In this decision of the Patents Court, Meade J dismissed an infringement claim brought by Promptu against Sky, on grounds that Promptu’s patent lacked inventive step. The decision is interesting as it grapples with issues of infringement where some steps of a claimed method are performed outside the UK. The decision also provides helpful guidance on the instruction of experts in cases involving lengthy prior art documents.
The claimant (“Promptu”) brought patent infringement proceedings against the defendants (referred to collectively as “Sky”) alleging that the voice control functionality in their Sky Q subscription television service infringed its European Patent (UK) No. 1,290,889 titled “System and Method of Voice Recognition Near a Wireline Node of Network Supporting Cable Television and/or Video Delivery”.
At trial, only a single claim (claim 13 as amended) remained in issue. Sky denied infringement both as a matter of claim construction and on the basis that certain parts of the claimed method were performed outside the UK (the “territoriality issue”). Sky also counterclaimed that the patent was invalid for added matter and lack of inventive step over a prior art citation referred to as “Houser”.
Ultimately Sky succeeded in revoking the patent on the ground of lack of inventive step. However, the Court dismissed Sky’s claim construction points, added matter attacks and held that Sky would have infringed the patent had it been valid.
There are two aspects of the decision which, in this author’s view, are likely to be of particular interest to patent practitioners:
- The judge’s findings on the territoriality issue.
- The judge’s comments on guiding experts in their reading of lengthy prior art documents.
The Territoriality Issue
The invention related to a remote control and set top box communicating in a network with a central server referred to as a “wireline node”. Each of several end users can input speech commands using a microphone in a remote control. Each microphone has a “push-to-talk” button which begins a speech recognition process at the wireline node by informing the system that the subscriber is about to speak and also provides information about the user’s location. The speech from the user (and any other users sending voice commands) is then transmitted on a “back channel” to the wireline node which processes the speech, and responds to the user with customised entertainment and/or information from a “content engine”.
The relevant part of the Sky Q system comprised three sub-systems connected via the internet:
- A VREX voice platform hosted on Amazon Web Servers (“AWS”) in the UK.
- An automatic speech recognition (“ASR”) function provided by Google which turns voice data into text. This may or may not take place in the UK but Sky does not know for any individual instance whether it does or not.
- Sky Search: a content engine containing searchable metadata for the content available to the user. This is also hosted on AWS, but in Ireland.
In the Sky Q system, when the user presses the voice button on their remote control, a WebSocket (a persistent bidirectional connection between the client and server) to that user is opened. This process involves sending, among other things, the user’s IP address to the VREX platform. If multiple users activate voice commands at the same time, the VREX will receive data from all of them and separate out each user’s speech using their IP address. The VREX then sends a response back to each user, once it has received what it needs from the ASR and content engine.
Subject to two points of construction, Sky accepted that it performed the claimed method but denied that the method was performed in the UK because the ASR was sometimes performed abroad and the Sky Search was performed in Ireland.
In considering this argument, the Court reviewed the previous cases of Menashe v. William Hill  EWCA 1702, RIM v. Motorola  EWHC 188 (Pat), and Illumina Inc v Premaitha Health Plc  EWHC 2930 (Pat), and derived the following principles:
“(a) the Court’s task is to identify by whom and where, in substance, the method is being used; and (b) it is relevant to take into account that for some steps it simply may not matter where processing power is located.”
In the present case, the court considered that the claim was in substance to a “method of using a back channel…” which takes place at the wireline node (i.e. a server), and involves partitioning the incoming signals on the back channel and subsequently providing unique recognised speech content responses and individual delivery of services accordingly. Whilst user input triggers part of the method, (i.e. sending data from the user’s end identifying the user), it was not the user who put the method into effect. The court further considered that the content engine access and speech recognition in the Sky Q system were subordinate and essentially sub-contracted functions, and it did not matter where those two functions took place. This was consistent with the fact that Sky did not know where the ASR took place for any given user interaction.
The Court therefore concluded that the method was performed, in substance, by Sky, in the UK.
It is interesting to note Sky’s submission that the present case was most similar to the RIM case, where it was found that there was no infringement, and was unlike the Menashe case, where the method was focused on the user’s end (“a gaming system for playing an interactive casino game”). Although Meade J did not think that making comparisons with the facts of these cases was helpful, he considered that the submission did not help Sky in any event because in RIM, the server was in Canada, whereas in the present case, the VREX server (which he considered to be “the important one”) was located in the UK.
This aspect of the decision is likely to be of considerable interest to technology companies involved in developing and maintaining distributed systems. It highlights that one cannot avoid infringement simply by sub-contracting certain functions to servers outside of the jurisdiction. A question not answered in the judgment is whether Sky would have been able to avoid infringement by moving its VREX server to Ireland. Given the judge’s conclusion that the VREX platform was performing the substance of the claimed method, on one view, it might have done so, despite the fact that UK end users would continue to benefit from the services delivered by it. Patentees may therefore want to consider crafting claims covering the end user equipment where possible to avoid this problem in the future.
Instruction of experts
Many UK practitioners will be familiar with the difficulties in guiding experts in their reading of lengthy prior art documents, particularly where only a small part of the document is relevant to the envisaged inventive step attack. Often, legal advisors will be pursuing discussions with a number of experts in parallel and it is necessary to limit the expert’s reading in a sensible way to manage the costs of the action. However, by directing an expert’s attention to a specific part of a prior art document, solicitors risk introducing hindsight bias if there is no reason why a skilled person would have focussed on that section of the document.
In this case, Promptu alleged that Sky’s expert had been directed by Sky’s solicitors to the key part of Houser (a 46 page US patent) which formed the foundation for their obviousness attack. Whilst this criticism was largely directed at the evidence on amended claim 1 which was no longer in issue by the time the case got to trial, it is interesting to note the judge’s observations:
“It is sometimes necessary for purely practical reasons to ask a witness to look at a particular part of a prior art citation, otherwise if they are asked to give all their thoughts about many different passages the task is too big and too diffuse. If it presents a risk of hindsight the Court may have to take it into account, but it is not necessarily fatal or even serious, especially if, as in this case, the witness acknowledges the pointer and gives evidence about why that part of the prior art would be of interest (which as it turned out was common ground at trial). A similar issue arises when experts are given, as they often have to be, guidance about what aspects of CGK to explain.”
These observations are likely to provide some comfort to practitioners looking to take a pragmatic approach in their discussions with experts. They support a commonly held view that it is permissible to direct an expert’s attention to certain parts of a prior art document, provided that the direction is acknowledged in the expert’s report and the expert gives reasons as to why that part of the prior art would have been of interest to the skilled person. The approach is not risk free, however, and practitioners may still wish to avoid being overly specific in their instructions regarding the review of prior art documents where possible.
A copy of the judgment can be found here.