In the field of distributed computing, user devices with limited computing power may be connected to a remote computing system where the main purpose of the user device is to receive commands from the user and to display processed data provided by the remote computing system. The main computational processing takes place at the remote computing system, typically formed by a “cloud” of network-connected computer resources. As the computational processing is shifted from the user device to the remote computing system, it is possible to provide the user with computing services that are typically not available in a stand-alone user device due to factors such as cost, computational power and licensing issues. This provides an opportunity to completely rethink how and where to place your company’s computing resources. It also allows you to decide whether you should upgrade your internal servers or outsource the computational power and storage to an external provider.
Accordingly, a new service-based business model is made available, which however introduces problems in regards to how to handle the intellectual property of the software providing the service. The main concern typically relates to the possibility of geographically positioning the servers in another country than in which the user is active. Geographical distribution of for example patented software, in different countries and jurisdictions put an undesirable spin on the enforceability of patent rights. It is easy to see how this could potentially allow a wilful infringer to position servers in a country or jurisdiction that provides good bandwidth and accessibility, but where no patent rights are enforceable.
However, based on fairly recent case law (e.g. Football Dataco & others v Sportradar GmbH & another (Case C-173/11) as well as NTP, Inc. v. Research-in-Motion, Ltd. (RIM), 418 F.3d 1282, 75 U.S.P.Q.2d 1763 (Fed. Cir. 2005)), it has been made clear that the courts generally try to look to where the beneficial use of the invention is obtained. This provides at least some relief in view of the above identified geographical disparity between servers running patented software and the possibility of patent enforcement.
This is however only half of the truth. The fact is that the courts (i.e. in relation to the RIM case) prefer to look at patent infringement from a “system” perspective, i.e. where the beneficial use of a system may be obtained even though servers are placed in one country and the users in another. It is a different story with method claims where it for example has been determined that each step of the process must be performed within the same country (i.e. the USA) in order to be seen as infringing.
Based on the above, care must thus be taken in crafting your intellectual property strategy when deciding to protect, for example, new computational and/or social media related software-based methods having the possibility to be executed within the cloud, where geographical disparity between user devices and the computing system is apparent. Obviously, it is of great importance to conduct a general review of the possibility of including patent claims from all different perspectives (providing they are considered patentable). This means including both device and method claims drafted from the perspective of the different computing nodes, the system including both the computing node(s) and the user device, as well as from the perspective of the user/user device.
In addition, the choice of where to file patent applications may differ slightly from the normal filing strategy (where filing a patent application in the USA is typically the first priority for software related inventions). Accordingly, it may be a good idea to also consider filing patent applications in countries where it may be suitable to place computer servers. An example of such a country is Sweden, where for example Facebook at present builds a 28 000 m2 server farm, their first outside of the USA. The reason for placing the server farm in Sweden is in this case due to the high reliability in regards to electricity and network connectivity.
To summarize, achieving a desired level of patent protection for cloud-based inventions is more complicated than for a physical object where the provision (manufacture and/or sale) of the object as well as the use generally take place in a single jurisdiction at any given time. The same, of course, applies where all steps of a patented method are performed by a single user in a single location. It is therefore essential to review your intellectual property strategy taking into account possibilities with regard to claim construction as well as the choice of countries in which to file patent applications.
Magnus Nilsson, European Patent Attorney