Infringement by Overseas Servers Revisited
Since the UK Court of Appeal judgment in Menashe v William Hill, it has not been safe to assume that infringement of a patent claim including a processor or a processing step can be avoided by performing the processing on a server outside the UK. A third judgement on this topic has recently been issued – making the score two for infringement and one for non-infringement – so it is instructive to consider what factors affect a finding of infringement. We review the relevant points of the three cases and draw some practical conclusions.
In Menashe v William Hill the claimed invention was an on-line gaming system that included a host computer and a terminal computer. The defendants operated a host computer in Antigua and supplied software to customers in the UK which allowed the customers to place bets via the host computer in Antigua. The claimant asserted that the supply of the software to UK customers was an indirect (or contributory) infringement under Section 60(2) of the Patents Act 1977 which says it is an infringement to “supply … means, relating to an essential element of the invention … for putting the invention into effect in the UK”. The defendants asserted that there could be no infringement because the invention was not put into effect in the UK.
The question of whether the location of the server in Antigua was sufficient in itself to avoid infringement was tried in a preliminary hearing on an agreed set of facts. Both at first instance and on appeal, it was held that infringement was not avoided merely by the location of the server, albeit for different reasons. The reasons of the Court of Appeal looked at use of the invention and asked “who uses the claimed … system?” and “where does he use it?”.
On the agreed facts, the Court of Appeal decided that the user of the claimed system was the customer (punter) in the UK and stated:
it is not a misuse of language to say that he uses the host computer in the United Kingdom. It is the input to and output of the host computer that is important to the punter and in a real sense the punter uses the host computer in the United Kingdom even though it is situated in Antigua and operates in Antigua. In those circumstances it is not straining the word “use” to conclude that the United Kingdom punter will use the claimed gaming system in the United Kingdom, even if the host computer is situated in, say, Antigua. Thus the supply of the CD in the United Kingdom to the United Kingdom punter will be intended to put the invention into effect in the United Kingdom.
It was recognised that this conclusion could be extended to say that the punters in the UK committed a direct infringement by using the claimed system, although no end users had been sued. (In the UK it is not essential to a finding of indirect infringement that there has been a direct infringement.)
In the recent case of Premaitha v Illumina it was held that performing some data processing outside the jurisdiction did not prevent a finding of infringement. This case concerned several patents with a variety of claims to methods of, or useful in, diagnosis by analysing DNA in a blood sample. The Judgment is lengthy and mostly concerns issues relating to biotechnology which we will not address here. However, the alleged infringers sought a declaration of non-infringement in respect of a proposed alternative method. In the alternative method, a blood sample is to be taken from a patient in the UK, DNA sequencing carried out in the UK, “raw” sequence data sent to Taiwan for computer analysis and a report sent back to the UK where it is communicated to the patient. Although the analysis is automatic, at least some of the steps in communicating the data could or would be manual.
The Judge, following Menashe v William Hill, looked at where the invention would be “used” (skating over the question of who is the user) and concluded that it would be used in the UK “in substance”.
In this case, RIM (now Blackberry) were accused of infringement of four patents owned by Motorola. The relevant patent for present purposes included claims to a “method of operating a messaging gateway system” which acted as an intermediary between an email server and the end user’s handheld device. For certain users, the relevant gateway system was located in Canada. The judge held that this system was used by RIM and they used it in Canada – hence no infringement.
Lessons to be learnt
It has often been asserted that the patent proprietor should avoid the problem by careful drafting: including claims requiring only integers and/or steps that will be carried out in the jurisdiction. Of course patent applicants lack a crystal ball to allow them to reliably anticipate how an infringer will arrange matters to try to avoid infringement. In addition, it is often not possible to draft novel and inventive claims that cover only one side of a method involving two parties.
A potentially more reliable approach to maximising coverage is to draft claims aimed at different users. In particular, it is advisable to include claims that cover the activities of users who cannot easily move out of the jurisdiction, often end users, i.e. the consumers. This is consistent with a strategy that seeks protection in the largest and/or most valuable markets, rather than the locations of competitors.
In the RIM v Motorola case, claims to a method of sending or receiving messages might have served the patent proprietor better. In Europe, even where the invention lies in operation of a server, such claims could be presented as dependent claims adding extra integers at the user’s end or steps carried out by the user, even if such integers or steps were known.
The issue of who is the user can often depend on who has control, who provides inputs and who receives outputs. In seeking to draft a claim that covers the actions of only one party, it is common to draft features in terms like “receiving a user input”. However, the user of such a claim might be seen to be the person running the server and so a feature in terms like “providing a user input” might better indicate that the user of the claim is the end user. For example, some apps using machine learning algorithms for recognition of objects off-load the main processing to a server, often seamlessly to the user. This would seem a clear case where an invention claimed primarily referring to user actions would be seen as being used in the jurisdiction where the user is even if the server is elsewhere.
The approaches of claiming actions performed by only one actor and claiming from the perspective of the user often present somewhat contradictory requirements. This requires careful management within the EPO’s constraints on the number of independent claims and the total number of claims, necessitating judicious use of dependent claims. To get around the one independent claim limit, it is sometimes possible to present the invention as alternative solutions to a common problem (server-side and client-side) or as inter-related products. It is likely that claims that are optimal for UK and European considerations will diverge from those best for the US, where NTP, Inc. v Research in Motion, Ltd reached different conclusions for method and system clams based on a similar fact pattern to the UK case of RIM v Motorola.