Multiple-actor method claims lurk in patents, waiting to surprise patent owners that move to enforce their rights. Rather than simply identifying one entity that performs all of the steps of a claimed method and asserting infringement, patent owners could find themselves forced to establish the existence of a joint enterprise among multiple actors or otherwise to attribute performance of method steps to an entity that did not actually perform them. These method claims are also troubling to technology users that wish to perform only a portion of a claimed method, but fear that a patent owner will come knocking on their door when some other entity performs the remaining method steps. In the wake of the en banc decision of the Federal Circuit in Akamai Technologies, Inc v Limelight Networks, Inc, one actor’s conduct can be attributed to another – even in the absence of a joint enterprise, an agency relationship or a contractual obligation governing that conduct. After a decade-long wrestling match in the US federal courts in Akamai, the state of the law for multi-actor infringement of method claims seems less clear than ever.
Even in this shadow of uncertainty, the fact remains that the ease or difficulty of enforcing a method claim may largely depend on how carefully the drafting attorney limited the claim to acts attributable to one entity more than any other factor. Mindful claim drafting can remove much of the uncertainty associated with the enforcement of method claims and help patent owners to avoid the additional time, cost and effort required to establish infringement.
Akamai v Limelight – a tale of four reversals
The uncertainty associated with multiple-actor method claims can be appreciated from the epic litigious journey of one humble method claim at issue in Akamai: the final claim of US Patent 6,108,703. This journey began in a federal district court and continued all the way up to the Supreme Court, and then went part-way back. The answer as to whether this claim was infringed changed from yes to no, back to yes, back to no and finally back to yes again. The defendant did not dispute that every step of the patented claim was performed. The question to be answered by the courts was whether any party could be held liable for patent infringement when no one entity actually performed every step of the claim.
The subject of the Akamai patent was a scheme of distribution and delivery of webpage content (eg, images and videos) among a number of physically separate computer file storage locations. The claimed method required the following steps:
- distributing objects across a network of servers;
- tagging embedded objects of a page (eg, a webpage) in a particular manner;
- resolving a client request for one of the embedded objects; and
- returning to the client an Internet Protocol address of a server where the requested object was hosted.
Limelight, Inc owned and operated a content delivery network (CDN) and provided a service to customers, allowing them to use the network to host webpage content remote to the customers’ own servers. Akamai accused Limelight of infringing the patented method by providing this service. Limelight did not dispute that it performed three of the four steps, but denied infringing the claim based on the fact that Limelight did not perform the tagging step. Limelight’s customers performed the tagging step, which generally involved modifying a customer-operated webpage to access the Limelight CDN when retrieving embedded objects. Akamai acknowledged that Limelight’s customers were the actors actually performing the tagging step, but asserted that the actions of those customers should be attributed to Limelight as if Limelight itself had performed the tagging step, thereby making Limelight liable for patent infringement.
A jury found Limelight liable for infringement based on evidence that Limelight “directed or controlled” its customers’ performance of the tagging step. The jury instruction explained the ‘direct or control’ requirement for attributing a different actor’s conduct to the accused infringer using the rule set out by the Federal Circuit in BMC Resources. The jury applied this rule and relied on evidence that included:
- a service contract between Limelight and its customers;
- instructions provided to the customers for using the Limelight CDN; and
- the availability of Limelight customer service personnel to assist customers with the service.
The service contract included steps that must be taken by the customer to use the service, including performance of the tagging step, and the written instructions provided information on content tagging.
The district court vacated the jury decision, entering a judgment of non-infringement as a matter of law after the Federal Circuit issued its opinion in Muniauction, Inc v Thomson Corp, in which the owner of an auction website was not held liable for infringement of a method that required the step of placing a bid – a step performed by Muniauction’s customers, who paid fees to Muniauction and who received bidding instructions from Muniauction. The judgment of non-infringement as a matter of law opinion reasoned that Muniauction required something more than a service contract between parties and instructions on using the service to establish conduct rising to the level of direction or control of another entity by the accused infringer.
On appeal, a Federal Circuit panel affirmed the judgment of non-infringement, holding “as a matter of Federal Circuit law that there can only be joint infringement when there is an agency relationship between the parties who perform the method steps or when one party is contractually obligated to the other to perform the steps”. The Federal Circuit reasoned that Limelight’s customers were under no contractual obligation to perform the tagging step; rather, the customers performed the tagging step only under circumstances through which the customer wanted to use the service and on webpages identified by the customer – decisions controlled by the customer, not by the accused infringer.
Limelight’s victory was short-lived. The Federal Circuit vacated the panel decision and reheard the case en banc, this time holding Limelight liable for patent infringement under 35 US Code Section 271(b) on a theory of inducement. The en banc court held that a defendant may be held liable for induced infringement of a method if the defendant performs some of the method steps and induces others to perform the remaining steps, also holding that no direct infringer is required for a party to be liable for induced infringement.
Soon after, Akamai’s fortunes reversed when the Supreme Court stepped in and unanimously reversed the en banc decision. The court held that there can be no inducement without direct infringement and reminded the Federal Circuit of its own case law in Muniauction that requires a single entity to direct or control the performance of all steps of a method for there to be direct infringement. On remand, the Federal Circuit panel again affirmed the district court’s judgment of non-infringement, holding that Limelight’s customers did not perform the step of tagging as an agent of Limelight, as part of a joint enterprise with Limelight or under a contractual obligation to do so.
Just when it seemed that the dust had finally settled, the Federal Circuit issued a second en banc opinion – issued per curiam – again vacating the panel opinion and reinstating the jury’s finding that Limelight directly infringed the patented method by attributing the actions of Limelight’s customers to Limelight. The Supreme Court decided not to revisit the Federal Circuit’s analysis the second time around. After a decade of patiently hoping for the law of divided infringement to be settled, patent owners and practitioners are left with what is arguably a less defined doctrine.
The more things change, the more they remain the same – or vice versa?
In the end, the original Akamai jury verdict was upheld and the Federal Circuit did not expressly overrule its precedents in BMC Resources or Muniauction. So did anything change? And if so, to what effect?
While the legal standard for attributing one actor’s conduct to another in the context of a patented method claim remains unchanged, the type of conduct that rises to the level of directing or controlling another entity’s actions appears to have broadened – albeit to an unknown extent; the effect is different for patent owners than for accused infringers. Now, according to the Federal Circuit, “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”. The Federal Circuit explained its legal framework for direct infringement by expanding its notion of vicarious liability. Specifically, in addition to the traditional agency and contractual relationships between actors, vicarious liability now also encompasses scenarios in which an accused infringer conditions participation in an activity or receipt of a benefit on performance of a step or steps of a patented method and establishes the manner or timing of that performance. These combined actions are sufficient to find that the accused infringer directed or controlled the acts of a third party.
The effect is that liability largely depends on the fact-based circumstances under which the method is performed. The Federal Circuit expressly invited litigants to test scenarios under which actions of one can be attributed to another, recognising that “[i]n the future, other factual scenarios may arise which warrant attributing others’ performance of method steps to a single actor. Going forward, principles of attribution are to be considered in the context of the particular facts presented”. The conditions of participation and the benefits received as part of a collectively performed patented method are open to interpretation, as is what constitutes the establishment of the manner or timing of the performance of its individual steps. The result is more uncertainty for both patent owners and those performing portions of patented methods.
Prudent patent applicants can continue to take steps to avoid future problems and the uncertainty associated with divided infringement by pursuing method claims that are written to be performed by a single entity. This is not a new concept and was recognised by the BMC Resources court, which commented that “[a] patentee can usually structure a claim to capture infringement by a single party” – for example, by drafting its claims to focus on one entity. This prevents clever defendants from attempting to escape liability by asserting that some other entity performs one or more of the method steps and forcing the patent owner to present extensive fact-specific evidence of vicarious liability.
Patent practitioners or other claim drafters can implement this technique by first identifying the preferred actor that will perform the claimed method steps. This actor may be a person or entity, such as a business competitor of the patent applicant. Each method step can then be written from that actor’s perspective and may include actions performed by the actor, by an apparatus controlled by the actor or at a location where the action is carried out. Also, claim passages can be written to avoid positively recited steps or actions where unnecessary, or where such steps are unlikely to be performed by the same actor as the other steps. As a real-world example, the relevant passage from the Akamai method claim discussed above could have easily been written differently to state:“wherein, for a given page normally served from the content provider domain, tagging at least some of the embedded objects of the page are tagged so that requests for the objects resolve to the domain instead of the content provider domain.”
In this example, the verb ‘tagging’ is omitted in favour of the adjective ‘tagged’, which is used to characterise the embedded objects. This simple change effectively eliminates a method step, resulting in a method claim in which all of the steps would be performed by the same actor – the CDN operator.
Even if patent owners find themselves with previously issued patents containing multiple actor claims without a path to amend these claims, the Federal Circuit has provided a lifeline with which they can assert direct infringement. However, a pre-litigation investigation may not conclusively determine whether a potential defendant conditioned participation or received a benefit on a third party’s performance of a step, or whether the potential defendant established the manner or timing of that performance. These fact-intensive enquiries and proofs are sure to be time consuming, expensive, difficult and uncertain.
Potential defendants that perform some, but not all, of the steps of a method claim have a much foggier path. While patent owners at least have some control over claim scope to help to avoid the complexities associated with divided infringement, defendants are left with more questions than answers. Potential defendants must determine whether their behaviour towards another entity performing other method steps conditions participation in an activity or receipt of a benefit on that performance, and whether their own behaviour establishes the manner or timing of that performance. Given the lack of identifiable pointers to what these various indicators of ‘direct or control’ mean, uncertainty reigns.
One subsequent district court case has found the latest Akamai test met when a doctor prescribes a treatment to a patient. In that case, administration of a medication to the patient was attributed to the doctor even though the patient self-administered the medication, because the doctor conditioned the benefit of the treatment on the patient’s performance of taking the medication and established the manner or timing of taking the medication. Otherwise, despite the fact that Akamai was finally put to rest, potential defendants must continue to wait to better understand their position, as interpretation of the broader application of divided infringement unfolds in the coming years.
Reising Ethington PC
755 W Big Beaver Rd
Troy MI 48084
Tel +1 248 689 3500
Fax +1 248 689 4071
Colin Cicotte is a partner whose practice integrates multiple disciplines of IP law, including patent enforcement, analysis and prosecution. He has a BS in electrical engineering from Wayne State University, a BA in economics from the University of Michigan and a JD from Wayne State University. He has extensive experience with patent infringement, trademark infringement and trade secret litigation, and has been involved with complex IP litigation in a variety of US federal jurisdictions. In addition to patent litigation, he renders opinions on infringement and invalidity issues. He also prepares and prosecutes patents in both the United States and abroad. His fields of expertise include electrical, mechanical and computer technologies.
Scott Hogan is a partner whose practice includes working with clients to help to identify and protect their intellectual property. With hands-on work experience in a number of fields – from biopolymers and polymer solution chemistry to electric generators and automotive components – he is as at home at a university symposium as he is in a manufacturing plant, and has several US and foreign patents to his name. Mr Hogan has expertise in a variety of technologies, including laser-based manufacturing processes, nanomaterials and nanotechnology, microfabricated devices and automotive components. His practice also encompasses other aspects of intellectual property, with experience in trademark prosecution, opposition and litigation, as well as copyright registration and licensing.