United States District Court, N.D. California
May 1, 2014
JOBSCIENCE, INC., Plaintiff,
CVPARTNERS, INC., a California Corporation; SKIPAN, LLC, a Delaware Limited Liability Company; SKIPAN SAAS, LLC, d.b.a. TALENT ROVER, a Delaware Limited Liability Company; BRANDON METCALF, an individual; KENT GRAY, an individual; and DOES 1-100, Defendants.
ORDER GRANTING MOTION TO STRIKE MISAPPROPRIATION OF TRADE SECRETS CLAIM AND ORDER RE SOURCE-CODE COMPARISON
WILLIAM ALSUP, District Judge.
In this action involving allegations of copyright infringement and misappropriation of trade secrets for recruiting software, defendants move to strike the misappropriation of trade secrets claim and object to plaintiff's identification of alleged trade secrets. For the reasons stated herein, the motion is GRANTED.
Plaintiff Jobscience, Inc. develops and licenses recruiting software. Defendants are individuals and entities alleged to have copied Jobscience's software application, which was licensed to defendant CVPartners, Inc. Jobscience has obtained copyrights for versions of the code.
In September 2013, Jobscience commenced this action. In November 2013, Jobscience filed a first amended complaint alleging thirteen claims for relief. A January 2014 order granted in part and denied in part defendants' motion to dismiss. Notably, the misappropriation of trade secrets claim was dismissed because it was preempted by the Copyright Act and the complaint relied on nothing more than a vague reference to "software code, methods and other trade secrets" (Dkt. No. 29).
A February 2014 order then (among other things) granted plaintiff's motion to amend the first amended complaint to add a trade-secrets claim. The claim, however, could not cover software code because of the preemption problem. Because defendants raised serious challenges to the existence of trade secrets, Jobscience was ordered to file a detailed description of the alleged misappropriated trade secrets. At oral argument, counsel was warned to "be specific. If it's not specific, then [Defendants] can bring motions" (Feb. 27, 2014 Hr'g. Tr. 21-22). A companion order instructed defendants to turn over every version of the allegedly copied source code for a source-code comparison. Plaintiff chose and paid for an expert to analyze the code.
Plaintiff then filed a second amended complaint alleging copyright infringement, breach of contract, misappropriation of trade secrets, unfair competition, and constructive fraud (Dkt. No. 45).
Three weeks later, plaintiff filed a statement of trade secrets, organized as a three-page narrative with three schematics. Contrary to the instructions in the February 2014 order, plaintiff did not include a numbered list of the precise trade secrets with the specific elements for each trade secret. Instead, Jobscience alleged that it had "[REDACTED/], " derived "[REDACTED/]." The process was "not known to the general public" and Jobscience maintains its secrecy by requiring confidentiality agreements to access its products. Jobscience stated that the process has independent economic value as demonstrated through customers "paying to use it, " but the secrets "are not evident in a reading of [the] code." Jobscience does not claim that the existence of relational databases or their use of supporting business processes are trade secrets (Dkt. No. 48-4).
The instant defense motion seeks to strike plaintiff's misappropriation of trade secrets claim and objects to plaintiff's statement of confidential trade secrets.
On April 7, Jobscience filed a letter, appending the expert report of Kenneth Amron. Jobscience asked Mr. Amron to "determine whether [defendants'] Talent Rover code base exhibits evidence of code copying." Of the files Mr. Amron analyzed, he found "strong evidence of copying" (Dkt. No. 54).
On April 18, defendants filed a letter identifying deficiencies with Mr. Amron's report and qualifications. Defendants argued that Mr. Amron failed to perform a proper comparison of source code and argued that the "Court should either strike the copyright claim as a sanction for violating the February 28 Order or require Plaintiff to complete an actual source code comparison using CodeSuite" (Dkt. No. 62).
The fact discovery deadline is October 31, 2014. This order follows full briefing and oral argument.
1. MOTION TO STRIKE.
Even though defendants styled this motion as a "motion to strike, " they essentially move to dismiss the misappropriation of trade secrets claim based on inadequate disclosure in Jobscience's statement of trade secrets. Section 3426.1(d) of the California Civil Code defines trade secret as:
information, including a formula, pattern, compilation, program, device, method, technique, or process, that:
(1) Derives independent economic value, actual or potential, from not being generally known to the public or to other persons who can obtain economic value from its disclosure or use; and
(2) Is the subject of efforts that are reasonable under the circumstances to maintain its secrecy.
The February 2014 order stated (Dkt. No. 43):
Experience has shown that it is easy to allege theft of trade secrets with vagueness, then take discovery into the defendants' files, and then cleverly specify what ever happens to be there as having been trade secrets stolen from plaintiff. A true trade secret plaintiff ought to be able to identify, up front, and with specificity the particulars of the trade secrets without any discovery. This order will not allow this old trick of vague pleading with the blanks to be artfully filled in only after discovery.
Given the dubious nature of Jobscience's alleged trade secrets, Jobscience was required to provide a detailed statement of trade secrets. Section 2019.210 of the California Code of Civil Procedure states that (emphasis added): "the party alleging the misappropriation shall identity the trade secret with reasonable particidarity. " The order also stated (Did. No. 43):
For each trade secret, plaintiff must file, and serve on counsel, a statement, under seal, that should include:
(1) a summary of the specific trade secret;
(2) the background of the trade secret and a description of how each secret has derived independent, actual or potential economic value by virtue of not being generally known to the public;
(3) a description of how each secret has been the subject of reasonable efforts to maintain its secrecy; and finally
(4) each of the precise claimed trade secrets, numbered, with a list of the specific elements for each, as claims would appear at the end of a patent.
Jobscience did not come close to doing this. Rather, Jobscience provided a three-page narrative and three barely-legible schematics. No explanation was provided for the schematics other than to say a "representation of Jobscience's proprietary process and efficient solution is reflected in the following schematic diagram and screen shot examples." In opposition, Jobscience tried to cure the deficiency by providing a numbered list of allegedly misappropriated trade secrets:
(Opp. 3-4). Jobscience stated that it "is not claiming a trade secret on the use of relational databases to support business processes, but rather on [REDACTED/] (Opp. 4". The mysterious process is never identified. This is too little, too late. Two prior orders warned Jobscience that if it could not identify its trade secrets, the claim would be distnissed. That time has come.
The statement of trade secrets falls short of the trade secrets promised in the second amended complaint (filed after the February 2014 order):
Plaintiffs trade secrets include Plaintiffs manuals, training materials. contract tenns and conditions of service, its business model and pricing information, its marketing strategies, product design, the strategy, process and methods of integrating Plaintiffs product and services with the Salesforce.com platfotm, the strategy and functioning of Plaintiffs business in its partnership with other software providers, its methods of designing and constructing work flows and information sharing to create efficiencies for recruiting and corporate clients.
* * *
It is Plaintiffs know how. carefully guarded by Plaintiff, that Defendants have misappropriated for their own gain.
* * *
Defendants also gained access to additional trade secrets of Plaintiff, including the strategy and functioning of Plaintiffs business in its partnership with other software providers, its methods of designing and constructing work flows and information sharing to create efficiencies for recruiting and corporate clients and related non-secret but confidential information.
(Second Amd. Compd. ¶¶ 18, 19, 30). Jobscience promised it could disclose trade secrets, but when it came time to show us the money, its wallet was empty. At oral argument, Jobscience conceded that nine of the eleven items promised in the complaint were not disclosed in its statement of alleged trade secrets. This experience has been nothing more than a fishing expedition.
Instead of defining with particularity its trade secrets, Jobscience used a hodgepodge of ambiguous phrases like "[REDACTED/], " "[REDACTED/], " "[REDACTED/], " and "[REDACTED/]." Jobscience stated that these:
(Dh. No. 48-4 at 2). Such blather makes it virtually impossible to distinguish the alleged trade secrets from lalowledge in the field. Defendants cannot reasonably prepare its defenses and search for art in the field if the boundaries of the trade secret are so undetermined.
The closest Jobscience comes to specificity is the desigii of its recruiting software:
( id. at 3). Nevertheless, these are mere criteria and in no way tell the reader the secret allow it is done. No details are provided for how Jobscience allegedly achieves the desired design and functionality.
Jobscience's reliance on Silvaeo Data System v. Intel Corp., 184 Cal.App.4th 210, 221 (Cal.Ct.App. 2010), disapproved on other grounds by Kwikset Coip. v. Superior Court 51 Cal.4th 310 (2011), is misplaced. In Silvaco, the court found that the design itself could not constitute a trade secret because it was evident to anyone running the program. Other than the source code (preempted here), the design information failed to describe a trade secret. See also Agency Solutions.Com, LLC v. TriZetto Grp., Inc., 819 F.Supp.2d 1001, 1021 (E.D. Cal. 2011) (Judge Anthony Ishii) (diagram and functional description failed to constitute a trade secret).
In defense of its vagueness, Jobscience argues that the Copyright Act does not preempt its "[REDACTED/] "because "copyright does not extend to methods and processes" (Opp. 6) (emphasis in original). This misses the point. Just because something is not copyrightable (or patentable), does not make it a trade secret. Some information, like the vague information referenced in Jobscience's statement, is simply too well-known in the field to provide the basis for an intellectual-property claim. That Jobscience was unable to capture its "15 years of research and development work leading to the current Jobscience product" in a copyright does not somehow render it a trade secret.
Jobscience cites Advanced Modular Sputtering, Inc. v. Superior Court, 132 Cal.App.4th 826, 835-36 (2005), for the proposition that "reasonable particularity" does not mean that the plaintiff needs to "define every minute detail of its claimed trade secret at the outset of the litigation." But the plaintiff in Advanced Modular Sputtering was able to identify eight alleged trade secrets:
each with several discreet [sic] features that, in combination with one another, formed the alleged trade secrets. It has described how it believes the combination of these features distinguish the alleged trade secrets from the prior art, or matters within the general knowledge of persons in the sputtering industry.
Id. at 836. By contrast, Jobscience's statement is woefully deficient. We cannot tell where the claim ends and the known art begins. We cannot tell what the secret is to achieving the criteria for the recruiting software.
In the alternative, Jobscience "requests that the Court allow Jobscience to amend the [trade secret] Statement to provide further detail or cure any deficiencies." At this point, Jobscience has now had two opportunities to state a claim for the misappropriation of trade secrets. Each time, it has failed. No more tries will be allowed, especially now that plaintiff's counsel have seen the accused source code. The trade secrets claim is DISMISSED.
2. SOURCE-CODE COMPARISON.
Defendants object to the expert report of Kenneth Amron based on inadequate qualifications, failure to compare at least 668 files of Defendants' 701 files, and failure to use a source-code comparison program like CodeSuite (Dkt. No. 62).
In February 2014, plaintiff was allowed to conduct a source-code comparison. On April 7, plaintiff filed a letter, appending the expert report of Kenneth Amron. Plaintiff asked Mr. Amron to "determine whether [defendants'] Talent Rover code base exhibits evidence of code copying." Mr. Amron stated the "review was requested as an effort to identify evidence of copying'.... Other comparisons of source code copying were not performed, owing to limited time available for this review." Mr. Amron reviewed the code using defense counsel's laptop which was pre-installed with "Beyond Compare" a "file and directory comparison utility" and "a trial version of slickEdit 2013, " a "software project editing tool." He found that plaintiff's source code contained 2, 072 files and defendant Skipan SAAS LLC's source code had 701 files. He found 33 identical file names. He then looked at the code within the files with the identical file names and opined that he found "strong evidence of copying." Rather than isolate any actual copying, Mr. Amron identified similarities in file names (Dkt. No. 54 at ¶¶ 1, 31, 50, 51, 204).
Names cannot be copyrighted. Mr. Amron found no passages of copying. He merely concluded "strong evidence of copying, " cherry-picking similar file names. This falls short. There either is or is not identical passages of source code. This kind of comparison is possible and often done. Why hasn't Mr. Amron done it?
On April 18, defendants filed a letter identifying deficiencies with Mr. Amron's report and qualifications. Defendants appended the declaration of Robert Zeidman, the founder of Zeidman Consulting and the creator of a tool called "CodeSuite, " which includes a function called CodeMatch. The tool "conducts source code comparisons by calculating source code correlation. CodeMatch analyzes each source code file using specific algorithms and finds similarities." It has been used in "roughly seventy lawsuits" (Zeidman Decl. ¶¶ 1, 5, 6, 7).
Defendants argue that Mr. Amron has not performed a proper comparison of source code and suggests that the "Court should either strike the copyright claim as a sanction for violating the February 28 Order or require Plaintiff to complete an actual source code comparison using CodeSuite" (Dkt. No. 62). Possibly this will be done eventually, but for now, both sides will be given a chance to retain experts to render expert reports on source-code comparisons. The reports should address the issues raised at the May 1 hearing, including identification of any passages of copied code. Defendants should then bring a summary judgment motion on the copyright-infringement claim. The opposition should append any expert report relied upon. Defendants may then depose plaintiff's expert on the issue. Both sides should put forth their best case on the copying issue.
For the reasons stated herein, the motion to strike is GRANTED. The misappropriation of trade secret claim (third claim for relief) is DISMISSED. Leave to amend will not be allowed. The temporary discovery stay is LIFTED. Even though there is no trade secrets claim, redactions have been applied to this order to preserve any appeal of the misappropriation of trade secret claim.
The copyright-infringement issue will be addressed at summary judgment. All existing deadlines remain in place (Dkt. No. 30).
IT IS SO ORDERED.