By now you’ve heard about a decision called Akamai v. Limelight. Your patent attorney probably sent you an email, newsletter or some kind of brief synopsis of the legal issues the CAFC ruled on. You may have heard the phrase “joint infringement” or “divided infringement.” Ultimately, what you really want to know is, ‘what can I, as a patent owner learn from this case?’ The most important thing you can learn is how to read the claims of that patent application your attorney just sent you. No, you don’t have to understand every detail and nuance of the claim language, but you definitely need to understand the overall operation of the method, the overall layout of the system, or the overall structure of the device.
The significance of Akamai is that it reminds all of us that work with patents that there is a world of difference between the use or existence of a patented process, system, or product, and whether that use or existence actually infringes the patent. This dichotomy is actually nothing new, despite what your attorneys might be telling you. Like many before it, Akamai lost despite the fact that the patented method was actually being performed. The same thing was true when BMC Resources lost its infringement case against Paymentech, and when Muniauction lost its case against Thompson. But the court didn’t invent this doctrine out of whole cloth. The rule that direct infringement requires the accused product or process to satisfy every single limitation of at least one claims has existed throughout the entirety of modern patent law. The rule in Akamia, Muniauction and BMC is simply an extension of that. In each of these cases, the defendants avoided liability because, quite simply, the processes they were performing didn’t satisfy each limitation. You’re thinking, ‘Didn’t you just say that the patented method was being performed?’ Yes, I did say that. But the patents weren’t infringed because the steps that these defendants were performing failed to include one of the method steps. In each case, the court agreed that some third party not under a contractual obligation was, essentially voluntarily, performing the missing step. This scenario has been described as divided infringement, and also joint infringement.
The last three cases to address this divided infringement phenomenon happen to have dealt specifically with method claims. Unfortunately, this has led to another common fallacy being propagated by patent attorneys. I’ve heard more than one patent attorney mention (to clients, no less) that you only need to worry about a divided infringement when analyzing method claims. This is flatly, and demonstrably, false. Consider a patent on a gaming system that includes elements of a ‘game console, at least one game controller, and a display device for displaying to the user images related to the game.’ Would a typical gaming manufacturer be infringing this claim? Consider the fact that, in most cases, the manufacturer makes consoles, game cartridges, and controllers, but the consumers combine these devices with their own televisions so they can actually see the games. Am I crying wolf? Well, before 2005, Cross Medical Products might have thought so … unfortunately for them, they ultimately lost a patent infringement case against Medtronic over a patent involving a surgical implant. The patent claim at issue was an apparatus claim containing a specific structural limitation that it have an anchor seat in contact with bone. Medtronic, the manufacturer of the implants, escaped liability for an otherwise infringing product because the court understood that “[t]he anchor seat of the device does not contact bone until the surgeon implants it.” Check out Cross Med. Prods. v. Medtronic Sofamor Danek, Inc., 424 F.3d 1293 if you don’t believe me.
No matter what your patent covers, you need to be aware of these nuanced issues. The claims need to be written in a way that covers your products at the time they leave your warehouse (and more importantly, your competitors’ warehouses), not at the time their customers actually use the products. For method claims, think carefully about what types of entities you want to exclude from using your technology (e.g. service providers vs users of services) and make sure the claim steps are written only to require the steps that would typically be carried out by those types of entities.
If you’re wondering what to do if your patent has already been issued, the court did suggest using a procedure known as a “reissue” application. While not a guarantee, this process might allow you to revise your patent claims, provided you’re willing to tell the patent office your patent is “inoperable.” However, thanks to Akamai’s example, at least you can learn this lesson the “easy way.”
If you don’t want to read the full case, check out the following blogs with a detailed breakdown:
- PatentlyO – Akamai v. Limelight: Joint Infringement Requires an Agency Relationship or a Contractual Obligation
- Patent Hawk – Bollocks to the Dollops