NOT RECOMMENDED FOR PUBLICATION File Name: 25a0438n.06
No. 24-3726
UNITED STATES COURT OF APPEALS FILED FOR THE SIXTH CIRCUIT Sep 26, 2025 KELLY L. STEPHENS, Clerk
)
CINCOM SYSTEMS, INC., )
Plaintiff-Appellant, )
) ON APPEAL FROM THE v. ) UNITED STATES DISTRICT ) COURT FOR THE SOUTH-
LABWARE, INC., ) ERN DISTRICT OF OHIO )
Defendant-Appellee. ) OPINION )
Before: BOGGS, McKEAGUE, and MATHIS, Circuit Judges.
BOGGS, Circuit Judge. This case concerns federal copyright and state trade-secret claims brought by Plaintiff-Appellant Cincom Systems, Inc. (“Cincom”) against Defendant-Ap- pellee LabWare, Inc. (“Labware”) concerning the source code of VSE, a software-development tool written in the Smalltalk programming language. In 2019, Cincom discovered that LabWare was modifying and distributing VSE’s source code as part of LabWare’s own software product.
Cincom sued in federal district court on a number of claims, two of which are at issue here:
a trade-secret claim under the Ohio Uniform Trade Secrets Act (OUTSA) and a federal copyright claim. LabWare moved for summary judgment on both claims and the district court granted Lab- Ware’s motion. This appeal then followed.
For the reasons below, we affirm the judgment of the district court in its entirety.
BACKGROUND
VisualSmalltalk Enterprise (“VSE”) is a software-development tool—a software program that enables users to write their own programs in the Smalltalk language, which they can then sell to other users. One of VSE’s component pieces of software is a “virtual machine,” a common type of software that helps programs like VSE run on a user’s computer. Other varieties of Smalltalk software have their own unique virtual machines, which are distinct from VSE’s. VSE was origi- nally created by ObjectShare, Inc. (“ObjectShare”).
In 1999, ObjectShare sold all its rights in VSE to Seagull Software, Inc., along with related assets (including several copyright registrations). Seagull immediately licensed back certain rights in the program to the seller, ObjectShare. These included the exclusive, perpetual right to “market and license [VSE] directly or indirectly . . . to third party end-users for the express purpose of building Smalltalk-based applications . . . .” Shortly thereafter, Ohio software company Cincom Systems, Inc. (“Cincom”) purchased ObjectShare and all of its assets. At that time, ObjectShare assigned its rights in VSE to Cincom. Cincom thus possessed the exclusive right to market and license VSE, even though Seagull technically retained ownership of the source code.
Prior to its acquisition by Cincom, ObjectShare sold VSE using “shrinkwrap” licenses.
Under this model, VSE users received a printed software license with every physical copy of the software they purchased. This license permitted a single developer to use that copy of VSE “in developing end-user software applications.” Under the shrinkwrap license, individual developers could also transfer their copy of VSE to another user, but would have to themselves “permanently cease use . . . and destroy[] all copies” of the program. If a company wanted its software developers to use VSE, then, it would need to pay for copies of VSE (and accompanying shrinkwrap licenses) for each one of its developers. And under the terms of the license, each developer was forbidden
3
to reverse engineer or distribute copies of VSE itself—developers could only distribute their own original software, which they could use VSE to create.
After ObjectShare sold its VSE licensing rights to Cincom in 1999, Cincom abandoned the shrinkwrap-license model. Under Cincom’s new model, VSE licensees would pay Cincom a yearly subscription fee to use the product, plus a percentage of any revenue the licensees earned from selling any software created through VSE. Cincom continued to forbid licensees from modifying VSE’s underlying source code. And crucially, VSE customers who obtained a license to use the product did not receive a copy of, or gain access to, any of the underlying source code, including the source code to the virtual machine.
Accordingly, there are only two ways that a non-Cincom entity can now obtain a copy of VSE’s source code in violation of Cincom’s exclusive distribution right. First, that entity could reverse-engineer the VSE program; however, this process would be, as Cincom’s UK managing director Jason Ayers noted, “very time-consuming and difficult” as well as forbidden. Second, that entity could obtain a copy of the source code from someone else who possessed it, such as Rocket Software, the company that acquired Seagull and owns the underlying source code to VSE.
As the largest commercial purveyor of Smalltalk, Cincom markets other Smalltalk variants in addition to VSE. Cincom has apparently suggested to some VSE customers that they transition to other, more modern Smalltalk variants that Cincom offers. However, as of 2022, Cincom still maintained between thirteen and thirty VSE licensees as customers. Cincom has also engaged in litigation on several occasions to protect its licensing and distribution rights in VSE.
Since purchasing its interest in VSE, Cincom has also had some limited contact with the owners of the program’s source code, Seagull Software. In 2008, Cincom discovered that Seagull had subsequently sold a copy of VSE’s source code, despite Cincom’s exclusive right to license
4
and distribute the product. Cincom was able to definitively confirm Seagull’s sale to one company, named APIS, and suspected further sales to at least one other person, a man named Frank Lesser. After Cincom confronted Seagull, Seagull agreed to no longer sell VSE source code to any other party. However, Seagull apparently did not verify to Cincom whether it had sold source code to any entities beyond APIS (including Frank Lesser or any other yet-unknown company).
LabWare, Inc., a Delaware company, is the North American subsidiary of parent company LabWare Holdings, Inc. (collectively, “LabWare”). Appellee’s Br. at 2. LabWare develops soft- ware for managing laboratory data, including the Smalltalk software program LabWare Laboratory Information Management System (“LIMS”). Ibid. LabWare has been a VSE user since the 1990s. LabWare first acquired copies of VSE from ObjectShare, under that company’s shrinkwrap-li- cense model. When Cincom bought ObjectShare in 1999, Cincom apparently had no way of knowing the identities of preexisting ObjectShare customers such as LabWare.
As time went on, LabWare continued to use VSE under its existing shrinkwrap license.
Eventually, LabWare determined that VSE’s virtual-machine software was outdated, and in need of modernization. The terms of LabWare’s shrinkwrap license to use VSE forbade it from modi- fying the underlying source code, including the code to VSE’s virtual machine. However, in 2006, LabWare contacted Seagull, which at that time owned the VSE source code. Seagull agreed to illegally sell LabWare a license to the source code for VSE’s virtual machine, in violation of Cin- com’s trade-secret rights. Pursuant to its license from Seagull, LabWare then modified its copies of VSE’s virtual machine, to use and distribute with its LIMS products. At the time, LabWare was apparently aware that Cincom, and not Seagull, held the exclusive licensing rights to VSE—how- ever, LabWare wanted to avoid paying Cincom’s annual licensing fee and continuing royalty pay- ments, and preferred to make a onetime payment to Seagull.
5
In 2014, one of Cincom’s software engineers left the company for a position at LabWare.
The circumstances of the engineer’s departure caused Cincom’s executives to worry that he might take confidential information to his new job at LabWare. However, after some initial internal dis- cussion at the company, Cincom eventually dismissed this possibility as unlikely.
But five years later, in 2019, Cincom manager Suzanne Fortman attended a software con-
ference at which LabWare was presenting. The presentation triggered Cincom’s suspicions that LabWare was not only using VSE, but had accessed and modified the program’s source code.
After further investigation, Cincom filed suit against LabWare in the Southern District of Ohio, on January 30, 2020. LabWare successfully moved to dismiss Cincom’s claims for conver- sion and unjust enrichment, but the district court allowed Cincom’s federal copyright-infringement and state trade-secret claims (under the Ohio Uniform Trade Secrets Act) to proceed to discovery. LabWare initially filed a third-party complaint against Rocket Software, the successor to Seagull, which had sold LabWare the relevant source code to VSE in 2006. However, a month later, Lab- Ware and Rocket apparently reached a settlement. LabWare voluntarily dismissed its complaint against Rocket, stating that it “believes Cincom’s claims against LabWare to be without merit and shares a common interest with third-party defendant Rocket Software, Inc. (‘Rocket’), in defend- ing against and defeating them.”
After several years of discovery, both Cincom and LabWare moved for summary judg-
ment, which the district court then granted for LabWare in 2024. This timely appeal then followed.
ANALYSIS
On appeal, this court reviews the district court’s denial of summary judgment de novo.
Gillis v. Miller, 845 F.3d 677, 683 (6th Cir. 2017). Summary judgment is proper where “the mo- vant shows that there is no genuine dispute as to any material fact and the movant is entitled to
6
judgment as a matter of law.” FED. R. CIV. P. 56(a). On review, we “view all facts and inferences drawn therefrom in the light most favorable to the nonmoving party.” Kraft v. United States, 991 F.2d 292, 296 (6th Cir. 1993).
Initially, the party moving for summary judgment bears the burden of explaining the basis for its motion and identifying the portions of the record that “it believes demonstrate the absence of a genuine issue of material fact.” Celotex Corp. v. Catrett, 477 U.S. 317, 323 (1986). Next, the burden shifts to the nonmoving party, who must point to specific facts in the record that indicate a genuine factual issue remains for trial. Id. at 324; see FED. R. CIV. P. 56(e). The nonmoving party must offer more than a “scintilla” of evidence; in other words, it must show that, based on the record at summary judgment, a jury could reasonably find in its favor. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 252 (1986). There must be “evidence . . . upon which a jury could properly proceed to find a verdict for the party producing it, upon whom the onus of proof is imposed.” Id. at 251 (citation modified).
A. Ohio Uniform Trade Secrets Act (OUTSA) Claim Cincom first appeals the district court’s grant of summary judgment on its trade-secret claim. Cincom’s complaint claims that LabWare misappropriated Cincom’s trade secret in VSE, by “acquiring, using, marketing, licensing, and distributing the VSE Software . . . without a license and without authorization from Cincom.” For purposes of this appeal, neither party appears to dispute that the VSE software is a trade secret.
Cincom’s trade-secret claim arises under the Ohio Uniform Trade Secrets Act (OUTSA).
Where a plaintiff seeks to sue for trade-secret misappropriation, OUTSA requires that he bring his action “within four years after the misappropriation is discovered or by the exercise of reasonable diligence should have been discovered.” Ohio Rev. Code § 1333.66. A single continuing
7
misappropriation “constitutes a single claim” under OUTSA. Ibid. The Sixth Circuit has explained that “misappropriation is discoverable if the owner has knowledge of such facts as would lead a fair and prudent man, using ordinary care and thoughtfulness, to make further inquiry.” Campfield v. Safelite Grp., Inc., 91 F.4th 401, 415 (6th Cir. 2024) (citation modified). Because the statute of limitations is an affirmative defense, the defendant (LabWare) bears the burden of proving the limitations period has run. Metron Nutraceuticals, LLC v. Cook, No. 23-3596, 2024 WL 3877388 at *12 (6th Cir. Aug. 20, 2024) (citing Campbell v. Grand Trunk W. R.R. Co., 238 F.3d 772, 775 (6th Cir. 2001)).
In 2019, Cincom investigated LabWare’s use of VSE and modification of the software.
Appellant’s Br. at 21–23. That investigation resulted in Cincom bringing this action in January 2020. However, LabWare argues that Cincom was on inquiry notice of alleged trade-secret mis- appropriation in 2014. Appellee’s Br. at 23. The district court agreed, granting summary judgment for LabWare on this claim after determining that Cincom’s suit was untimely, because it fell out- side OUTSA’s four-year limitations period.
If “a fair and prudent man, using ordinary care and thoughtfulness” would have conducted a deeper investigation in 2014 than Cincom did, then the district court correctly determined that OUTSA’s four-year time-bar provision applies. Campfield, 91 F.4th at 415. But if, in 2014, Cin- com had no reason to further investigate whether LabWare was modifying VSE’s source code, then the statute of limitations would not have begun to run until Cincom received such notice in 2019. Therefore, we turn now to Cincom’s knowledge in 2014.
1. Cincom’s Actions in 2014 The relevant period in 2014 began when Andres Valloud, a virtual-machine engineer at Cincom, posted on his personal blog that he was leaving his job at Cincom for a position with
8
LabWare. Suzanne Fortman, Valloud’s manager, considered Valloud a “disgruntled” employee, and stated at deposition that Valloud had a difficult relationship with both her and his other coworkers. When Valloud announced his imminent departure, Fortman suggested to other Cincom management that the company limit Valloud’s access to assets on the company network. While at Cincom, Valloud did not work on VSE; instead, he was a virtual-machine engineer on another Cincom Smalltalk product, VisualWorks. Valloud’s posts did not describe in detail the work he would be doing at LabWare; they stated only that he would be “doing Smalltalk VM [virtual ma- chine] work.” On March 21, Fortman forwarded the post to her boss, Brian Bish, the Managing Director of Smalltalk at Cincom. Bish responded, “Hmm, interesting. Wonder which version of Smalltalk LabWare is using for their applications.” Eight minutes later, Bish emailed again, noting that he had done “some surfing on the internet and saw a reference to [VSE].”
Because the Smalltalk community is fairly close-knit and “well-networked,” other mem-
bers of the community outside of Cincom responded to Valloud’s post. In early April, Henrik Hoyer, an individual who ran an online forum for Smalltalk users, reposted a link to Valloud’s blogpost, and added a new allegation: “[Valloud] will be working on the [virtual machine] for Visual Smalltalk Enterprise (VSE) at LabWare.” The subject line of Hoyer’s message was “A great [virtual machine] engineer is turning his knowledge towards VSE.” At least one member of the forum responded to Hoyer’s post with congratulations for Valloud.
In response to Hoyer’s post, Bish emailed Jason Ayers, the European Sales Director for Cincom Smalltalk. He noted that “From everything I have been able to find, LabWare is using [VSE]. We have no record of them being a VSE customer on a Cincom [i.e., non-shrinkwrap] contract.” Bish added that Cincom might reach out to both Valloud and Hoyer, “to remind [them] that Cincom has exclusive distribution and support rights for VSE.”
9
Ayers then responded to Bish, and speculated about how LabWare might have obtained a copy of VSE without becoming a Cincom customer. First, he (correctly) acknowledged that Lab- Ware might have been one of ObjectShare’s customers back in the 1990s, who continued operating under one of the leftover, valid shrinkwrap licenses that predated Cincom’s interest in VSE. How- ever, Ayers also wrote that, even if LabWare had obtained a licensed copy of VSE, there was no way they could have legally obtained a copy of the underlying source code. Neither Cincom nor its predecessor, ObjectShare, had ever sold a copy of the program’s source code (including the source code for its virtual machine). Accordingly, if LabWare had a copy of the code, they must have gotten it either “by decompiling it or from Seagull—who does not have the rights to sell VSE in any form including [its virtual machine]. But I am guessing that is who they got it from.” Finally, Ayers noted that if Seagull had gone behind Cincom’s back and sold LabWare a copy of the source code, it would not be the first time this happened: “[I]f that is the case then it looks like [Seagull] sold it to three companies[,] not the two they told us about.” Bish responded “Perhaps. Probably difficult to go after LabWare but I can go after Andres [Valloud] if he decides to post any of his work to the VSE community.”
But Ayers and Bish’s initial suspicion soon cooled. Upon reviewing Valloud’s blog post, Ayers and Bish realized that Valloud’s original post and Hoyer’s subsequent repost said two dif- ferent things. To be sure, Hoyer’s repost mentioned VSE by name. But in his original post, Valloud said only that he was going to be working on a Smalltalk-based virtual machine at LabWare. And crucially, he did not specify what variety of Smalltalk he would be working on—whether a pro- prietary Smalltalk variant owned by Cincom (like VSE), or an open-source variant that was free to use.
10
Upon reflection, Ayers concluded that it was far more likely that LabWare would be using an open-source Smalltalk variant—particularly a variant called Pharo—than it would be to use the older VSE in its software products. Although Hoyer had specifically said Valloud would be work- ing on VSE, Valloud’s original post (which Hoyer cited) did not.
Ayers and Bish therefore concluded that Hoyer had merely misstated the nature of Val-
loud’s new position in his repost of the Valloud’s blog, and essentially ended their investigation there. Likewise, Fortman said that instead of directly confronting Valloud about any potential mis- appropriation, she “would rather wait for him to post something [to Hoyer’s forum], then we have evidence that he and/or his company are in violation.” In response, Bish wrote “Agreed.” Later, when Fortman said she was still waiting “to see something come up [on Hoyer’s forum] and then we’ll have proof,” Bish responded “We can only hope!”
At no point in 2014 did Cincom’s employees reach out to LabWare directly, to Valloud, or to Seagull’s successor company Rocket Software, with the aim of ascertaining whether anyone was violating Cincom’s trade-secret rights in VSE.
Months later in 2014, Cincom’s executives learned that LabWare had donated $50,000 to the Pharo Consortium, the nonprofit organization that owns the intellectual property to the open- source Smalltalk variant Pharo. In Ayers’s view, this was a “shitload of money” and “a significant act” that further indicated that LabWare was using Pharo in its software products, rather than VSE. Cincom would not become suspicious of LabWare again until 2019.
2. Analysis As the nonmovant at summary judgment, Cincom is entitled to have all inferences drawn in its favor. Kraft, 991 F.2d at 296. Nonetheless, a fair and prudent individual in Cincom’s position would have knowledge of all of the following facts in 2014: 1) Andres Valloud was going to work
11
on a Smalltalk virtual machine at LabWare; 2) Henrik Hoyer reported that Valloud was going to work on VSE at LabWare; 3) Bish had found online references to LabWare using VSE; 4) If Lab- Ware was working on the VSE virtual machine, it must have either violated its shrinkwrap license by reverse-engineering VSE’s source code, or violated Cincom’s trade-secret rights by buying the source code from Seagull; and 5) Seagull had already gone behind Cincom’s back, by selling third parties a copy of the VSE source code. Taken together, a fair and prudent individual would have conducted at least some further investigation into LabWare.
Cincom argues that it did not need to conduct any investigation, or that, if it did, its inves-
tigation was reasonable. First, it asserts that Hoyer was “wrong” when he announced that Valloud was going to work on VSE at LabWare. Appellant’s Br. at 15. But in fact, Hoyer was not “wrong”; although his repost made an assertion that the cited source (Valloud’s blog) did not include, he did not misquote or contradict the blog either. His announcement would otherwise have appeared en- tirely accurate to Cincom’s executives: he correctly summarized Valloud’s previous position at Cincom (an engineer for the VisualWorks virtual machine), and also correctly stated that Valloud would be working on a virtual machine at his new job. It is entirely plausible that Hoyer had other, unstated sources in addition to Valloud’s blog for the (extremely specific) claim that Valloud would be working on VSE’s virtual machine at LabWare.
Moreover, even if Hoyer misquoted Valloud, Cincom’s executive Brian Bish had seen other indications online that LabWare was using VSE as its chosen Smalltalk product. Given that Lab- Ware was potentially using VSE, that Valloud was a virtual-machine engineer, and (crucially) that there was no way for Valloud to do VSE virtual-machine work at LabWare without violating Cin- com’s license, these facts should have raised some suspicion that Valloud and LabWare were mis- appropriating Cincom’s trade secret. However, Bish apparently refused to consider at the time that
12
LabWare would behave illegally: “So therefore, if we believe that LabWare is honorable to their [shrinkwrap] license, and Andres [Valloud] is going to go work there, it has to be on something other than VSE because he doesn’t have access to the source code.” (emphasis added). And alt- hough Cincom had repeatedly clashed with unlicensed VSE users in the past, Bish testified that “I believe that most companies do not violate their contract, especially in the United States, where we are fairly litigious, right, as a society.”
Cincom’s real problem is that its executives hoped to wait for hard evidence of LabWare’s wrongdoing before beginning any investigation. For example, Brian Bish stated at deposition that Cincom had not really taken “any actions to investigate LabWare in 2014.” Suzanne Fortman, Valloud’s manager at Cincom, testified that the company didn’t reach out to inquire with LabWare directly because “we didn’t have confirmation at that point that . . . LabWare was using VSE.” Fortman also advised Jason Ayers to not “stir the pot” until Cincom had proof of a license violation. Critically, Bish noted that he had “no evidence that [Valloud] had done anything wrong, only sus- picion or concern that he might.” (emphasis added).
A plaintiff’s four-year window to sue under OUTSA may commence even before it receives ironclad proof of wrongdoing. The Sixth Circuit has stated that trade-secret owners need only learn of “possible misappropriation” before they must begin to take “affirmative steps” to investigate. Adcor Indus., Inc. v. Bevcorp, LLC, 252 F. App’x 55, 62 (6th Cir. 2007). Even “rumors” can some- times suffice to trigger the discovery rule. Ibid.
Cincom cites three Sixth Circuit cases in support of its position: Campfield v. Safelite Grp., Inc., 91 F.4th 401 (6th Cir. 2024), Metron Nutraceuticals, LLC v. Cook, No. 23-3596, 2024 WL 3877388 (6th Cir. Aug. 20, 2024), and B&P Littleford, LLC v. Prescott Mach., LLC, Nos. 20- 1449/1451, 2021 WL 3732313 (6th Cir. Aug. 24, 2021). None are availing. In Campfield, the
13
plaintiff’s only notice of potential misappropriation was a letter sent by the former employee claim- ing he had “documents from a trash dumpster,” though he did not provide the documents or de- scribe their contents. Id. at 415. The Sixth Circuit held that this vague reference was insufficient as “[n]othing in the letter suggested that the information . . . had been stolen, and a reasonable person would not assume that trade secrets are being thrown in the trash.” Ibid. The letter had therefore failed to trigger OUTSA’s discovery rule, and the plaintiff’s suit eight years later was not untimely. See ibid.
Cincom, however, had substantially more information than a single isolated reference to unspecified “documents from a trash dumpster.” By contrast, Cincom’s executive Brian Bish had seen multiple references online to LabWare using his company’s proprietary software. He had also just read Hoyer’s blog indicating that one of his company’s former employees might be doing virtual-machine work on VSE at LabWare, which would implicate VSE’s protected source code. And he knew that the owner of the source code, Seagull/Rocket, had previously sold the code in violation of Cincom’s rights.
In Metron, the Sixth Circuit denied summary judgment to a defendant who raised OUTSA’s statute of limitations, reversing the district court’s ruling. 2024 WL 3877388 at *13. Metron alleged that defendant Thomas had conspired with a Metron employee and “copied a specific Metron doc- ument, used that document to draft [his] own patent, and used that patent to develop and market products.” Id. at *13. Thomas argued that Metron’s 2020 lawsuit was time-barred by a previous lawsuit from February 2016, in which Metron accused Thomas of unsuccessfully trying to obtain information from Metron’s investors and attorneys in order to “knock off” Metron’s patent. Id. at *12.
14
We rejected this, however, because Thomas had participated in two wholly distinct schemes to misappropriate trade secrets. Id. at *13. The fact that Thomas had tried to obtain information from investors did not give Metron notice that he would later conspire with an employee to copy documents. Ibid. And most critically, neither Thomas nor the court could identify any affirmative steps that Metron should have taken to discover Thomas’s misappropriation sooner. Ibid.
Cincom’s situation is distinguishable from Metron because, first, it did not encounter two separate instances of trade-secret misappropriation: LabWare began violating its license when, un- beknownst to Cincom, it purchased the source code from Seagull in 2006 and continued to do so until this suit. See Ohio Rev. Code § 1333.66. While Cincom’s suspicions were piqued in both 2014 and 2019, it was responding to the same conduct. A confidential relationship “once rent is not torn anew with each added use or disclosure,” meaning Cincom’s time limit to bring its claim did not refresh in 2019 alongside its refreshed suspicions. Kehoe Component Sales Inc. v. Best Lighting Prods., Inc., 796 F.3d 576, 583 (6th Cir. 2015) (quoting Monolith Portland Midwest Co. v. Kaiser Aluminum & Chem. Corp., 407 F.2d 288, 293 (9th Cir. 1969)). To rely on Metron, Cincom would have had to bring a claim that LabWare engaged in a wholly unique misappropriation scheme in 2019. Additionally, unlike Metron, it is clear what affirmative steps Cincom could have taken to discover LabWare’s misappropriation in 2014, because it took them in 2019.
In 2019, the company again developed suspicion that LabWare was modifying and distrib-
uting VSE as part of its LIMS software—this time, after viewing a slideshow by LabWare at a trade conference. Again, as in 2014, Cincom lacked affirmative proof of wrongdoing. But this time, Cincom’s suspicion led to the company pursuing multiple leads to find its proof. Following the presentation at the trade conference, Suzanne Fortman discussed Cincom’s rights in VSE with a LabWare employee named John McIntosh. McIntosh mentioned that LabWare had obtained
15
some sort of license from Seagull/Rocket, which gave rise to further suspicion. Meanwhile, Jason Ayers searched various online sources (Google, LabWare’s website, LinkedIn, GitHub, LabWare’s user manuals, etc.), and eventually found a reference to VSE in a piece of LabWare’s LIMS source code. Upon further digging, Ayers found part of LabWare’s code online, and determine that “the entire . . . VSE development environment, was sitting in the main executable of LabWare 7.0.2, which is pretty gobsmacking actually.” It was at that point that Cincom decided to retain outside legal counsel. In summary, this rapid, successful investigation stands in sharp contrast to Cincom’s seeming reluctance to investigate in 2014, and provides a model of how the company might have more diligently policed its trade-secret right.
Third and finally, Cincom relies on B&P Littleford, LLC v. Prescott Mach., LLC, Nos. 20-
1449/1451, 2021 WL 3732313 (6th Cir. Aug. 24, 2021), in which this court considered a trade- secret claim arising under the Michigan Uniform Trade Secrets Act (MUTSA). Like OUTSA, MUTSA is based on the Uniform Trade Secrets Act. See id. at *5; Stolle Mach. Co. v. RAM Pre- cision Indus., 605 F. App’x 473, 484 (6th Cir. 2015). In B&P Littleford, the Sixth Circuit again reversed the district court’s grant of summary judgment against a trade-secret plaintiff, on the grounds that a jury could find MUTSA’s limitations period had not run against the plaintiff. 2021 WL 3732313, at *8. The court reasoned that although the plaintiff had some suspicions of misap- propriation more than four years before bringing its suit, it had worked diligently to investigate those suspicions, and failed to turn up evidence through no fault of its own. Ibid. In its decision, the Sixth Circuit distinguished two prior cases, Stolle and Adcor, where plaintiffs had received word from customers or employees of possible misappropriation but failed to diligently follow up to seek further information. Ibid. (discussing 605 F. App’x at 483; 252 F. App’x at 60–62). In B&P Littleford, however, the plaintiff had reached out to vendors to ask if they had heard about the
16
alleged misappropriation, or might be participating in it. The Sixth Circuit reasoned that a jury could find this investigation reasonable, and allowed the case to proceed past summary judgment. 2021 WL 3732313, at *8.
But as the district court noted, Cincom’s behavior in 2014 is “readily distinguishable” from the plaintiff’s diligent investigation in B&P Littleford. Cincom Sys., Inc. v. Labware, Inc., 2024 U.S. Dist. LEXIS 151357, at *38. Cincom was not “stymied by a lack of documentation and [oth- ers’] responses,” because it did not seek documentation or responses in 2014—not from Andres Valloud, not from LabWare itself, and not even from Seagull/Rocket, which it had previously contacted under similar circumstances. 2021 WL 3732313, at *8. If Cincom had made at least some showing of chasing down these leads, then it would at least arguably have behaved as a reasonably prudent person should, and LabWare would not be entitled to summary judgment based on the OUTSA statutory time limit. See ibid.
Because Cincom was on inquiry notice of LabWare’s trade-secret misappropriation in 2014, its claim is time-barred by OUTSA’s statute of limitations. B. Copyright Infringement Claim Cincom’s second claim on appeal is that the district court erred by granting summary judg-
ment against its federal copyright-infringement claim. Cincom alleges that LabWare engaged in various forms of unauthorized copying of the VSE Version 3.1 software, without obtaining Cin- com’s permission despite Cincom’s exclusive copyright in the software. However, Cincom has no copyright registration on VSE Version 3.1, the software that LabWare allegedly copied. Instead, Cincom alleges infringement of a copyright registration for Registration TX 3-733-100 (“the ‘100 Registration”), issued in 1994, which covers Version 1.3 of the VSE Software. Two other regis- trations, TXu 192-061 and TX 951-742 (“the Prior Registrations”), issued in 1984 and 1985,
17
apparently cover even older versions of VSE. However, Cincom alleges infringement of only the ‘100 Registration, not the Prior Registrations.
Consequently, to show that LabWare has infringed on the ‘100 Registration by copying Version 3.1, Cincom must be able to show that in doing so, they copied code protected by the ‘100 Registration. However, there is scant documentation in the record pertaining to the ‘100 Registra- tion. The only evidence Cincom has produced is a single-page printout from the Copyright Office’s official database entry for the ‘100 Registration. The printout identifies the subject of the registra- tion as “Smalltalk/V [i.e., VSE] PM (version 1.3),” and describes the basis of the registrations as “New Matter: rev. & supplemental text.” Cincom has not produced a copy of the actual code pro- tected by the ‘100 Registration, which the original owner (ObjectShare) was required to deposit with the Copyright Office to obtain the registration for Version 1.3. See 17 U.S.C. § 408(b). Ac- cording to Cincom, the deposit is “unfortunately not available for review.” Appellant’s Br. at 45.
To win a copyright-infringement claim, “the claimant must prove two elements: ‘(1) own-
ership of a valid copyright, and (2) copying of constituent elements of the work that are original.’” RJ Control Consultants, Inc. v. Multiject, LLC, 100 F.4th 659, 667 (6th Cir. 2024) (quoting Feist Publ’ns, Inc. v. Rural Tel. Serv. Co., 499 U.S. 340, 361 (1991)). At both the district court and now on appeal, the parties do not dispute that Cincom has a valid interest in the ‘100 Registration, even though the registration itself is now owned by Rocket Software (Seagull Software’s successor).1 But as to the second element of the infringement analysis, Cincom’s inability to produce any of the purportedly protected code leaves the panel unable to “test[] whether any copying oc- curred (a factual matter) and whether the portions of the work copied were entitled to copyright
1 Under the Copyright Act, exclusive licensees like Cincom are generally understood to be able to bring claims for infringement. See Gardner v. Nike, Inc., 279 F.3d 774, 779 (9th Cir. 2002).
18
protection (a legal matter).” ECIMOS, LLC v. Carrier Corp., 971 F.3d 616, 628 (6th Cir. 2020) (quoting Lexmark Int’l, Inc. v. Static Control Components, Inc., 387 F.3d 522, 534 (6th Cir. 2004)).
Cincom hired an expert, George Heeg, to compare the code of Version 3.1 (which LabWare is alleged to have copied) and Version 1.3. The expert’s report states that 68% of Version 1.3’s code can be found within the later Version 3.1. Therefore, if the ‘100 Registration protected every line of code in Version 1.3, Cincom would have credibly alleged that LabWare infringed on cop- yrighted code when it copied Version 3.1, because 68% of Version 1.3 is contained in Version 3.1. However, while Cincom argues that the ‘100 Registration covers every line of code in Version 1.3, LabWare argues that it covers only the new lines of code added after the Prior Registrations issued (in 1984 and 1985).
LabWare’s argument is better supported by fact, statute, and precedent. The ‘100 Registra-
tion describes itself as covering only “New Matter: rev. & supplemental text,” which supports LabWare’s understanding of the registration’s coverage. Moreover, the Copyright Act cautions registrants that “copyright in a compilation or derivative work extends only to the material con- tributed by the author of such work, as distinguished from the preexisting material employed in the work, and does not imply any exclusive right in the preexisting material.” 17 U.S.C. § 103(b) (emphasis added). The Copyright Office understands this statute to apply specifically to successive releases of computer programs, and advises registrants that only new material is protected in each successive registration of software. See SimplexGrinnell LP v. Integrated Sys. & Power, Inc., 642 F. Supp. 2d 206, 213 (S.D.N.Y. 2009).
Cincom cites two cases that it claims hold that a later registration (i.e., the ‘100 Registra-
tion) can incorporate by reference the earlier registrations (i.e., the Prior Registration). Appellant’s Br. at 45-46. However, neither is availing. Cincom’s primary authority is a single line in our
19
opinion in Murray Hill Publ’ns, Inc. v. ABC Commc’ns., Inc., 264 F.3d 622, 632 (6th Cir. 2001), abrogated on other grounds by Reed Elsevier, Inc. v. Muchnick, 559 U.S. 154 (2010). In Murray Hill, the plaintiffs registered their copyright for the original song “Jeanette” but did not register a copyright in a subsequent derivative song, “J.P.’s Theme.” 264 F.3d at 629–30. We held that, despite plaintiffs’ ownership of the original work, without more, they could not sue for infringe- ment of the derivative work after failing to register that work as well. Id. at 632. In the process, the court acknowledged that in the inverse situation—where a derivative is registered but the orig- inal is not—the original can nonetheless gain some protection. Id. at 631–32. For example, where the author L. Ron Hubbard failed to register various Scientological texts he wrote (the originals), then registered a compendium volume that collected those writings (the derivative), a district court determined that the originals were also protected by the registration. Id. at 631 (citing Religious Tech. Ctr. v. Netcom On-Line Comm. Servs., Inc., 923 F. Supp. 1231, 1241–42 (N.D. Cal. 1995)).
In drawing this distinction, the Sixth Circuit stated: “Because a derivative work is cumula-
tive of the earlier work, it is logical that the registration of the derivative work would relate back to include the original work . . . .” Id. at 632. Drawing on this one line, Cincom reasons that it, too, should be permitted to relate its derivative work (the ‘100 Registration) back to the original work (the Prior Registrations). However, Murray Hill was discussing an entirely different fact pattern (an unregistered original work, and a registered derivative work) than the instant case (a registered original and a registered derivative). Placed in its surrounding context, the quotation from Murray Hill is not applicable to Cincom’s situation, which involves a plaintiff attempting to tack prior registrations’ copyright coverage onto a subsequently registered derivative.
The same is true for Cincom’s citations to Gamma Audio & Video, Inc. v. Ean-Chea, 11 F.3d 1106 (1st Cir. 1993). In that case, the infringed derivative works (Cambodian-language
20
episodes of a TV series) were unregistered, while the original works (Chinese-language episodes of the same series) were registered. Id. at 1111. Nonetheless, the First Circuit held that the plaintiffs could use the original works’ registrations to pursue the defendants for infringing the derivative works. Id. at 1112. By comparison, however, Cincom is not asserting violations of the Prior Reg- istrations in the instant case; it did not name those registrations in its motion for summary judg- ment, nor does it discuss them in its appeal. Contrary to Cincom’s position, the Gamma Audio court explained that “the copyright in a derivative work only protects the original elements con- tributed by the author of the derivative work . . . . Any elements that the author of the derivative work borrowed from the underlying work . . . remain protected by the copyrights in the underlying work.” Ibid. (emphasis added). And again, Cincom has not asserted or sought the protection of the copyrights in the underlying work (the Prior Registrations) in its suit.
Cincom cannot show which, if any, portions of Version 3.1’s code come from the protected portion of Version 1.3. There is no evidence in the record that any of the code in Version 3.1 is “new matter” protected by the ‘100 Registration, so we cannot engage in the required substantial- similarity analysis to determine whether any copyright infringement occurred. See Automated Sols. Corp. v. Paragon Data Sys., 756 F.3d 504, 518 (6th Cir. 2014).
Therefore, summary judgment in favor of LabWare on the copyright-infringement claim was appropriate.
CONCLUSION
For the reasons above, the judgment of the district court is AFFIRMED.