Joint patent infringement has become a common defense for retailers and their vendors in patent litigations.  Particularly with internet and software technologies, the retailer or the vendor alone cannot perform all of the steps of a method or do not have all of the hardware required of the system.  Of course, joint infringement has been a dynamic area of the law over the last several years, and recent developments — including that customers may not be agents — have been positive for retailers and their supply chain.  In June, Scott Alter (of Faegre & Benson) gave an excellent presentation about the state and history of joint and divided infringement at the 9th Annual Rocky Mountain IP & Technology Institute in Denver.  Here are some of the highlights:

• BMC v. Paymentech (2007): This case involved method claims directed to separate parties — caller, debit processor & bank. The parties agreed that no single party performed each step. The Federal Circuit said Section 271 required one party to perform every step, unless one defendant played a “mastermind” role. BMC had argued that plaintiff just needed to show participation and combined actions. Court noted that patentee could have drafted the claims to apply to a single entity.
• Muniauction v. Thomson (2008): Relying upon BMC, the Federal Circuit required that a single party perform every step of a method for direct infringement. Again, the parties agreed that no signal party performed all steps. Thomson controlled access by bidders to its system and instructed bidders about how to use the system. The Federal Circuit held that one party must exercise “control and direction” over the other parties to complete all steps. It was not enough to give bidders instructions as Thomson did to the bidders in this case. Therefore, no joint infringement.
• Golden Hour v. emsCharts & Softtech (2009): emsCharts provided billing software and Softtech provided dispatch software. They partnered together and sold their software as a single unit. The district court found no control or direction of one party over the other. But there was no discussion of indirect infringement. Regardless of joint infringement issues, the joint product may have infringed the system claims, but the system claims were not put before the jury. So, the Federal Circuit did not decide the issue.
• Centillion v. Qwest (2011): The technology at issue was a system for collecting and sending service provider (telephone) information to a customer. The claims required a data processing means and a home computer. The district court held that no single party practiced each limitation of any asserted claims and there was insufficient control. The Federal Circuit overturned the district court, holding that to “use” a system a party must put the invention into service by controling the system and benefiting from it. Without customers’ actions (using their computers to interact with the Qwest system), the system would never be put into service. Qwest did not “use” because it did not put the system into service and could not do so without a personal computer. The Court also looked at vicarious liability, but Qwest’s customers were not agents.
• Akamai v. Limelight (2010): The patent was for a web content delivery method, embedded objects (pictures) were stored on a server and tagged so that when a user requested the page, the embedded objects could be found by their tag. Limelight instructed customers to tag. Limelight’s contract also obligated customers to tag. The Federal Circuit held that when a party is contractually obligated to perform a method step, joint infringement exists. The Court also required an agency relationship between the parties or a contractual requirement before joint infringement could be found.  And the Court held that Limelight’s customers, who were the alleged joint infringers, did not perform steps as Limelight’s agents.  That is a key holding for retailers and their supply chain.
• McKesson v. Epic Systems (2011): Epic did not “use,” but licensed its accused MyChart software. Epic was charged with induced infringement, for allegedly inducing its licensee healthcare providers to infringe. The first step of the method claim requires “initiating a communication by [a user/patient].” Parties agreed healthcare providers did not initiate the communication. There was no direct infringement, unless there was joint infringement. The only issue was whether the relationship between Epic’s customers and patients was sufficient to create an agency relationship. The Federal Circuit said no. Patients chose whether to communicate and were not obligated to do so. The doctor-patient relationship did not create either agency or a contractual obligation. The Federal Circuit also held that indirect infringement addresses any concern about joint tortfeasors that are not jointly infringing. Judge Newman’s dissent said that the majority decision made joint infringement of the claim “impossible.” Newman also argued that McKesson conflicted with pre-BMC decisions.
• Akamai and McKesson are both set for rehearing en banc.  The key issue on each appeal is whether if multiple entities perform the steps of a method, under what circumstances can they be liable for indirect infringement. A majority of the judges — all but Newman & Bryson, who see inconsistency in the cases, and O’Malley and Meara, who have not decided the issue — appear to believe that the pre and post-BMC case law is consistent. So, the en banc decisions may not change the current state of the law so much as clarify it.

For those who are looking for some high quality continuing legal education, that is in a beautiful location, you should come to the 10th Annual Rocky Mountain IP & Technology Institute next June. It is one of the two best CLE programs I have ever attended.

This post was adapted to focus upon retail issues from a post at my Chicago IP Litigation blog.