OPINION & ORDER

JAMES D. PETERSON, District Judge.

This is a patent infringement case involving electronic cattle monitoring systems. The court granted summary judgment of non-infringement of both patents-in-suit to defendant Agis Automatisering B.V. Dkt. 289, at 1. The court also granted summary judgment against plaintiffs on their allegation of willful infringement. Id. On September 1, 2016, the Federal Circuit affirmed. Dkt. 336-1.

Before the case went up on appeal, defendant moved for attorney fees pursuant to 35 U.S.C. § 285. Dkt. 304. Defendant contends that plaintiffs' infringement claims relied on baseless claim constructions, and that the suit was motivated by plaintiffs' desire to impede competition rather than a sincere interest in its intellectual property rights. The question is a close one, but the court is not convinced the case is "exceptional" under Octane Fitness, LLC v. ICON Health & Fitness, Inc., 134 S.Ct. 1749 (2014), or that fee shifting is warranted. The court will deny the motion.

ANALYSIS

Under § 285 of the Patent Act, "[t]he court in exceptional cases may award reasonable attorney fees to the prevailing party." 35 U.S.C. § 285. For nearly a decade, under Brooks Furniture Manufacturing, Inc. v. Dutailier International, Inc., 393 F.3d 1378 (Fed. Cir. 2005), an "exceptional" case was one that involved some actionable misconduct. But now, under Octane, an "exceptional" case is "simply one that stands out from others with respect to the substantive strength of a party's litigating position (considering both the governing law and the facts of the case) or the unreasonable manner in which the case was litigated." Octane, 134 S. Ct. at 1756; see also Highmark Inc. v. Allcare Health Mgmt. Sys., Inc., 134 S.Ct. 1744, 1748 (2014). District courts determine whether a case is exceptional on a case-by-case basis, considering the totality of the circumstances. Octane, 134 S. Ct. at 1756. The moving party must demonstrate that the case is exceptional by a preponderance of the evidence. Id. at 1758. The determination is a discretionary one for the court.

The question is not governed by clear-cut rules, but the court's discretion is guided by the purpose of fee-shifting under § 285, which is to dissuade unreasonable lawsuits and litigation tactics by compensating the prevailing party for enduring them. See, e.g., Commil USA, LLC v. Cisco Sys., Inc., 135 S.Ct. 1920, 1930-31 (2015). The court should not award fees to punish a patent holder simply for failing to prove infringement. Gaymar Indus., Inc. v. Cincinnati Sub-Zero Products, Inc., 790 F.3d 1369, 1373 (Fed. Cir. 2015) (citing Octane, 134 S. Ct. at 1753). A litigation position that seems obviously wrong after a decision on the merits might have seemed reasonable when judged without the benefit of hindsight. And fee-shifting in routine cases would deter legitimate efforts to enforce patent rights. So the court will award fees only if there has been some unreasonable conduct that warrants imposing on the losing party the additional burden of paying the other side's attorney fees. See Intellect Wireless, Inc. v. Sharp Corp., 45 F.Supp.3d 839, 848 (N.D. Ill. 2014). The court need not find that the losing case was frivolous, or that it was brought in bad faith for an illegitimate purpose (although that would clearly warrant fee-shifting). See Lumen View Tech., LLC v. Findthebest.com, Inc., 24 F.Supp.3d 329, 335 (S.D.N.Y. 2014), aff'd, 811 F.3d 479 (Fed. Cir. 2016). Such clearly inappropriate cases and tactics would be sanctionable pursuant to Rule 11 or the court's inherent powers even without § 285.

Because the purpose of § 285 is to compensate parties forced to defend against unreasonable litigation and to deter improper conduct, a good way to think of the question is this: was the losing party's case so unusually weak on the merits that it suggests an improper purpose or demonstrates irresponsible conduct that should be deterred? Or, is there other evidence, besides substantive weakness on the merits, that suggests an improper purpose? With these questions in mind, the court turns to the parties' arguments.

A. Evidence of copying by defendant

The court begins with plaintiffs' contention that they received important evidence of copying by defendant only after the parties had briefed the motion for summary judgment. The purported evidence of copying and its allegedly late disclosure are immaterial. At best, the information would have informed the parties' validity positions, not their infringement positions. See Allen Eng'g Corp. v. Bartell Indus., Inc., 299 F.3d 1336, 1351 (Fed. Cir. 2002) ("[C]opying may have relevance not on infringement but on validity as one of a number of `indicia of obviousness or nonobviousness.'" (citations and internal quotation marks omitted)). The court did not reach the issue of validity in its summary judgment decision.

Plaintiffs do not adequately explain how the "late" discovery would have had any bearing on claim construction, plaintiffs' knowledge of the accused system, or plaintiffs' infringement positions. As defendant aptly phrases it, "plaintiffs' scattershot allegations of copying cannot be used as a proxy for establishing infringement." Dkt. 330, at 28. The court concludes that the copying evidence has no bearing on whether this case is exceptional.

B. Bad faith by plaintiffs

Defendant contends that plaintiffs litigated this case in bad faith to prevent competition. In support, defendant cites two actions by plaintiffs: a press release that plaintiffs circulated a few months after filing suit; and a meeting that plaintiffs had with one of SCR's U.S. distributors, who had indicated that it would begin distributing defendant's products. Defendant's showing is not persuasive.

Plaintiffs' press release was a simple announcement that they had sued defendant for infringing their patents and that they were committed to vigorously protecting their intellectual property rights. See Dkt. 305-6. The statements about the litigation were accurate, and plaintiffs were within their rights to publicly proclaim their intent to enforce their patents. Defendant surmises that the press release was calculated to discourage customers from dealing with defendant. But any proclamation by a patent owner that it will enforce its patent rights, or that it has sued a competitor, carries that implication. Factually accurate press releases about litigation, such as the one issued by plaintiffs here, do not suggest any bad faith or illegitimate anti-competitive motive on the patent owner's part.

The distributor meeting does not demonstrate bad faith on plaintiffs' part either. Based on the record before the court, sometime during the litigation, defendant reached an exclusive marketing agreement with Select Sires, a major U.S. distributor. But Select Sires was already distributing plaintiffs' products, and under its agreement with plaintiffs, Select Sires was required to inform plaintiffs before dealing in competitive products. After being so notified, plaintiffs requested a meeting with Select Sires, at which plaintiffs asked Select Sires to forestall distributing defendant's products until the patent litigation was resolved. Defendant does not address the issue in its reply brief, so maybe it has conceded the point. But nothing about this situation suggests that plaintiffs were using the litigation in a bad-faith attempt to interfere with competition.

C. Substantive strength of plaintiffs' case

That leaves the substantive strength of plaintiffs' litigation position. "In Octane Fitness, the Supreme Court made clear that it is the substantive strength of the party's litigating position that is relevant to an exceptional case determination, not the correctness or eventual success of that position." SFA Sys., LLC v. Newegg Inc., 793 F.3d 1344, 1348 (Fed. Cir. 2015) (citation and internal quotation marks omitted). Defendant contends that plaintiffs' proposed claim constructions and related infringement positions were totally baseless. The court considers plaintiffs' case for each patent-in-suit.

1. '481 patent (the rumination patent)

At summary judgment, the court construed two terms with respect to the '481 patent: "sensor for sensing chewing actions" and "data processor." Dkt. 289, at 7. The court determined that both terms were means-plus-function limitations because the inventor had expressly acknowledged that they were during prosecution, and that under the court's construction of the terms, defendant was entitled to summary judgment of non-infringement of the '481 patent.

Plaintiffs contend that they were entitled to argue that the terms were not means-plus-function limitations. Plaintiffs are correct that, under the applicable law at the time the court issued its summary judgment opinion, the absence of "means for" phrasing created a strong presumption that the terms were not means-plus-function limitations. See id. at 8. But plaintiffs were not entitled to rely on the presumption. During prosecution, the inventor expressly stated that amendments to the relevant claim created means-plus-function limitations. See Dkt. 41-1, at 261 ("Thus, the recitation in Claim 18 of a `sensor for sensing' means exactly the same as `means for sensing,' and therefore is clearly a recitation of structure to be given patentable weight . . . Similarly, the recitation in Claim 18 of a data processor accumulating both the time of each of said sensed chewing actions and the number of such chewing actions per unit time interval, for determining the chewing rhythm of the animal, is also a recitation of `means plus function' expressly permitted under 35 U.S.C. [§] 112[.]"). These statements barred plaintiffs from reversing course during litigation: the statements should have stopped plaintiffs from arguing that "sensor for sensing chewing actions" and "data processor" are not means-plus-function limitations. But, to be fair, plaintiffs did not stake their whole infringement case on this point. To the contrary, plaintiffs appeared to recognize that this was not a winning point, and they argued in the alternative that even if the terms were means-plus-function limitations, the patent disclosed a structure that would read on defendant's product. The fact that plaintiffs maintained a feeble secondary argument does not make this case exceptional.

The decisive issue at summary judgment was whether structures disclosed in the '481 patent read on defendant's product. In short, plaintiffs relied on overly broad constructions that the court rejected.1 The court construed "sensor for sensing chewing actions" to require the corresponding structure of a "sound sensor." Plaintiffs argued that the accelerometer in the accused product was the equivalent of a microphone, relying on their infringement expert, Daniel J. Aneshansley, PhD. The doctrine of equivalents would have been available to plaintiffs, because the scope of a means-plus-function limitation includes the corresponding structure plus equivalents. 35 U.S.C. § 112(f); see also Noah Sys., Inc. v. Intuit Inc., 675 F.3d 1302, 1318 (Fed. Cir. 2012). But the court rejected plaintiffs' equivalents argument as "implausible, undeveloped, and supported only by inadmissible evidence." Dkt. 289, at 15. The court continues to view this argument as particularly weak.

But that was not plaintiffs' only argument about the construction of "sensor for sensing." The patent states that "[t]he sensors for sensing chewing actions are preferably sound sensors . . . However, other types of sensors could be used." '481 patent, at 2:43-46. The court determined that only sound sensors had been adequately disclosed as corresponding structures, but that point is not beyond reasonable debate. The court rejected plaintiffs' arguments about the scope of the claim, and the related infringement arguments, but the court concludes that plaintiffs' arguments were not so weak as to make the case exceptional.

In connection with the analysis of the "data processor" element, the court determined that defendant's system does not detect rumination using any of the algorithms disclosed as corresponding structures. The critical issue was whether defendant's system measured the time of each chew or counted individual chews. As the court explained in its summary judgment opinion, plaintiffs misunderstood Paul Rump's testimony: Rump explicitly testified that the accused system is "never capable of detecting a frequency." Dkt. 289, at 16 (citing Dkt. 101 (Rump Dep. 131:13-14)). Plaintiffs' expert, Dr. Aneshansley, had strong credentials, but the court was not persuaded by his analysis of the operation of defendant's system under the terms of the '481 patent. The court continues to regard plaintiffs' arguments about the "data processor" element as weak ones, but they are not so unusually weak as to make this case exceptional.

In sum, for the '481 patent, plaintiffs put up a weak infringement case, predicated on marginal claim constructions. The court would hope that a patent owner would think twice about pressing a case like this. But the court must balance the concern for discouraging unreasonable conduct against the concern that the court not discourage good-faith, even if ultimately unsuccessful, enforcement efforts. Accordingly, the court concludes that the case for the '481 patent was not so beyond the pale that the case is exceptional or that fee-shifting is warranted.

2. '149 patent (the estrus patent)

On claim construction of the '149 patent, the court rejected aspects of both sides' claim constructions of three terms. The decisive issue was the meaning of the "attenuation" step. Plaintiffs advocated for a broad definition that would have applied if the accused device discounted evidence of animal motion whenever there was an indication of eating behavior. Defendant advocated for a narrower construction. The court did not completely adopt either side's construction, but the court concluded that plaintiffs' construction was foreclosed by the prosecution history, ultimately adopting a construction closer to defendant's. Plaintiffs' infringement case was predicated on Dr. Aneshansley's catalog of all the ways in which defendant's system could be said to "attenuate" data. The court was not persuaded because Dr. Aneshansley did not show how defendant's system attenuated the acceleration signal in response to an increase in the indication of eating, as the claims required.2

Plaintiffs' infringement case was weak on the critical "attenuating" element. But plaintiffs' arguments based on Dr. Aneshansley's opinions about the various forms of attenuation that he found were at least arguable. These positions were not so weak that they demonstrated the kind of irresponsible litigation that should be deterred by fee-shifting.

CONCLUSION

Summary judgment was not a close call. Whether plaintiffs' case was so weak on the merits that it was "exceptional" under § 285 is a closer question. But plaintiffs' case was not so far outside the norms of patent litigation that it demonstrated irresponsible litigation or suggested that plaintiffs maintained the suit for some improper purpose. Plaintiffs presented arguable positions supported by qualified experts. Even if some of plaintiffs' positions were weak ones, fee-shifting under the circumstances of this case would discourage aggressive but good-faith patent enforcement.

ORDER

IT IS ORDERED that defendant Agis Automatisering B.V.'s motion for attorney fees under 35 U.S.C. § 285, Dkt. 304, is DENIED.