On August 12-14, 2015, the Chisum Patent Academy held a three-day seminar at the historic Mayflower Park Hotel in downtown Seattle, Washington to discuss and debate current developments in U.S. patent law. The roundtable seminar group was limited to ten persons; sessions were led by treatise authors and educators Donald Chisum and Janice Mueller.
Our attendees included experienced patent professionals from law firms, corporations, and non-profit research centers in Seattle, Philadelphia, California, Wisconsin, France, and Mexico. Participants enjoyed the Mayflower Park's central location, friendly staff, gracious accommodations, and lively seminar discussion.
Here’s our recap of the takeaways from the seminar:
▪ 2015 Blockbuster Supreme Court and Federal Circuit Decisions: In our opening “blockbuster” session, we reviewed 2015 Supreme Court cases and Federal Circuit en banc cases, decided and pending, as well as those in which a denial of en banc review drew dissenting and concurring opinions.
Compared to 2014, there were fewer Supreme Court decisions. There was no Alice, but the triumvirate of Teva (appellate review of claim construction), Commil (active inducement) and Kimble (post-expiration royalties) is nonetheless highly significant.
The Supreme Court’s B&B Hardware decision gave issue preclusion (collateral estoppel) effect between the parties to a PTO TTAB trademark cancellation decision. An intriguing question is: to what extent will issue preclusion apply to PTAB decisions in inter partes review (IPR) and post-grant review (PGR) of patents? PTAB IPR and PGR proceedings resemble, procedurally, those of the TTAB. Quite possibly a PTAB finding on a fact issue, such as what a particular prior art reference teaches, will bind the petitioner in future litigation and do so even beyond the limited estoppel provision of the IPR and PGR statutes. More complex is the question of issue preclusion as applied to claim construction. If the PTAB continues to apply a “broadest reasonable interpretation” rule, then its decisions on claim construction will not meet the requirement for issue preclusion that an issue in a first and subsequent proceeding be, in fact, the same.
Compared to the Supreme Court, the Federal Circuit in 2015 shifted into hyper drive. During the week of our Seattle seminar (August 10, 2015), the Circuit issued two en banc rulings, Suprema (8/10) (on active inducement as a theory for the ITC to exclude imports of noninfringing products that induce direct infringement of method claims), and Akamai (8/13) (on divided infringement). Our consensus was that practitioners face major challenges not only in keeping up with the case law developments, but also in projecting the effects of case law developments on past, pending, and future decisions.
Of uncertain status are two core standards that frequently arise in patent litigation: willful infringement and the laches defense (unreasonable, prejudicial delay as a bar to pre-suit damages for infringement). The Federal Circuit has questioned the viability of laches in light of the Supreme Court’s Petrella copyright case, taking the issue under en banc review in SCA Hygiene. On willful infringement, in Halo Electronics, the Circuit denied en banc review of the question whether the Supreme Court’s Octane Fitness and
Highmark decisions on attorney fee awards, which rejected the Federal Circuit’s rigid standard requiring an objective baselessness threshold, should extend to the Federal Circuit’s similar threshold for a finding of willful infringement.
The recent Federal Circuit panel mega-damages decision in Carnegie Mellon (8/4) highlighted the importance of both laches and willful infringement doctrine. In Carnegie Mellon, the panel rejected laches because of an infringer’s egregious copying, but threw out a $250 million enhancement for willfulness because the infringer developed an objectively reasonable obviousness defense at trial.
The Federal Circuit’s en banc ruling in Williamson (6/16/15) on mean-plus-function clauses in claims provoked lively discussion of the challenges that Williamson and other cases pose for specification and claim drafting, especially with computer-implemented technology. Uncertainty continues on when functional language and words other than “means” will evoke means-plus-function treatment (the example, use of “module” rather than “means”). That treatment triggers an obligation to find more detailed “corresponding structure” than is otherwise required by Section 112 for an enabling description. Uncertainty also continues concerning how much detail the Circuit’s algorithm disclosure rule requires. We agreed that full disclosure of source code would suffice for a computer implemented invention, but that strategy would severely limit the scope of a matching means clause in the claims. A recommended alternative strategy is a detailed flow chart. At a minimum, a specification’s description must provide at least one level of greater “granularity” than the language of the claim.
▪ Patent Claim Interpretation: SCOTUS Dropped the F[act]-Bomb but Did the Federal Circuit Hear? In January 2015, the Supreme Court (SCOTUS) decided Teva v. Sandoz, its most significant patent claim construction case since the 1996 blockbuster Markman v. Westview. In a 7-2 opinion authored by Justice Breyer, the Teva Court started from the unremarkable (but important) premise that patent claim construction does sometimes involve underlying factual disputes. It thus rejected the Federal Circuit's contrary protestations dating from at least the appellate court's 1998 Cybor en banc decision. In claim construction disputes involving "evidentiary underpinnings" (an unhelpful characterization, because all claim construction involves intrinsic evidence and sometimes extrinsic evidence), district courts will sometimes find it helpful to consult the extrinsic evidence (such as an expert's testimony concerning what a term of art meant to a skilled artisan as of the invention's effective filing date). When district court findings of fact result, the Federal Circuit must review them under the clear error standard of Fed. R. Civ. P. 52 (i.e., the finding must not be set aside unless clearly erroneous, and due regard must be given a trial court's opportunity to judge witness credibility). As with determinations of the "usage of trade or locality" in interpreting contracts, district courts may consult extrinsic evidence that is helpful to determine patent claim meaning. When district courts make findings of fact, those findings are subject to Rule 52 review like any other type of finding of fact. Notably, the Teva Court relied in part on a 1950 Second Circuit opinion by Judge Learned Hand that mandated deference to fact finding in patent claim construction. Nevertheless, the ultimate question of patent claim interpretation remains a legal question subject to de novo (no deference) review. The Teva majority doubted that fact findings based on extrinsic evidence would be required in many cases. When they are, courts will be capable of disentangling fact from law, and various doctrines and procedures such as collateral estoppel and case consolidation will help maintain uniformity in the construction of particular patents. On the merits, the Federal Circuit had erred in overturning the district court's finding that patentee Teva's proposed definition of the claim phrase "molecular weight" was correct. In particular, the appellate court improperly reviewed without deference the district court's decision to credit the Teva expert's explanation of why slight discrepancies in a molecular weight distribution curve did not undermine Teva's position.
Writing for himself and Justice Alito in dissent, Justice Thomas doubted the Teva majority's prediction that "subsidiary factfinding" is unlikely to loom large in future patent claim constructions. Rather, he charged that the majority's holding will spawn costly and meritless collateral litigation on the law/fact distinction, as parties prevailing at the district court will be incentivized to argue on appeal that the claim construction below involved findings of fact owed Rule 52 deference.
As of our August 2015 seminar date, the Federal Circuit had decided several precedential patent claim construction cases post-Teva. In most, the district court construed the claims relying only on intrinsic evidence. Hence, the Federal Circuit applied de novo review and Teva had no impact.
Notably, the Federal Circuit's June 2015 decision on remand from SCOTUS in Lighting Ballast v. Philips Elecs. (Reyna, J.) represents the only post-Teva Federal Circuit decision to date in which the appellate court specifically relied on and deferred to a district court's findings of fact in affirming the court's claim interpretation. Expert testimony in Lighting Ballast convinced the district court (and on appeal, the Federal Circuit) that the claim-recited "voltage source means" conveyed a class of structures known to skilled artisans, such that the claim element in dispute was not means-plus-function and was not indefinite for lack of supporting structure. The Federal Circuit reversed course from its original, pre-Teva, non-precedential panel opinion in which it had held the claims indefinite. In short, Teva made all the difference in Lighting Ballast.
In contrast, the Federal Circuit's June 2015 decision on remand from SCOTUS in Teva itself produced a 2-1 panel decision in which the majority sustained its original position that the claims were invalid for indefiniteness. The Teva remand majority (Moore, J.) relied on an aspect of the case not specifically dealt with by SCOTUS--inconsistent statements in the prosecution histories of related patents. Despite expert testimony which convinced the district court that skilled artisans would not have relied on a plainly scientifically erroneous statement in the prosecution history, the Federal Circuit majority held that the patentee was bound by its inaccurate statement. In the majority's view, the matter presented a legal question of interpreting the prosecution history rather than a question of fact about what a skilled artisan would understand from that history. In dissent, Judge Mayer charged that the relied-upon prosecution statement was "nonsensical, and a skilled artisan would not rely upon any part of it." In deciding otherwise, the majority had erroneously ignored the district court's extensive fact-finding about the background science involved, including the consideration of competing expert declarations, deposition testimony, and two hearings.
We debated whether Lighting Ballast signals that future claim interpretation cases will routinely present underlying fact questions to which Federal Circuit panels will defer, or whether the appellate court will consider such findings unnecessary because (in its view) the intrinsic evidence resolved any ambiguities. Given the current trend of its decisions, it is likely that the Federal Circuit will continue to rely on the Teva holding that the "ultimate" question of patent claim construction remains one of law, not fact.
▪ Patent Eligibility and the Fall-Out from Alice. In a session devoted to Section 101 patent eligible subject matter, we reviewed the eleven Federal Circuit cases on the topic since Alice.
In the nine “software/business method” cases, Federal Circuit panels acknowledged, at least implicitly, the interpretation of Alice's “abstract idea” focus that permits claims to technological solutions to technological problems, including improvements in computing and manipulation of how network communications operate. But only once, in DDR Holdings, did the Federal Circuit find that claims met that test. The invalidating decisions applied the Mayo/Alice “two step.” Most cases fit what Judge Chen in DDR Holdings described and distinguished: a "commonplace business method aimed at processing business information, applying a known business process to the particular technological environment of the Internet, or creating or altering contractual relations using generic computer functions and conventional network operations.”
In finding that steps or elements added to an “idea” did not meet the second step (i.e., does the claim add more than what was “well-understood, routine, and conventional activity”?, a phrase repeated three times by the Supreme Court in Mayo), the invalidating decisions often pointed to a characterization of steps and elements as “conventional” in the patent specification or prosecution. Thus, the Circuit used the patent as evidence that a step or element was “conventional” within the meaning of Alice/Mayo step two. That seemed unfair. In the past, court decisions encouraged brevity in patents, urging drafters to disclose the invention and not burden the system and the public with well-known details required for enabling the invention. Now, under Mayo and Alice, such statements are treated as fatal admissions. We concluded that in the future, drafters should avoid words like “conventional,” using carefully framed language to refer to sources for techniques not set forth explicitly in a specification.
Our detailed review of the post-Alice cases confirmed the imprecise and potentially illogical character of the Alice/Mayo second step. The Supreme Court’s decisions do not pinpoint a time framework for assessing whether an implementing step or element was (or is) “conventional.” Patent law fundamentals teach that what is new and obvious (Sections 102 and 103) changes with time. In contrast, determining whether an invention fits within Section 101 should be a threshold determination, not time dependent. Upon filing and issue and for a time thereafter, should a claim be patent-eligible because it entailed a technique that was not conventional, but later become ineligible because the technique became conventional? Surely not.
Nothing good could be said about the Federal Circuit’s decision on “natural phenomenon,” Ariosa v. Sequenom (6/12/15). Concurring, Judge Linn identified a clear point of distinction between the Supreme Court’s Mayo decision and the claims at issue in Ariosa. In Mayo, the level of certain substances (metabolites) had been assayed in the past. The claimed invention was based on the alleged discovery of the significance of the assay levels for drug efficacy and toxicity. In Ariosa, in contrast to
Mayo, the natural phenomenon (presence of certain DNA in certain blood) was previously unknown and had, presumably and unlike in Mayo, never been measured, assayed or amplified. The claim was to amplifying that DNA. Amplifying DNA, in general, was conventional, but it was not and could not have been applied to the unknown DNA in Ariosa. We concluded that Ariosa demonstrates the Federal Circuit's over-zealousness in appearing to “play ball” according to the Supreme Court’s rules. Judge Linn should have dissented, not concurred. As he pointed out, the invention opened the door to a new generation of better fetal testing. The Circuit should have focused on the fact that the claimed method resulted in a novel, physical result.
That said, we agreed that the claim in Ariosa could have been redrafted to strengthen the case against Section 101 ineligibility.
▪ Inter Partes Review: Where Are We So Far? Is the Broadest Reasonable Interpretation Rule Fair in IPRs? Hundreds of IPR petitions are now flooding into the PTO each month. Initial readings of PTAB IPR decisions suggest that the Board responds to obviousness challenges far more favorably would than lay jurors. For example, the Board gives minimal weight to secondary considerations evidence. We observed that the PTAB is quite comfortable in construing a patentee’s statements in foreign prosecution or litigation proceedings (parallel, related, or not) as admissions against interest, even when such statements do not appear to qualify as prior art. The PTAB is heavily influenced by patentee concessions that all elements of a claimed invention can be found in the prior art (despite the lack of persuasive reasons for combining the prior art teachings).
In In re Cuozzo (Feb. 4, 2015) (Dyk, J.), the first Federal Circuit review on the merits of an IPR final written decision, a two-judge panel majority of the Circuit bent over backwards to approve the USPTO's controversial use of the "broadest reasonable construction" (BRC) rule for interpreting patent claims in IPRs, despite admitting that opportunities to amend claims in IPRs are "cabined." Although dissenting Judge Newman observed that the PTAB has granted motions to amend claims in only two cases as of January 2015, the majority countered that an opportunity to amend is "nonetheless available" and the facts that the patent owner's presumptively single amendment must address the grounds of unpatentability "involved in the trial" and can not enlarge claim scope are not "material difference[s]" from pre-AIA proceedings involving BRC. Moreover, the BRC rule was used in interferences, another type of adjudicatory proceeding. In the Cuozzo majority's view, Congress was aware of the BRC rule when it passed the AIA, thus legislatively ratifying it. Even if the AIA did not explicitly incorporate a BRC standard for IPR claim constructions, the agency's regulation mandating BRC was properly adopted under Chevron. Given its approval of the PTAB's broad construction of the claim phrase "integrally attached" in a claim to a speed limit indicator, the Cuozzo majority had no difficulty affirming the Board's obviousness determination on the merits.
On July 8, 2015, the Federal Circuit denied rehearing en banc in Cuozzo. At the same time, it issued a revised panel opinion that superseded the February panel opinion (with no reference to the withdrawn February opinion and no introductory explanation of how or why the majority's reasoning had changed). Our reading concluded that the July panel majority (Dyk, J.) maintained its prior position that application of BRC was proper in the case at bar, but significantly narrowed its reasoning. The PTAB had denied Cuozzo's motion to amend his claims on the ground that the proposed amendment would have improperly enlarged the scope of the claims (a conclusion with which Cuozzo strenuously disagreed). The pre-AIA procedures, reexamination and reissue, likewise prohibited amendments that enlarged claim scope. Because both reexam and reissue used the BRC rule, it was also proper to apply it in the case at bar. Limiting its holding to the facts before it, the July panel majority observed that "[i]f there are challenges to be brought against other restrictions on amendment opportunities as incompatible with using the broadest reasonable interpretation standard, they must await another case."
The Cuozzo panel majority also held (with no substantive change between the withdrawn February opinion and the revised July opinion) that the Federal Circuit does not have authority to directly review under 35 USC 314(d) the propriety of a PTAB institution decision, even after a final written decision has been issued and even when the grounds asserted by the petitioner for institution were less than precisely correct from the standpoint of the statutory "particularity" requirement. In the majority's view, the fact that the petition was defective is not relevant because "a proper petition could have been drafted." A writ of mandamus to review an institution decision after issuance of a final written decision remains a slim possibility, but only when the USPTO has "clearly and indisputably" exceeded its authority.
▪ Method and System Claims: Active Inducement: In this session we repeated a topic from prior seminars. But it is a topic that keeps on giving! In 2015, the Supreme Court gave us Commilon the knowledge requirement for active inducement. The Court sharply distinguished what an accused inducer must have known and believed about infringement, on the one hand, and validity, on the other hand. A reasonable no-liability position on infringement (such as by a narrow claim interpretation) is a defense; a reasonable invalidity defense is not.
During the course of our seminar, the Federal Circuit graciously released its mercifully short en banc decision in Akamai/Limelight on “divided infringement.” A divided infringement arises when a method claim requires step A and then step B, and an accused infringer performs step A but instructs its customers on how to perform step B. Alternatively, an accused infringer may perform no step but encourage or facilitate a customer to perform step A with the customer arranging with another for performance of step B. Can direct infringement liability exist when no single entity performs all the method steps? Can there be indirect infringement when all the claims steps were performed but no entity is liable for direct infringement?
In a 2012 divided en banc decision, the Federal Circuit held in Akamai that inducement liability could exist even when no entity was liable for direct infringement. In its 2014 Limelight decision, the Supreme Court reversed that holding but remanded, suggesting that the Federal Circuit might wish to revisit its strict “direction or control” test in applying the single entity rule to determine direct infringement under Section 271(a).
On remand, in its initial May 18, 2015 decision, a Federal Circuit panel majority maintained the Circuit’s prior test for direction and control. It found no liability under the facts because the accused infringer performed step A in its system, gave instructions to its customers to take step B if they used the system, but did not control whether or when the customers should use the system. The May decision produced a heated and lengthy exchange between Judge Linn for the majority and Judge Moore in dissent. Each characterized in extreme terms the consequences of the other’s position. It was an ugly exchange that could have led to further uncertainty and confusion. Petitions for rehearing en banc probably would have followed and likely spawned a petition for Supreme Court review.
Just a few months later, on August 13, 2015, the Federal Circuit, sua sponte (on its own motion), vacated the panel opinion and issued a short en banc opinion in Akamai. The August en banc opinion contained a concise statement of the Circuit’s law on “divided infringement.” It affirmed the single entity rule. That rule required that “all steps of a claimed method [be] performed by or attributable to a single entity.” In determining attribution, the Circuit affirmed the prior “direction and control” test and added a “joint enterprise” test. For direction and control, it confirmed the existing subtests of agency and contracting. Significantly, the court announced a new subtest: an accused infringer can be found liable for another’s performance of method steps if the accused infringer (1) “conditions participation in an activity or receipt of a benefit upon performance of a step or steps of a patented method,” and (2) “establishes the manner or timing of that performance." The Circuit emphasized that there could be additional subtests, which would depend upon the particular facts of a case.
On the facts of the Akamai case, the en banc court held that the “conditions/establishes” subtest was met. A jury had already found that the accused infringer had controlled its customer’s performance of step B. There was evidence to support findings on both prongs of the new subtest. On conditioning, the accused infringer had a standard contract obligating a customer to perform step B if it wished to use the accused infringer’s system. On establishing the manner or timing of the performance, the accused infringer provided detailed instructions and its “engineers continuously engage[d] with customers' activities."
We agreed that Akamai was an amazing, positive switch by the Federal Circuit. Some of us speculated: did the Federal Circuit judges hold a retreat with a facilitator and yoga instructor on how to achieve internal harmony and produce simpler, clearer guidance to the patent law world? Will they use this calm, centered approach in future cases to resolve “hot” disputes? Will this newly collegial style of decisionmaking be better received at the Supreme Court?
Nothing in the new Akamai decision would justify a patent drafter deliberately drafting a claim to a method or system requiring participation by multiple entities, however. As we concluded in prior eminars, clear and thoughtful drafting of both method and system claims, with a view to the "who" as well as the "what," can avoid problems with "divided" infringement and the necessity of showing intent to infringe. The objective is to define "direct" infringement in terms of the actions of a single entity. Cases approving claims reciting relatively complex "environment" limitations in preambles vividly illustrate one drafting technique that can achieve this goal (see, e.g. Advanced Software Design Corp. v. Fiserv, Inc., 641 F.3d 1368 (Fed. Cir. 2011), and HTC Corp. v. IPCom GmbH & Co., KG, 667 F.3d 1270 (Fed. Cir. 2012)).
▪ Exhaustion of Method Claims: Uncertainties Spawned by Quanta. Our final seminar topic was (appropriately) “Exhaustion.” The topic is of heightened interest after the Federal Circuit’s en banc order in Lexmark indicating that it would address whether to overrule its prior decisions on two exhaustion issues. The first is: can a sale of a patented item outside the United States ever give rise to patent exhaustion? The Federal Circuit’s prior Jazz Photo case held that there was no such thing (at least in U.S. patent law) as international patent exhaustion. But the Supreme Court’s copyright decision (Kirtsaeng) casts doubt on that rule. The second issue is: does “a sale of a patented article, when the sale is made under a restriction that is otherwise lawful and within the scope of the patent grant, ... give rise to patent exhaustion?" The Federal Circuit’s 1992 Mallinckrodt decision held that no exhaustion arose from conditioned sale as to uses in violation of the condition. But the Supreme Court’s Quanta decision casts doubt on the Mallinckrodt rule. Quanta Computer held that sales of components of a patented invention could exhaust a method claim as well as a system claim if the sold components "substantially embodied" the invention.
We next reviewed three recent Federal Circuit exhaustion decisions, Keurig (2013), Lifescan (2013), and Helferich (2015). Our consensus view was that the Supreme Court’s Quanta decision was wrong and has created many uncertainties. For example, which components "embody" an invention?
In Lifescan, a patent was on an assay method that used special test strips and a meter. The patent owner sold meters below cost or even gave them away to promote sales of its strips. An accused infringer so ld “generic” strips for use with the patent owner’s meters, thereby directly undermining the patent owner’s primary revenue source for its patented technology. The majority found exhaustion because the meters “embodied” the method. But in dissent, Judge Reyna argued persuasively that the strips, not the meter, embodied the method because, inter alia, several of the novel features of the method were physically located on the strip. Thus, there should be exhaustion of the method claim only as to each test strip sold with the patent owner’s authority. We did agree: the patent owner had erred by failing to get separate claims to the strip.
In Keurig, the Federal Circuit created further confusion by suggesting that exhaustion should be analyzed on a "whole patent" rather than claim-by-claim basis.
The most recent decision, Helferich, dispelled a proposed extreme extension of exhaustion to cover not only whole patents but related patents that covered complementary services and products. Judge Taranto's opinion in Helferich provides a lesson on the use of Supreme Court precedent, which has lately become fashionable. In this instance, dicta in an 1894 Supreme Court decision worked in a patent owner's favor!
Upcoming Chisum Patent Academy Seminars
A single seat remains open for our next seminar, to be held September 24-25, 2015 at the offices of Kaye Scholer LLP in downtown Washington, DC. For additional details, see our website or email info@chisum.com.