Case 1:18-cv-01519-MN Document 190 Filed 09/29/20 Page 1 of 12 PageID #: 6607
`
`Plaintiff,
`
`
`
`v.
`
`
`
`
`
`C.A. No. 18-1519-MN
`
`
`FINJAN, INC.,
`
`
`
`
`
`RAPID7, INC., et al.,
`
`
`
`
`
`
`
`
`
`IN THE UNITED STATES DISTRICT COURT
`FOR THE DISTRICT OF DELAWARE
`
`)
`)
`)
`)
`)
`)
`)
`)
`)
`
`
`
`MEMORANDUM OPINION
`
`On September 17, 2020, Special Master Order #4 (D.I. 189) granted-in-part and denied-
`
`
`
`Defendants.
`
`
`
`in-part Defendants Rapid7, Inc. and Rapid7 LLC’s (collectively, Rapid7) Motion to Strike
`
`Portions of the Expert Report of Dr. Mitzenmacher (D.I. 183). Order #4 was issued on an
`
`expedited basis because Dr. Mitzenmacher’s deposition was scheduled for September 21, 2020.
`
`This memorandum opinion explains the reasoning for the rulings in Order #4.
`
`BACKGROUND
`
`
`
`At the scheduling conference in this case, Plaintiff Finjan, Inc. (Finjan) requested, and the
`
`Court ordered, that Rapid7 was to make its source code available for inspection at the time of
`
`core technical document production, which is earlier than usual. In doing so, the Court warned
`
`Finjan that it was “going to expect that the infringement contentions be pretty full and pretty
`
`informative” and “[if] they’re giving you the source code on these products, I expect you
`
`[Finjan] to incorporate that and explain how the products infringe.” 1/30/19 Teleconference
`
`Transcript at 9:21-24; 9:25-10:2.
`
`
`
`After producing its source code for inspection and receiving Finjan’s preliminary
`
`infringement contentions, Rapid7 believed the contentions were lacking and moved to compel
`
`

`

`Case 1:18-cv-01519-MN Document 190 Filed 09/29/20 Page 2 of 12 PageID #: 6608
`
`more detailed contentions. At a December 13, 2019 hearing on Rapid7’s motion to compel,
`
`Finjan’s counsel distinguished its contentions from the greater level of detail offered in an expert
`
`report. 12/13/2019 Hearing Transcript at 11. Judge Noreika then ruled as follows:
`
`It’s up to Finjan, you can by January 10th supplement again to address some of
`these concerns. It’s up to you. But if you don’t, you’re not going to be able to
`raise new issues. Like when you said this is not an expert report, you’re not going
`to suddenly be able to get into a lot more detail in actually fleshing out your
`contentions in an expert report if you haven’t fairly disclosed them here. If you
`think this is sufficient to give notice of what your contentions are, I’m not going
`to make you supplement because I’m not sure exactly what you need, but if you
`don’t supplement and get more specific with what you’re claiming, what aspects
`of this meet the claim elements, you’re not going to be able to do [that] for the
`first time and surprise them in the expert reports. I’m not saying that’s what
`you’re intending to do, it just caught my attention when you said this is not an
`expert report. The expert reports may want to flesh out a little bit more, but they
`shouldn’t for the first time be giving what specific aspects of this [] meet the
`element.
`
`Id. at 12.
`
`
`
`Finjan proceeded to supplement its infringement contentions and then served Final
`
`Infringement Contentions (FICs) on February 26, 2020. See Rapid7’s 9/3/2020 Letter Brief at
`
`Ex. D (excerpts from the FICs). On June 17, 2020, almost four months later, Finjan served the
`
`opening expert report of Dr. Mitzenmacher regarding infringement. See id. at Ex. C (excerpts
`
`from Dr. Mitzenmacher’s report).
`
`
`
`Rapid7 moves to strike allegedly new infringement theories disclosed for the first time in
`
`Dr. Mitzenmacher’s report. These new infringement theories fall into two categories: (1) a
`
`literal infringement theory that a “Universal Translator” capability meets the requirement of “an
`
`input modifier” in claim 41 of the ‘289 patent for the InsightAppSec and AppSpider products;
`
`and (2) doctrine of equivalents (DOE) infringement theories on several elements of claims in the
`
`‘154, ‘289, and ‘408 patents.
`
`
`
`2
`
`

`

`Case 1:18-cv-01519-MN Document 190 Filed 09/29/20 Page 3 of 12 PageID #: 6609
`
`Finjan’s Allegedly New Universal Translator Literal Infringement Theory
`
`
`
`As to the allegedly new Universal Translator capability theory, Rapid7 argues that
`
`Finjan’s FICs do not reference the Universal Translator or its functionality. Rapid7 argues that
`
`while Finjan’s FICs refer to a “scan engine,” this disclosure did not put Rapid7 on notice that the
`
`more specific Universal Translator capability allegedly meets element 41(c). Rapid7 also points
`
`out that Finjan was aware of the Universal Translator capability well before its FICs, and Finjan
`
`even pointed to the Universal Translator capability in its FICs for another patent. See Rapid7’s
`
`Opening Letter at Ex. D, Appendix F-5 at 7.
`
`
`
`Finjan’s letter brief admits that “the level of detail is greater in Dr. Mitzenmacher’s report
`
`than in the FICs” and it did not use the term “Universal Translator” in the FICs with regard to
`
`element 41(c). 9/10/2020 Letter Brief at 2. But Finjan argues that it disclosed the same
`
`functionality as the Universal Translator when it disclosed “scan engines” and described their
`
`function. Finjan also argues that Rapid7 belatedly produced documents regarding the Universal
`
`Translator, which Dr. Mitzenmacher relies on in his report.
`
`Finjan’s Allegedly New DOE Theories
`
`Rapid7 argues that Dr. Mitzenmacher disclosed new DOE theories for claim element
`
`
`
`
`1(b) of the ‘154 patent, element 1(d) of the ‘154 patent, elements 41(c) and 41(e) of the ‘289
`
`patent, and element 1(b) of the ‘408 patent. Finjan’s FICs for element 1(d) of the ‘154 patent,
`
`elements 41(c) and 41(e) of the ‘289 patent, and element 1(b) of the ‘408 patent do not identify a
`
`DOE theory of infringement. For element 1(b) of the ‘154 patent, however, the FICs disclose a
`
`DOE theory. Rapid7 argues that Finjan’s DOE theory in its FICs relates to only one part of
`
`element 1(b) (“Only if…such invocation is safe”), while Dr. Mitzenmacher’s report sets forth a
`
`
`
`3
`
`

`

`Case 1:18-cv-01519-MN Document 190 Filed 09/29/20 Page 4 of 12 PageID #: 6610
`
`DOE theory on another part of element 1(b) (“processing content…including a call to a first
`
`function”).
`
`
`
`Finjan argues that Dr. Mitzenmacher’s DOE opinions are not new but rather “modest
`
`elucidations” of the functionalities accused in the FICs. Finjan also argues that each challenged
`
`DOE opinion corresponds to an equivalent literal infringement theory in the FICs. That is,
`
`according to Finjan, its literal infringement theories in the FICs put Rapid7 on notice of the
`
`accused functionalities and the DOE theories in the expert report should have come as no
`
`surprise given the functionalities accused in the FICs.
`
`LEGAL STANDARDS
`
`Fed. R. Civ. P. 37(c)(1) provides that “[i]f a party fails to provide information . . . as
`
`required by Rule 26(a) or (e), the party is not allowed to use that information . . . to supply
`
`evidence on a motion, at a hearing, or at trial, unless the failure was substantially justified or is
`
`harmless.” Infringement contentions are considered to be “initial disclosures” under Rule 26(a).
`
`Intellectual Ventures I LLC v. AT&T Mobility LLC, 2017 WL 658469, at *1 (D. Del. Feb. 17,
`
`2017).
`
`To determine whether a failure to disclose is harmless, courts in the Third Circuit
`
`consider the “Pennypack” factors, which include: (1) the prejudice or surprise to the party
`
`against whom the evidence is offered; (2) the possibility of curing the prejudice; (3) the potential
`
`disruption of an orderly and efficient trial; (4) the presence of bad faith or willfulness in failing
`
`to disclose the evidence; and (5) the importance of the information withheld. See
`
`Konstantopoulos v. Westvaco Corp., 112 F.3d 710, 719 (3d Cir. 1997) (citing Meyers v.
`
`Pennypack Woods Home Ownership Ass’n, 559 F.2d 894, 904-05 (3d Cir. 1977)). The
`
`
`
`4
`
`

`

`Case 1:18-cv-01519-MN Document 190 Filed 09/29/20 Page 5 of 12 PageID #: 6611
`
`determination of whether to exclude evidence is within the discretion of the court. See In re
`
`Paoli, 35 F.3d 717, 749 (3d Cir. 1994).
`
`DISCUSSION
`
`I.
`
`
`
`Rapid7’s motion to strike the portions of paragraphs 552-554 of Dr.
`Mitzenmacher’s report relating to the “Universal Translator” is GRANTED.
`
`In its FICs, Finjan accused the broad functionality of the “scan engine” of meeting the
`
`claim element “an input modifier” in claim 41 of the ‘289 patent for the InsightAppSec and
`
`AppSpider products. The parties agree that the broad functionality accused in the FICs includes
`
`the Universal Translator capability. But Finjan failed to accuse the Universal Translator
`
`capability until its opening expert report. Finjan provides no reason why it failed to accuse the
`
`Universal Translator capability in its FICs. In fact, Finjan’s counsel admitted “if I could go back
`
`in time to moot this issue, would I have pointed that [Universal Translator] out there, of course.”
`
`Tr. 46:1-3.
`
`If those were the only relevant facts, this would be a closer call. Here, however, Finjan
`
`received at least two warnings from the Court to provide specificity in its infringement
`
`contentions. The second warning addressed the situation here: “if you don’t supplement and get
`
`more specific with what you’re claiming, what aspects of this meet the claim elements, you’re
`
`not going to be able to do [that] for the first time and surprise them in the expert reports.”
`
`12/13/2019 Hearing Transcript at 12. Because Finjan’s new reliance on the Universal Translator
`
`capability violates the Court’s warning, the new disclosure must be stricken.
`
`Finjan’s letter brief cites several decisions where an expert was allowed to provide more
`
`specific infringement opinions based on broad theories disclosed in infringement contentions.
`
`This case is different from those decisions in two important respects. First, none of these
`
`decisions involved prior, direct warnings from the Court like the ones made here. Second, Finjan
`
`
`
`5
`
`

`

`Case 1:18-cv-01519-MN Document 190 Filed 09/29/20 Page 6 of 12 PageID #: 6612
`
`knew enough about the Universal Translator capability to specifically accuse it as part of its FICs
`
`for another patent in the case. This shows that Finjan knew about the Universal Translator
`
`capability at the time of its FICs and chose not to include it as meeting the “an input modifier”
`
`element.1 Based on the specific facts at issue here, Finjan’s new reliance on the Universal
`
`Translator capability is untimely.
`
`The five Pennypack factors do not excuse Finjan’s untimely disclosure.
`
`(1)
`
`Surprise or Prejudice. Rapid7 is prejudiced by the new disclosure of the
`
`Universal Translator capability at this advanced stage of the case. Fact discovery has been
`
`closed for over three months. While this new disclosure affects only one of many claim
`
`elements, every claim element is important because every element must be met for a product to
`
`infringe. Rapid7 struggled to define its prejudice with precision, noting that its claim
`
`construction positions might have been different, it might have developed the factual record in
`
`different ways, and its expert reports might have been different. While a more precise prejudice
`
`argument would have been more compelling, Rapid7 is correct that there is no way to know
`
`whether its expert’s opinions are the same as they would have been if Finjan had timely accused
`
`the Universal Translator capability. See Galderma Labs, L.P. v. Amneal Pharms., LLC, C.A. No.
`
`16-207-LPS, 2018 WL 508876, *2 (D. Del. Jan. 22, 2018). Finjan also notes that Rapid7’s
`
`expert was able to respond to the new disclosure within the existing expert disclosure guidelines.
`
`While true, it would be unjust to penalize Rapid7 for doing its best under difficult circumstances.
`
`
`1 Finjan attempts to blame the absence of the Universal Translator capability from the FICs on
`Rapid7’s late production of documents about the Universal Translator. 9/10/2020 Letter Brief at
`3. This argument has no force because Finjan knew enough about the Universal Translator
`capability to accuse it of meeting another claim element for another patent in its FICs. In fact, the
`graphic Finjan included regarding the Universal Translator capability for this other patent is
`essentially the same as the graphic Dr. Mitzenmacher included in the challenged paragraphs of his
`report. Compare Rapid7’s Opening Letter at Ex. D, Appendix F-5 at 7 with Ex. C at ¶ 552.
`6
`
`
`
`

`

`Case 1:18-cv-01519-MN Document 190 Filed 09/29/20 Page 7 of 12 PageID #: 6613
`
`Id. (noting that a party in this position is faced with “two bad options: either scramble to have an
`
`expert respond…or offer no response and risk not preserving an opinion for trial if the [] Motion
`
`to Strike [is] denied.”).
`
`(2)
`
`and
`
` (3) The possibility of curing the prejudice and potential disruption of an
`
`orderly and efficient trial. While the prejudice to Rapid7 is partially curable through responsive
`
`expert reports and expert depositions, at least some of the prejudice remains because fact
`
`discovery is closed and there is no time to reopen fact discovery under the schedule. Thus, this
`
`factor favors striking the new disclosure.
`
`(4)
`
`Bad faith or willfulness. The Special Master does not believe that Finjan acted in
`
`bad faith. The Court’s repeated warnings about providing complete contentions and Finjan’s
`
`repeated supplementation of its contentions, however, cannot be ignored. Nor can it be ignored
`
`that Finjan has had the relevant source code since May 2019. Thus, this factor slightly favors
`
`striking the new disclosure.
`
`(5)
`
`Importance of the information withheld. Striking the new disclosures will not
`
`leave Finjan without an infringement case. At the hearing, Finjan admitted that “[t]his is a few
`
`pages of the whole treatise on infringement.” Hearing Tr. 62:7-8. This is consistent with
`
`Finjan’s letter brief, which admitted that “the challenged opinions are of modest importance.”
`
`9/10/2020 Letter Brief at 5. Thus, this factor favors striking the new disclosure.
`
`Taken as a whole, the Pennypack factors show that Finjan’s untimely disclosure was not
`
`harmless. Further, in “sophisticated, complex litigation involving parties represented by
`
`competent counsel” (a hurdle this case easily clears), courts have “been less indulgent” in
`
`applying the Pennypack factors and “more willing to exclude evidence without a strict showing
`
`that each of the Pennypack factors has been satisfied.” Bridgestone Sports Co. v. Acushnet Co.,
`
`
`
`7
`
`

`

`Case 1:18-cv-01519-MN Document 190 Filed 09/29/20 Page 8 of 12 PageID #: 6614
`
`2007 WL 521894, at *4 (D. Del. Feb. 15, 2007). Thus, even without a strict showing that each
`
`Pennypack factor is satisfied, the result here would be the same.
`
`II.
`
`Rapid7’s motion to strike paragraphs 434-438, 449-453, 464-468, 477, 483, 546-
`550, 564-568, 581-585, 645-649, 658-662, 670-674, 749-753, and 766-770 of Dr.
`Mitzenmacher’s report, which contain doctrine of equivalents opinions as to
`element 1(d) of the ‘154 patent, elements 41(c) and 41(e) of the ‘289 patent, and
`element 1(b) of the ‘408 patent, is GRANTED.
`
`
`Finjan’s FICs did not disclose a DOE theory of infringement for element 1(d) of the ‘154
`
`patent, elements 41(c) and 41(e) of the ‘289 patent, or element 1(b) of the ‘408 patent. Thus, Dr.
`
`Mitzenmacher’s DOE opinions on these claim elements were new. Finjan provides no reason
`
`why it did not disclose these new DOE theories earlier in the case. Hearing Tr. 61:18-62:1.
`
`Disclosing a new theory of infringement in the opening expert report violates the Court’s earlier
`
`warnings discussed above. Therefore, the Special Master grants Rapid7’s motion to strike Dr.
`
`Mitzenmacher’s new DOE theories.
`
`Finjan argues that its new DOE theories are supported by its previously disclosed literal
`
`infringement theories accusing the same functionality. But DOE theories of infringement are
`
`separate and distinct from theories of literal infringement. See TQ Delta, LLC v. Adtran, Inc.,
`
`2020 U.S. Dist. LEXIS 136536, at *7-8 (D. Del. July 31, 2020) (“Plaintiff just cannot raise the
`
`entirely new argument that the products infringe under a DOE theory.”).
`
`Finjan extensively cites to HSM Portfolio LLC v. Elpida Memory, Inc., 2016 U.S. Dist.
`
`LEXIS 16786, at *13-15 (D. Del. Feb. 11, 2016). In HSM, the earlier infringement contentions
`
`accused a certain component “or its equivalents.” HSM, 2016 U.S. Dist. LEXIS 16786, at *13.
`
`The court’s ruling in HSM focused on the “or its equivalents” language. Here, however, the
`
`contentions fail to mention “equivalents.” Thus, HSM is distinguishable. Judge Noreika’s prior
`
`
`
`8
`
`

`

`Case 1:18-cv-01519-MN Document 190 Filed 09/29/20 Page 9 of 12 PageID #: 6615
`
`warnings to provide complete contentions further distinguishes the present situation from the
`
`facts of HSM.
`
`The Pennypack factors again show Finjan’s failure to disclose was not harmless. The
`
`discussion of the Pennypack factors above in Section I applies with equal force here. In addition,
`
`re-opening discovery to cure the prejudice from new DOE theories would disrupt an already tight
`
`case schedule. See TQ Delta, 2020 U.S. Dist. LEXIS 136536, at *7 (“re-opening discovery to
`
`give Defendant an opportunity to cure the prejudice [from new DOE theory] would only disrupt
`
`the schedule further”). As to the importance of the DOE theories, as in TQ Delta, this factor
`
`favors striking the new DOE theories because Finjan will still be able to assert its literal
`
`infringement theories, which it claims are a mirror image of the stricken DOE theories. See id. at
`
`*7-8 (Plaintiff can still rely on that opening report at trial. Plaintiff just cannot raise the entirely
`
`new argument that the products infringe under a DOE theory.”).
`
`III. Rapid7’s motion to strike paragraphs 324-328, 352-356, 378-382, 410-414, and
`424 of Dr. Mitzenmacher’s report, which contain doctrine of equivalents
`opinions as to element 1(b) of the ‘154 patent, is DENIED.
`
`
`For element 1(b) of the ‘154 patent, Finjan’s FICs disclosed a DOE theory of
`
`infringement. See Rapid7’s 9/3/2020 Letter Brief at Ex. D, Appendix C. But Rapid7 argues that
`
`the DOE theories in Finjan’s FICs are for only a subpart of element 1(b) while Dr. Mitzenmacher
`
`put forth a different DOE theory as to a different subpart of element 1(b). Rapid7 is correct that
`
`the DOE theories for element 1(b) in Dr. Mitzenmacher’s report are not identical to the DOE
`
`theories in the FICs. The parties dispute the magnitude and importance of the differences.
`
`After a thorough review of the parties’ arguments, the differences between the FICs and
`
`the expert report on this element appear relatively insignificant. Rapid7 certainly cannot argue
`
`that a DOE theory on this element in Dr. Mitzenmacher’s report was a surprise. Also, any
`
`
`
`9
`
`

`

`Case 1:18-cv-01519-MN Document 190 Filed 09/29/20 Page 10 of 12 PageID #: 6616
`
`prejudice to Rapid7 is slight. Rapid7 does not point to any discovery it would have taken had it
`
`received the specific formulation of Finjan’s DOE position in Dr. Mitzenmacher’s report during
`
`fact discovery. Rapid7 argues that it might have sought to construe an additional claim element.
`
`But Rapid7 has not asked to reopen claim construction. Rapid7 also argues it might have
`
`asserted an ensnarement defense. Rapid7 did not assert an ensnarement defense as to this
`
`element during fact discovery, however, despite knowing that Finjan was asserting a DOE theory
`
`on this element. And since receiving Dr. Mitzenmacher’s report, Rapid7 has not asked to assert
`
`an ensnarement defense. There is therefore no reason to believe that Rapid7 would have asserted
`
`ensnarement as to this element had it known of Dr. Mitzenmacher’s specific formulation of
`
`Finjan’s DOE theory earlier.
`
`The facts here, as opposed to the facts in Section II above, line up more closely with
`
`HSM. As in HSM, “equivalents” was mentioned in the earlier infringement contention, and the
`
`opposing expert fully responded to the DOE theory in a rebuttal report.
`
`For these reasons, the Special Master denies Rapid7’s motion to strike the paragraphs of
`
`Dr. Mitzenmacher’s report with a DOE theory of infringement on element 1(b) of the ‘154
`
`patent.
`
`IV.
`
`The Result Would Be the Same under a Good Cause Standard
`
`This motion could be viewed through the lens of a scheduling order violation rather than
`
`as a motion to strike. See, e.g., British Telecommunications PLC v. IAC/INTERACTIVECORP,
`
`C.A. No. 18-366-WCB, 2020 WL 3047989 (D. Del. June 8, 2020). Rapid7 at one point suggests
`
`this. 9/3/2020 Letter Brief at 2. Viewed through this lens, Finjan must show good cause. See
`
`12/13/2019 Hearing Transcript at 16 (“And then only if you show good cause after that for
`
`learning something new can you supplement again.”). The key factor in demonstrating good
`
`
`
`10
`
`

`

`Case 1:18-cv-01519-MN Document 190 Filed 09/29/20 Page 11 of 12 PageID #: 6617
`
`cause is diligence. British Telecommunications, 2020 WL 3047989 at *2 (“In determining
`
`whether a party has demonstrated good cause to amend its contentions, the key factor that courts
`
`have considered is whether that party has shown diligence both in discovering that an
`
`amendment was necessary and in moving to amend after that discovery.”). “If the moving party
`
`can establish diligence, other considerations pertinent to the good cause inquiry come into play,
`
`including the importance of the new information, the difficulty of locating the new information,
`
`any gamesmanship that is evident from the untimely disclosure, and the potential prejudice to the
`
`opposing party that would result from permitting the belated amendment.” Id.
`
`Applying the good cause standard would not change the result here. Because Finjan
`
`knew about the Universal Translator capability well before its FICs—and even cited the
`
`Universal Translator capability in a chart for a different patent in its FICs—Finjan could not
`
`show diligence and thus could not meet the good cause standard. As to the previously
`
`undisclosed DOE theories in Dr. Mitzenmacher’s report, the record does not show that Finjan
`
`acted diligently to disclose these theories or that they are based on recently discovered
`
`information. Finjan has had the relevant source code since May 2019. For element 1(b) of the
`
`‘154 patent, where Finjan previously disclosed a DOE theory in its FICs, Finjan can establish
`
`diligence more easily because it is merely supplementing an existing DOE theory rather than
`
`asserting a DOE theory on an element for the first time. Other considerations such as lack of
`
`gamesmanship and lack of prejudice also support finding good cause for the supplementation as
`
`to element 1(b).
`
`
`
`11
`
`

`

`Case 1:18-cv-01519-MN Document 190 Filed 09/29/20 Page 12 of 12 PageID #: 6618
`
`CONCLUSION
`
`For these reasons, Rapid7’s motion to strike (D.I. 183) is GRANTED-IN-PART and
`
`DENIED-IN-PART.
`
`
`
`
`
`Dated: September 28, 2020
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Special Master Chad S.C. Stover
`
`
`
`
`
`12
`
`