# A Legal Tangle of Secrets and Disclosures in Trade

**Authors:** Jeanne C. Fromer
**Citation:** A Legal Tangle of Secrets and Disclosures in Trade, in *Intellectual Property at the Edge: The Contested Contours of IP* 271 (Rochelle C. Dreyfuss and Jane C. Ginsburg, eds., Cambridge University Press, 2014)
**Source:** https://ssrn.com/abstract=2291726

*p. 3*
produce the pump, Tabor made thirty-six pieces of patterns for the pump's parts, from which the pump could be manufactured. 5 Making these patterns for the pump correctly, according to Tabor, "required a good deal of time, study, thought, labor and money." 6 As per Tabor, the reason for this significant expenditure of time and money to translate the pumps into corresponding patterns is because the patterns do not match up precisely to the pump's pieces because there are both shrinkages and expansions of the pump's metals under different conditions of hot or cold liquid being pumped through it. 7 Getting the patterns right, on Tabor's account, requires a "series of experiments," rather than simple calculations as to shrinkage and expansions. 8 Tabor insisted that he kept these patterns secret and in his possession. 9 Yet he admittedly gave the patterns to Frank Collingnon, a machinist, for the sole purpose that he make pump castings for Tabor. 10 According to Tabor, without Tabor's or Collingnon's permission or knowledge, Francis Walz, a pattern maker, measured Tabor's patterns to make a copy. 11 Walz got Tabor's patterns from Collingnon's possession. 12 He was able to do so due to his access to Collingnon's working space: Walz had a shop in Collingnon's building and Collingnon sometimes gave him permission to take other patterns. 13 Tabor claimed that William Hoffman had paid Walz to take these patterns and then Hoffman used a copy of them to produce pumps. 14 In Tabor's view, Walz and Hoffman had copied his patterns to avoid the significant expenditure required to use the pumps themselves to derive corresponding patterns. 15 Tabor professed that owing both to the shrinkages and expansions of the pump's metals and brass castings hiding from view parts of the pump, "[a] perfect set of patterns could not be made from seeing a finished pump." 16 Hoffman told a different story. He claimed that he had "employed … Walz to make a set of patterns for castings for pumps, like the one manufactured by [Tabor], … but there was no agreement or arrangement between … Walz and [Hoffman] that … Walz should take any measurements from [Tabor's] patterns." 17 According to Hoffman, Walz got ahold of Tabor's patterns when they were in his possession for repairs. 18 Walz used them on Hoffman's behalf but without Hoffman's knowledge until Tabor complained, at which point he stopped and "took all his measurements from a pump which [Tabor] had sold to a third party." 19 Hoffman maintained that using the pumps to come up with patterns necessitated "little more expense and trouble than [to do so] from measurements taken from [Tabor's] patterns." 20 Similarly, Walz indicated that he had used Tabor's patterns only because "it was [just] a little less trouble than to go outside and measure a machine." 21 In 1885, Tabor sued Hoffman seeking $1,000 in damages, interest, and an injunction restraining Hoffman from making, selling, or otherwise disposing of any pumps made from Tabor's patterns. 22 After Hoffman denied all of the allegations, 23 the case proceeded to a bench trial in the Erie County division of the Supreme Court of the State of New York.

*p. 4*
Hoffman attempted to use patent law to shield himself from liability for misappropriating Tabor's trade secrets. He maintained that Tabor's pump was a mere improvement on a previous pump invented by Tabor, for which he had received a patent in 1861. 24 Figure 1 shows a drawing of the patented pump. Walz testified to that effect. 25 The patent had expired in 1878, several years before the events at issue here. 26 Hoffman argued on that basis that "[t]he invention which [Tabor] had patented became, after the expiration of the patent, public property, so that thereafter [Hoffman] had the right to make pumps just like those made by [Tabor] under his patent and to sell them in the market," including from Tabor's patterns based on these pumps. 27 Tabor, however, testified at trial that the patented pump was an entirely distinct invention from the one at issue in this case. 28 He maintained that the patented pump was "not a pump for pumping liquids in the proper sense of the term; it was a pump for pumping solids, liquids and solids combined." 29 Moreover, Tabor claimed never to have sold the patented pump or used the patterns corresponding to that pump after the patent had expired. 30 19 Id. at 12. Tabor insisted that Walz copied thirty-five of his thirty-six patterns and ascertained the last one, for a stand, easily only because it was "a perfectly immaterial piece, as any kind of a stand would answer the purpose." Respondent's Points, supra note 5, at 7 (emphasis in original). 20 Defendant's Requested Findings, supra note 17, at 12. 21 Trial Transcript, supra note 3, at 41. 22 Complaint, supra note 15, at 4. 23 Answer to the Complaint, Tabor v. Hoffman, 118 N.Y. 30 (1889). 24 Defendant's Requested Findings, supra note 17, at 10-11; Trial Transcript, supra note 3, at 56-57. Although not identified in the trial record, that patent is apparently U.S. Patent No. 33,550 (patented Oct. 22, 1861) (Improvement in Rotary Pumps). 25 Trial Transcript, supra note 3, at 40. 26 Defendant's Requested Findings, supra note 17, at 11; Trial Transcript, supra note 3, at 25. 27 Defendant's Requested Findings, supra note 17, at 13. On subsequent appeal to the New York Court of Appeals, Hoffman additionally argued that after the patent's expiration, Tabor "sought to impress the public with the belief that these pumps were still protected by patent by casting upon each the well known abbreviation 'Pat.,' in violation of [the law]." Appellant's Points at 2, Tabor v. Hoffman, 118 N.Y. 30 (1889). Tabor disputed this possibility, maintaining that his pumps were marked "Special Pat. 8 x 8" with a date, with "Pat." standing for "pattern" and not "patent." Respondent's Points, supra note 5, at 14. The Court of Appeals did not take up this particular issue. 28 Trial Transcript, supra note 3, at 26. 29 Id. at 25. 30 Id. at 25-26. The trial court sided with Tabor on this issue. It declined to find that the pump at issue here involved "substantially the same principle and improvement as the pumps so manufactured under said letters patent, and being substantially the same pump with some changes in some parts of it." 31 The court stated that it was unnecessary to issue a finding on the differences between the patented invention and the pump at issue in the suit, "because both parties agree that it was necessary to have patterns and castings to manufacture the new pump. "32 In so doing, it rejected Hoffman's defense. The court's reasoning is cryptic, but it seems the judge thought that the current pump was different from the patented pump because one needed patterns to make the current pump. Implicitly, the court seems to be linking the premise to the conclusion on the basis that were the pumps the same invention, the patent would have disclosed the current pump's pattern and there would therefore be no need for a new set of patterns.

*p. 6*
The Supreme Court proceeded to rule in Tabor's favor on his claim. 33 It credited Tabor's account of why and how Walz used Tabor's patterns for Hoffman. 34 The court then determined that even though Hoffman had every right to make pumps like Tabor's using Hoffman's own patterns, Hoffman "has no right to manufacture pumps from [the] patterns made from measurements taken by Francis Walz from [Tabor's] patterns," reasoning that Tabor "has an exclusive right of property in said patterns." 35 Accordingly, the court enjoined Hoffman from making, selling, or otherwise disposing of both the purloined patterns and any pumps made from the patterns. 36 In coming to its conclusion, the Supreme Court ruled that "the evidence shows that [Tabor] prepared patterns and spent much time and labor in their production, and also incurred considerable expense." 37 Interestingly, the trial court relied merely on Tabor's own difficulty in making patterns from the pump to conclude both that "the[] patterns were the property of [Tabor], the product and creation of his own reflections, investigations and industry," and that Walz and Hoffman's unauthorized taking was wrongful. 38 There had been ample testimony at trial from six pattern makers and one molder that a pattern maker like Walz could, without much effort, estimate shrinkages and expansions in pump metals based on certain known rules. 39 Moreover, Tabor himself testified that "there is a rule I believe, known to pattern makers by which they estimate shrinkage; I don't know anything about the rule; I never tried it; I didn't use any rule of that description when I made patterns; I used my judgment in the matter." 40 Even though the court explicitly found that "[s]uch patterns can be made from measurements taken … from the original patterns … and that a competent pattern maker can make a set of patterns from measurements taken from the pump itself without the aid of plaintiff's patterns," 41 the court gave that finding no legal weight. The court maintained that such 32 Supreme Court Opinion at 67, Tabor v. Hoffman, 118 N.Y. 30 (1889). 33 Supreme Court Findings, supra note 3, at 9. 34 Id. at 7-8. 35 Id. at 8; Supreme Court Opinion, supra note 32, at 68. 36 Supreme Court Findings, supra note 3, at 9. 37 Supreme Court Opinion, supra note 32, at 67. 38 Id. at 68-69. That said, the court did ambiguously find that "the plaintiff spent, and it required a good deal of time, study, thought, labor and money, to make and perfect said patterns." Supreme Court Findings, supra note 3, at 7. 39 Trial Transcript, supra note 3, at 31, 37, 39-40, 45, 48-50, 52, 63. But cf. id. at 31 (testimony by a molder that different pattern makers have different rules that produce varying patterns from the same pump pieces). 40 Id. at 26. 41 Defendant's Requested Findings, supra note 17, at 12. evidence was irrelevant because "[i]t is certainly unjust and inequitable that [Hoffman] should, without [Tabor's] consent, be allowed to use duplicates and measurements from the creations and inventions of [Tabor]-his property, and never intended for public use, to compete with him in manufacturing the same kind of pumps." 42 The court concluded that if Hoffman is correct that it is easy to reverse engineer the pump's patterns from the pump itself, "he will not suffer seriously [from the court's imposition of an injunction], for he can manufacture [patterns] from [Tabor's] pumps." 43 Hoffman appealed to the General Term of the Supreme Court, Erie County, which affirmed the trial court's judgment. The panel reasoned that:

*p. 7*
[T]he manufacture and sale of the pump in question was not a publication of the plans, specifications and patterns from which the pumps were manufactured. That so far the improved pump was concerned [Tabor] had the right to keep the plans, specifications and patterns from the public, and thus secure to himself the benefit of the business in which he was engaged, that [Hoffman] had no right to procure copies surreptitiously without the knowledge of [Tabor] by employing one of his pattern-makers or repairers to so make them, and that the Court has power to restrain the use of the pattern so acquired. 44 The General Term also affirmed the trial court's rejection of Hoffman's patent-law defense. It reasoned that "[s]o far as the patented pump is concerned, [Tabor] has doubtless no further exclusive property in the invention." For that pump, the court asserted that "[t]he plans and specifications from which the machine was constructed were placed on file in the patent office where they became the property of the public, and all persons desiring may have copies thereof." 45 By contrast, the court continued, "[A]s to the improved machine, the subsequent inventions, a different question is presented. These improvements have never been patented, and consequently [Tabor] is unable to claim any protection through the patent laws. He has invented and constructed patterns from which he has manufactured the improved pump." 46 To the court, there was no requirement-grounded in patent law-that Tabor reveal the patterns, even if he had made the pump commercially available. 47 Hoffman appealed again, to the New York Court of Appeals, therein pressing the argument that it does not make sense to provide a legal remedy for any competitive injury to Tabor's business, because the same injury would result from selling pumps made from reverse engineering the pumps themselves, a situation which is legally permissible. 48 Moreover, Hoffman insisted that the patterns were not truly secret because pattern makers could readily construct the commercially available pump's patterns by applying known rules to the pump's measurements. To confer legal protection here would, in 42 Supreme Court Opinion, supra note 32, at 70-71. 43 Id. at 70. 44 General Term of the Supreme Court Opinion at 76, Tabor v. Hoffman, 118 N.Y. 30 (1889). 45 Id. at 74. 46 Id. at 75. 47 Id. 48 Appellant's Points, supra note 27, at 13. Hoffman's view, "protect a secret when there is no secret." 49 Tabor disputed this understanding, replying that it was difficult to use the pumps to make patterns. Moreover, Tabor argued that "[t]he fact that Hoffman could have made a set of patterns from a finished pump does not by any means give him the right to have a duplicate set made from [Tabor's] patterns while in the hands of [Tabor's] employees for repairs, or to make use of such duplicates." 50 The New York Court of Appeals affirmed the General Term. 51 The court emphasized that once Tabor made his pumps publicly available by marketing them, without securing patent protection, he relinquished any exclusive property to them. Nonetheless, the court continued, Tabor had kept secret the patterns, "which greatly aided, if they were not indispensable, in the manufacture of the pumps." To the court, the relevant legal question was whether by making public the pumps, Tabor thereby constructively also made public the patterns. 52 If not, Tabor had a valid claim for misappropriation of trade secrets against Hoffman.

*p. 8*
The court answered this question in the negative. Because of the discrepancies between the pump and its patterns, for the reasons discussed by the trial court, the patterns' sizes could not be discovered by merely investigating the different pump pieces. The court credited Tabor's account of how much work it took to yield a set of patterns from a pump. Therefore, the court concluded that the pumps' public availability did not undermine the patterns' status as a secret that belonged to Tabor, unless Tabor were to publish the patterns or someone else were to reverse engineer them from the pumps. 53 Moreover, the possibility that someone might undertake the experimentation necessary to reverse engineer the pumps to derive their patterns-something the court termed "discovery … by fair means"-does not imply that someone like Hoffman could take Tabor's patterns for his use-what the court calls "discovery by unfair means, such as bribery of a [secretholder]." 54 The court therefore concluded that "[w]hile [Hoffman] could lawfully copy the pump, because it had been published to the world, he could not lawfully copy the patterns, because they had not been published, but were still, in every sense, the property of [Tabor], who owned not only the material substance, but also the discovery which they embodied." 55 Chief Judge Follett dissented from the Court of Appeals's decision and would have reversed the trial court's judgment. Unlike the majority, he focused on Tabor's expired patent:

*p. 8*
The patent on the original invention having expired, and the plaintiff having voluntarily made the subsequent improvements public by 49 Appellant's Supplemental Points at 18-22, Tabor v. Hoffman, 118 N.Y. 30 (1889). 50 Respondent's Points, supra note 5, at 4, 9. 51 Tabor v. Hoffman, 118 N.Y. 30 (1889). 52 Id. at 34-35. 53 Id. at 35-36. 54 Id. at 35-37. 55 Id. at 37. selling the improved article, he lost his right to their exclusive use…. The invention was not the patterns, but the idea represented by them, to which the plaintiff had lost his exclusive right. 56 At the core of the Tabor case are important ways in which trade secrecy intersects and sometimes falls into tension with patent law, particularly with regard to the disclosure and secrecy aims of each. After outlining contemporary trade secrecy and patent laws, I turn to a general discussion of those intersections and tensions.

*p. 9*
Current trade secrecy protection looks much like the protection conferred long ago in Tabor. Pertinently, the Uniform Trade Secrets Act, adopted in forty-six states and the District of Columbia, shields as a trade secret from certain types of misappropriation information that "derives independent economic value, actual or potential, from not being generally known to, and not being readily ascertainable by proper means by, other persons" and is "the subject of efforts that are reasonable under the circumstances to maintain its secrecy." 57 Misappropriation occurs principally when the acquirer of another's trade secret contravenes a pre-existing contractual or other obligation to the secret holder not to disclose the secret or commits an improper act of industrial espionage (often quasi-tortious, if not tortious) to get it. 58 So long as a secret remains unrevealed, legal protection is everlasting. 59 Of course, the information comprising the secret might always be properly revealed through acts that do not constitute misappropriation, principally independent discovery or reverse engineering of an available product. 60 With this avenue of sanctioned uses of secrets so long as they are independently found, as Rochelle Dreyfuss explains, "trade secrecy laws … safeguard public access." 61 As the Supreme Court has reasoned, there are thought to be two key purposes to trade secrecy laws: "maintenance of standards of commercial ethics" and "the encouragement of invention." 62 With regard to commercial ethics, the general idea is that trade secrecy liability will deter people from carrying out certain illicit forms of commercial behavior. 63 Additionally, trade secrecy protection encourages investment in 56 Id. at 37-38 (Follett, C.J., dissenting). 57 Unif. Trade Secrets Act § 1(4) (amended 1985). 58 scientific and technological research, according to Mark Lemley, by "giv[ing] the developer of new and valuable information the right to restrict others from using it, and therefore the prospect of deriving supracompetitive profits from the information." 64 Both motivations have served to protect trade and innovation, 65 particularly in an era of mass production, in which commercialization of innovation depends on access to trade secrets by a large number of employees and third parties. 66 Now consider patent law. Utilitarianism is the dominant purpose of American patent law. 67 According to utilitarian theory, patent law provides the incentive of exclusive rights for a limited duration to inventors to motivate them to create technologically or scientifically valuable inventions. In exchange for this incentive, patent law also requires patentees to disclose their inventions to the public. 68 Without the patent incentive, the theory goes, inventors might not invest the time, energy, or money necessary to create the works because such works might be copied cheaply and easily by free riders, thereby eliminating inventors' ability to profit from their labors. According to utilitarian thinking, public benefits accrue by rewarding inventors for taking two steps they likely would not otherwise have taken: first, to invent, and possibly commercialize; and second, to reveal information to the public about their inventions that serves to stimulate further innovation. 69 Consistent with utilitarianism, the rights conferred by patent laws are designed to be limited in time and scope. 70 The reason for providing patent protection to creators is to encourage them to produce socially valuable works, thereby maximizing social welfare. 71 If the provided rights were exceedingly extensive, society would be hurt and social welfare diminished. 72 Exclusive rights in patent law prevent competition in protected works, allowing the patent holder to charge a premium for access and ultimately limiting these valuable works' diffusion into society. 73 Moreover, given that knowledge is frequently cumulative, society benefits when subsequent creators are not prevented from building on previous scientific and technological creations to generate new works. 74 Therefore, patent law ensures both that the works that it protects will fall into the public domain in due course and that third parties will be free to use the protected works for certain socially valuable purposes. 75 64 Id. at 330. 65 Id. at 319-20. 66 72 Lemley, supra note 70, at 996-97. 73 Id. 74 Id. at 997-98. 75 As one example of socially valuable uses of patented works, patent law excuses doctors from infringement liability for performing patented medical activity. See 35 U.S.C. § 287(c); Katherine J.

*p. 11*
Moreover, patent law's requirement of disclosure of information about inventions can stimulate productivity, as I have previously explored, in two ways: First, it permits society at large to apply the information by freely making or using the patented invention after the expiration of the patent. Second, the disclosure can stimulate others to design around the invention or conceive of new inventions-either by improving upon the invention or by being inspired by it-even during the patent term. Otherwise, the patent system would not require disclosure earlier than the expiration of the patent term, as it does here by requiring disclosure at the time of the patent grant, at the latest, and typically much sooner.

*p. 11*
…. Disclosure of an invention sets out what others have already accomplished, thereby both revealing information about those discoveries-enabling the avoidance of wasteful duplication of the original inventor's research-and noting, usually implicitly by omission, what has yet to be done. Patent disclosures act, as one commentator labels it, as an "invisible college of technology." Use of these disclosures, in turn, speeds the rate of innovation in society, which is central to economic growth. 76 There are a number of reasons an innovator might prefer trade secrecy protection to patent protection. First, protection vests without any need for government approval, which is typically time-consuming and expensive to obtain. 77 Second, because trade secrecy laws protect all economically valuable information from misappropriation, they cover inventions that are either unpatentable or of dubious patentability. 78 Third, protection can last for a longer time, and potentially forever, if the chances are great that the invention will not be independently discovered by a third party for a time exceeding patent duration (or the time in which the invention is commercially important). 79 That is, trade secrecy protection is particularly attractive for inventions that are likely to stay secret, such as chemical or mechanical processes, as they can be shielded from public view and are hard to reverse engineer even as the products they produce are commercialized. 80 When those conditions do not hold, patent protection is likely to be favored over trade secrecy. An inventor is likely to be happy to forego trade secrecy protection in exchange for the more certain but time-limited protection that patent law confers if he is willing and able to spend the money and time to obtain patent protection for a patentable invention that is likely to be independently discoverable or reverse engineered once it is commercialized. 81 More extremely, the Supreme Court, in ruling that the federal patent laws do not preempt Ohio's trade secrecy laws, thought that patent protection would always be preferred to trade secrecy when an invention is patentable. 82 Reasoning that trade secrets are at risk of honest discovery and failed lawsuits even when misappropriated, the Court stated that "[t]he possibility that an inventor who believes his invention meets the standards of patentability will sit back, rely on trade secret law, and … forfeit any right to patent protection is remote indeed." 83 Empirical evidence shows that the Supreme Court's conclusion is categorically wrong, 84 but there is a reasonable set of situations in which patent protection is preferable.

*p. 12*
From this comparative description, it might seem that trade secrecy is focused heavily on keeping inventions secret from others for as long as possible, while patent law is centered on ensuring that inventions are disclosed widely to the public. 85 In broad strokes, that is true. 86 But it neglects the disclosures about inventions that trade secrecy enables and the secrets about inventions that patent law permits, sometimes in conflicting ways. I turn now to explore the legal tangle of disclosures and secrets in trade secrecy and patent laws, and in particular, the many ways these issues lay underexplored in Tabor.

*p. 12*
For one thing, Tabor suggests that a trade secret can be much less than a true secret, in that it can be something relatively publicly accessible to third parties. The lynchpin for a conclusion that there was misappropriation of a trade secret in Tabor was that Hoffman, through Walz, used Tabor's secret and laboriously created patterns to make pump patterns, even though Walz might have used his pattern-making rules to reverse engineer patterns solely from Tabor's commercially available pump. The courts considering Tabor's case apparently situated their ruling in the unfairness to Tabor of using his secret patterns without permission, patterns on which Tabor spent heavily in 81 Id. 82 Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470, 484-91 (1974). 83 Id. at 490 (citation omitted). 84 Fromer, supra note 78; Lemley, supra note 58, at 338-41; Sharon K. Sandeen, Kewanee Revisited: Returning to First Principles of Intellectual Property Law To Determine the Issue of Federal Preemption, 12 MARQ. INTELL. PROP. L. REV. 299, 345-46 (2008). 85 In fact, courts hold that information contained in a published patent application destroys the possibility of trade secrecy because there is no longer a secret. E.g., Tewari De-Ox Sys., Inc. v. Mountain States/Rosen, L.L.C., 637 F.3d 604, 611-12 (5 th Cir. 2011). 86 This analysis leaves aside the dissemination effects of inventions protected under each regime. As Katherine Strandburg describes, "[t]rade secrecy effectively ties disclosure and dissemination together because it permits reverse-engineering and independent invention. When an invention is reverseengineered or independently invented, it becomes known to others (disclosed) at the same time that it becomes available to others to use (disseminated). Patenting is different, however. Where an inventor seeks to use an invention exclusively rather than to sell embodiments of it, patenting not only raises consumer prices during the patent term, but also separates disclosure (which occurs at the time the patent or application is published) from dissemination (which occurs only when the patent expires at the end of its twenty-year term)." Katherine J. Strandburg, What If There Were a Business Method Use Exemption to Patent Infringement?, 2008 MICH. ST. L. REV. 245, 271. time and money. In focusing on how hard it had been for Tabor to make the pump's patterns, the courts disregarded the significant, and uncontradicted, trial testimony that pattern makers could have devised patterns corresponding to Tabor's pump using patternmaking rules to account for shrinkages and expansions with significantly greater ease and lesser expense than it took Tabor himself to design the patterns for his pump, as he did so using a less rigorous system of trial and error. For this reason, although nothing of the sort was decided, the full record of the Tabor case exposes the possibly erroreous widespread understanding that this case represents the factual situation of the impracticality of reverse engineering a commercially available product to derive otherwise secret information. 87 The implication of Tabor is that secret information cannot be used directly by an unauthorized third party, even if it can be discovered or discoverable by third parties via reverse engineering of public information or products, whether with as much ease as using the secret information directly or with some greater hardship. This observation exposes a puzzle in trade secrecy law: How can information be considered secret, and thus qualify for protection in the first place, if it is reverse-engineerable? Tabor suggests that it can be considered to be a secret until it is actually reverse engineered, even if it is beforehand conceivable that the information could be obtained through reverse engineering. The 1939 Restatement of Torts is consistent with this view. 88 By contrast, most states' current laws-modeled on the more recent Uniform Trade Secrets Act-take a different approach by considering as secret only information that is "not being generally known to, and not being readily ascertainable by proper means by, other persons." 89 These laws thus make it harder for information to qualify as secret. They raise the question whether information that might be obtained through plausible and established techniques of reverse engineering is truly secret (no matter the significant efforts the secret holder put into getting that information and keeping it directly from the public). Just as one must use one's senses to comprehend the observable aspects of a publicly available product, one must comparatively use more abstract "senses" to reverse engineer less readily observed aspects of a publicly available product, as with the patterns in Tabor's pump. The information derived from the former form of perception is not considered to be a trade secret, while the latter might be. 90 89 Supra text accompanying note 57. 90 See, e.g., Data Gen. Corp. v. Grumman Sys. Support Corp., 825 F. Supp. 340, 359 (D. Mass. 1993) (holding that software distributed only in object code does not destroy the secrecy of the information contained in the corresponding source code); cf. Robert G. Bone, A New Look at Trade Secret Law: Doctrine in Search of Justification, 86 CAL. L. REV. 241, 257 n.81 (1998) ("[H]ow was it possible for anyone to have a secret in information that was part of a publicly marketed product?" (citing Tabor)).

*p. 14*
secrecy law presumes without much analysis that it is a difference in kind by making protection hinge on whether reverse engineering to derive the relevant information is hard to accomplish. 91 As Tabor attests, trade secrecy protection can vest even in information that is reverse-engineerable if it was wrongfully obtained. That said, the possible philosophical oddness of distinguishing the two situations not infrequently gives some courts pause, and they will not enforce a trade secret that is reverse-engineerable, on the ground that it is not a true secret in the first place. One such court reasoned:

*p. 14*
The unrestricted sale of a product embodying a trade secret places the secret in the public domain. This rule applies even where the defendant acquired the secret in confidence. The rule also applies even though the product would have to be rendered inoperative and studied for 3 or 4 weeks to divine the "secret." The plaintiff in the instant case has sold and leased cameras embodying its alleged secrets without any attempt to restrict customers' use of the product, and the cameras could be fully understood by an engineer in a couple of days of study. The antiquity of the comparable rule in patent law is illustrated by … Egbert v. Lippmann, 104 U.S. 333 (1881), as holding that a "public use" was effective even if the product had to be taken apart. The trade secret owner should not be held to a lower standard of proof than the patent owner. 92 As another example, some courts find that the chemical analysis of publicly available products is not a trade secret, even though such analysis might be expensive to undertake, because it is common. 93 Query further the efficiency of trade secrecy's rule treating information that can plausibly be derived through reverse engineering of a publicly available product as a trade secret until reverse engineering happens. This rule forces a trade secret holder's competitors to expend resources to derive independently what might sometimes be procured more cheaply by using the secret directly. 94 Viewed alone, this expenditure is wasteful and might be better directed elsewhere. On the other hand, it diminishes any first-mover advantage the secret holder might otherwise have, thereby muting the secret holder's incentive to innovate in the first place. 95 Additionally, requiring third parties to reverse engineer-rather than use the secret directly-might also be helpful to the third 91 parties (and society at large) by teaching them more about the information, its uses, and further refinements. 96 The Tabor courts did not much emphasize this sort of analysis, instead focusing on the unfairness-and therefore the illegality-of Hoffman's appropriation.

*p. 15*
There is another way in which trade secrecy protects already disclosed information. As is clear from Tabor, Tabor's trade secrecy right was not destroyed by virtue of him sharing his patterns with the machinist, Collingnon. Generally speaking, trade secrecy protection is not lost if a secret holder discloses the secret confidentially to third parties "to profit from its secrets in its business dealings, not to reveal its secrets to the public." 97 That is, a trade secret holder need not maintain perfect secrecy to retain protection. Unless more widely disseminated, information can qualify for trade secrecy protection even if shared with third parties, such as employees, independent contractors, or others, so long as they hold themselves to a confidential relationship. 98 The reason for permitting such disclosures is that without the help of others, a secret holder typically cannot fully effectuate the information's commercial value. 99 In that sense, a trade secret holder can decide to share his or her secret with select others confidentially, all the while still maintaining secrecy legally.

*p. 15*
The trade secret holder is likely to choose to share that information confidentially with a small circle whom he or she already trusts-such as trusted employees, those with whom the secret holder has done business, and established companies-so that they can help the secret holder employ the information to commercial advantage without destroying the secrecy. 100 These are precisely the set of people and entities that already have a competitive advantage, thus making the rich that much informationally richer. Trade secrecy's allowance of targeted disclosures is less democratic in sharing the informational wealth than is patent law's disclosure regime. Requiring patentees to disclose their inventions to the public at large in the patent document in exchange for the patent right democratizes innovation. As I have written elsewhere in describing how an effective patent disclosure regime compares with one that is ineffective, effective disclosure in a patent system should tend to equalize the positions of the initial innovator and potential competitors by granting 96 See, e.g., Mathew Schwartz, Reverse-Engineering, COMPUTERWORLD, Nov. 100 Cf. Sandeen, supra note 84, at 344 (observing that "[t]he sharing and licensing of trade secret information" will likely be targeted at "faithful employees" and trusted "vendors, contractors, business partners, or licensees"); Katherine J. Strandburg, User Innovation Community Norms: At the Boundary Between Academic and Industry Research, 77 FORDHAM L. REV. 2237, 2261 (2009) ("Industry groups that have developed sharing norms historically seemingly have relied on personal relationships, opening their facilities to visits from competitors, movement of personnel within the industry, and publications to develop the means to enforce them."). the latter the information needed to innovate subsequently in the field. Without successful disclosure, the same inventor will be more likely to continue building up on his original invention because he will be the one with the best information to do so. In fact, inventors appear to innovate based only on the information they already have when other information is difficult to acquire. Ineffective disclosure, by extension, can also prolong the patent right beyond its stated expiration because more of the useful information about an invention remains only in the patentee's hands. Innovative rivalry, despite creating some inefficiencies, is more beneficial to society-both economically speaking and as a matter of distributive justice-than a prospecting system that fully concentrates the investment in a technological area in the hands of the initial innovator. History has shown that "most technological change … comes through the small contributions of ordinary, anonymous workers and tinkerers." That is, more minds are able to effect that much more technological progress-both in quantitative terms and in terms of the breadth of creativity-which benefits both society and a broader set of innovators, including newcomers and those in the developing world. 101 Trade secrecy-even with its allowance of targeted disclosures-is much like an ineffective patent disclosure regime, in that those not among the chosen few to whom innovation details have been disclosed have less of a chance to contribute further to innovation. Not only is this harmful to those not in the circle of secrecy, but it is likely detrimental to the public at large, due to a tapered path of innovation, constricted economic growth, and diminished human flourishing. 102 Mark Lemley, however, argues that "trade secret law actually encourages broader disclosure and use of information, not secrecy." 103 For one thing, he suggests that trade secrecy protection encourages people not to overinvest in protecting secrets, because the law will do it. 104 Were it not for trade secrecy, secret holders would be less likely to involve third parties or even employees in commercial production of products and services related to their secrets out of fear that the secret will get out in a way that will not be protected by the law. 105 As Lemley puts it, trade secrecy "encourages disclosure of information that companies might otherwise be reluctant to share for fear of losing the competitive advantage it provides." 106 He is correct to observe that this situation is better than the alternative-real secrecy-particularly for inventions that inventors reasonably think can be kept secret and thus prefer secrecy to patent protection and concomitant 101 Fromer, supra note 69, at 551 (quoting ROBERT FRIEDEL, A CULTURE OF IMPROVEMENT: TECHNOLOGY AND THE WESTERN MILLENNIUM 3 (2007)). 102 Nonetheless, as Jonas Anderson points out, because of the "time pressure to maximize the value of an innovation" protected through trade secrecy (as protection can end at any time), "inventors who maintain inventions as trade secrets likely have more incentive to efficiently disclose their inventions to the proper individuals." J. Jonas Anderson, Secret Inventions, 26 BERKELEY TECH. L.J. 917, 946 (2011). 103 Lemley, supra note 58, at 333. 104 Id. at 333-36. 105 Id. at 334-36. 106 Id. at 335-36. Lemley also maintains that trade secrecy protection provides a semi-solution to Arrow's Information Principle, by enabling bargaining over information without worry that it will lose its secrecy. Id. at 336-37. public disclosure. 107 Yet there is evidence that even with trade secrecy laws, at least in certain industries like the candy business, that secret holders will take excessive precautions to protect their secrets because they "are far from indifferent between the legal remedies for trade secrecy misappropriations and avoiding an actual misappropriation in the first instance, in that they are of the opinion that the former under-compensate for misappropriation." 108 Moreover, as Lemley recognizes, there is a non-negligible class of inventions-those that are neither inherently self-disclosing nor completely obscured, such as computer software-for which inventors might reasonably choose either trade secrecy or patent protection. 109 For these inventions, opting for trade secrecy over patent law is inferior with regard to effectuation of disclosure. Despite Lemley's helpful analysis, there is nonetheless an important tension between the limited disclosures that trade secrecy permits and the thicker public disclosures that patent law requires. 110 There is another important tension between trade secrecy and patent law with regard to disclosure. Patent law, in its current instantiation, requires patentees to enable their inventions. To satisfy the enablement requirement and be granted a valid patent, the patent applicant must provide a description of "the manner and process of making and using [the invention], in such full, clear, concise, and exact terms as to enable any person skilled in the [relevant] art [how] … to make and use the [invention]." 111 This standard requires patentees to reveal plenty of information about how to make use their invention, but it need not be more than a person having ordinary skill in the art needs to do so without undue experimentation. 112 That means that even though patent law is tasked with promoting disclosure, it permits the patentholder to keep secret any information short of undue experimentation for a person having ordinary skill in the art to make and use the 107 See supra text accompanying notes 77-80 (suggesting conditions under which inventors might prefer trade secrecy to patent protection). Lemley reasons further that the secrecy requirement the law imposes acts to channel to trade secrecy protection those whose inventions are not self-disclosing, with the rest being channeled to patent protection, if anything. Lemley, supra note 58, at 335-36. As Lemley sees it, "[t]he traditional conception of the trade-off between patents and trade secrets views the disclosure function of the patent system as one of its great advantages over trade secret law. And indeed the law operates in various ways to encourage inventors to choose patent over trade secret protection where both are possible. But for certain types of inventions we may actually get more useful 'disclosure' at less cost from trade secret than from patent law." Id. at 314; cf. Katherine J. Strandburg, What Does the Public Get? Experimental Use and the Patent Bargain, 2004 WIS. L. REV. 81, 105, 111, 119 (proposing that patent disclosure is relatively unhelpful for the many inventions that are self-disclosing, as they are best divulged by the commercialization of the invention, but suggesting that energy should be focused on permitting experimental use on patented inventions to effectuate the disclosure function for non-selfdisclosing inventions). 108 Fromer, supra note 78, at 15-16. 109 Lemley, supra note 58, at 338-40. 110 Cf. Sandeen, supra note 84, at 343-45. Interestingly, both patent and trade secrecy laws provide protection to inventors against misuse by the class of people to whom the relevant invention has been disclosed. Patent law requires disclosures to the public at large, and thus, the public is on constructive notice not to make use of the patented invention. Trade secrecy law principally counts as misappropriations unauthorized use of the secret information by the limited class to whom the information has been disclosed. 111 invention. This information is then construed as being a complete disclosure of the invention, even though not all relevant information was spelled out. In that sense, patent law permits the keeping of some secrets related to a patented invention.

*p. 18*
Compare patent law's enablement standard with trade secrecy's standard of relative secrecy. Patent law seeks to provide the public-in particular, the person having ordinary skill in the art-with enough information to make and use the invention, all the while conferring the protection of patent law's exclusive rights to the inventor. By comparison, trade secrecy laws provide that if an innovator provides the public with enough material to derive protected information (be it a publicly available product that can be reverse engineered or leakages of the protected information itself), the innovator will lose the shelter of trade secrecy protection. 113 The incentives to disclose are thus different: Inventors naturally want to disclose as little as possible about their creations, 114 but patent law's enablement requirement will push them to disclose a minimally sufficient amount. Choosing trade secrecy's protection, on the other hand, will lead inventors to disclose as little as possible about their creation and to make whatever is publicly available hard to reverse engineer, because once the cat is out of the bag, trade secrecy protection evaporates.

*p. 18*
What happens, then, when the worlds of trade secrecy and patent laws collide, as they might have in Tabor? We do not have enough information about the pump at issue in the litigation to evaluate whether Hoffman was right that it fell within the scope of Tabor's expired patent. However, suppose for the moment that Hoffman was correct. Recall that Tabor asserted in the context of litigation that it was not possible to derive the patterns corresponding to the pump's pieces without heavy and costly experimentation. 115 Holding Tabor to that claim suggests that the failure to disclose his pump's patterns in the patent-something it seems he did not do, at least completely, in his patent 116 -would contravene today's enablement standard 117 because a person having ordinary skill in the art would have to undertake undue experimentation to make the pumps without also having the patterns, just as Tabor did. Yet if Tabor gravely overestimated how much experimentation it would take a pattern maker to derive patterns for his pump, his patent 113 See supra text accompanying notes 97-100. 114 Fromer, supra note 69, at 552-53. 115 Supra text accompanying notes 6-8. 116 A perusal of Tabor's patent, U.S. Patent No. 33,550 (patented Oct. 22, 1861) (Improvement in Rotary Pumps), reveals that he provided no express patterns for his invention. However, Tabor did disclose some relative measurements for some, but not all, of the patented pump's parts. E.g., id. ("A part of the shell on the side of the discharge-orifice is the quadrant of a circle concentric with the valve-cylinder and of equal radius."). Nonetheless, it is unclear whether those relative measurements specify how the pump ought to be constructed or an approximation of how it ought to be should the pump parts expand or contract in use. 117 When Tabor secured his patent in 1861, an earlier version of patent law's current enablement standard was securely in place. For example, in 1832, the Supreme Court observed that patent law's requirement that "a correct specification and description of the thing discovered" be filed in a patent "is necessary in order to give the public, after the privilege shall expire, the advantage for which the privilege is allowed, and is the foundation of the power to issue the patent." Grant v. Raymond, 31 U.S. 218, 247 (1832). And in the Supreme Court's 1854 resolution of the validity of Samuel Morse's patent related to his invention of the telegraph, the Court held that Morse could not claim a broader invention than that which he discovered, and thus, enabled in the patent. O'Reilly v. Morse, 56 U.S. 62, 117 (1854).

*p. 19*
would be sufficiently enabled absent the patterns but his trade secrecy claim would be on shaky ground given the ease of reverse engineering. 118 All in all, it is hard to see why Tabor ought to be able to have it both ways: having an enabled patent without disclosing the pump's patterns, while having a valid trade secret in the secret patterns because they are difficult to reverse engineer from the commercially available pump. Modern case law makes much the same point. 119 For example, in Christianson v. Colt Industries Operating Corp., the litigants quarreled over whether information alleged to be a trade secret was required to have been disclosed in a patent to satisfy the enablement requirement. 120 Although the Seventh Circuit held that the information was outside of the patent's scope and therefore did not have to be revealed, 121 the opposite conclusion on patent scope would have likely necessitated application of the rule that the district court had invoked: "a patentee cannot, in equity, claim trade secrecy for conventional information which could readily have been discovered had the required disclosures been made in its patent." 122 Although there are difficulties applying this rule in the context of contemporary patent law, 123 it was that much harder in Tabor and Hoffman's time because then, peripheral patent claims were not well established. 124 As is evidenced by Tabor's patent itself, his patent claims were not peripheral but principally and loosely referred to the patent specification. 125 This central claiming in his patent made it hard to ascertain readily whether the current pump was the same, a question the Tabor courts never expressly engaged.

*p. 19*
There is a further possible collision between trade secrecy and patent laws in Tabor. Suppose that Hoffman was right that the pump being litigated ought to have been in the public domain by virtue of it falling within the scope of the expired patent and not being a significant enough improvement to constitute a new and separate invention. Should the patterns for that pump have landed in the public domain, capable of being copied so long as they are not obtained in a prohibited way, such as through theft? That 125 Tabor's patent has 2 claims: (1) "The formation of the shell with the concentric parts e g and f h and eccentric parts e h and g f, as and for the purposes described," and (2) "So constructing and arranging the valves, valve-cylinder, and shell as that the valves will not be subjected to a transverse movement while doing their work, for the purposes and substantially as set forth." U.S. Patent No. 33,550 (patented Oct. 22, 1861). is, are a pump and its corresponding patterns part and parcel of a single invention? 126 If the pumps and their patterns constitute a single invention, allowing the pump's patterns to be protected by trade secrecy when the pump is in the public domain (both through expiration of the patent, if Hoffman was correct, and even if not, through commercial sale of the pump without patent protection) might constitute an impermissible conflict between the trade secrecy and patent laws. Although its reasoning is hard to parse, it seems the Tabor trial court disagreed with this assessment when it stated that it was unnecessary to issue a finding on the differences between the patented invention and the pump at issue in the suit, "because both parties agree that it was necessary to have patterns and castings to manufacture the new pump." 127 If so, the court's unspoken assumption that the pump and its pattern are distinct inventions is unjustified without more firm reasoning.

*p. 20*
Given these multiple possible clashes between trade secrecy and patent laws in Tabor, perhaps it is not inconceivable to read sympathy with these concerns into Chief Judge Follett's succinct dissent in the Court of Appeals.

*p. 20*
The same issues that occupy a robust understanding of Tabor and trade secrecy law-the secrecy of information that is reverse-engineerable and how much disclosure of trade secrets is permissible before protection dissolves-is one that will likely arise anew in patent law. Until recently, the effect of maintaining a trade secret would have had a relatively clear effect on the possibility of obtaining patent protection. Under the patent law in effect until March 16, 2013, a patent was statutorily barred if, among other things, "the invention was … in public use … in this country, more than one year prior to the date of the application for patent in the United States." 128 A patent applicant's sale of a product produced by a secret method would have constituted a "public use" of the method, possibly barring the patentability of the method depending on whether the statutory-bar period had passed. 129 By contrast, a third party that had independently discovered that method could have patented it-assuming it had met the other patentability requirements-in spite of the other's prior secret use of that method, which would not have constituted a "public use" as against the third party. 130 The different treatment seemed to have been driven by the policy of encouraging patent applicants to file early rather than sit on their rights to extend their potential term of patent protection. 131 Since March 16, 2013, patent law is different: A patent is now statutorily barred instead if, among other things, "the claimed invention was in public use … or 126 The reasoning would go that they are essentially linked, in a way similar to the secrecy (or lack thereof) of private information about a publicly available product that is reverse-engineerable but has not yet been reverse engineered. See supra text accompanying notes 95-102. 127 otherwise available to the public" more than one year before the patent's filing date. 132 It is an open question whether a third party's trade secret linked to a commercially available product, like in Tabor, would be "otherwise available to the public," even if not "in public use." Are Tabor's patterns "available to the public" if they are theoretically reverse-engineerable by pattern makers? Or must the patterns be more immediately accessible to the public to qualify? These are principally reformulations of the issues posed above about secrecies and disclosures in trade secrecy law. Are the disclosures that invalidate a patent the same as the disclosures that render a trade secret public? 133 In sum, a review of the record in Tabor provides a helpful specimen with which to explore the tangle of disclosures and secrecies that trade secrecy and patent laws each promote, or at least tolerate. This examination chips away at the case's usual reading, that the pump's patterns were protectable as a trade secret because it was significantly hard to reverse engineer them. This evidence impels the questions of how secret information must truly be to qualify as a trade secret, both with regard to reverseengineerability and disclosures to third parties. The law tends to cut a good deal of slack for trade secrets to not be truly secret in this regard. On this metric, by comparison, patent law does a better job at disclosing inventions more widely. Yet there are ways in which patent law permits secret-keeping, with regard to the enablement standard and its statutory bars, in ways that come into tension with trade secrecy provisions. Moreover, there are important questions about what the patent law puts into the public domain that cannot later be maintained as a trade secret. All in all, there is an interconnected jumble of disclosures and secrets in trade secrecy and patent laws, sometimes moving in the same direction and sometimes cutting against one another. These tensions have been long present and still remain mostly unresolved.

## Footnotes

> Defendant's Requested Findings, supra note 17, at 13.Electronic copy available at: https://ssrn.com/abstract=2291726 Electronic copy available at: https://ssrn.com/abstract=2291726

> 62 Kewanee, 416 U.S. at 481; accord Lemley, supra note 58, at 329-32.63 See Lemley, supra note 58, at 319 (criticizing this theory for "presuppos[ing] a wrong without offering any substantive definition of what that wrong is").