Opinion
18 C 8175
03-08-2022
KOVE IO, INC., Plaintiff, v. AMAZON WEB SERVICES, INC., Defendant.
MEMORANDUM OPINION AND ORDER
REBECCA R. PALLMEYER, United States District Judge.
Plaintiff Kove IO, Inc. alleges that Amazon Web Services, Inc. (AWS) infringed three patents that disclose a distributed-data-storage technology used for large-scale cloud storage. In this motion, AWS asks the court to stay the litigation because the United States Patent and Trademark Office (USPTO) has initiated reexamination of all three patents in suit. The court grants the motion for the reasons provided below.
BACKGROUND
Kove filed this case in December 2018, alleging that two of AWS's products (S3 and DynamoDB) infringed three of Kove's patents (U.S. Patent Nos. 7, 103, 640, 7, 814, 170, and 7, 233, 978). Since then, the parties have engaged in extensive fact discovery and energetic motions practice; the court has denied a motion to dismiss, issued a Markman order, and resolved several discovery disputes. Magistrate Judge Finnegan has devoted substantial time to the litigation as well. Despite these efforts, plenty of work remains, including more fact discovery, the entirety of expert discovery, dispositive motions, and trial.
AWS served its final non-infringement contentions on the deadline of August 13, 2020. (See Am. Scheduling Order [180]; Final Noninfringement Contentions of Def. Amazon Web Services, Inc. [183-2].) Fifteen months later, on November 19, 2021, AWS petitioned the USPTO for ex parte reexamination (EPR) of all three patents in suit. (See AWS's Mot. to Stay Case Pending Outcome of Ex Parte Reexaminations [490] (hereinafter “AWS Mot.”) at 3.) EPR is an administrative procedure by which the USPTO evaluates the validity of a patent based on prior-art references. See 35 U.S.C. §§ 301-302. A petition for reexamination may be filed by “[a]ny person at any time, ” id. § 302, and the USPTO will initiate EPR if it concludes that the petition raises a “substantial new question of patentability, ” see Id. §§ 303-304. Generally, a patent owner engaged in EPR may propose amendments to a patent so long as those amendments would not “enlarge[] the scope” of any claim. Id. § 305. But where the patent under reexamination has already expired, “no amendment, other than the cancellation of claims, will be incorporated into the patent” as a result of the EPR. 37 C.F.R. § 1.530(j); see also In re Rambus, Inc., 753 F.3d 1253, 1255-56 (Fed. Cir. 2014). And, of course, “when a [patent] claim is cancelled, the patentee loses any cause of action based on that claim, and any pending litigation in which the claims are asserted becomes moot.” Fresenius USA, Inc. v. Baxter Int'l, Inc., 721 F.3d 1330, 1340 (Fed. Cir. 2013).
AWS's petitions for EPR identify prior-art references that were not considered during the initial patent prosecution and that AWS says it did not uncover until the fall of 2021. (AWS Mot. at 3-4; AWS's Reply in Further Support of Mot. to Stay Case Pending Outcome of Ex Parte Reexaminations [507] (hereinafter “AWS Reply”) at 12-13.) The USPTO granted the petitions in December 2021 and January 2022. (AWS Mot. at 3.) On January 13, 2022, AWS filed this motion to stay the case pending the EPR [490].
DISCUSSION
A district court is entitled to “substantial discretion” in managing its docket. Grayson v. O'Neill, 308 F.3d 808, 817 (7th Cir. 2002) (internal quotation marks omitted). In patent cases, that discretion extends to the decision whether to grant a stay pending the reexamination of a patent in suit. Procter & Gamble Co. v. Kraft Foods Glob., Inc., 549 F.3d 842, 848-49 (Fed. Cir. 2008); Viskase Corp. v. Am. Nat'l Can Co., 261 F.3d 1316, 1328 (Fed. Cir. 2001). It also extends to the interpretation and enforcement of local patent rules. Keranos, LLC v. Silicon Storage Tech., Inc., 797 F.3d 1025, 1035 (Fed. Cir. 2015) (“Local patent rules are essentially a series of case management orders that fall within a district court's broad power to control its docket and enforce its order.”).
When evaluating a motion for stay, a court typically considers “(1) whether a stay will unduly prejudice or tactically disadvantage the non-moving party; (2) whether a stay will simplify the issues in question and streamline the trial; and (3) whether a stay will reduce the burden of litigation on the parties and the court.” Norix Grp., Inc. v. Corr. Techs., Inc., 542 F.Supp.3d 824, 830 (N.D. Ill. 2021) (internal quotation marks omitted). As part of its administrative discretion, however, the court retains the “prerogative to balance considerations beyond those captured by the three-factor stay test.” Murata Mach. USA v. Daifuku Co., 830 F.3d 1357, 1362 (Fed. Cir. 2016). It ultimately seeks to “balance the competing interests of the parties and the interest of the judicial system.” Markel Am. Ins. Co. v. Dolan, 787 F.Supp.2d 776, 779 (N.D. Ill. 2011) (citing Landis v. N. Am. Co., 299 U.S. 248, 254-55 (1936)).
In this district, one additional rule bears on a request like AWS's. Local Patent Rule 3.5(b) provides that “[a]bsent exceptional circumstances, no party may file a motion to stay the lawsuit pending any proceeding in the U.S. Patent and Trademark Office after the due date for service of that party's Final Contentions pursuant to LPR 3.2.” LPR 3.5(b).
As explained below, the court finds that the three stay factors, taken together, support a stay pending inter partes review of the patents.
I. Stay Factors
A. Undue Prejudice and Tactical Disadvantage
The first factor is whether granting a stay will “unduly prejudice or tactically disadvantage the non-moving party.” See Norix Grp., 542 F.Supp.3d at 830. Given the indefinite but potentially considerable length of ex parte reexaminations, Kove argues that it would likely “suffer from lost or misplaced documents and faded memories” when a stay is eventually lifted. (Kove IO, Inc.'s Opp. to AWS's Motion to Stay [502] (hereinafter “Kove Resp.”) at 8-10.) Kove bears the burden of proving infringement in this case, and because most of the relevant evidence is in AWS's possession, Kove believes that it would be disproportionately harmed by any further delays in discovery. (Id. at 10.) AWS responds by pointing to various delays of Kove's own making. For example, Kove did not file this case until several years after AWS had begun the alleged infringement, and Kove has either requested or agreed to several extensions of the case schedule. (AWS Mot. at 10-11; AWS Reply at 6-8.) In any event, AWS says, the latest statistics suggest that the USPTO should complete the EPR in little more than one year. (AWS Mot at 10; AWS Reply at 7-9.) Finally, AWS notes that Kove's patents have expired, so there is no possibility of future damages. (AWS Mot. at 9-10; AWS Reply at 5-6.)
The parties have completed substantial fact discovery, but they agree that much more remains. (See, e.g., AWS Reply at 3 & n.13 (noting that less than half of fact depositions have been taken).) The court therefore recognizes the possibility that a stay could compound the problems of “stale evidence, faded memories, and lost documents.” See Cooper Notification, Inc. v. Twitter, Inc., No. 09-cv-865, 2010 WL 5149351, at *4 (D. Del. Dec. 13, 2010). That said, Kove's sense of urgency appears to be significantly exaggerated, and Kove conveniently minimizes its own role in hindering the discovery process by waiting so many years to bring this case. Kove never sought a preliminary injunction, nor did it ever sell products that practiced the patents-which have now expired anyway. These factors all suggest that any prejudice resulting from a delay will not be undue prejudice. See, e.g., Putco, Inc. v. Carjamz Com Inc., No. 20-cv-50109, 2021 WL 4926499, at *2 (N.D. Ill. Mar. 2, 2021) (“Putco's own four-year delay in filing this action, and failure to seek a preliminary injunction, demonstrate that Putco's concerns about potential delay or strategic disadvantage are overstated.”); Ignite USA, LLC v. Pac. Mkt. Int'l, LLC, No. 14 C 856, 2014 WL 2505166, at *3 (N.D. Ill. May 29, 2014) (“Ignite's own delay, waiting from March 2012 to April 2014 to file the Complaint, further indicates that Ignite will not suffer undue prejudice from a stay.” (citation omitted)); Baxter Int'l, Inc. v. CareFusion Corp., No. 15-cv-09986, 2017 WL 11624669, at *4 (N.D. Ill. May 31, 2017) (finding no undue prejudice in granting a stay where the plaintiff had “show[n] no sense of urgency in enforcing its patents, ” declined to seek a preliminary injunction, and did not practice a patent in question).
It is true, as Kove says, that we do not know how long the reexamination will take, but that fact is always unknown when a defendant seeks a stay in such circumstances. Court after court has found that the delay inherent to a stay pending EPR is not, by itself, unduly prejudicial to the party attempting to enforce its patent rights. See British Telecomms. PLC v. IAC/InterActiveCorp, No. 18-cv-366, 2020 WL 5517283, at *6-7 (D. Del. Sept. 11, 2020) (collecting cases). This factor favors AWS.
B. Simplification of Issues for Trial
The second factor is whether granting a stay will “simplify the issues in question and streamline the trial.” See Norix Grp., 542 F.Supp.3d at 830; see also British Telecomms., 2020 WL 5517283, at *9 (describing this factor as the “most important”). AWS emphasizes that (1) the USPTO is reexamining every patent claim that Kove is asserting, and (2) Kove's patents have already expired, so an adverse finding by the USPTO could lead only to cancellation, not amendment. (AWS Mot. at 6-7; AWS Reply at 2-3.) AWS believes it is likely that the USPTO will cancel at least some claims, necessarily simplifying the infringement analysis in expert discovery, in dispositive motions, and at trial. (AWS Mot. at 6-7; AWS Reply at 2-3.) Kove has no direct answer to that hypothetical scenario, as it apparently assumes that the USPTO will uphold its patents in full. Because the patents have already expired, Kove says, there is no risk that “resources will have been wasted litigating claims that get amended during reexamination.” (Kove Resp. at 12 (emphasis added).) Kove also points out that EPR has no estoppel effect on litigation, so even if the USPTO rejects AWS's invalidity arguments, AWS could, after the stay is lifted, continue to press those arguments before this court. (Id.)
This factor clearly favors AWS. While the parties do not thoroughly discuss the effects of EPR on this litigation, it is indisputable that the issues would be simplified if the USPTO cancelled any of Kove's patent claims. That simplification, which the USPTO's statistics suggest is quite plausible, could have various streamlining effects on this case. See JAB Distribs., LLC v. London Luxury, LLC, No. 09C5831, 2010 WL 1882010, at *2 (N.D. Ill. May 11, 2010) (noting that a stay pending EPR “may have numerous benefits, including streamlining the case, simplifying or eliminating remaining issues, promoting settlement, alleviating discovery problems, providing the Court with the PTO's expert review of prior art, and reducing the length and expense of litigation”); see also Oil-Dri Corp. of Am. v. Nestle Purina Petcare Co., No. 15-cv-1067, 2015 WL 13650951, at *2 (N.D. Ill. May 5, 2015) (finding that this factor strongly favors a stay “where all the asserted patents are under review”).
C. Reduction of the Burden of Litigation
The third factor is whether granting a stay will “reduce the burden of litigation on the parties and the court.” See Norix Grp., 542 F.Supp.3d at 830. Kove contends that the EPR would materially reduce the burden of litigation only in the unlikely event that the USPTO “invalidates all or almost all of Kove's 19 asserted claims.” (Kove Resp. at 13.) After all, Kove says, “many of Kove's asserted claims cover the same accused products and features, ” so removing only a few claims would not significantly narrow the scope of fact discovery, expert discovery, or summary judgment. (Id. at 13-14.) AWS does not respond to that argument directly, but it maintains the broader position that any narrowing of the issues by the USPTO would inevitably reduce this case's burden on the parties and the court. (AWS Mot. at 8-9; AWS Reply at 3-5.)
To the extent that a USPTO decision will simplify the legal issues in this case (as discussed above), the court finds it plausible that the decision will simultaneously reduce the burden of the litigation. AWS has provided few specific details about that scenario. As Kove argues, the scope of discovery might be only minimally affected by a USPTO decision that cancels only some claims, because many patent claims concern the same products and features. In other words, the parties (particularly Kove) may well need to take wide-ranging discovery about those products even if there are fewer claims asserted. The court recognizes the heavy burdens imposed on the parties by ongoing discovery, but unless the pending EPR would significantly reduce those burdens-which is not clear-the court does not find that this factor strongly favors a stay.
II. Exceptional Circumstances
As just discussed, the court finds that the three stay factors, taken together, support a stay pending EPR. But there remains the question of whether AWS's request is timely. Local Patent Rule 3.5(b) provides that “[a]bsent exceptional circumstances, ” a party may not, after its deadline for filing final infringement contentions, move for a stay pending reexamination of a patent in suit. See LPR 3.5(b). AWS does not deny that it filed this motion more than 15 months after its deadline for serving final non-infringement contentions. Thus, the operative question now is whether these circumstances are sufficiently exceptional to justify the motion's untimeliness. The court concludes that they are, though not for all the reasons that AWS urges.
Although Local Patent Rule 3.5(b) was adopted in 2009, it has received virtually no judicial attention. The court has found no instance of any court discussing it, and the parties have cited only one out-of-district case applying an analogous rule. See ClearPlay v. Dish Network LLC, No. 2:14-cv-00191, 2021 WL 2515008 (D. Utah June 18, 2021). That one case, however, is on all fours with the facts here. In ClearPlay, the defendant successfully petitioned the USPTO for EPR of all the claims in suit. Id. at *1. In its petition, the defendant cited prior-art references that it had not discovered until several years after serving its final non-infringement contentions on the plaintiff. Id. at *3. On a motion for stay pending EPR, the court found that “the USPTO's decision to reexamine all the claims in this case renders the circumstances here exceptional because of the greater likelihood of reexamination proceedings nearing this case towards its complete resolution.” Id. The court also reiterated that the defendant had not discovered the critical prior art until well after the deadline for infringement contentions. Id.
Kove attempts to fill in the gap by citing a law review article about this district's Local Patent Rules that was published shortly after their adoption. In that piece, Judge Kennelly and another member of the drafting committee explained that “[t]he view of the Court and the working group was that [Local Patent Rule 3.5(b)] would, among other things, discourage parties from strategically requesting a stay in belated fashion.” Matthew F. Kennelly & Edward D. Manzo, Northern District of Illinois Adopts Local Patent Rules, 9 J. Marshall Rev. Intell. Prop. L. 202, 217 (2010). As the authors noted, however, “[n]othing stated in this article has any official imprimatur.” Id. at 202.
This court takes a somewhat narrower view of “exceptional circumstances.” For one thing, the USPTO's decision to reexamine all claims in suit is statistically unexceptional. Since 1981, when EPR was introduced, the USPTO has granted more than 92% of all petitions for this procedure. (Ex. B to AWS Mot. [490-2] at 1.) That aggregate number was no different in 2009, when Local Patent Rule 3.5(b) was adopted. See Honeywell Int'l, Inc. v. Furuno Elec. Co., No. CIV. 09-3601, 2010 WL 3023529, at *1 (D. Minn. July 30, 2010) (noting that, as of 2009, the USPTO had granted 92% of EPR requests since 1981). And while AWS emphasizes that some 78% of EPRs requested by third parties result in the amendment or cancellation of at least one patent claim (see Ex. B to AWS Mot. at 2), that number, too, is little different than it was in 2009, see Honeywell, 2010 WL 3023529, at *1 (noting that, as of 2009, patents had been upheld without amendment in only 25% of EPRs since 1981). In other words, this district chose to adopt the strict time bar in Local Patent Rule 3.5(b) despite the frequency with which the USPTO has historically granted petitions for EPR and modified patents through that procedure. The statistical likelihood that the USPTO will modify a patent is thus not an “exceptional circumstance”; it is the backdrop against which the local rule was enacted to limit the availability of stays.
That leaves the timing of the prior-art discovery as the only significant factor that might render these circumstances exceptional. In ClearPlay, the defendant's petition for EPR relied on prior-art references that the defendant did not discover until long after its deadline for serving infringement contentions. ClearPlay, 2021 WL 2515008, at *3. The court found that such timing contributed to the exceptional circumstances justifying the apparent tardiness of the motion for stay. See Id. Here, AWS's petitions to the USPTO relied on four pieces of prior art (two patents and two printed publications) that AWS says it did not discover until the fall of 2021. (AWS Mot. at 3, 12; AWS Reply at 11-13.) As an initial matter, the court notes that the belated discovery of prior-art patents is not especially favorable to AWS's position. The two patents contained in AWS's petitions for EPR were both listed as “References Cited” in other patents that AWS had disclosed in this litigation well over a year before it served its final non-infringement contentions in August 2020. (See Kove Resp. at 5 (citing Exs. 4-5 to Kove Resp. [502-5, 502-6]).) Thus, to the extent that AWS is now contending that it did not discover these patents until recently, the court notes that it could have found them through a minimally more thorough search.
But the two printed publications do complicate the court's assessment of AWS's diligence and motives. The two publications are Joseph B. Greene, Oracle DBA Survival Guide (1st ed. 1995), and Oracle Names Administrator's Guide, Release 2.0 (1996). According to AWS, it stumbled upon the first in a used bookstore in August 2021, and it identified the second after contacting the author of the first. (AWS Reply at 12-13.) Just a few months later, AWS filed its petitions for EPR, which the USPTO granted largely on the basis of these two publications. (See Ex. A to AWS's Notice of Related Proceedings [487-1] (order granting EPR request for '978 Patent); Ex. B. to AWS's Notice of Related Proceedings [487-2] ('640 Patent); Ex. C to AWS Mot. [490-3] ('170 Patent).) AWS now explains that finding software references from the 1990s is a “Herculean task.” (AWS Reply at 11.) “Software development typically happens at a fast pace, in a competitive environment, using development tools that aren't optimized for record keeping.” (Id.) AWS points out, further, that many software releases are never recorded, and if they are recorded, that record is often online, where information is easily lost to time. (Id. at 11-12.) Given these constraints, AWS asserts, Kove itself hindered the prior-art search by waiting so many years to file suit. (Id. at 12.)
Disappointingly, AWS has not identified the individuals involved nor provided any other details about the circumstances of this discovery.
The court is persuaded that the timing of AWS's discovery is an “exceptional circumstance” that justifies departure from the time bar in Local Patent Rule 3.5(b). There is little evidence that AWS is engaged in improper gamesmanship in attempting to stay the case now. This case has been, and may well continue to be, a costly one to litigate, and it obviously hinges on the validity of Kove's patents. Given the possibility that the prior-art references could undermine those patents, AWS had very reason to present the references to the USPTO sooner-if it knew about them.
CONCLUSION
For the reasons provided above, AWS's motion to stay the case [490] is granted. Pending motions are stricken without prejudice. The parties are directed to submit a written status report no later than September 7, 2022.