Opinion
Case No. 3:18-cv-02848-WHO
02-11-2021
INTEL CORPORATION, Plaintiff, v. TELA INNOVATIONS, INC., Defendant.
ORDER ON MOTION FOR LEAVE TO FILE A MOTION TO RECONSIDER
Re: Dkt. Nos. 324, 325
Among many issues addressed in my order of December 22, 2020 (Dkt. No. 316) ("Prior Order"), I granted summary judgment to Intel Corporation ("Intel") that several of its products did not literally infringe the patents of Tela Innovations, Inc. ("Tela"). Tela moves for leave to file a motion to reconsider several aspects of that order. Its motion is DENIED.
I will exercise my discretion to make several minor modifications and clarifications to the order. See City of Los Angeles, Harbor Div. v. Santa Monica Baykeeper, 254 F.3d 882, 889 (9th Cir. 2001) ("[A]s long as a district court has jurisdiction over the case, then it possesses the inherent procedural power to reconsider, rescind, or modify an interlocutory order for cause seen by it to be sufficient."). I therefore concurrently file an amended order that supersedes the Prior Order. None of the modifications alters the order's conclusions or reasoning.
For the parties' ease of reference, I will email counsel a version of track changes showing the amendments to the Prior Order. The amended order filed on the docket accepts those changes.
Each party has moved to certify the summary judgment determination for appeal and a hearing on those motions is set for February 17, 2021. Because the amended order does not alter the grants and denials of summary judgment that I previously entered, I do not expect it to make a difference to the parties' certification motions. However, if either party wishes to withdraw its motion in light of these few alterations and clarifications, it should provide notice prior to the hearing.
I. LEAVE TO FILE A MOTION TO RECONSIDER
Civil Local Rule 7-9 governs motions for reconsideration of interlocutory orders prior to "the entry of a judgment adjudicating all of the claims and the rights and liabilities of all the parties in a case." CIV. L. R. 7-9(a). Under that rule, "any party may make a motion before a Judge requesting that the Judge grant the party leave to file a motion for reconsideration of any interlocutory order on any ground set forth in Civil L.R. 7-9(b). No party may notice a motion for reconsideration without first obtaining leave of Court to file the motion." Id. Under Rule 7-9(b),
The moving party must specifically show reasonable diligence in bringing the motion and one of the following: (1) That at the time of the motion for leave, a material difference in fact or law exists from that which was presented to the Court before entry of the interlocutory order for which reconsideration is sought. The party also must show that in the exercise of reasonable diligence the party applying for reconsideration did not know such fact or law at the time of the interlocutory order; or (2) The emergence of new material facts or a change of law occurring after the time of such order; or (3) A manifest failure by the Court to consider material facts or dispositive legal arguments which were presented to the Court before such interlocutory order.Id. 7-9(b).
Even if a motion for leave to file a motion to reconsider does not satisfy Rule 7-9, district courts have the inherent authority to modify interlocutory orders prior to entry of final judgment. Amarel v. Connell, 102 F.3d 1494 (9th Cir. 1996), as amended (Jan. 15, 1997). But generally, "[r]econsideration is appropriate if the district court (1) is presented with newly discovered evidence, (2) committed clear error or the initial decision was manifestly unjust, or (3) if there is an intervening change in controlling law" and any other circumstances warranting reconsideration would be "highly unusual." Sch. Dist. No. 1J, Multnomah Cty., Or. v. ACandS, Inc., 5 F.3d 1255, 1263 (9th Cir. 1993).
A. Diffusion Regions
This order assumes familiarity with my Prior Order. I previously found that none of the asserted claims were literally infringed because all required "diffusion regions" and that the Accused Products lacked "diffusion regions." See Prior Order 7-11. Tela makes two arguments that this determination is incorrect.
First, Tela argues that I narrowed the parties' agreed construction of "diffusion regions" and applied that narrowed construction. See Motion for Leave to File a Motion for Reconsideration ("Mot.") [Dkt. No. 325] 5-9. The parties adopted the construction of diffusion regions as "selected portions of the substrate within which impurities have been introduced to form the source or drain of a transistor." Prior Order 8. I found that, in the Accused Products, "the regions that contain impurities that form sources and drains on the Accused Products are not, as they must be, selected portions 'of the substrate.'" Id. The essential reason was that Intel's sources and drains are not "substrate" "within" which impurities have been introduced. Instead, the sources and drains of the Accused Products are created by [Redacted] . Then, [Redacted] . Id. 8-9. That contains the impurities that form the sources and drains. Id. 9.
Despite this, Tela argues that the Prior Order "alters the parties' agreed construction and—for the first time—narrows 'diffusion region' by further construing the phrase 'portions of the substrate' to be limited to only the individual source or drain for each transistor, as opposed to a broader area of the substrate in which those sources and drains are formed." Mot. 4. The Prior Order did not alter the construction. The language of the claim construction imposes no requirement about how much of the substrate a diffusion region must cover or how many sources and drains it must include. It merely requires, as it says, "portions of the substrate" that meet the other requirements. This aside, even if Tela is correct that diffusion regions must encompass some area that includes multiple or even all sources and chains—it is unclear exactly where those boundaries would lie given that the impurities are in [Redacted] rather than in diffused areas of the substrate itself—it does not mean that Accused Products literally infringe because the impurities are still not introduced within selected portions of the substrate for the reasons explained.
Intel relied on a statement by Tela's counsel in the International Trade Commission Markman hearing that "the parties have all agreed and stipulated that the diffusion region in the physical semiconductor is the source and drain of a transistor," which cuts against Tela's position here. See, e.g., Dkt. No. 234-4 at 6. At the hearing on the motions addressed in the Prior Order, Tela's counsel described this statement as "maybe inartful[]." Dkt. No. 313 at 15:12. I did not rely on it as a formal stipulation in the Prior Order, in part because it was not made before me and the evidence of the Accused Products themselves settled the issue.
Relatedly, Tela argues that my findings were inconsistent with the Asserted Patents' specifications concerning diffusion regions because those specifications contemplate layout shapes for diffusion regions that extend over parts of the substrate that include multiple sources and drains. Mot. 6-7. That does not show inconsistency, largely for the reasons explained above. Moreover, resort to the specification actually weakens Tela's argument rather than strengthening it. The specifications show that "diffusion regions" extend over that portion of the substrate because impurities have been diffused within it. Accordingly, an area of the substrate could be identified as having those diffused impurities and contrasted with substrate that did not. In the however, [Redacted] . As a result, reliance on the specification underlines how different the Accused Products are, not that they literally infringe.
Tela further argues that the Prior Order diverted from the plain meaning of "portion," which it defines from a dictionary as "any part of a whole, either separate from or integrated with it." Mot. 8. It is unclear how that helps Tela. Replacing "portion" with this definition in the claim construction leads to something like, "selected parts of the whole substrate, either separate from or integrated with it, within which impurities have been introduced to form the source or drain of a transistor." Again, there would be no "part of the whole" substrate within which impurities have been introduced.
Tela also argues that the Prior Order erred by finding that there was no genuine dispute of material fact that would preclude summary judgment. Mot. 10-12. It first argues that its technical expert, Dr. Daniel Foty, opined that the Accused Products' sources and drains are part of the substrate. Tela contends that his fundamental reason for doing so was that, when the [Redacted] . Mot. 10. According to Tela, this shows that [Redacted] becomes a part of the substrate itself. This evidence, however, was never raised in Tela's Opposition. Tela relied on Foty for a number of assertions, including when it came to the diffusion regions, and had other arguments about why Intel's sources and drains qualify as substrate. But Tela never referenced this. Consequently, Intel never addressed it before me, nor did it have the chance to.
It was, of course, Tela's burden to raise those facts or opinions that it would use to meet its burden. That would be the case even if previous set of motions did not concern more than 500 pages of substantive briefing, many more pages of record, and highly technical evidence. "Judges are not like pigs, hunting for truffles buried in briefs." United States v. Dunkel, 927 F.2d 955, 956 (7th Cir. 1991). Similarly, judges do not hunt through voluminous records in search of some fact that would preclude summary judgment. See, e.g., Fed. R. Civ. P. 56(c)(1) ("A party asserting that a fact . . . is genuinely disputed must support the assertion by . . . citing to particular parts of materials in the record."); Keenan v. Allan, 91 F.3d 1275, 1279 (9th Cir. 1996) ("It is not our task, or that of the district court, to scour the record in search of a genuine issue of triable fact. We rely on the nonmoving party to identify with reasonable particularity the evidence that precludes summary judgment.") (internal quotation marks and alteration omitted).
I do not know whether this opinion of Foty's would have precluded summary judgment had Tela raised it appropriately and Intel had a chance to respond, but I will not reopen an intensive, finalized briefing process based on Tela belatedly raising it for the first time now. Despite relying on other opinions of Foty's with particularity, Tela never mentioned this one. I also note—though I take no position because Intel has not had a fair opportunity to brief the issue—that it appeals that Tela makes more of Foty's testimony than is there. In particular, it appears to be Tela's brief, not Foty, that provides the ultimate link between this [Redacted] and a substance being "substrate."
Tela does not make this argument in its Motion—indeed, it does not address that the evidence was not raised before at all—but it would have been incorrect to argue that the evidence was properly raised merely because Tela argued generally that Foty's report created a factual dispute. Even if this were not a highly technical case, Tela's primary role at summary judgment was to raise the specific facts that would preclude it. It could not simply have said that all of Foty's report raised a factual question without elaboration. And, in fact, it appears that this evidence and argument is largely derived from pieces of deposition testimony by Foty, not his report.
Next, Tela contends that the Prior Order resolved disputed facts in Intel's favor. It only points, however, to Foty's general opinions that the epitaxially deposited sources and drains meet the construction of diffusion regions. Mot. 11-12. For the reasons explained in the Prior Order, his opinions that were raised do not create a dispute of material fact. Tela's argument elides Foty's opinions with its overall definitional argument about what can fairly be called "substrate."
Tela also asserts that internal Intel documents use the phrase "diffusion regions" to describe the area in question on its products. For the reasons adequately explained in the Prior Order, that (likely holdover) internal nomenclature does not create a genuine dispute of material fact about the Accused Products themselves. Prior Order 9-10.
B. Gate Contact Structures
The Prior Order found that two Intel products, called the 14 nm and 10 nm products, did not literally infringe claims that required "contact structure(s)" or "gate contact structure(s)." Prior Order 11-14. Tela makes similar arguments as above: that the Prior Order narrowed the claim construction and ignored genuine disputes of material fact. I previously construed both terms to mean "conductive structure(s) in a gate contact layer above and separate from the gate layer and below and separate from interconnect layers." Prior Order 11.
First, I agree that some language in the Prior Order, if read in a vacuum, might miscommunicate the scope of the construction. In particular, the Prior Order sometimes discussed the contact structures as being separate from other layers when it should have discussed contact structures in a layer separate from the other layers. While the substance of the Order made clear that there was no narrowing of the claim construction, I will amend those discrete phrases to remove any doubt.
Second, Tela argues that the Prior Order ignored genuine disputes of fact about the 14 nm products. The Prior Order found that the accused gate contract structure was not in a gate contact layer "separate from the interconnect layers" because [Redacted] . Prior Order 11-12. In its Motion, Tela's single argument on the issue of what evidence the Prior Order missed from its experts about the 14 nm products is Foty's "reli[ance] upon fabrication-step documents" to explain to show that the construction was met. Mot. 14-15. The Prior Order, however, rejected Tela's argument that because openings for different parts of the structure were patterned separately, that creates a genuine dispute of material fact that the layers were merged in the 14 run context. See Prior Order 12. The Prior Order also rejected a similar type of argument in the 10 nm context. See id. 14.
In a different vein, Tela contends that Intel did not meet its initial summary judgment burden because there is no evidence that the [Redacted] . That is inaccurate. As noted in the Order, Intel's Ph.D.-holding vice president, who worked for many years on the technical aspects of the products, submitted a sworn declaration saying just this. Id. 12. Tela did not file a Daubert motion or motion to strike those statements. In its present Motion, Tela now seeks to discount this statement because he is an "interested witness" and "inferences should not be drawn in favor of his testimony." Mot. 15. But credibility determinations are not appropriate at summary judgment and no inference was drawn in his favor. Instead, he put forward that evidence and the burden shifted to Tela to show a genuine dispute of material fact, which it did not for the reasons described in the Prior Order and here.
Third, Tela argues that the Prior Order ignored genuine disputes of fact about the 10 nm products. As explained in the Prior Order, the accused gate contacts are not in a layer "separate from" interconnect because [Redacted] . Prior Order 13-14. Again, Tela offers fact that it contends would preclude summary judgment: that the term is not met from the perspective of a "top view." Mot. 17. But there is no record evidence that a structure becomes physically "separate" just because it is viewed from the top.
Tela also asserts that the Prior Order effectively rewrote the claim construction in a different way for the 10 nm products, by imposing a requirement that the gate contact structures and other relevant structures not be "multiple structures that are separate." Mot. 17. The claim construction, again, was "conductive structure(s) in a gate contact layer above and separate from the gate layer and below and separate from interconnect layers." If the gate contact layer was not separate from those other layers, it does not literally infringe; if there are combined layers, they are by definition not separate and cannot literally infringe. There was no alteration.
C. Design Stage Abstractions and Layout Shapes
In ruling on one of several dozen of the parties' discrete arguments on motions to strike and exclude evidence, I partially granted Intel's motion to strike certain upstream design-stage abstractions and layout shapes on which Tela relied (though I denied it with respect to certain fabrication steps). Prior Order 43-46. The fundamental reason was that the parties had before them the Accused Products themselves, yet many of Tela's arguments were not geared toward showing that their structure infringed but instead that these abstractions and layout shapes demonstrated infringement in spite of any facts about the structure. As a result, I concluded that reliance on them ran afoul of the Federal Circuit's instructions, would alternatively be excluded on evidentiary grounds, and would not create a genuine dispute of material fact if admitted. Id.
Tela contends that the Prior Order effectively established a rule that only direct evidence, but not circumstantial evidence, could show infringement, in violation of the rules to the contrary. Mot. 18-19. That is incorrect. I did not exclude that evidence based on a distinction between circumstantial and direct evidence. The Prior Order explicitly recognized that there is no special rule excluding circumstantial evidence of infringement. Prior Order 43. The Prior Order instead held that there was not sufficient evidence that these abstractions or layout shapes were reflective of the products but that the physical products were accessible to and understood by both parties.
Even if exclusion of this evidence were erroneous, it would not alter the summary judgment decision because, as the Prior Order recognized, these abstractions and layout shapes would not create a genuine dispute about any of the essential facts if admitted.
Tela also asserts that "the Court ignored the material fact that Tela is not relying on the design documents and layouts in isolation, but rather in addition to direct evidence of the Accused Products through teardown images." Mot. 19. It is unclear how that would alter the decision on the motion to strike and exclude, which only excluded reliance on the design documents and layouts and explicitly acknowledged Foty's reliance on teardowns. Prior Order 45.
D. Pre-Suit Damages
I also found that Tela had not adequately disclosed the basis for pre-suit damages that its damages expert awarded. See Prior Order 53-54. I noted that Intel has served interrogatories that requested such disclosure but that Tela's responses were only objections or incorporations of its contentions. Id. 53.
Instead of pointing to a damages disclosure or interrogatory response, Tela's Opposition cited a sentence from Section VI of its Patent Local Rule 3-1 disclosures that I discussed in the Prior Order and that Intel responded to in its Reply. Dkt. No. 259-3 at 23. Section VI dealt with the basis for Tela's allegations of willful infringement (not damages). Tela represented that this disclosure "in and of itself, moots Intel's claim." Id. The sentence stated that "Intel itself alleges that it became aware of an assertion by Tela of . . . the asserted patents . . . in August 2014 when Tela sent Intel a list of certain patents for a licensing discussion." Id. In the Prior Order, I explained that "I fail to see how this disclosure would put Intel on notice concerning even the basic facts required." Prior Order 53. That is especially so as Tela's damages expert assessed pre-suit damages from earlier than August 2014 and because it is unclear how being sent a list of patents one has rights in would be sufficient.
Tela now relies on two parts of those 3-1 disclosures that it did not cite. First, it cites a different section of those disclosures than Section VI in which the cited statement appeared. That section explicitly concerns the damages period and does contemplate "[d]amages for infringement" as beginning in 2012 (at least for those patents issued by then). See Mot. 21-22. Although those disclosures were in an infringement contention, they do appear to plausibly put Intel on notice that the damages period would extend to the pre-suit damages assessed by the expert. Second, Tela cites some statements from the section in which the above statement did appear, on willful infringement. It is far less clear that those statements would put Intel on adequate notice that Tela was pursuing pre-suit damages on this basis. While they do concern Intel's purported knowledge prior to the suit, none disclose that pre-suit damages will be pursued on that basis. The point of our disclosure requirements, though, is to ensure that the other party can fairly defend against contentions, not just that some underlying facts are disclosed unconnected from the basic theories they will support. The disclosure would therefore have to fairly inform Intel that pre-suit damages were being sought and on what broad basis, which this second disclosure does not. Accordingly, it appears that the first disclosure but not the second may have been adequate.
The problem with the more plausible first disclosure is that Tela did not raise it. Again, the only concrete statement Tela raised was the one about "August 2014" described above. Tela seeks to get around this by now calling that statement "exemplary" and pointing out that, in its Opposition, it said that the disclosure included, "among numerous bases," the statement. Mot. 22-23. Tela therefore argues that the Prior Order ignored the remainder of the 3-1 disclosure. Reading Tela's Opposition liberally, it fairly directed me and Intel to Section VI of the disclosure, because that is the section in which the cited statement falls. (Even in that case, Tela did not actually cite anything further from it, but I nonetheless reviewed it in ruling on the motion.) But, as explained above, it is not in itself a sufficient disclosure.
Because the more plausible disclosure was not cited, Tela argues that the failure of the Prior Order was not "consider[ing] the entirety of Tela's P.L.R. 3-1 disclosure." Mot. 23 (emphasis added). But Tela cannot vaguely gesture to the entirety of a 3-1 disclosure and call it a day. It is not the role of a court to sift through 56 pages of disclosures (especially when this particular dispute was only one of several dozen) merely because a party cites a single statement, does not cite any others, and now brands that statement "exemplary." Tela is intimately familiar with its disclosures; I am not. Moreover, its failure to cite other statements prejudiced Intel, which responded to the cited statement in particular, as would be natural, but had no opportunity to respond to the paragraphs Tela now cites. See Dkt. No. 270-4 at 13. Because the adequacy of this disclosure, even on Tela's account, had to be derived from documents in which such disclosures do not usually appear, its lack of citation is even less helpful. If this disclosure is so on point, it is unclear why Tela did not point to it. Tela says that "space constraints" required it to omit this information in favor of the "exemplary" statement, Mot. 23, but (1) adding a few extra sentences to cite the pertinent information is not onerous or overly space-consuming; (2) the statement is not truly exemplary as it is pulled from a different section of the disclosure; and (3) in any event, parties must always contend with space constraints, which are not an excuse to raise something for the first time in a motion to reconsider.
Tela is incorrect that it adequately raised this at the hearing. Its slides do not reference the other portions of the 3-1 disclosures with any particularity. Tela relies on portions of the hearing when its counsel discussed the adequacy of its disclosures, but the statements it cites are all general statements such as, "[i]n the Patent Local Rule 3-1 disclosures, we set up all the ways that Intel was on notice about Tela's patents and their infringement" and "[i]f we look at those 3-1 and 3-8 disclosures, that explicitly says that Tela will be relying on the time from December of 2012, depending on the patents." See Mot. 21 n.4 (citing Dkt. No. 313 at 79:9-81:7). The most reasonable understanding is that this argument was not intended to do anything but reflect what Tela said in its brief. Counsel at the hearing did not point to the separate timing section of the disclosures that was not cited in its brief.
I do not know what the outcome would have been if Tela had raised this portion of its disclosure earlier and Intel had a chance to respond. But reopening the issue would countenance a party's ability to reference specific portions of a document, omit reference to other portions, and then successfully move to reconsider because the court did not scour the entire document, no matter how long or complex, to find the omitted information and recognize its alleged relevance. That dog won't hunt.
Tela is correct that one footnote in the Prior Order was erroneous. Footnote 15 treated a footnote in Tela's Opposition as being related to the discussion on pre-suit notice when, in reality, Tela's footnote was from the preceding section of the Opposition. That explains why it appeared so irrelevant to the issue, as the Prior Order observed. I will remove the footnote (which was not material) from the amended order. --------
II. MOTION TO SEAL
Tela moves to seal certain information in its filings. Separately pending before me is the parties' joint renewed motion to seal related to the Prior Order. I will rule on the current motion to seal when that comprehensive renewed motion is decided because it covers the same ground. I will issue an unredacted version of this order if necessary.
IT IS SO ORDERED.
Dated: February 11, 2021
/s/_________
William H. Orrick
United States District Judge