This article was first published in IPWatchdog.com on July 14, 2020.
“The court’s analysis is flawed, and it relies on authority that is of limited value.”
Part 1of this article addressed the Eleventh Circuit’s decision in Compulife Software, Inc. v. Newman, __ F.3d __, 2020 WL 2549505, (11th Cir. May 20, 2020) and the court’s dubious conclusion that information “scraped” from a public website could be a trade secret. In particular, on this issue, the court held that even if the “scraped quotes were not individually protectable trade secrets because each is readily available to the public…taking enough of them must amount to misappropriation of the underlying secret at some point. Otherwise, there would be no substance to trade-secret protections for ‘compilations,’ which the law clearly provides.” Id. at *16 (emphasis in original). Part I concludes that this is contrary to a basic understanding of trade secret law for a number of reasons including that the information was freely accessible to the public, and that how the information was accessed is completely immaterial to the inquiry of whether such information meets the requirement that for information to be considered a trade secret that it must be subject to efforts that are reasonable under the circumstances to maintain its secrecy.
Part II of this article will address the understanding of “improper means” under trade secret law and whether the Eleventh Circuit was correct in determining that the use of bots to scrape a very large amount of information from a website can constitute “improper means” for acquiring such information.
As noted in Part I, Compulife and the defendants are direct competitors in the business of generating life insurance quotes. Compulife obtains rate tables from insurance companies, that are publicly available, compiles this information and makes them publicly available on a website that allows visitors to obtain life insurance quotes at no cost. Defendants using bots “scrapped” every possible combination of demographic data, totaling more than 42 million quotes. This is far more than the amount of information that could have been obtained through any other process, which the court found significant. Compulife alleged that the defendants then used the scrapped data as a basis for generating quotes on their own websites. According to Compulife, these facts establish a claim of misappropriation under the Florida Uniform Trade Secrets Act (FUTSA).
With regard to the “improper means” holding, the Eleventh Circuit “paused to unpack” this concept and noted that the FUTSA “may apply in both the acquisition and use contexts” and “’improper means’ is defined to include ‘theft, bribery, misrepresentation, breach or inducement or a breach of a duty to maintain secrecy or other means.” Id. at *14. The court went on to state that “[i]n the law of trade secrets more generally, ‘theft, wiretapping, or even aerial reconnaissance’ can constitute improper means ….” Id. (citing Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470, 476, 94 S.Ct 1879, 40 L.Ed. 2d 315 (1974). Moreover, the court found that “[a]ctions may be ‘improper’ for trade-secret purposes even if not independently unlawful.” (citing duPont deNemours & Co. v. Christopher, 431 F.2d 1012, 1014 (5th Cir. 1970).
The court explained that in Christopher, DuPont asserted that trade secrets had been misappropriated by defendants who took pictures of its methanol plant from a plane, which according to DuPont threatened to reveal its secret method of manufacturing methanol. The defendants moved for summary judgment asserting that “’they conducted all of their activities in public airspace, violated no government aviation standard, did not breach any confidential relation and did not engage in fraudulent or illegal conduct.’” Compulife 2020 WL 2549505 at *14, (citing Christopher 431 F.2d at 1014.)
However, the Fifth Circuit in Christopher rejected defendants “contention and held that the aerial photography constituted improper means even though DuPont had left the facility open to inspection from the air. According to Compulife, under the broad definition adopted in Christopher, misappropriation occurs whenever a defendant acquires the secret from its owner ‘without his permission at a time when his taking reasonable precautions to maintain its secrecy.’” Id. (citing Christopher 431 F.2d at 1015).
The Compulife court then went on to find even though the insurance quotes were taken from a publicly accessible does not necessary mean that the taking was proper:
“Although Compulife has plainly given the world implicit permission to access as many quotes as is humanly possible, a robot can collect more quotes than any human practicably could. So, while manually accessing quotes from Compulife’s database is unlikely ever to constitute improper means, using a bot to collect an otherwise infeasible [sic] amount of data may well be—in the same way that using aerial photography may be improper when a secret is exposed to view from above.”
The court also noted that in the “most closely analogous case of which we are aware, a district court held that hacking a public-facing website with a bot amounted ‘improper means’” (sic), citing Physicians Interactive v. Lathian Sys., Inc., No. CA 03-1193-A, 2003 WL 2301820, at *8 (E.D.Va. Dec. 5, 2003). According to the Eleventh Circuit “[i]n that case, the trade secret owner’s ‘failure to place a usage restriction on its website’ did not automatically render the hacking proper. So too here.” Id. at *16 (citation omitted).
Based on the above understanding, the court concluded that the magistrate judge was incorrect in concluding that the taking through the use of bots could not be considered improper and that this issue and other must be determined on remand. Id. at *17.
The court’s analysis is flawed, and it relies on authority that is of limited value. In reaching its decision that defendants’ conduct could constitute “improper means” the court relied primarily on Christopher, which, as described above, held that although the aerial photographs had not violated any law, espionage would signify a form of improper means, as the conduct practiced by the defendant could not reasonably be considered in accordance with the standards of commercial morality. In other words, pursuant to this analysis, a court should engage in a factual analysis in which the conduct charged will be considered relying on some undefined standard of commercial morality.
However, “improper means” under the Uniform Trade Secrets Act is defined as “theft, bribery, misrepresentation, breach or inducement of a breach of duty to maintain secrecy or espionage through electronic or other means.” The definition of “improper means” is identical under the FUTSA. While a limited number of courts have held that this does not provide an exhaustive list of what constitutes improper means, the better understanding is the opposite. By not limiting the “improper means” to the list in the UTSA, courts are free to determine whether defendant’s conduct constitutes improper means. And what is improper to one court maybe entirely improper to another court. This flexible standard is likely to lead to inconsistent outcomes among courts, whereas a more uniform standard will create greater certainty, which is beneficial to all parties. Competitive intelligence can play an important role in companies’ business decisions and large companies may spend millions of dollars on collecting competitive intelligence. The lack of clarity as to what constitutes permissible competitive intelligence, and what does not, can create tremendous problems for a company which believes that how it goes collecting competitive intelligence is permissible when it turns out that it is not.
Next, while Christopher has become widely cited in textbooks, scholarly commentary, and treaties, the Compulife decision is arguably the first appellate decision in more than 50 years that has relied upon Christopher, and which defendants were liable despite having broken no law and having breached no contract or confidential relationship. Surely, if other appellate decisions had involved similar circumstances to Christopher, the court in Compulife would have cited such a case. Further, the court admits that the “most closely analogous case of which we are aware” is the aforementioned Physicians Interactive, which turns out is not analogous at all. (Full disclosure, the author was lead counsel for the plaintiff Physicians Interactive).
The Compulife court quotes from Physicians Interactive that “[t]here can be no doubt that the use of a computer software robot to hack into a computer system and to take or copy proprietary information is an improper means to obtain a trade secret, and thus is misappropriation ….” Id. at 16. The Eleventh Circuit should have realized from just this quotation that Physicians is completely inapposite to the Compulife facts: Physicians involved “hacking” into a computer system without authorization to obtain confidential information. In contrast, in Compulife, there is no allegation that defendants hacked into the plaintiff’s computer system in order to obtain confidential information or trade secrets. The defendants in Compulife obtained publicly available information from a public website that at the time the information was obtained did not contain any use restrictions. The case did not involve illegal hacking and there was no allegation that confidential information was at issue. The two cases involve completely different factual situations, and Physicians does not support the Eleventh Circuit’s Compulife decision that there is a question of whether defendants had acquired the insurance quotes through improper means.
Courts Should Be Suspicious
In short, the Eleventh Circuit addressed two issues with regard to trade secret law. First, as described in Part I of this article, the court erroneously concluded that information “scraped” from a public website could be a trade secret. Second, as described herein, the court erroneously also concluded that the UTSA, and by extension other state statutes based on the UTSA, does not provide an exhaustive list of what constitutes improper means. In light of the prominence of this decision and that a number of commentaries that have already suggested that this case may offer website publishers a basis to combat website scrapers who rely on a public data argument to justify scraping, it is likely that the decision will lead to plaintiffs using this opinion to support their case. However, courts should be extremely suspicious of following this “very dubious” Eleventh Circuit decision.