Closing the Akamai II Loophole: An Alternative Approach to Determining Liability for the Joint Infringement of Interactive Method Claims

 

 

By: Ehsun Forghany

 

 

 

 

 

Table of Contents

 

Introduction................................................................................................................................. 2

I.    Unified Patent Infringement Law........................................................................................ 3

II.    Divided Patent Infringement Law...................................................................................... 6

A.    Divided Infringement Before Akamai II: The Single-Entity Rule............................................ 8

B.    Divided Infringement After Akamai II: The Inducement-Only Rule...................................... 10

III.     Identification of the Legal Problem: Akamai II Failed to Close the Loophole for Actors Jointly Infringing Interactive Method Claims.......................................................................... 13

A.    Patentees Cannot Enforce Their Interactive Method Claims Under the Single-Entity Rule... 14

B.    The Inducement-Only Rule Creates An Almost Impenetrable Barrier for Patentees Seeking to Enforce Interactive Method Claims.................................................................................................................... 16

C.    Patentees Cannot Cure Divided Infringement Issues by Drafting Unitary Claims................. 18

IV.     Proposed Solution: An Alternative Approach to Determining Liability for the Joint Infringement of Interactive Method Claims.............................................................................................. 21

A.    Step 1: Find a Nexus Between Interactivity and Patentability.............................................. 22

B.    Step 2: Apply the Golden Hour Factors............................................................................... 22

C.    Advantages of the Proposed Approach................................................................................ 24

V.     Conclusion.......................................................................................................................... 25

 

 


 

Introduction

            The patent system exists to incentivize inventors to create and share innovative, useful inventions with the public by giving them exclusive rights to practice their inventions for a limited time.  Such an incentive only exists, however, if inventors can readily and reliably enforce their exclusive rights.  A loophole in patent infringement law that allows others to practice a patented invention overrides this incentive.  It instead discourages inventors from sharing their inventions, and as a result, prevents society from receiving all their benefits.  The doctrine of divided infringement created such a loophole by allowing entities to jointly infringe a patented method without any liability.  By entering into arms-length agreements, infringing entities collectively performed different steps of a patented method and avoided both direct and indirect infringement liability. The Federal Circuit historically struggled to close this loophole because it was unable to establish a standard that both enforced patented methods and shielded innocent infringers from liability.

In the recent Akamai II[1] decision, the Federal Circuit found its second wind and attempted to finally close this loophole by broadening its joint infringement standard to include inducement liability.  Under the court’s newly minted standard, a “single entity” must still perform an underlying act of infringement to be liable for direct infringement, but inducers will not be liable for indirect infringement by inducing others to jointly infringe the patented method. While Akamai II certainly closed the loophole for most patentees, it failed to completely close it for those seeking to enforce interactive method claims.  Methods involving interactive technologies require the cooperation of multiple entities to complete each and every step of the method.  These methods are interactive by their very nature and their claims must be directed to or at least mention more than one actor to properly claim the invention.  Even under the court’s new joint infringement standard, infringing entities can still divide the steps of an interactive method claim without one actor controlling or even inducing the other actors.  Thus, the Akamai II standard prevents patentees from enforcing their interactive method claims.

In order to understand the new joint infringement standard under Akamai II, Part I briefly outlines the law of unified patent infringement, while Part II explains the new standard for determining joint infringement liability, specifically the single-entity and inducement-only rules introduced in Akamai II.  Part III identifies the legal problems this standard poses for enforcing interactive method claims.  Part IV proposes an alternative approach exclusively designed to enforce interactive method claims, discourage poorly drafted claims, and shield innocent infringers from liability.

I.               Unified Patent Infringement Law

The law generally divides patent infringement into two distinct yet interconnected categories: direct and indirect infringement.  Section 271(a) under Title 35 defines direct infringement as follows:

(a) Except as otherwise provided in this title, whoever without authority makes, uses, offers to sell, or sells any patented invention, within the United States or imports in the United States any patented invention during the term of the patent thereof, infringes the patent.[2]

Direct infringement is a strict liability offense – i.e., it does not require that the accused infringer possess knowledge of the patent.  It instead only requires that the accused infringer perform each and every step of a claimed method.[3]  For a party to be liable for direct infringement, that party must directly infringe the patent by performing all the necessary steps of a patented method, either personally or vicariously.[4]   

Indirect infringement, on the other hand, occurs when a party participates in or encourages infringement, but does not directly infringe the patent.[5]  There are two types of statutorily defined indirect infringement: inducement and contributory infringement.[6]  Inducement refers to the active and knowing aiding and abetting of another’s direct infringement.[7]  Section 271(b) states that “[w]hoever actively induces infringement of a patent shall be liable as an infringer.”[8]  Liability under Section 271(b) exists if an inducer specifically intends to cause, urge, encourage, or aid another to infringe a patent.[9]  In other words, the accused infringer must have knowledge of the patent’s existence, know that the induced acts constitute patent infringement, and take affirmative steps to lead, prevail, or influence another to infringe the patent.[10]

Induced infringement is both narrower and broader than direct infringement for several reasons.[11]  First, induced infringement, unlike direct infringement, has an intent requirement; the inducer must 1) know that the induced acts constitute patent infringement,[12] and 2) possess the specific intent to encourage another’s infringement.[13]  Second, induced infringement only attaches liability if the inducement results in actual infringement.[14] In other words, even though a single entity does not have to perform all the steps of a claimed process, every step must still be performed by at least one entity.  Because patent law does not recognize attempted patent infringement, actual infringement is required to establish liability under inducement infringement.[15]  Lastly, a party directly infringes a patent by either performing all the steps of the claim method directly, or engaging in an agency-type relationship with the actors performing the steps.[16]  By contrast, inducement does not require any special relationship between the parties beyond performing the inducing acts themselves.[17]

An example may help illustrate the nuances between direct and indirect infringement.  Suppose, for instance, that Inventor patents a method for manufacturing steel more efficiently.  Shortly thereafter, Company A produces a chemical compound that can be used in the claimed method.  To increase revenue, Company A decides to sell the chemical with a manual that instructs customers to use its product in the claimed method, knowing full well that Inventor holds the aforementioned patent.  Customer innocently purchases the product and personally performs each and every step of the claimed method.  Under these facts, only Customer would be liable for directly infringing the patent because she was the only one to practice each and every claimed step.  Company A, however, would still be liable for inducement because 1) it knew the patent existed, 2) it specifically intended for Customer to infringe the patent; and 3) Customer performed each and every step of the patented method.
            Contributory infringement, on the other hand, occurs when “a party offers to sell or imports a component that does not have a non-infringing use.”[18]  Section 271(c) states that:

(c) Whoever offers to sell or sells within the United States or imports into the United States a component of a patented machine, manufacture, combination or composition, or a material or apparatus for use in practicing a patented process, constituting a material part of the invention, knowing the same to be especially made or especially adapted for use in an infringement of such patent, and not a staple article or commodity of commerce suitable for substantial noninfringing uses, shall be liable as a contributory infringer.[19]

 

Liability exists under Section 271(c) if a seller 1) knows that the component was “especially made or especially adapted for use in an infringement of such patent,” and 2) the component does not have any substantial noninfringing uses.[20]  Like inducement, contributory infringement also requires the accused infringer know the patent exists.[21]  

II.             Divided Patent Infringement Law

Divided infringement occurs when the conduct of multiple parties infringes a patent.   This poses a particular problem for patentees seeking to enforce patented methods.  Unlike claimed methods, whenever a claimed product is made, used, or sold, there is always a direct infringer: the entity who completes the product.[22]  By contrast, patented methods may only be infringed by combining the steps each party takes to complete the claim.[23]  In these situations, direct infringement does not provide patentees with a remedy because, as explained above, “when one person does not perform each and every step of the claimed process, no person directly infringes the claim.”[24]

Unsurprisingly, courts have historically struggled to adopt a joint infringement standard that strikes a reasonable balance between protecting patented methods and shielding innocent actors from strict liability.[25]  Establishing the requisite relationship between the entities that collectively perform a claimed method before attaching liability has been central to this issue.  On the one hand, raising the joint infringement standard effectively renders some valid method claims unenforceable, which discourages companies from seeking patent protection and as a result, stifles innovation.[26]  This outcome directly conflicts with the constitutional mandate that Congress “promote the Progress of Science and useful Arts.”[27]  On the other hand, broadening rules for applying joint infringement captures customers of an alleged infringing service that innocently perform a step in the method.[28]  A standard that finds liability without regard to any circumstances surrounding the infringement ensnares innocent actors, and once caught, places an undue burden on the courts to cut them free.[29] 

Prior to the Federal Circuit’s decision in Akamai II, courts applied a more stringent joint infringement standard. In order for a court to find entities liable for either direct or indirect infringement, at least one actor had to exert enough “direction or control” over for all the infringing actors to be viewed as a “single entity.”  In other words, direct infringement was a predicate to finding joint infringement liability, even under inducement.  This became known as the “single-entity” rule.  Unfortunately, this rule created a loophole that many defendants exploited to avoid liability.  Infringers, knowing of patented processes, essentially left the last step to be performed by another actor.  By engaging in what appeared to be an arms-length transaction, infringing entities were able to claim that they were not controlling or directing the actions of the other entities.  Because no single person was performing each and every step of a claimed invention, many infringers avoided liability. 

Fortunately, the Akamai II court recognized this loophole and expanded joint infringement liability to include inducement.  This newly minted “inducement-only” rule no longer requires the inducer to “direct or control” the induced actors.  Courts instead find inducement liability even where a single actor does not perform all the steps of a claimed method.  Under this new joint infringement standard, the single-entity rule still governs liability under direct infringement, but the inducement-only rule now determines liability under indirect infringement.

A.             Divided Infringement Before Akamai II: The Single-Entity Rule

            In the 2007 decision BMC v. Paymentech,[30] the Federal Circuit sought to clarify “the proper standard for whether a method claim is directly infringed by the combined actions of multiple parties.”[31]  There, plaintiff BMC asserted two patents that disclosed a method requiring the combined action of several distinct participants against defendant Paymentech, claiming both direct and indirect infringement.  Paymentech moved for summary judgment, arguing that it did not perform all of the steps of the claimed method either alone or in concert with its customers.[32]  BMC responded by arguing that under the Federal Circuit’s pervious decisions, a plaintiff must only show “participation and combined action” on the part of the alleged infringer to prove joint infringement.[33] 

On appeal, the Federal Circuit rejected BMC’s joint infringement argument.[34] The court instead stated that “when a defendant . .  . does not directly infringe a patent, the normal course under the law is for the court to apply the standards for liability under indirect infringement.”[35]  The court further explained that a finding of indirect infringement requires that a party amongst the accused actors commit the entire act of direct infringement.[36]  Relying on past decisions, the court refused to find joint liability where one party did not control or direct each step of the claimed method.[37]  To succeed in an action for inducement or contributory infringement, BMC further held that a plaintiff must first show that direct infringement under Section 271(a) occurred.[38]  The Federal Circuit therefore reaffirmed the long-standing principle that “inducement requires a predicate finding of direct infringement.”[39]  In cases where a mastermind instructed a third party to carry out one or more steps of a claim, the court stated that the controlling party would not be liable for indirect infringement either.[40]

This single-entity rule created a loophole that many infringing entities exploited to avoid liability.  By entering into arms-length agreements, they could practice the claimed method without violating the single-entity rule.  Revisiting an earlier example may help explain this situation.  Suppose, for instance, that Company A only performed the first step of the claimed method by producing its chemical compound.[41] To avoid liability, Company A sold this compound to Customer and included detailed instructions on how to perform the remaining steps of the patented method.  Under the single-entity rule, neither Company A nor Customer are liable because neither one 1) performed each and every step of the claimed method, or 2) exerted the necessary direction or control over the other.

The BMC opinion acknowledged this loophole, but expressly decided that such situations did not warrant expanding the law of direct infringement.[42]  Instead of judicially closing this loophole, the court suggested that this concern could be “offset by proper claim drafting.”[43]  Specifically, the Federal Circuit stated that patentees could avoid joint infringement issues by drafting claims directed to a single infringer.[44]  Through this reasoning, the BMC court concluded that Paymentech did not infringe the patented method.[45] 

B.             Divided Infringement After Akamai II: The Inducement-Only Rule

            After BMC, many commentators became concerned that the single-entity rule would effectively render many patented methods unenforceable.[46]  To close the loophole created by the single-entity rule, the Akamai II opinion created a new judicial doctrine under Section 271(b) to find liability when multiple entities collectively performed a method claim.[47]  While many believed that the Federal Circuit would address the continued use of the single-entity rule, the court declined to do so.[48]  Instead, the decision focused on whether an entity may be liable for induced infringement if it has induced other actors to collectively perform the steps without a single direct infringer (e.g., the Akamai fact pattern) or has only partially performed the method itself and induced another to commit the remaining steps (e.g., the McKesson fact pattern).

            The Akamai II en banc opinion rendered a combined decision for Akamai v. Limelight[49] and McKesson v. Epic Systems[50].  The patents at issue in Akamai recited methods for delivery of web content by placing some of a provider’s content on servers and modifying the provider’s website to instruct users to retrieve the content from the server.[51]  Akamai alleged that Limelight was liable for both direct and induced infringement through the maintenance of a network of servers that allow delivery of web content placed on its servers.[52]  Limelight instructed its users on how to modify their websites to access the server based information, but did not modify the sites directly.[53]  Along the same lines, the asserted patent in McKesson covered a method of electronic communication between a healthcare provider and the patient.[54]  McKesson claimed that Epic Systems induced infringement of the patent by licensing its “My Chart” software to healthcare providers, which enabled them to communicate electronically with patients.[55]  Epic Systems, however, did not complete any of the steps of the patent method.[56]  Instead, the patients that initiated the communications and the healthcare providers that performed the remaining steps completed those steps.[57]

            In the en banc opinion, the Federal Circuit preserved the “single-entity rule” for direct infringement but expressly overruled its application to inducement.  The court instead held that for a party to be found liable for inducement, all the steps of the claimed method must still be performed, but every step does not have to be performed by a single entity.[58]  Under this new, so-called “inducement-only” rule, an entity can be liable for indirect infringement when it induces one or more entities to perform all the steps in a method claim, regardless of whether the inducer performs some of the steps itself or whether there is control or direction between the inducer and the induced entities or between the induced entities themselves.[59]  Relying on legislative history, the majority reasoned that inducement applies to joint infringement cases because one who aids or abets infringement is “likewise an infringer.”[60]  Applying its newly minted inducement-only rule, the Akamai II court reversed and remanded both cases down for further proceedings on the theory of induced infringement.[61]

            The Federal Circuit’s reasoning rests on sound policy.  Requiring a single entity to commit the underlying act of direct infringement as a prerequisite to liability for inducement served purpose as it “invite[d] evasion of the principles of patent infringement.”[62]  There is no reason to hold a party that induces multiple actors to collectively commit an act of direct infringement differently from one who induces a single actor.  Either way, the patentee suffers the same impact.  An actor who directly infringes a patented method might be considered more culpable than one who does not perform any infringing acts, but as long as the infringement occurs, the harm to the patentee is identical.  It would therefore be nonsensical to immunize an inducer who completes some of the steps of the claimed method, but induces others to complete the remaining steps.  Moreover, the court’s new inducement-only rule also protects innocent infringers.  Liability for induced infringement requires the accused party to “knowingly induce infringement and possess specific intent to encourage another’s infringement.”  This specific intent requirement protects those who may be innocently unaware of the existence of a patent or unaware that there are other parties committing other steps of a patented method.  Accordingly, the Akamai II court seems to have struck a balance between enforcing patented methods and shielding innocent infringers from liability.   

In essence, the Akamai II decision places direct and induced infringement into distinct categories of joint infringement.  Direct infringement still requires that the infringing entities jointly perform all of the steps of the claimed method through a vicarious relationship – i.e., the single-entity rule.  Induced infringement, however, will now arise when multiple actors, without a vicarious relationship, combine to commit direct infringement, meaning the performance of all of the steps of a method claim, as a separate liability.  As the Federal Circuit succinctly stated, “Congress did not intend to create a regime in which parties could knowingly sidestep infringement liability simply by arranging to divide the steps of a method claim between them.”[63]  The inducement-only rule therefore tightens the loophole by holding liable entities that knowingly induce infringement of a method claim.

III.           Identification of the Legal Problem: Akamai II Failed to Close the Loophole for Actors Jointly Infringing Interactive Method Claims

While Akamai II closed the divided infringement loophole for most patented methods, it failed to do so for interactive method claims.  Methods involving interactive technologies require the cooperation of multiple entities to complete each and every step of the method.[64]  A person may, for example, invent a new and useful process that requires steps (a) and (b) of a claimed process to be performed by one person and step (c) to be performed by another person.  These interactive method claims are surprisingly common, particularly in the field of computer networking, where a patented method may involve some steps performed on the client side and other performed on the server side.[65]  Since these methods distribute the claimed steps amongst different actors, they create significant problems for patentees seeking enforcement.

The Akamai II court’s joint infringement standard is, at least in theory, a clever solution for solving the BMC problem.  By simultaneously preserving the single-entity rule and creating the inducement-only rule, the court found a reasonable balance between protecting most method claims and shielding innocent infringers from liability.  In terms of practicality, however, this new standard obstructs the attempts of patentees seeking to enforce interactive method claims.  As demonstrated below, the single-entity and inducement-only rules fail to completely close the loophole for patentees of interactive methods, enabling actors to reap the benefits of patents that cannot be easily enforced.  Until the court adopts a new approach, patentees will continue facing barriers to enforcing their interactive method claims.

A.             Patentees Cannot Enforce Their Interactive Method Claims Under the Single-Entity Rule. 

The Akamai II court’s decision to preserve the single-entity rule renders most interactive method claims unenforceable under the doctrine of direct infringement.  Pursuant to the single-entity rule, “multiple entities could only be held liable for infringement when an underlying act of direct infringement under Section 271(a) occurred.”[66]  Entities meeting the same relationship requirement for direct infringement must perform this underlying act.[67]  In other words, one actor must “control or direct” the others to perform each and every step of a claimed method before a court can attach direct infringement liability.[68]  This rule presents a unique challenge for patentees seeking to enforce interactive method claims.  These methods are interactive by their very nature and their claims must be directed to or at least mention more than one actor to properly claim the invention.  Infringing entities can therefore more easily divide the steps of an interactive method claim without one actor controls the others.   As a result, many patentees can no longer enforce their interactive method claims through the doctrine of direct infringement.

This outcome may lead to severe consequences for interactive technology providers, especially those involved in the computer networking, software, or biotechnology industries.  For these firms, a competitor can easily replicate an innovative method containing steps performed by more than one entity without infringing the corresponding patent.  Amazon.com’s patent covering its 1-Click purchasing method,[69] for example, falls victim to this gamesmanship.  This patent covers an interactive method claims containing, inter alia, two steps for a client system and two other steps for a server system.[70]  Since a single actor cannot complete the method, a competitor can create a server system with all the same functions as the 1-Click purchasing method without directly infringing Amazon’s method claim because the end user must complete the remaining steps of the method.  The single-entity rule therefore removes direct infringement liability as a potential remedy for such patentees.  As the BMC court acknowledged, “the standard requiring ‘control or direction’ for a finding of joint infringement may in some circumstances allow parties to enter into arms-length negotiations to avoid infringement.”[71] This statement rings particularly true where multiple entities jointly infringe an interactive method.

B.             The Inducement-Only Rule Creates An Almost Impenetrable Barrier for Patentees Seeking to Enforce Interactive Method Claims.

Although Akamai II broadened joint infringement liability by creating the inducement-only rule, the additional requirements for inducement, specifically the intent requirement, may pose an additional barrier to enforcement.  Inducement requires not only an act of direct infringement, but also that the inducer possess the affirmative intent to cause the direct infringement.[72]  As a result, an infringing entity will only be held liable for inducement if the patentee proves that the inducer: 1) knew the patent existed, 2) knowingly induced the completion of all of the steps of the claimed method, and 3) possessed the specific intent to encourage the completion of all the steps of the claimed method.[73]  This scienter requirement presents a significant challenge for enforcing interactive methods because an inducer may not intend for the induced actor to perform the specific limitation of the claim.   Accordingly, this intent requirement places patentees of interactive methods in an unfortunate Catch-22 situation: while drafting a narrower claim limitation likely makes the claim more patentable,[74] it also makes it more difficult to prove the intent prong of inducement.

Comparing two sample claims of an interactive medical process method from the recent Supreme Court case Mayo Collaborative Servs. v. Prometheus Labs, Inc.[75] may help explain this precarious situation.  The claims at issue in Mayo cover a medical process that informs a doctor using thiopurine drugs to treat patients with sleep apnea about the efficacy of a given dosage of the drug.[76]  Asserted independent Claim 1 reads:

A method for assessing the effectiveness of a treatment for sleep apnea, wherein said method comprises: administering the CPA treatment for sleep apnea to a mammal before sleep, determining whether or not the level of expression of a nucleic acid in a mammal being treated for sleep apnea changes during sleep, wherein a change in said level during sleep indicates that the treatment is ineffective, and modifying the treatment based on the level of change identified.

 

Assuming that the diagnostic test provider is the alleged infringer, intent could likely be proven for the administering step because the test provider must intend for the doctor to administer the treatment in order to provide a proper sample for the diagnostic test.  The test provider would likely complete the determining step itself with the provided sample.  Therefore, the main issue in this claim involves proving intent for the final step, modifying the treatment.  Since this claim describes the final step in fairly simple terms, the patentee can easily prove that the test provider intended for the doctor to make some change in his practice based on the results provided.  The patentee could argue, for example, that the doctor would not send the sample for testing unless he had been told what to do with the results.  Thus, while a patentee can enforce broad claims like Claim 1 under the inducement only rule, not all method claims will be patentable once drafted in this manner.

            By contrast, Claim 3 depends from Claim 1 and describes more specific administering and treatment steps.[77]  While this specificity increases the likelihood of patentability, it nevertheless becomes more difficult to prove the test provider possessed the requisite intent for all of the limitations in the claim.  Asserted dependent Claim 3 reads:              

Administering CPAP treatment comprising two three hour intervals of at least 6 cm H20 to a mammal before sleep for sleep apnea, determining whether the level of expression of a nucleic acid in a mammal being treated for sleep apnea increases, wherein an increase of at least .2% in nucleic acid level during sleep indicates that CPAP is ineffective, and wherein determining whether the level of expression of a nucleic acid in a mammal being treated for sleep apnea increases during sleep comprises performing assay 1 and then performing assay 2, and modifying CPAP treatment administered to a mammal being treated for sleep apnea who exhibits an increase of at least .2% in nucleic acid level, the modified CPAP treatment comprising three 2.5 hour intervals of at least 10 cm H20.[78]
 

Compared to Claim 1, this claim contains much more detailed information about administering the CPA treatment and modifying the treatment after the lab provides results.  Since the alleged inducer must possess an affirmative intent to cause the direct infringement, he must intend for the doctor to complete the treatment step with every claim limitation, no matter how specific.  In these circumstances, it would be significantly harder for the patentee to prove that the test provider possessed the specific intent for the doctor to modify the treatment in such a particular manner.  Thus, while drafting a narrower claim like Claim 3 makes the method more patentable, it also makes it more difficult to enforce under the inducement-only rule.

C.            Patentees Cannot Cure Divided Infringement Issues by Drafting Unitary Claims

The Federal Circuit and several commentators have suggested that patentees of interactive method claims can avoid divided infringement issues by drafting broader, unitary claims.[79]  Unitary claims are claims in which all the steps of the method are illustrated from the perspective of one entity.[80]  In many instances, a patentee can describe a method in unitary form by “structur[ing] a claim to capture infringement by a single party.”[81]  Compare, for example, two different claims[82] directed to a method commonly employed in electronic commerce to secure communication between browsers and websites. Claim 1 discloses:

1. A method for negotiating a secure communication session, comprising: (a) transmitting a request to a server; (b) in response to the request, supplying the server a server certificate, the server certificate including the server’s public key; (c) generating at the client a unique key and communicating the unique client key to the server using the server’s public key; and (d) thereafter communicating information using a crypto-algorithm that employs a derivative of the unique client key and the server’s public key.[83]

 

Claim 2 covers the same invention, but reads:

2. A method for negotiating a secure communication session, comprising (a) receiving a request from a client; (b) in response to the request, supplying a server certificate, the server certificate including a public key; (c) receiving from the client a unique client key communicated using the server’s public key; and (d) thereafter communicating information using a crypto-algorithm that employs a derivative of the unique client key and the server’s public key.[84]

 

Despite the fact that both claims seek to cover the same invention, only the second claim is unitary because the first requires both the client and the server to perform the claimed steps.[85]  As a result, the divided infringement issues arise for the first claim but not the second.[86]

Truly interactive methods, unlike most methods, cannot be drafted in unitary form because they cannot be properly claimed unless they are directed to multiple actors.  Unitary claim drafting only works where the method claim involves a “hub” – i.e., an actor for at least one of the claimed steps that interacts with other actors for all the remaining steps, so that the remaining steps can be written from the perspective of the hub.[87]  Since most interactive methods do not involve a hub, they cannot be drafted in unitary form. 

Medical diagnostics patents like those previously discussed in Mayo, for instance, often involve a method that contains steps such as data gathering, analysis, correlation, and treatment with a drug in the end.[88]  These steps are typically carried out by different parties, such as a diagnostic laboratory that performs the analysis, a pharmaceutical company that provides the drug, and a physician that collects the data and provides treatment.[89]  It is therefore impossible to draft so-called better claims directed to a single infringer for these interactive diagnostic patents.  As a result, a rigid rule that relies on inducement makes many patents directed to interactive technologies unenforceable.  As Judge Newman warned in her dissenting opinion, “[a] patent that cannot be enforced on any theory of infringement is not a statutory patent right.”[90]  Rendering valid interactive method claims unenforceable will inevitably “eliminate the patent incentive from such interactive procedures,” leaving these types of methods vulnerable to infringement and deter innovation.[91] 

Neither the single-entity rule nor the inducement-only rule properly equips the judiciary to handle the joint infringement fact patterns of the Information Age.  Patentees seeking to enforce interactive claims are therefore at a server disadvantage.[92]  As one amicus curie explained, interactive claim language may be the best way to capture the interactive nature of the invention and the patentee should not be forced to draft unitary claims so that those claims can be enforced.[93]  Accordingly, a more flexible approach for determining joint infringement liability is desperately needed to make interactive patents easier to enforce.

IV.           Proposed Solution: An Alternative Approach to Determining Liability for the Joint Infringement of Interactive Method Claims

This section proposes an alternative standard designed to close the Akamai II loophole for patentees of interactive method claims.  The proposed approach undertakes a two-step procedure for determining joint-infringement liability adapted from the Golden Hour[94] decision.  First, the patentee would provide a rationale for an asserted claim’s interactivity – i.e., prove a nexus between interactivity and patentability.  Second, the court would selectively apply a certain test for joint infringement based on the result of the interactive inquiry.  By providing patentees with a flexible alternative to the newly minted single-entity and inducement-only rules, the proposed standard accomplishes three objectives: 1) it does not force patentees to drafted unitary claims, 2) it protects interactive technologies to encourage further innovation, and 3) it shields innocent actors from liability.  Incorporating this test into a judicial framework will afford courts more flexibility in analyzing joint infringement cases.  In sum, an alternative, more flexible standard for enforcement will encourage patentees to continue to file applications for methods covering interactive technologies.

A.             Step 1: Find a Nexus Between Interactivity and Patentability

The proposed test first requires the patentee to prove that a nexus exists between the “interactivity” of a claimed method – i.e., the reason why the claims require more than one actor – and the patentability of the claim.  In this context, “nexus” would mean that the claimed interactivity is “the reason for,” “reasonably related,” or “substantially related to” the patentability of the claim.  In order to make the evidentiary showing of a nexus, the patentee would primarily rely on intrinsic evidence, taking into account statements in the specification, the preamble, and the prosecution history.  By providing evidence that an asserted claim’s interactivity was a factor in its patentability, the patentee can show that the claim’s interactivity made it novel and nonobvious.[95] 

B.             Step 2: Apply the Golden Hour Factors

The second step of the proposed test requires the court to evaluate the patentee’s evidentiary showing of a nexus.  If a nexus does exist, then the court would not apply the inducement-only test.  Instead, the court would apply the three factors originally outlined in the Golden Hour opinion to determine joint infringement liability.  First, the court would determine whether a “joint venture” exists between the actors performing the claimed method.[96]  This determination would depend on the facts of the case.[97]  Evidence such as joint sales activities, joint promotional activities, and another other joint activity would weigh in favor of a finding of joint infringement.[98] 

Second, the court would determine whether all the steps of the claimed method were performed by the “concerted action” of all the relevant actors.  In Golden Hour, for example, the jury found that the two accused entities, emsCharts and Softech, enabled their respective programs to work together such that their combined system, which they sold as a package, met all the limitations of the asserted claims.[99]  Such evidence would rise to the level of “concerted action.”  Lastly, the court would examine any “benefit” an accused entity received from performing the claimed method.[100]  Evidence of an entity obtaining a commercial, competitive, or financial benefit would weigh in favor of finding joint infringement.[101] In Golden Hour, for instance, Softech and emsCharts both benefited economically and competitively by selling their combined system.[102]  This factor essentially prevents parties not engaged in a “joint venture” from undermining the patent system by misappropriating the benefit of an invention.

Alternatively, if the court finds that a nexus does not exist, it should instead focus on the nature of the relationship between the relevant actors and apply either the single-entity or inducement-only rule.  The lack of a nexus may suggest that 1) interactivity is secondary to other limitations in the claim, or 2) the claim could have been drafted in unitary form.  Either way, a patentee’s ability to enforce an interactive method will no longer be frustrated by applying the inducement-only rule because truly interactive claims will easily exhibit the requisite nexus.  

C.            Advantages of the Proposed Approach

The proposed approach has several advantages.  First and foremost, it would finally close the Akamai II loophole by lowering the hurdle of enforcement for truly innovative interactive patents.  By making it easier for patentees of interactive methods to enforce their patents, the proposed approach would foster further innovation of interactive technologies.  Second, it would support the Federal Circuit’s goal for patentees to draft “proper” claims by discouraging poorly drafted ones.  The proposed test would not apply to the type of claims that courts typically deem poorly drafted unless a nexus between the interactivity of a claim and its patentability existed.  To the contrary, the proposed standard may incentivize patentees to proffer more robust specifications and more directed arguments during prosecution. 

Lastly, the proposed approach would protect innocent actors from liability.  As previously discussed, advocates for a higher standard such as the inducement-only rule are primarily concerned that broadening the joint infringement rules would ensnare innocent infringers.[103]  Most innocent actors are customers of an alleged infringing service and typically perform only one step in the alleged infringed claim.[104]  Under the proposed standard, however, truly innocent parties will not be subject to liability because courts will have the leeway under this flexible standard to deny relief altogether.  For example, a court could refuse to extend liability to innocent infringers because they neither engage in “joint ventures” nor infringe through “concerted action.”

Moreover, innocent infringers have rarely been involved in joint infringement cases.[105] If the innocently induced are individuals, a patentee will not bring suit either because they are not cost-effective litigation targets or they are constituents of the patentee’s market – e.g., Internet users or doctors.  If the innocently induced are more sophisticated parties, such as corporations, they can incorporate an indemnification clause in their business contract placing liability solely on the accused entity.  In sum, the proposed approach would grant patentees a better opportunity to enforce their claims, support the Federal Circuit’s position on drafting “proper” claims, and shield innocent actors from liability.

V.             Conclusion

The Federal Circuit’s new joint infringement standard does not properly equip the judiciary to resolve the divided infringement issues surrounding interactive method claims.  It instead focuses on the inducement inquiry and completely ignores other practical factors that should be considered in determining joint infringement liability.  Absent evidence of intent, the inducement-only rule does nothing to close the judicial loophole.  Accordingly, courts should adopt an alternative, more flexible standard for determining divided infringement liability for interactive method claims.

The proposed two-step standard for determining joint infringement applies to claims that possess a nexus between their interactivity and patentability.  Rather than subject these claims to the Akamai II court’s rigid joint infringement standard, the proposed approach determines liability by weighing the Golden Hour factors.  In doing so, it prevents misappropriation of interactive technologies, encourages properly drafted claims, and shields innocent actors from strict liability.



[1] Akamai Techs., Inc. v. Limelight Networks, Inc., 692 F.3rd 1306 (Fed. Cir. 2012) (en banc).

[2] 35 U.S.C. § 271(a) (2013).

[3] BMC Res., Inc. v. Paymentech, L.P., 498 F.3d 1373, 1378 (Fed. Cir. 2007), overruled in part by Akamai, 692 F.3rd 1306.  

[4] Akamai, 692 F.3d at 1305.

[5] DynaCore Holdings Corp. v. U.S. Philips Corp., 363 F.3d 1263, 1272 (Fed. Cir. 2004).

[6] See Mark A. Lemley, Inducing Patent Infringement, 39 U.C. Davis L. Rev. 225, 227 (2005).

[7] Walter Techs. Corp. v. Calco, Ltd., 850 F.2d 660, 668 (Fed. Cir. 1988); See also Lemley, supra note 8, at 229.

[8] 35 U.S.C. § 271(b).

[9] Chisum, supra note 4,  § 17.01, at 17-2.

[10] Global-Tech Appliances, Inc. v. SEB S.A., 131 S. Ct. 2060, 2065 (2011) (interpreting Section 271(b) despite absence of an explicit “intent” requirement but rather an “induce” requirement).

[11] Akamai, 692 F.3d at 1308.

[12] Global-Tech Appliances, 131 S. Ct. at 2068.

[13] DSU Med. Corp. v. JMS Co., 471 F.3d 1293, 1306 (Fed. Cir. 2006) (en banc).

[14] Akamai, 692 F.3d at 1308.

[15] See in re Seagate Tech., LLC, F.3d 1360, 1368 (Fed. Cir. 2007).

[16] See Cross Med. Prod., Inc. v. Medtronic Sofamor Danek, Inc., 424 F.3d 1293, 1311 (Fed. Cir. 2005).

[17] See Akamai, 692 F.3d at 1308 (“[I]nducement does not require that the induced party be an agent of the inducer or be acting under the inducer’s direct or control . . . It is enough that the inducer ‘cause[s], urge[s], encourage[s], or aid[s]’ the infringing conduct and that the induced conduct is carried out.”).

[18] Lemley, supra note 6, at 230; 35 U.S.C. § 271(c).

[19] 35 U.S.C. § 271(b)-(c).

[20] Chisum, supra note 9,  § 17.01, at 17-2.

[21] Id.; See also Global-Tech Appliances, 131 S. Ct. at 2068

[22] Akamai, 692 F.3d at 1316.

[23] See Mark A. Lemley et al., Divided Infringement Claims, 6 Sedona Conf. J. 117 (2005).

[24] Id. at 58.

[25] See Id. By contrast, most courts did not question applying a higher joint infringement standard when multiple entities jointly infringed a non-interactive patent because they did not want to “bailout” patentees with claims that could have been drafted in unitary form to avoid divided infringememt. Id.

[26] See McKesson Techs., Inc. v. Epic Sys. Corp., 98 U.S.P.Q. 2d (BNA) 1281, 1291 (Fed. Cir. 2011) (Newman, J., dissenting), rev’d sub nom. Akamai, 692 F.3d 1301.

[27] U.S. Const. art. I, § 8, cl. 8.

[28] See Lemley, supra note 6, at 261 (discussing cases where alleged infringers were customers of the defendant).

[29] See Akamai, 692 F.3d at 1350 (Linn, J., dissenting) (describing Judge Newman’s approach as permitting “joint actor infringement liability whenever independent parties collectively infringe a patent”).

[30] BMC Resources, Inc. v. Paymentech, L.P., 498 F.3d 1373, 1376 (2007).

[31] Muniauction, Inc. v. Thomson Corp., 532 F.3d 1318, 1329 (Fed. Cir. 2008) (reviewing the analysis in BMC).

[32] BMC, 498 F.3d at 1377.

[33] Id. at 1380 (quoting On Demand Machine Corp. v. Ingram Industries, Inc., 442 F.3d 1331 (Fed. Cir. 2006)).

[34] See id. at 1380 (concluding that On Demand did not change the standard for joint infringement).

[35] Id. at 1379.

[36] Id.

[37] See id. at 1380 (citing district court decisions that did not find infringement in arms-length agreements and divided performance situations).

[38] Id. at 1380.

[39] See BMC, 498 F.3d at 1380 (citing Dynacore Holdings Corp., 363 F.3d at 1272).

[40] See id. at 1381 (arguing that contracting out steps to specifically avoid infringement would still result in liability for the mastermind).

[41] See supra pg. 5.

[42] BMC, 498 F.3d at 1381.

[43] Id.

[44] Id. (stating that “[a] patentee can usually structure a claim to capture infringement by a single party”).

[45] Id.  The Court noted that (1) there was no evidence that Paymentech provided its customers or financial institutions with instructions or directions for the use of the data it  provided, and (2) the record contained no evidence of a contractual relationship between Paymentech and the financial institutions.  Id. at 1382.

[46] See McKesson Techs., 98 U.S.P.Q. 2d (BNA) at 1291  (Newman, J., dissenting).

[47] See generally Akamai, 692 F.3d 1301.

[48] Id. at 1306.

[49] Akamai Techs., Inc. v. Limelight Networks, Inc., 629 F.3d 1311, 1319 (Fed. Cir. 2010).

[50] McKesson Techs. Inc. v. Epic Sys. Corp., No. 2010-1291, 2011 U.S. App. LEXIS 7531, at 1-2 (Fed. Cir. Apr. 12, 2011).

[51] Akamai, 692 F.3d at 1306.

[52] Id.

[53] Id.

[54] Id.

[55] Id.

[56] Id.

[57] Akamai, 692 F.3d at 1306.

[58] Id.

[59] Id.

[60] Id. at 1309 (quoting H.R. Rep. No. 82-1923, at 9 (1952)).

[61] Id. at 1319.

[62] Id. at 1317-18.

[63] Akamai, 692 F.3d at 1318.

[64] Lemley, supra note 23, at 125.

[65] Id. at 117.

[66] BMC, 498 F.3d at 1379-81.

[67] Id.

[68] See id.

[69] See U.S. Patent No. 5,960,411 (filed Sept. 12, 1997).

[70] Id. col. 10.  The two steps for the client systems are “displaying information identifying the item” and “in response to only a single action being performed, sending a request to order the item along with an identifier of a purchaser of the item to a server system.”  The two steps for the server are “receiving the request” and “retrieving additional information previously stored for the purchaser identified by the identifier in the received request.”  Id.

[71] Id. at 1381.

[72] DSU Med. Corp., 471 F.3d at 1306.

[73] See Vita-Mix Corp. v. Basic Holding, Inc., 581 F.3d 1317, 1328 (Fed. Cir. 2009).

[74] Applicants draft narrower claims as part of target claiming strategies to retain some measure of protection against infringement even if the prior art comes to the fore that would invalidate the broadest claim.  Chisum, supra note 9,  § 17.01, at 17-23.

[75] Mayo Collaborative Servs. v. Prometheus Labs., Inc., 132 S. Ct. 1289, 1294 (2012).

[76] Id. at 1295.

[77] Mayo, 132 S. Ct. at 1295.

[78] Id. at 1298. 

[79] See BMC, 498 F.3d at 1381 (stating that “[a] patentee can usually structure a claim to capture infringement by a single party”), overruled by Akamai, 692 F.3d 1301; See also Lemley et al., supra note 24, at 272 (“Most inventions that involve cooperation of multiple parties can be covered using claims in unitary form simply by focusing on one entity and whether it supplies or receives any given element.”).

[80] See Lemley et al., supra note 23, at 272 (providing a comparison between non-unitary and unitary claims for a method).  

[81] BMC, 498 F.3d at 1381. 

[82] See Lemley et al., supra note 23, at 272.  The claimed subject mater is an abstraction and simplification of secure sockers layer (SSL) techniques developed by Netscape Communications to enable secure, authenticated communications across the Internet using public key encryption.  SSL support is included as part of both the Microsoft and Netscape browsers and most Web server products.

[83] Id.

[84] Id.

[85] Id.

[86] Id.

[87] See Lemley et al., supra note 23, at 272.

[88] See supra pgs. 17-18.

[89] See, e.g., U.S. Patent No. 8,287,869 (filed Jan. 7, 2009) (claiming a method that comprises a step of measuring a patient’s level of an intercellular signaling molecule, which is typically performed by a physician and a diagnostics laboratory, and a step of treating the patient with an antibody, which is usually provided by a pharmaceutical company and administered by the physician).

[90] McKesson, 98 U.S.P.Q. 2d at 1291 (Newman, J., dissenting), rev’d sub nom. Akamai, 692 F.3d 1301.

[91] Id. at 1286.

[92] Compare Muniauction, 532 F.3d at 1380 (finding no liability for direct infringement and no legal theory to support joint infringement where no evidence of direction or control was presented), with BMC, 498 F.3d at 1381-82 (stressing the importance of the direction or control test in determining whether infringement has occurred).

[93] See generally Corrected Brief of Amici Curiae Aristocrat Technologies Australia Pty Lt. & Aristocrat Technologies, Inc. in Support of Appellants Akamai Technologies, Inc. & the Massachusetts Institute of Technology & Supporting Reversal, Akamai, 692 F.3d 1301 (Nos. 2009-1372, -1380, -1417), 2011 WL 3101891.

[94] Golden Hour Data Sys. v. emsCharts, Inc., 614 F.3d 1367, 1369 (Fed. Cir. 2010).

[95] See 35 U.S.C. § 103(c) (2013) (setting forth the conditions of patentability).

[96] See Golden Hour, 614 F.3d at 1371 (discussing defendants’ collaboration to combine their systems into one unit).

[97] See id. at 1370. 

[98] See id. at 1370-71.

[99] Id. at 1383 (Newman, J., dissenting).

[100] See id.  Courts have found parties liable for contributory infringement to prevent them from obtaining the benefit of a patent.  See Mobil Oil Corp. v. W.R. Grace & Co., 367 F. Supp. 207, 253 (D. Conn. 1973) (allowing plaintiff’s claim for contributory infringement where the defendant was able to achieve all benefits of its patent by making each of its customers its agents in completing the infringing step).

[101] See Golden Hour, 614 F.3d at 1383 (Newman, J., dissenting).

[102] Id.

[103] See Lemley, supra note 6, at 262 (stating that joint infringement unreasonably expands liability).

[104] Id. at 261 (discussing cases where alleged infringers were customers of the defendant).

[105] See id.