By Peter E. Heuser and Ryan C. Fox-Lee

Akamai Technologies – A unanimous en banc Circuit holds that liability for direct infringement under § 271(a) is not limited to principal-agent or contractual relationships and joint enterprises but can also be based on conditioning participation in an activity upon performance of steps of a patented method. This decision makes it considerably easier to establish direct infringement than was previously the law under Muniauction. Here the Circuit reinstates a $45 million verdict based upon substantial evidence of such conditioning of participation upon performing steps that infringed the asserted patents.

Power Integrations The Circuit vacates and remands a Board decision reached in an ex parte patent reexamination. In a case filed over 10 years ago, Power Integrations sued Fairchild for infringement of patents directed to energy efficient power supplies. The patents were found valid and infringed, damages were assessed at $105 million, and a permanent injunction was issued. While the district court proceedings were ongoing, Fairchild filed for ex parte reexamination, and the PTO found that several of the claims were anticipated. The present decision instructs the Board on remand to fully consider Power Integrations’ arguments and explain the reasoning supporting its decision.

Keranos – Keranos has standing to sue for infringement of expired patents based on its exclusive license of all substantial rights in the patents. This is true regardless of the fact that the license was acquired after expiration of the patents. The Circuit also finds that the district court did not necessarily abuse its discretion in denying Keranos’ motions for leave to amend infringement contentions, but vacates and remands for party-by-party consideration of Keranos’ motions for leave to amend.

Akamai Technologies, Inc. v. Limelight Networks, Inc., Fed. Cir. Case 2009-1372; -1380; -1414;
-1417 (August 13, 2015)

This case was returned to us by the Supreme Court, noting “the possibility that we erred by too narrowly circumscribing the scope of § 271(a).” Sitting en banc, we unanimously set forth the law of divided infringement, and conclude that substantial evidence supports the jury’s finding that Limelight directly infringes under § 271(a).

I. Divided Infringement

Direct infringement under § 271(a) occurs where all steps of a claimed method are performed by or attributable to a single entity. See BMC Res., Inc. v. Paymentech, L.P., 498 F.3d 1373 (Fed. Cir. 2007). Where more than one actor is involved in practicing the steps, a court must determine whether the acts of one are attributable to the other such that a single entity is responsible for the infringement. We will hold an entity responsible for others’ performance of method steps in two sets of circumstances: (1) where that entity directs or controls others’ performance, and (2) where the actors form a joint enterprise.

To determine if a single entity directs or controls the acts of another, we continue to consider general principles of vicarious liability. In the past, we have held that an actor is liable for infringement under § 271(a) if it acts through an agent or contracts with another to perform one or more steps of a claimed method. See BMC. We conclude, on the facts of this case, that liability under § 271(a) can also be found when an alleged infringer conditions participation in an activity or receipt of a benefit upon performance of a step or steps of a patented method and establishes the manner or timing of that performance. In those instances, the third party’s actions are attributed to the alleged infringer such that the alleged infringer becomes the single actor chargeable with direct infringement. Whether a single actor directed or controlled the acts of one or more third parties is a question of fact, reviewable on appeal for substantial evidence, when tried to a jury.

Alternatively, where two or more actors form a joint enterprise, all can be charged with the acts of the other, rendering each liable for the steps performed by the other as if each is a single actor. A joint enterprise requires proof of four elements:

(1) an agreement, express or implied, among the members of the group;

(2) a common purpose to be carried out by the group;

(3) a community of pecuniary interest in that purpose, among the members; and

(4) an equal right to a voice in the direction of the enterprise, which gives an equal right of control.

Restatement (Second) of Torts § 491 cmt. c. As with direction or control, whether actors entered into a joint enterprise is a question of fact, reviewable on appeal for substantial evidence. Section 271(a) is not limited solely to principal-agent relationships, contractual arrangements, and joint enterprise, as the vacated panel decision held. Rather, to determine direct infringement, we consider whether all method steps can be attributed to a single entity.

II. Application to the Facts of This Case

Akamai filed an infringement action alleging infringement of several patents that claim methods for delivering content over the Internet. The parties agreed that Limelight’s customers, not Limelight, perform the “tagging” and “serving” steps in the claimed methods. For example, Limelight performs every step save the “tagging” step, in which Limelight’s customers tag the content to be hosted and delivered by Limelight’s content delivery network. The district judge instructed the jury that Limelight is responsible for its customers’ performance of the tagging and serving method steps if Limelight directs or controls its customers’ activities. The jury found that Limelight infringed. The district court initially denied Limelight’s JMOL for noninfringement, ruling that Akamai had presented evidence that Limelight directed or controlled its customers. Following our decision in Muniauction, Inc. v. Thomson Corp., 532 F.3d 1318 (Fed. Cir. 2008), the district court granted Limelight’s motion for reconsideration, holding as a matter of law that there could be no liability.

We reviewed the evidence that Limelight directs or controls its customers’ performance of each remaining method step. Specifically, we reviewed the evidence that Limelight conditions its customers’ use of its content delivery network upon its customers’ performance of the tagging and serving steps, and that Limelight establishes the manner or timing of its customers’ performance.

First, the jury heard evidence that Limelight requires all of its customers to sign a contract delineating the steps customers must perform to use the Limelight service. As to tagging, Limelight’s form contract provides: “Customer shall be responsible for identifying via the then current Limelight process all URLs of the Customer Content to enable such Customer Content to be delivered by the Limelight network.” As for the serving step, the form contract states that Limelight is not responsible for failures in its content delivery network caused by its customers’ failure to serve content. Thus, if Limelight’s customers wish to use Limelight’s product, they must tag and serve content. Accordingly, substantial evidence indicates that Limelight conditions customers’ use of its content delivery network upon its customers’ performance of the tagging and serving method steps.

Substantial evidence also supports finding that Limelight established the manner or timing of its customers’ performance. Upon completing a deal with Limelight, Limelight sends its customer a welcome letter instructing the customer how to use Limelight’s service. The welcome letter contains a hostname assigned by Limelight that the customer “integrates into its webpages.” This integration process includes the tagging step. Moreover, Limelight provides step-by-step instructions to its customers telling them how to integrate Limelight’s hostname into its webpages if the customer wants to act as the origin for content. If Limelight’s customers do not follow these precise steps, Limelight’s service will not be available. Lastly, the jury heard evidence that Limelight’s engineers continuously engage with customers’ activities. The engineers remain available if the customer experiences any problems. In sum, Limelight’s customers do not merely take Limelight’s guidance and act independently on their own. Rather, Limelight establishes the manner and timing of its customers’ performance so that customers can only avail themselves of the service upon their performance of the method steps. Therefore, Limelight is liable for direct infringement.

Comment: This case has a long history that is important to understand in order to appreciate the significance of this ruling. Soon after a jury returned a $45 million verdict for Akamai in 2006, the Circuit issued its Muniauction decision, which required that, for direct infringement, all the steps of a patent be performed by a single party or multiple parties in a formal relationship. The district court therefore issued JMOL, holding that Limelight could not have infringed. On appeal, the Circuit held that Limelight could be held liable for induced infringement but not direct infringement. The Supreme Court reversed that ruling, holding that there could be no induced infringement unless there is direct infringement, but instructed the Circuit to address the issue of direct infringement. The panel followed Muniauction, holding that there was no formal relationship and so there could be no direct infringement. In the present decision, the Circuit makes it considerably easier for direct infringement to be found based on something less than a formal arrangement; for example, if the accused infringer conditions participation in an activity upon performance of steps of a patented method.

Power Integrations, Inc. v. Michelle K. Lee, Fed. Cir. Case 2014-1123
(August 12, 2015)

In 2004, Power Integrations brought suit against Fairchild Semiconductors, alleging that it infringed four of its patents. During claim construction proceedings, Power Integrations argued that the term “coupled” in claim 1 of the ‘876 patent, when read in light of the specification and surrounding claim language, required that two circuits be connected in a manner “such that voltage, current, or control signals pass from one to another.” It further contended that the “recited coupling” between the counter and the digital to analog converter must be “present for the purposes of control.” Power Integrations made clear, however, that its proposed construction did not require a direct connection between circuit elements. The district court adopted Power Integrations’ proposed claim construction, concluding that it was “consistent with the claim language and the context of the specification which describes the purpose for which various parts of the claimed invention are coupled.” The court emphasized, moreover, that its construction of the term “coupled” did not “require a direct connection or. . . . preclude the use of intermediate circuit elements.” In the wake of the trial court’s claim construction, Fairchild withdrew its anticipation defense, instead arguing at trial that a patent to Martin rendered claim 1 obvious.

In 2006, the trial court bifurcated the litigation, separating issues of infringement and damages from issues related to patent validity. A first jury found that Fairchild had willfully infringed claim 1 of the ‘876 patent, as well as several claims of the other asserted patents. After a trial on validity, a second jury returned a verdict that claim 1 of the ‘876 patent was not obvious in view of Martin. On appeal, this court affirmed the jury’s finding that claim 1 of the ‘876 patent was not invalid for obviousness.

In December 2006, while district court proceedings were pending, the PTO granted Fairchild’s request for ex parte reexamination of claims 1, 17, 18, and 19 of the ‘876 patent. The board affirmed the examiner’s rejection of claim 1 as anticipated by Martin, stating that Power Integrations “appeared to argue that one of ordinary skill in the art would understand the term ‘coupled to’ to restrict device connections to exclude intervening components.” The board concluded, however, that the term meant simply to “join two devices into a single circuit,” and did not preclude the presence of intervening components. In light of its construction of the term “coupled to” in claim 1, the board rejected Power Integrations’ argument that Habetler did not anticipate claims 17, 18, and 19 because it contains an EPROM between the counter and the digital to analog converter. The board likewise rejected Power Integrations’ argument that Habetler fails to disclose primary and secondary voltage sources. According to the board, both the output from the digital to analog converter and the “average slope” of Habetler are voltage signals.

After unsuccessfully asked for rehearing, Power Integrations filed suit in District Court for the District of Columbia challenging the board’s decision. After correctly determining that it lacked subject matter jurisdiction, the district court transferred the case to this court pursuant to 28 U.S.C. § 1631.

The Board’s Anticipation Rejections

Proceedings of the board are governed by the Administrative Procedure Act (“APA”), which “establishes a scheme of reasoned decision making.” Under the APA, the board is obligated not only to come to a sound decision, but to fully and particularly set out the bases upon which it reached that decision. To permit effective appellate review, the board’s patentability analyses must be both “clearly disclosed and adequately sustained.” Here, however, the board fundamentally misconstrued Power Integrations’ principal claim construction argument and failed to provide a full and reasoned explanation of its decision to reject claim 1 of the ‘876 patent as anticipated.

Before this court, the district court, and the board, Power Integrations has consistently argued that claim 1, when read in light of the specification and surrounding claim language, requires that the counter itself—not a pre-programmed memory—controls the digital to analog converter’s output to vary the switching frequency. In its view, the “coupled” limitation in claim 1 requires that the counter be connected to the digital to analog converter in a manner that allows it to pass voltage, current, or control signals to instruct the digital to analog converter. Power Integrations has repeatedly made clear, however, that its proposed claim construction does not preclude the presence of intervening components between the counter and the digital to analog converter. The district court adopted Power Integrations’ proposed construction of the term “coupled,” concluding that it was “consistent with the claim language and the context of the specification which describes the purpose for which various parts of the claimed invention are coupled.”

During reexamination, however, the board failed to acknowledge the district court’s claim construction or to assess whether its interpretation of the term “coupled” was consistent with the broadest reasonable construction of the term. Instead, the board devoted a substantial portion of its analysis to resolving the question of whether the term “coupled” requires a direct connection between the counter and the digital to analog converter. Relying on a generalist dictionary definition, the board concluded that no such direct connection is required because “the plain and customary meaning” of the term “couple” is simply to “join two devices into a single circuit.”

We do not hold that the board must in all cases assess a previous judicial interpretation of a disputed claim term. Nor do we express any view on the merits of Power Integrations’ proposed construction of the term “coupled to.” We hold only that the board on remand should carefully and fully assess whether the disputed claims of the ‘876 patent are anticipated by the prior art, setting out its reasoning in sufficient detail to permit meaningful appellate review.

Keranos, LLC v. Silicon Storage Technology, Inc., Fed. Cir. Case 2014-1360; -1500
(August 13, 2015)

Keranos is the exclusive licensee of three patents owned by United Module Corporation (“UMC”) relating to split-gate flash memory. The patents expired in 2006 and 2008, but Keranos obtained rights to the patents in 2010; UMC continued to hold legal title to the patents. Keranos sued forty-nine parties in the Eastern District of Texas for infringing the patents, contending that the defendants used a specific type of flash memory technology, called “SuperFlash,” that implements a split-gate memory design.

Keranos did not join UMC as co-plaintiff so defendants moved to dismiss the actions for lack of standing, because the asserted patents had expired before Keranos acquired rights to them and filed suit. The district court concluded that the license agreement between UMC and Keranos transferred all substantial rights in the patents, giving Keranos standing without joining UMC.

When Keranos served its Local Patent Rule 3-1 Disclosure of Asserted Claims and Infringement Contentions, the original contentions identified some of defendants’ products with specificity, by product name or number, and others more generally, by product family (e.g., “SuperFlash”). After some discovery, Keranos moved for leave to amend the infringement contentions, but its motion was denied. The district court explained that the local patent rules required Keranos to identify infringing products by specific product numbers, not simply product families, and that Keranos had not demonstrated that it was diligent in searching for and identifying infringing products to the extent possible based on publicly available information. Eventually only two products specifically identified by product number were left, and Keranos agreed to dismiss its infringement claims, given that damages as to those items would be de minimis. The district court then granted summary judgment for the manufacturers.

On appeal, Keranos contended that the district court abused its discretion in denying its motions for leave to amend infringement contentions. Appellees contended that Keranos lacked standing to sue for infringement of the asserted patents in the first place.

Standing

When a legal title holder to a patent makes an assignment of all substantial rights under the patent, then that assignee has standing. An exclusive license may be tantamount to an assignment for standing purposes if it conveys to the licensee all substantial rights to the patent at issue. Otherwise, the licensee must join the owner of legal title to satisfy the standing requirement.

UMC transferred to Keranos the exclusive past, present, and future rights to sue for infringement, to make, use, import, and sell products covered by the patents, and to negotiate and grant sublicenses. UMC did not retain the right to sue accused infringers and indeed the agreement expressly prohibited UMC from suing or from negotiating or granting further licenses to the patents, absent written authorization from Keranos. The Circuit finds that the agreement as a whole indicated that UMC intended to transfer all substantial rights to Keranos.

The defendants cited Mars, Inc. v. Coin Acceptors, Inc., 527 F.3d 1359 (Fed. Cir. 2008) for the proposition that substantial rights no longer exist once a patent expires, so Keranos lacked standing without full legal title. The panel notes that Mars held that a transfer of only “rights in or to past infringement” in a then-expired patent did not convey standing on the transferee. However, Mars did not create separate standing rules for expired and unexpired patents, and the panel notes that Mars only involved a limited transfer of rights.

The Circuit concludes that Keranos had standing to sue for infringement of the asserted patents and declined to create separate tests for establishing standing for unexpired patents and expired patents.

Compliance with Infringement Contention Rule

The district court held that Keranos’ amendments to its infringement contentions did not meet the requirements of the local patent rules. Keranos argued that it needed discovery to obtain required product information, because publicly available information was insufficient for it to identify products and that, by identifying “SuperFlash”, certain product families, and certain product numbers in the original infringement contentions, Keranos gave appellees sufficient notice of its theory of infringement.

The district court disagreed that public information was insufficient to identify products and found that Keranos did not satisfy local rules by failing to identify products by specific product name or number. The district court concluded that Keranos “failed to demonstrate that it acted diligently in searching for and naming the additional products that incorporate the accused technology”. The panel holds that placing this burden of proof on Keranos was not an abuse of discretion.

Keranos attempted to show that it spent many hours reviewing datasheets, reverse-engineering products, and searching public information sources to identify potentially infringing products and that much of the publicly available information was insufficient to identify accused products or did not exist for some of appellees’ products. Because the record indicated that publicly available information might not have been available for the products of some appellees, the Circuit court vacates and remands for the district court to consider, on a party-by-party basis, whether Keranos had shown good cause to amend its infringement contentions.

Sign up

Ideas & Insights