrets. . . . and the harm that would stem from the loss of those trades secrets can be measured, in part, in the loss of effective protection for the copyrighted works on DVD. But that measure of harm does not become the purpose of the statute for First Amendment purposes. Instead, . . . the statute was applied in this case to preserve DVD CCA's trade secrets, not to serve as a preemptive strike against copyright infringers.” [DVDCCA's Reply to Amicus Briefs at 7-8]   (See footnote 1) 


    The Attorney General proposes that this Court adopt “a multi-factor test to determine whether the dissemination of a trade secret on the Internet by independent third parties has fairly destroyed its status.” [AG at 7] Regardless of the propriety of writing this novel test into trade secret law instead of following the Legislature's plain command, the Attorney General makes several errors in applying it here. Indeed each of the four “factors” upon which the Attorney General bases its conclusion that the trade secret has survived its widespread dissemination are false. [AG at 8]

        1.     DVDCCA Made No Effort to Remove the Trade Secret from the Public Domain until it Filed this Lawsuit Almost Three Months after the Trade Secret First Appeared on the Internet

The Attorney General relies on “DVDCCA's widespread efforts to contact web sites nationwide and request they cease posting DeCSS.” [AG at 8] However, the pages of the record cited [AA344-47] contain no evidence of any effort by DVDCCA to contact web sites. Rather the record reflects only an intention, in the future, to contact the defendants who had already removed DeCSS from their websites. [AA346 (“We will contact, via electronic mail, each defendant who appears to have removed the DeCSS program from their web site . . .”); AA347 (“We will contact, via electronic mail, each Doe defendant who appears to have removed the DeCSS program from their web site . . .”)] DVDCCA did not attempt to contact any of the defendants prior to serving them process on December 27, 1999, almost three months after DeCSS first appeared on the Internet. [AA082]   (See footnote 2) 
        2.     There Is No Evidence in the Record of a “widespread awareness on the Internet in November of 1999 that the posting of DeCSS had become illegal and that plaintiffs were now contacting web sites to request that they remove DeCSS.” [AG at 8]

    Rather than evidencing a “widespread awareness,” the evidence shows at most a heated public debate among anonymous authors as to the legality of the reverse engineering that allegedly was at the origin of DeCSS. Within that debate, the majority sentiment was that there was no illegal or improper conduct. [AA348-470] There is no evidence of a “widespread awareness” that DVDCCA was contacting web sites, presumably because, as discussed above, DVDCCA did no such thing.
        3.     The Evidence Demonstrates That the CSS Encryption Algorithm or Master Key Could “be discovered by independent efforts aside from a review of DeCSS code.” [AG at 8]

    The Attorney General asserts that “testimony from Bunner's own expert” supports a contrary finding. [AG at 8] However, the Attorney General cites not to the testimony of Bunner's expert but to the testimony of the President of DVDCCA. [AA484-86] Bunner's expert actually testified that DeCSS could have been created by independent invention and not reverse engineering. [AA224-56] He testified that “the mere publication of a valid key player cannot be taken as proof that misappropriation must have occurred,” and that “there are a variety of methods that can achieve the cracking of the encryption scheme without ever seeing or agreeing to a Xing license agreement.” [AA227-28]
        4.     The Record Reflects That “the actual trade secrets contained in DeCSS, the proprietary CSS encryption algorithm or master keys, had been widely disseminated or become widely known among the public.”[AG at 8]
    The Attorney General, citing only to DVDCCA's reply brief, relies on the “absence of any indication that” the trade secrets had been widely disseminated, and claims that the trade secret was not destroyed “merely because it made an appearance on the Internet.” [AG at 7] However, it is undisputed that DeCSS appeared on over 118 web pages based in at least 11 states and 11 countries. [AA079] Nor was the “appearance” of DeCSS as fleeting as the Attorney General would have this Court believe. It is undisputed that DeCSS remained widely available for almost three months before DVDCCA filed this lawsuit. The San Jose Mercury News, reporting on December 28, 1999, on the filing of the lawsuit, provided a link to DeCSS and noted that “dozens of sites” continued to publish DeCSS. [AA216-17]   (See footnote 3)  Thus a consideration of the “time during which the secret has been exposed” [AG at 7-8] also weighs in favor of the secret no longer existing.
        5.     Bunner Has Never Claimed That His Own Republication of DeCSS “Destroyed” the Trade Secret by Making it Public Knowledge

The Attorney General, knocking down a straw man of his own construction, contends that “Bunner's wrongful acts in publishing the code on the Internet did not strip the code of its trade secret status” [AG at 3] and concludes that CSS' trade secret status was not destroyed by Bunner's publication of DeCSS on the Internet.” [AG at 8] But this is not the issue. Bunner has never argued that his own republication made DeCSS publicly known; rather, the publication and distribution of DeCSS by hundreds of other web sites, including CNET's “,” made DeCSS public knowledge long before the trial court issued the preliminary injunction.

    The Attorney General's argument that the trade secrets were in fact improperly acquired is based wholly on several critical misreadings of the record.
        1.     The Record Contains Uncontradicted Evidence That the Motivation for Both the Reverse Engineering of CSSs and Bunner's Republication of DeCSS Was to Make DVD Movies Interoperable with the Linux Computer Operating System

    The Attorney General asserts without support that the purpose of the reverse engineering that created DeCSS “was to create a decryption program whose sole use is to bypass the encryption program on DVDs and pirate their digital content.” [AG at 10] But the record contains no evidence that DeCSS, despite its widespread distribution, has ever been used to commit piracy, much less that this is its sole use. To the contrary, the evidence is uncontradicted that the motivation for both the reverse engineering of CSS and for Bunner's subsequent republication of DeCSS was to make DVD movies interoperable with the Linux computer operating system. University of California-Berkeley Computer Science Professor David Wagner testified that “My observation of the process-I studied the communications back and forth between researchers as the work was unfolding-was that DeCSS was built, not with the purpose to violate copyright or other intellectual property protections on DVDs, but rather to allow users of Linux and other publicly supported operating system to play legitimately obtained DVDs.” Bunner testified: “When I learned of the 'deCSS' program, I believed it would be useful to others in that it would enable 'Linux' users to use and enjoy 'DVDs' available for purchase or rental in video stores.” [AA287]   (See footnote 4) 
        2.     The Record Does Not Support a Finding That Alleged Reverse-Engineering of the Xing Program by the Creator of DeCSS Was an Actionable Breach of an Agreement That Amounted to “improper means.” [AG at 9,19]

    As the trial court noted when addressing the “improper means” issue in its preliminary injunction order, “Plaintiff's case is problematic at this pre- discovery stage.” [AA712] The record is not just contested on this point. In many instances it is utterly silent on a multitude of facts that are essential to support any determination that the alleged reverse engineering of CSS was accomplished by improper means. Among these are the following:
.     The 1998 version   (See footnote 5)  of Xing license agreement provided only that “The Product in source code is confidential and Xing's protected trade secret.” [AA339 ¶5 (emphasis added)] But there is no evidence that Xing distributed source code to consumers. Xing claimed no trade secrets in the object code version that was distributed to consumers and allegedly used to create DeCSS. The agreement sets forth no trade secret interest by DVDCCA. DVDCCA presented no evidence or argument that it has standing to enforce Xing's trade secret rights in this action, or that it has standing to enforce the Xing agreement.
.     Xing sold its program to computer original equipment manufacturers (OEMs), who then sold the Xing software preinstalled on their machines. [AA339-40 ¶ 6] Because the Xing click-through license agreement appears only at the time of initial installation, [AA340 ¶ 7] a consumer purchasing a computer with Xing software preinstalled by an OEM would not see the click-through license agreement, much less assent to it. Although Xing purportedly as of 1998 required its OEMs to pass on the terms of the Xing license to their customers (presumably in writing), [AA339-40 ¶6] no evidence was presented that this in fact did occur. The former Xing employee whose testimony DVDCCA presented was careful not to opine on whether or not consumers were given any opportunity to assent to the terms of the agreement. [AA339 ¶6] No evidence was presented that whoever allegedly used the Xing program to create DeCSS ever assented to an agreement.

    There is no evidence that DeCSS has only “minimal expression.” [AG at 37] To the contrary the record contains evidence of the great expressive value of DeCSS to cryptographers, computer scientists and the general public. [AA259-61; 282-84] DVDCCA has never disputed that DeCSS “is speech.” [RT39:19-20]

    This Court has cautioned litigants about the very limited scope of a preliminary injunction appeal. “At this initial stage in the proceeding, the scope of our inquiry is narrow. . . . [A]n order granting or denying interlocutory relief reflects nothing more than the superior court's evaluation of the controversy on the record before it at the time of its ruling.” People ex rel. Gallo v. Acuna, 14 Cal. 4th 1090, 1109 (1997) (emphasis original). “[E]vents occurring after the superior court's issuance of the preliminary injunction” are “[i]rrelevant.” Thompson v. Dep't of Corrections, 25 Cal.4th 117, 128-29 n.5 (2001).
    The Attorney General ignores this limitation and seeks to import the entire record of a federal copyright case, Universal City Studios v. Corley, 273 F.3d 429 (2d. Cir. 2001), not only to obscure DVDCCA's factual presentation to the superior court but to impute to Bunner the knowledge, motives and conduct of the defendants in that case.
    The limited factual record that was before the trial court clearly does not support any of the legal positions taken by the Attorney General before this Court. This case is an inappropriate vehicle for the sweeping revisions of trade secret and reverse engineering law that the Attorney General proposes.

    Central to the Attorney General's brief is the extraordinary and unprecedented proposal that reverse engineering, if done for certain purposes he deems “bad,” should be an “improper means” of acquiring a trade secret. In advancing this re-writing of the California UTSA, the Attorney General proffers that the “statutory language is less than pellucid” [AG at 14] and that “neither an examination of the plain language nor the legislative history of the UTSA conclusively resolve the scope of the UTSA's reverse engineering exemption.” [AG at 14] Furthermore, “Neither the legislative history surrounding the adoption of the UTSA by the California Legislature in 1984, nor the original drafting history of the UTSA itself, sheds any light on this subject. Equally, no decisions apparently exist anywhere in the 44 states governed by the UTSA which speak to this issue.” [AG at 15]
    This Court has no occasion in this case to consider the Attorney General's proposal. The key determination this Court must make in this regard in independently reviewing the record is not whether the alleged reverse engineering of CSS was “improper” under the California UTSA, but whether Bunner “should have known” that it was “improper.”
    Given the uncertain and inadequate state of the law as described by the Attorney General, no court could reasonably find that Bunner, assuming he even knew of the alleged reverse engineering, “should have known” that it was “improper.” As the Attorney General acknowledges, “from the face of the statute, it may appear that the Norwegian hacker's conduct is exempted from the UTSA's prohibition.” [AG at 10] Bunner cannot be charged with anticipating that this Court would rewrite the reverse engineering authorization in a way contemplated by neither the statutory language, the legislative history nor any judicial interpretation of the UTSA in the 44 states in which it has been adopted.
    Nor need this Court resolve the choice of law question. [AG at 19] Under either Norwegian law, which the trial court was “not well-positioned to interpret,” [AA714] or California law, which the Attorney General asserts does not in its present form provide a definitive answer, Bunner should not have known that the alleged reverse engineering that allegedly led to the discernment of the CSS trade secrets was “improper.”


        1.     The Plain Language of the UTSA Authorizes Reverse Engineering as a Permissible Way of Discerning Trade Secrets.

    The UTSA expressly and unqualifiedly provides: “Reverse engineering . . . alone shall not be considered improper means.”   (See footnote 6)  Civil Code § 3426.1. The Attorney General concedes that “reverse engineering” has an unambiguous meaning. The UTSA does not distinguish between “good” and “bad” reverse engineering on the basis of the purpose for which the reverse engineering is performed. It merely describes a technical process.
    That should be the end of the question. “If [a statute] is clear and unambiguous our inquiry ends. There is no need for judicial construction and a court may not indulge in it.” Diamond Multimedia Systems, Inc. v. Superior Court, 19 Cal. 4th 1036, 1047 (1999).
    If more is needed, the legislative history makes clear that the Legislature used the term “reverse engineering” in its ordinary sense, without any limitation based on the purpose for which the reverse engineering is performed. “Proper means include: . . . Discovery by 'reverse engineering,' that is, by starting with the known product and working backward to find the method by which it was developed. The acquisition of the known product must of course, also be by fair and honest means, such as purchase of the item on the open market, for reverse engineering to be lawful;” Sen. Com. on Jud., Rep. on Assem. Bill 501, 8 Sen. Jour. (1983-1984 Reg. Sess.) p. 13883. The Attorney General denies the existence of this legislative history, asserting “the legislative history surrounding the adoption of the UTSA . . . [does not] shed any light on this subject.” [AG at 15]
    In proposing to rewrite the UTSA to read “Certain reverse engineering shall not be considered improper means,” the Attorney General frankly dispenses with any fiction that he is seeking the actual intent of the Legislature in enacting the UTSA. He admits that his proposal is not part of the UTSA because the “20-20 hindsight” necessary to have conjured it up was “lacking among the drafters of the UTSA.”   (See footnote 7)  [AG at 16]
        2.    Federal and State Intellectual Property Policy Also Favor An Unrestricted Reverse Engineering Authorization

    Neither federal nor state intellectual property law supports the Attorney General's proposed amendment to narrow reverse engineering rights in trade secret cases. Indeed such an unprecedented construction would be contrary to the delicate balance struck in federal and state intellectual property policies.
    Reverse engineering is an accepted check against monopolistic practices in a variety of contexts. In Sega v. Accolade, 977 F.2d 1510 (9th Cir. 1992), the court recognized that a prohibition on reverse engineering of a computer program would grant “a de facto monopoly over [the unprotected] ideas and functional concepts [in the program].” Id. at 1527. The court ruled that to get a monopoly on such ideas and functional concepts, a creator needs to seek patent protection. Id. at 1526. Accord Sony Computer Entertainment, Inc. v. Connectix Corp., 203 F.3d 596 (9th Cir. 2000) (upholding as fair use the reverse engineering of multi-media works in order to build a competing platform).    Indeed, the U.S. Supreme Court ruled that a state law was preempted by the patent clause because it “prohibit[ed] the entire public from engaging in a form of reverse engineering of a product in the public domain.” Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U.S. 141, 160 (1989). And in Kewanee v. Bicron Corp., 416 U.S. 470, 489-90 (1974), the high Court distinguished trade secret laws from patent laws on the basis, in part, that “trade secret law does not forbid the discovery of the trade secret by fair and honest means, e.g., independent creation or reverse engineering.” Thus, the Court held: “Where patent law acts as a barrier, trade secret law functions relatively as a sieve.” Id.
    The California trade secret law has been interpreted similarly. In Chicago Lock Co. v. Fanberg, 676 F.2d 400 (9th Cir. 1982), the court held that if state trade secret law did not allow reverse engineering, it “would, in effect, convert the Company's trade secret into a state conferred monopoly akin to the absolute protection that a federal patent affords. Such an extension of California trade secrets law would certainly be preempted by the federal scheme of patent regulation.” Id. at 404.
        3.     The Reverse Engineering Here Would Not Be An Improper Means Of Acquiring A Trade Secret Even Under The Attorney General's Proposal

    Even under the Attorney General's proposal, reverse engineering is legitimate if done to achieve interoperability between different computer systems. [AG at 11] As set forth in section I above, the undisputed evidence in this case is that both the initial reverse engineering of CSS and Bunner's subsequent republication of DeCSS were done for the purpose of achieving the interoperability of DVD movies with the Linux operating system.
        4.     DVDCCA Has Not Established That It Is Likely To Show That Improper Means Accompanied The Reverse Engineering Of CSS

“Improper means” of trade secret acquisition under the UTSA do not include reverse engineering but only the breach of an independent legal duty (e.g., theft, bribery) or breach of a voluntarily-assumed duty of secrecy. Civil Code § 3426.1, subd. (a). The Attorney General, like DVDCCA, asserts that the purported Xing agreement creates this duty. As discussed above, the record does not support this conclusion.
    Moreover, no court has yet decided the enforceability of the purported reverse engineering restrictions in the Xing license agreement, and no court could intelligently do so on the basis of the very limited record here. Any decision of this question, in addition to considering the public policies in favor of reverse engineering and free competition, must also involve consideration of the first sale doctrine of copyright law and patent law, the preemptive scope of federal intellectual property law, and many other issues, none of which have yet been considered by any court in this lawsuit. See, e.g., Vault v. Quaid, 847 F.2d 255, (5th Cir. 1988) (holding that a mass-market consumer software license provision purporting to prohibit reverse engineering was preempted by federal law).   (See footnote 8)  As the Attorney General concedes, [AG at 10 n. 9] this interlocutory appeal is an inappropriate vehicle for resolving such legal issues.

    The Attorney General proposes “a multi-factor test to determine whether the dissemination of a trade secret on the Internet by independent third parties has fairly destroyed its status.” [AG at 7] As set forth above, when the record is properly read, the purported trade secret here fails this test.
    Nevertheless, there is no authority or reason to adopt this test. The UTSA's prohibition of injunctive relief for information no longer secret is absolute and mandatory, and is not conditioned on the manner in which the information became public: “Upon application to the court, an injunction shall be terminated when the trade secret has ceased to exist . . . .” Civil Code § 3426.2(a). Accord Kewanee Oil Co., 416 U.S. at 475 (“The subject of a trade secret must be secret, and must not be of public knowledge or of a general knowledge in the trade or business.”); Vacco Industries, Inc. v. Van Den Berg, 5 Cal.App.4th 34, 50 (1992) (“a trade secret is protectible only so long as it is kept secret”). As the legislative history to California's UTSA explains, “an injunction accordingly should terminate when a former trade secret . . . becomes generally known.” Unif. Trade Secrets Act § 2 cmt., reprinted in 14 Unif. Laws Annot. 450 (West 1990); Sen. Com. on Jud., Rep. on Assem. Bill 501, 8 Sen. Jour. (1983-1984 Reg. Sess.) p. 13883 (“the comments of the national conference with respect to the act reflect the intent of the Senate Committee on Judiciary in approving Assembly Bill 501”).   (See footnote 9)  The Legislature intended that “obtaining a trade secret from published literature” be a “proper means” of discovery. Legislative Committee Comment_Senate (1984 Addition), Civil Code §3426.1.
    The Attorney General further fabricates a distinction between “the Internet community,” in which there was “widespread awareness,” and “the public,” who supposedly knew nothing. [AG at 8] This distinction also runs afoul of the Legislature's intent in enacting the UTSA. As the legislative history makes clear, “widespread awareness” within a particular “community” that falls short of an awareness by the general public still destroys the trade secret. “The language 'not being generally known to the public or to other persons'does not require that information be generally known to the public for trade secret rights to be lost. If the principal person who can obtain economic benefit from information is aware of it, there is no trade secret. A method for casting metal, for example, may be unknown to the general public but readily known within the foundry industry.” Legislative Committee Comment _Senate (1984 Addition), Civil Code 3426.1.
    Even so, there is no factual basis in the record or elsewhere for the Attorney General's distinction between the Internet and the public and certainly no evidence that Bunner was not merely a member of the public. To the extent that the Attorney General seeks to treat every reader of a public news website like or every one who uses the Internet as a “member of the Internet community” owing a legal duty to DVDCCA, the argument must be rejected. “Once a trade secret is posted on the Internet, it is effectively part of the public domain. . . . [T]he party who merely downloads Internet information cannot be liable for misappropriation because there is no misconduct involved in interacting with the Internet.” Religious Tech. Ctr. v. Lerma, 908 F. Supp. 1362, 1368 (E.D. Va. 1995).

    The Attorney General's argument that intermediate scrutiny should be applied relies completely on the preliminary injunction being something it isn't: a restriction solely on computer code for the purpose of preventing function and only incidentally restricting speech. He further relies on DeCSS being something it isn't: a computer program with minimal expressive content. When these mistakes are corrected, the argument collapses.
    Bunner has already addressed in detail in his previous briefs that the preliminary injunction is neither a restriction on conduct nor content-neutral. Those arguments need not be restated here.
    Departing from DVDCCA and the other amici, however, the Attorney General proposes that this Court apply the intermediate scrutiny test of Madsen v. Women's Health Center, Inc., 512 U.S. 753, 765 (1994). That test asks: does the injunction challenge no more speech than necessary to serve a significant governmental interest?   (See footnote 10)  The Attorney General, again mistakenly assuming that the preliminary injunction applies only to computer code, also proposes two “refinements” from Madsen specific to computer code. First, does the injunction burden the expressive features of code more than is necessary? [AG at 27] Second, has the defendant advanced any reasonably practical technological alternatives to the injunction? [AG at 28]
    The analysis of these “refinements” reveals exactly why intermediate scrutiny is not the proper test here. The preliminary injunction burdens only the expressive features of the information. The Attorney General's argument that “the expression contained in the DeCSS computer code is incidental to its functional purpose of directly executing computer code” is not only contrary to the undisputed evidence in this case, as discussed above, but an improper statement of the constitutional test. A court must look to whether the restriction incidentally burdens speech, not whether the expression is incidental to function. See Madsen, 512 U.S. at 763 n.2 (stating its focus on “injunctions that may incidentally affect expression”); United States v. O'Brien, 391 U.S. 367, 376 (1968) (focusing on “incidental limitations of First Amendment freedoms”).   (See footnote 11) 
    The inquiry into “reasonably practical technological alternatives” is also a fruitless diversion. The preliminary injunction itself, as the trial court recognized, is not a “reasonably practical” way of keeping the purported trade secret secret. The purported trade secrets have already been so widely published, as the undisputed record reveals, that there is no “reasonably practical” way to salvage them. The preliminary injunction is as ill-suited to stuff the genie back in the bottle as an any “technological alternative.”
    Other reasons compel this Court to reject the Madsen test. Madsen addressed a permanent injunction issued after a full trial on a fully developed record and after a previous injunction was found to be insufficient. 512 U.S. at 758-62. And Madsen was analyzing a time, place and manner restriction and thus focused its analysis on ensuring that the defendants had adequate avenues to disseminate their message. Id. at 763 & n.2, 764-65, 768-75.   (See footnote 12)  Had the defendants been barred completely from expressing their message, as Bunner was here, the Court would have treated the injunction as a prior restraint. Id. at 763 n.2.
    Moreover, even if this Court were to apply Madsen or any other version of intermediate scrutiny, the preliminary injunction cannot survive. As explained in Bunner's previous briefs, the state interest in protecting this information from further republication is not strong enough to counterbalance Bunner and the public's liberty of speech.

    As explained in Bunner's previous briefs, the preliminary injunction must be subject to strict scrutiny for two independent reasons: one, it is content-based and a restriction solely on expression that wholly prevents Bunner from communicating the information at any time, in any medium, and in any language, see Turner Broadcasting Systems v. FCC, 512 U.S. 622, 642- 43 (1994) (“Nor will the mere assertion of a content-neutral purpose be enough to save a law which, on its face, discriminates based on content.”); and two, it is a prior restraint. See Alexander v. United States, 509 U.S. 544, 550 (1993) (“injunctions . . . are classic examples of prior restraints”).
    The Attorney General simply concurs with DVDCCA's argument that the preliminary injunction was not a prior restraint. The Attorney General cursorily notes that prior restraint law is applied only to pure speech and to content-neutral restrictions.   (See footnote 13)  [AG at 5 n.2]
    As every other court that has directly considered whether a preliminary injunction against a third party re-publisher of a purported trade secret could survive a First Amendment defense has applied prior restraint analysis,   (See footnote 14)  this perfunctory argument does this Court a great disservice. But moreover, it is incorrect. Even content-neutral injunctions, see Organization for a Better Austin v. Keefe, 402 U.S. 415, 417 (1971) (striking down injunction against “passing out pamphlets, leaflets or literature of any kind”), and restrictions on mixed speech and conduct, see National Socialist Party of America v. Skokie, 432 U.S. 43 713 (1977) may be prior restraints. Compare Madsen, 512 U.S at 763 n.2 (noting implicitly that many injunctions “that may incidentally affect expression . . . are 'prior restraints'”).

    By advocating for Madsen intermediate scrutiny, the Attorney General now agrees with Bunner and the amici supporting him, and DVDCCA and amici curiae MPAA et al. and Microsoft et al. that the trial court erred in not applying any First Amendment analysis at all.


    Ideally, an amicus brief brings new light and a fresh perspective to the case that helps illuminate the issues before the Court. The Attorney General's brief utterly fails to fulfill that purpose. Instead, its confused and erroneous statements of fact and law only obscure the issues before this Court.
DATED: August 22, 2002        Respectfully submitted,
                     THE FIRST AMENDMENT PROJECT

                by:    ________________________________
                    Attorneys for Andrew Bunner


    I certify that this brief complies with the type-volume limitation of California Rule of Court 14(c)(1). This brief is printed in 13 point Times New Roman font and, exclusive of portions exempted by Rule 14(c)(3), contains 6688 words.

                        David A. Greene

Footnote: 1
    One might excuse the Attorney General's confusion on this point. DVDCCA previously asserted that “the purpose of enjoining DeCSS was not to stop the spread of any ideas contained in the program; it was to prevent the improper use of DVD CCA's trade secrets for decrypting and illegally copying DVDs.” [Opening Brief at 21]

Footnote: 2
     The record does reflects that the Motion Picture Association (MPA), which does not own the purported trade secrets and is not a party to this action, did request that web site operators and Internet Service Providers take down certain web sites that allegedly provided information on how make illegal copies of DVDs and infringe copyrights. [RA015-024] These requests assert no trade secrets nor purport to be an effort to preserve anybody's trade secrets. Bunner did not receive one of these requests.

Footnote: 3
    Moreover, during the three months between the initial appearance of DeCSS on the Internet and the filing of the lawsuit, only 25 of the 66 removal requests issued by MPA were honored. [AA064]

Footnote: 4
    The Attorney General claims that the reverse engineering must have been improper because it was done without first informing DVDCCA or seeking a license from it. [AG at 13] A policy authorizing only reverse engineering consented to by the trade secret owner would surely contravene the clear intent of the reverse engineering provision of the UTSA to encourage innovation and fair competition.

Footnote: 5
    Although this case was filed in December 1999, DVDCCA presented only the 1998 version of Xing's license agreement and only the testimony of a former Xing employee whose job there ended in 1998. DVDCCA presented no evidence regarding the terms of the Xing license agreement in use at the time of the alleged reverse engineering in 1999 or the preliminary injunction in January 2000.

Footnote: 6
    The reverse engineering authorization was not part of the model UTSA adopted by the National Conference of Commissioners on Uniform State Laws. It was specially added by the California Legislature in recognition of the importance of reverse engineering to California. Compare Civil Code 3426.1(a) with Unif. Trade Secrets Act § 1(1), reprinted in 14 Unif. Laws Annot. 437 (West 1990).

Footnote: 7

    The Attorney General notes that a distinction between “good” and “bad” reverse engineering was made in three federal intellectual property statutes and in a Council of the European Union directive. Each of these was promulgated after the California UTSA. Whatever Congress or the European Union may have done after the UTSA was enacted is irrelevant to what the California Legislature intended when it enacted it. Nevertheless, the European Union has also declared that anti-decompilation clauses in software user contracts are null and void. See European Software Directive, art. 9(1).

    To further support his proposed reduction of the reverse engineering authorization, the Attorney General relies heavily on Samuelson & Scotchner,

The Law and Economics of Reverse Engineering, 111 Yale L.J. 1575 (May 2002). It bears noting, however, that the co-author of the article, Prof. Pamela Samuelson, is on the brief of Amici Curiae Intellectual Property Law Professors et al. which directly advocates for undiminished protection for the act of reverse engineering, and for the application of the reverse engineering authorization to the facts of this case.

Footnote: 8
    Numerous commentators have opined that such restrictions are unenforceable. Some argue that they conflict with copyright law. See, e.g., Charles R. McManis, The Privitization (or “Shrinkwrapping”) of American Copyright Law, 87 Calif. L. Rev. 173 (1999); David A. Rice, Public Goods, Private Contract and Public Policy: Federal Preemption of Software License Prohibitions Against Reverse Engineering, 53 U. Pitt. L. Rev. 543 (1992). Others believe they conflict with patent law. See, e.g., John E. Mauk, Note, The Slippery Slope of Secrecy: Why Patent Law Preempts Reverse Engineering Clauses in Shrinkwrap Licenses, 43 Wm. & Mary L. Rev. 819, 843 (2001). Others deem them to be an abuse of intellectual property rights. See, e.g., Mark A. Lemley, Beyond Preemption: The Law of Intellectual Property Licensing, 87 Calif. L. Rev. 111, 129 (1999). Yet others contend that such restrictions may be enforceable but only in negotiated agreements. See, e.g., David Nimmer, Elliot Brown & Gary N. Frischling, The Metamorphosis of Contract Into Expand, 87 Calif. L. Rev 17, 68 (1999); J.H. Reichman & Jonathan A. Franklin, Privately Legislated Intellectual Property Rights: Reconciling Freedom of Contract with Public Good Uses of Information, 147 U. Pa. L. Rev. 875, 939 (1999).

Footnote: 9
    Moreover, even if the Legislature had not acted to prohibit injunctions suppressing the republication of publicly available information, that possibility is foreclosed not only by the liberty of speech (for the reasons stated in Bunner's previous briefs and that of amici curiae ACLU et al.) but also by the preemptive effect of federal intellectual property law. “[I]deas once placed before the public without the protection of a valid patent are subject to appropriation without significant restraint.” Bonito Boats, 489 U.S. at 156. “ '[A]ll ideas in general circulation [are] dedicated to the common good unless they are protected by a valid patent.” Id. at 159-60.

Footnote: 10
    Quizzically, the Attorney General ignores Madsen's test for content- neutrality: would “similar conduct directed at a target having nothing to do with” the subject matter at hand have been enjoined? Id. at 763. In this case, as distinct from Madsen, in which the protestors would have been restricted even if they had been protesting peanut butter instead of abortion, the answer is clearly no.

Footnote: 11
    The Attorney General claims that Madsen's “precision of regulation” standard permits courts to “carefully balance the heightened First Amendment interests involved in cases where a program's expressive content is not merely incidental to its functional features against the substantial state interest of preventing the dissemination of a trade secret.” [AG at 27-28] But how? The test provides no relief at all when the restriction, unlike a time, place and manner restriction, flat out prohibits all communication of the information. In such cases, “precision of regulation” is a fiction.

Footnote: 12
    The Attorney General's claim that Bunner is free to republish the haiku version of DeCSS is false, based as it is on the Attorney General's misperception of the scope of the preliminary injunction. [AG at 37] Nor is the ability to discuss and criticize CSS an adequate substitute for disclosing the information itself. See Wilson v. Superior Court, 13 Cal.3d 652, 659 (1975) (rejecting the assertion that a preliminary injunction that allowed for a general discussion of the plaintiff's background but barred the dissemination of specific statements about him was not a prior restraint).

Footnote: 13
    The Attorney General also argues that “Content-neutral regulations that may implicate speech . . . are subject only to 'intermediate scrutiny.'” [AG at 20] Here, the preliminary injunction sought directly, not incidentally, to restrict the informational capacity of Bunner's republication. Compare Turner Broadcasting Systems, 512 U.S. at 662 (applying intermediate scrutiny to content-neutral restrictions that impose an “incidental restriction on speech”). Intermediate scrutiny must not be applied where, as here, “the communication itself is thought to be harmful.” United States v. O'Brien, 391 U.S. 367, 382 (1968).

Footnote: 14
    CBS, Inc. v. Davis, 510 U.S. 1315, 1316-18 (1994) (Blackmun, Circuit Justice in chambers); Oregon ex rel. Sports Mgt. News v. Nachtigal, 921 P.2d 1304, 1308-08 (Ore. 1996); Ford Motor Co. v. Lane, 67 F. Supp. 2d 745, 750, 752-53 (E.D. Mich. 1999); Religious Tech. Ctr. v. Lerma, 897 F. Supp. 260, 266-67 (E.D. Va. 1995). See also Garth v. Staktek Corp., 876 S.W.2d 545, 549-50 (Tex. App. 1994) (applying prior restraint doctrine to preliminary injunction against first party publishers).