ORDER RE DISPUTE OVER "SOURCE CODE" CONFIDENTIALITY DESIGNATION

WILLIAM ALSUP, District Judge.

This is a dispute between defendants Toshiba Corporation, Toshiba America Electronic Components, Inc., and Toshiba America, Inc. (collectively, "Toshiba") and plaintiff Lone Star Silicon Innovations LLC about the proper scope of certain confidentiality designations under our district's Patent Local Rule 2-2 Interim Model Protective Order ("Protective Order"). The Protective Order provides for three confidentiality designations:

Toshiba seeks to designate twelve of its process flows and process recipes for making a semiconductor die as "HIGHLY CONFIDENTIAL — SOURCE CODE" pursuant to Section 2.9 of the Protective Order. These documents — which Toshiba had previously produced to Lone Star while this case was pending in the Eastern District of Texas but clawed back after the case transferred here — "describe the step-by-step process for making a device from a blank silicon wafer, and . . . set forth particular settings, conditions and/or parameters for a particular step within the process flow." Toshiba claims these documents are its "crown jewels that can be used in a similar fashion as computer codes in a software case to replicate . . . Toshiba's semiconductor manufacturing processes." Although the documents are not "computer code" or "associated with computer code," Toshiba contends they qualify as "formulas, engineering specifications, or schematics that define or otherwise describe in detail the algorithms or structure of . . . hardware designs" within the meaning of Section 2.9 (Dkt. No. 201-1 at 2-3).

Lone Star agrees that the documents at issue qualify as "HIGHLY CONFIDENTIAL — ATTORNEYS' EYES ONLY" under Section 2.8 but disputes that they qualify as "HIGHLY CONFIDENTIAL — SOURCE CODE" under Section 2.9. According to Lone Star, Section 2.9 should be read to cover only "computer code" and materials "associated with computer code" (see, e.g., Dkt. No. 203 at 4, 6). This order disagrees. The plain text of Section 2.9 lists "computer code and associated comments and revision histories" as a category separate and apart from "formulas," "engineering specifications," and "schematics." All four categories fall within the scope of "extremely sensitive `Confidential Information or Items'" covered by Section 2.9 so long as they "define or otherwise describe in detail the algorithms or structure of software or hardware designs, disclosure of which to another Party or Non-Party would create a substantial risk of serious harm that could not be avoided by less restrictive means." Contrary to Lone Star, Section 2.9's reference to "comments and revision histories" "associated" with computer code does not mean that "formulas, engineering specifications, or schematics" fall under Section 2.9 only if they are also "associated" with computer code.

Lone Star claims the security afforded to materials with the "source code" designation makes sense only in light of the "unique nature of computer code and its documentation," in that anyone "with an off-the-shelf compiler and the software source code . . . could easily replicate and distribute unlimited copies of the software product" (Dkt. No. 203 at 2). In a similar vein, Lone Star cites three other protective orders from other cases that have defined "source code" more narrowly, and protests that expanding the designation beyond computer code and "associated" materials will make discovery in patent cases unduly burdensome (id. at 2-4). None of these arguments are persuasive.

First, easy unauthorized copying and distribution is a concern particularly prominent in, but not exclusive to, computer code. It remains conceivable that materials not "associated" with computer code nevertheless describe the details of an easily replicable product in such a manner as to warrant the highest level of confidentiality. Section 2.9 clearly contemplates this possibility, as it defines such materials — regardless of category — by their ability to "define or otherwise describe in detail the algorithms or structure of software or hardware designs, disclosure of which to another Party or Non-Party would create a substantial risk of serious harm that could not be avoided by less restrictive means." This, combined with the explicit inclusion of "formulas, engineering specifications, [and] schematics" that meet the foregoing criteria, indicates that Section 2.9 encompasses confidentiality concerns both within and without the specific category of computer code and "associated" materials.

Second, the three other protective orders cited by Lone Star do not compel a contrary conclusion. For starters, those protective orders from other cases are not the Protective Order at issue here. Moreover, read in context, the passages cited by Lone Star are readily distinguishable. Two specifically defined "source code" and its attendant protections while the third specifically defined "Highly Confidential Code" (Dkt. Nos. 203-2-203-4 (emphasis added)). In contrast, Section 2.9 does not purport to define "code" but instead defines the universe of all "extremely sensitive `Confidential Information or Items'" that deserve a "HIGHLY CONFIDENTIAL — SOURCE CODE" designation. The inclusion of the phrase "source code" in the designation itself does not override the simple fact that the plain text of Section 2.9 contemplates non-source code material that nevertheless deserves protection on par with actual source code.

Third, and in a similar vein, Lone Star's warning that Section 2.9 will render discovery unduly burdensome in patent cases unless limited to computer code and "associated" materials is unfounded. As stated, Section 2.9 does not cover all "formulas, engineering specifications, or schematics" but only those that require protection on par with true source code because they "define or otherwise describe in detail the algorithms or structure of software or hardware designs, disclosure of which to another Party or Non-Party would create a substantial risk of serious harm that could not be avoided by less restrictive means." This important limitation is sufficiently stringent to both distinguish Section 2.9 from Section 2.8 (which does not require that designated materials "define or otherwise describe in detail the algorithms or structure of software or hardware designs") and avert the practical discovery problems Lone Star envisions. Indeed, neither side offered any evidence that the plain language of Section 2.9 has ever hampered discovery in another patent case in our district.

It is therefore unsurprising that the parties have not unearthed any other decision analyzing the issue of whether or not Section 2.9 extends beyond computer code and "associated" materials. During oral argument, counsel for Toshiba pointed to Telebuyer, LLC v. Amazon.com, Inc., No. 13-cv-1677, 2014 WL 5804334 (W.D. Wash. July 7, 2014) (Judge Barbara Rothstein), as the closest decision on point. In Telebuyer, Amazon sought a protective order on materials designated as "source code" and proposed defining that term as "computer code and associated comments and revision histories, formulas, engineering specifications, or schematics that define or otherwise describe in detail the algorithms or structure of software or hardware designs." Id. at *3. Telebuyer protested that the definition was overbroad because it encompassed "routinely disclosed documents that have nothing to do with source code." The district court concluded:

Ibid. (citation omitted).

It would be circuitous for this order to rely on Telebuyer as authoritative in interpreting the Protective Order, given that Telebuyer actually relied on the wording of the Protective Order (as representing the cumulative experience of our district) in reaching its outcome. To the extent Telebuyer is instructive, however, it cuts against Lone Star and in favor of adhering to the plain language of the Protective Order. True, Telebuyer arguably suggested that Section 2.9 required the materials at issue to be "relate[d] to source code" and apparently conflated the definition of "source code" itself with our definition of materials deserving of protection on par with source code. Significantly, however, Telebuyer nevertheless declined to limit the protection afforded under Section 2.9 to literal computer code, opting instead to "err on the side of caution when dealing with highly technical materials." In doing so, Telebuyer implicitly recognized the practical fact that, when it comes to confidentiality designations, what matters is not whether particular materials can accurately be described as literal "source code" but whether they deserve protection on par with source code.

Although this order rejects Lone Star's narrow reading of Section 2.9, it nevertheless concludes Toshiba has not shown that the "HIGHLY CONFIDENTIAL — SOURCE CODE" designation is warranted here. The plain language of Section 2.9 does not encompass all "formulas, engineering specifications, or schematics." Whether or not specific materials actually "define or otherwise describe in detail the algorithms or structure of software or hardware designs, disclosure of which to another Party or Non-Party would create a substantial risk of serious harm that could not be avoided by less restrictive means," is a question that must be answered on the facts of each particular case. Here, the undersigned judge personally reviewed a sample of the documents Toshiba seeks to designate as "HIGHLY CONFIDENTIAL — SOURCE CODE" and finds that they do not require protection on par with true source code, especially since they were previously produced and apparently remained in Lone Star's possession for seven months without raising any reason for concern that such possession created "a substantial risk of serious harm" to Toshiba.

This order therefore DENIES Toshiba's request for a ruling that it properly designated twelve process flows and process recipes as "HIGHLY CONFIDENTIAL — SOURCE CODE" under our Protective Order. At least on this record, the lesser designation of "HIGHLY CONFIDENTIAL — ATTORNEYS' EYES ONLY" is sufficient. This ruling is without prejudice to the possibility that Toshiba may show, on an improved record, that more stringent protections are required for specific documents at issue.

IT IS SO ORDERED.