Mr Justice Mann :
- This is a patent action in which the second claimant claims infringement of its patent against the first two defendants. The second claimant is a member of the Rovi group of companies and it will be convenient to call it Rovi. The first two defendants (together "Virgin") are members of the well-known Virgin group of companies and the claim arises in respect of one of its set top boxes ("STB") which is supplied to viewers and which sits between the cable which supplies the signal and the television which displays the entertainment. In basic terms it turns the signals received into meaningful data (sound, vision and data presentations on screen). The case against the third defendant in respect of its separate STB is no longer pursued and can be ignored. In fact the infringement issues are limited. Insofar as the claims are valid, Mr Mellor QC, who appeared for Virgin, accepted that his product infringes, save for claims 40 and 57 in respect of which no infringement allegation is made.
- Virgin counterclaims for revocation of the patent on grounds of anticipation and obviousness. Attacks on the basis of added matter and excluded matter were not pursued at the trial, the former because the issue had been removed by an amendment of the patent. Two of the claims (claims 1 and 18, using a pre-amendment number) were not defended by Rovi, and fall to be revoked.
- Rovi applied to amend the patent, and that issue was dealt with at beginning of the trial. I allowed the amendments. Their effect was, in part, to add additional claims, which required the re-numbering of the following claims. In order to avoid confusion I shall use only the new claim numbers throughout, and not show the old ones.
- The principal claims which are relevant to this action as it now stands are set out in Annex 1 to this judgment. They are set out there in their amended form without showing the amendments (in the interests of clarity). The first number against the claims is the post-amendment number which is used in this judgment. The second number is the pre-amendment number. All other claims are dependent claims or other claims as to which no relevant issue is likely to arise. The debate at the trial concerned the annexed claims only.
The field of the invention
- The description appearing below is an outline of the invention. If there are any slight inaccuracies or over-generalisations in the description then the description does, of course, give way to the actual wording and effect of the patent. Any such errors are perpetrated in the interests of trying to achieve intelligibility.
- The priority date for the patent in suit is 20th November 1995. At that date the facilities offered by set top boxes were nothing like as advanced as those which are seen today. In particular the facilities which are now often seen under the "red button" facility on various TV systems was not in operation. In modern sets responding to a "red button" prompt on the screen can be done by pressing a red button on the handset controller, giving access to various additional features such as programme information or even a menu giving access to another TV "channel". That particular feature was not present in that particular manifestation on televisions in November 1995 (and in saying that I do not intend to pre-judge the anticipation arguments that arise in this case). Its visible manifestation is in the form of a prompt put on the screen (by superimposition) inviting input from the viewer, and if that input is made then a further message is conveyed, or something else happens in the STB by way of the execution of some computer code. That is the essence of the invention. (For the avoidance of doubt, the "red button" feature is just one possible use of it).
- The patent concerns the signals that go on "behind the scenes" in order to achieve that sort of result. At the heart of the invention is the existence of two signals. The first signal is called the "identification signal". That signal is responsible for bringing about the indication symbol on the screen including (in some of the claims) controlling the duration of its appearance. The manifestation is a prompt, in the form of an icon or text. The second signal (the "information signal") is responsible for displaying the message which is intended to appear if the user responds to the prompt of the first signal (by pressing the relevant button on the remote control of the TV or STB), or for running computer code contained in the information signal or triggering a program already resident in the STB (according to one construction of the claims).
- The invention is said to be that this is achieved in two signals which, taken together and used as described, facilitate the provision of the information in the information signal and reduce the amount of storage that would otherwise have to be used in order to make the information in the information signal appear at the right time. The objects of the invention, described in the specification, are said to be (summarising them) to allow the viewer to receive information in real time about forthcoming programmes, and to interact by pressing a button on the remote and (by way of examples) getting information, paying in advance for pay-per-view (PPV) programmes, or electing to record a programme. According to Mr Johnson, Rovi's expert, the technical benefits of the invention (not set out in the patent itself) are to enable the broadcaster to send out information in advance of the event to which it wishes to alert the viewer rather than simply sending a message close to the event, inviting a response and responding to requests by sending out a second message at that time. Sending the messages in advance means that slower data channels can be accommodated, a return path for a request for the information can be avoided (the message has already been sent, should the viewer wish to see it) and (when compared with a system in which input from the viewer results in a request to which a broadcaster has to respond, for example by responding to a PPV request) it can avoid a "storm" of requests immediately before an event.
The specification and the claims in issue
- The background to the invention is set out in the first section of the specification. Paragraph 2 starts by expressing the then limitations of "programming guide information". It was said to be constrained by the need for a viewer to wait for information to scroll past, or manually navigate to the part of the guide in which the viewer is interested, leaving the viewer unaware of the current and future events which are available.
"It is, therefore, an object of this invention to provide a process and in-home video guide hardware by which a home viewer may be informed, in real-time, of events that are currently available or will in the future be available for viewing. Another object of this invention is to provide a process and in-home video guide hardware which alert a viewer to the availability of selected current or future program events. Yet another object of this invention is to provide a process and in-home video guide hardware which permit a viewer to access information about selected programming without changing channels. Another object of this invention is to provide a process and in-home video guide hardware which permit a viewer to interact with information and prompts respecting selected programming without changing channels ... Still another object of this invention is to provide a process and in-home video guide hardware which permits the viewer to choose whether a program will be recorded or not recorded.... Yet another object of this invention is to provide a process and in-home video guide hardware which permits the viewer to view a display of the cost of the event. And it is an object of this invention to provide a process and in-home video guide hardware which permits the viewer to authorize payment for access to a selected program."
- The "Summary of the invention" describes how:
"In accordance with the invention, a system interactively controlled by a TV viewer remote control transmitter displays current or future programming information superimposed over the normal programming displayed on any channel and accessible to the viewer's display screen."
- A tuner receives transmission signals and passes them to a signal combiner. A computer receives control signals from a remote control transmitter and also receives:
"from the tuner and stores data contained in a current and future programming identification signal and data contained in a current and future programming information signal. The identification signal contains data indicating the availability of the information signal. The information signal contains data descriptive of and/or facilitating interaction with current and future programming. Portions of the stored identification signal data are accessed and generated as an output identification signal by the computer for durations defined in the identification signal data either automatically at times defined in the identification signal data or in response to a selected one of the plurality of control signals from the TV viewer remote control transmitter. The signal combiner superimposes the identification output signal or the information output signal over the viewer usable signal of any selected channel from the tuner to provide a display signal for input to the viewer's display screen. The identification signal may include data defining an icon or textual matter alerting a viewer to the availability of information data or may include data enabling access to such icon or textual matter data as may be already stored elsewhere in the computer. The information signal may include program descriptive data, schedule data, pricing data and/or payment authorisation and recording prompt data and the like sequenced for interactive control of the computer by the viewer and may also include data programming the interactive operation of the computer or data initiating operation of an interactive operating program resident in the computer."
- The specification then cross refers to various diagrams and drawings. Figure 3 is said to be: "a representation of a video screen display of a typical special events output identification signal superimposed over a normal programming signal"; and figure 4 is: "a representation of a video screen display of a typical special events output information signal containing program description data superimposed over a normal programming signal". In fact they both seem to be the first steps in a chain which invites a viewer to consider buying a viewing of a future film. Figure 3 contains a rustic snowy scene as a background in which there is superimposed an icon on the right hand side and a text box on the left containing the words "Request Ch 36 8:30 p.m. Thursday Dracula". Figure 4 shows the same scene with a box giving a little more information about the film, its duration as 2 Hours and a price of $4.95. Below that there are two buttons labelled "Order" and "Cancel". Figure 5 has the same scene with a box containing the text "This event is $4.95. Please enter PIN ____". It therefore appears that one particular application of the invention would be in the PPV function offered by some broadcasters pursuant to which viewers can buy viewing rights for a given program in advance of the broadcast of that program. This expressed embodiment has an impact on the identity of the skilled addressee, as will appear. Other embodiments appear in the figures - Fig 6 contains a textual invitation to record a programme and Fig 9 contains information about other films on one or more channels which the viewer is not currently viewing.
- The relevant claims are, as indicated above, set out in Annex A. The claims have been divided into their integers for ease of exposition. Claims 27, 28 and 29 are essentially method claims equivalent of the apparatus claims in 2, 3 and 4. Claims 53 and 57 are the method equivalents of claims 36 and 40. Claims 40 and 57 are not said to be infringed but there is still an attack on their validity.
- In considering the claims it may be useful to keep in mind what they each achieve. I can do that by reference to the method claims. Claim 27 describes the receipt of two signals - the identification signal and the information signal, and their processing by relevant equipment (including a computer). The identification signal is to be superimposed at pre-selected times and for pre-determined time increments without any viewer input. In essence it invites input, at which point the viewer can provide that input and the information signal is displayed. Claim 28 adds the requirement of the identification signal defining an icon or textual matter (to appear on the screen), or enabling access (pointing to) an icon or textual matter already stored on the computer, and with the data also defining the display time and period. Rovi maintained that those matters were implicit in claim 27 anyway, and added claim 28 to make the position clear. Claim 29 adds a requirement that the information signal should comprise "computer program data" or "data initiating operation of an interactive program resident on the system". A point of construction arises as to the meaning of "computer program data".
- Claim 53 (and its equivalent apparatus claim 36) requires the two signals but the information signal comprises "computer program data", which is executed when the user interacts with the system as a result of the identification system. There is no requirement that the identification signal should contain or point to icons, and no express requirement that it should contain timing matters.
- Claim 57, and its matching apparatus claim 40, has the two signals, and differs from claim 53 in that the information signal is "program initiation data" rather than "computer program data" i.e. it initiates something rather than being computer code itself. There is no infringement claim in relation to these claims because it is accepted that in the signals sent to Virgin's STB the second signal do not initiate anything - the equivalent of the information signal is "computer program data", not "program initiation data".
- The principal issues in this case are those going to the validity of the claims identified above. It is said that they are all variously anticipated by, or obvious over, certain pieces of prior art.
Questions of construction
- There are limited questions of construction of the claims which arise. One particular point turns on the use of the words "suitable for" in the apparatus or system claims. I shall deal with that separately later on in this judgment. The other construction points which arise are as follows.
- First there is a dispute as to the meaning of "program initiation data for initiating execution of computer program data" (claims 40 and 57). Neither "program initiation data" nor "computer program data" is a term of art. Rovi's case (at least in relation to claims 40 and 57) is that this expression means "data which initiates the execution of a program resident on the apparatus", taking wording actually used by Mr Turner in his first report (para 155). Virgin's case is more subtle. It says the expression means data which initiates "computer program data" (i.e. the execution of executable code). This is said to be capable of including any instruction to a computer, from one which triggers the execution of a program to one merely requiring the appearance of a character on the screen, or an instruction to make a teletext character which already appears one of a different colour or size. It is of the essence of Virgin's submission that it includes any sort of trigger which makes the computer do anything at all. Any such form of instruction is likely to lead to the execution of an instruction, and it is Virgin's case that this is the execution of executable code.
- Virgin's case turned out to be a wide one. Its consequences were investigated in the cross-examination of Mr Turner, who then expressed the wider view rather than the view initially expressed in his report. He said (in effect) that anything which told the computer to do something that it had not done, or been doing, before, fell within the expression, whether that involved the running of a new program, the invocation of a sub-routine in a running program, some minor operation triggered by a control character or even the display of a single letter on a screen. In support of its wide case Virgin points to the use of the word "data" at the end of the expression "computer program data".
- I think that Virgin's case is too wide. If such simple interactions were intended I think that a different expression would have been used. The word "initiation" has some significance - it suggests some sort of de novo activity. Coupled with the expression "computer program data", it suggests the triggering of a program, or a particular routine within a program, and not merely some (any) activity within the computer, which is still executable code. Otherwise far too much would be caught. If any data is supplied to a computer it does something with it that was not doing before, even if that involves putting up an error code or getting itself into a state in which it crashes. The precise boundary may be hard to draw in a fashion which would be applicable to every case, but it excludes something of which the only thing can be said is that the computer was not doing it before signal was sent. I therefore find that Virgin's wider case on construction fails and Rovi's case on construction is the correct one. "Program initiation data" in claims 40 and 57 means something in the form of data which initiates the execution of something that can be called a program (including a routine within a program) which is already resident on the apparatus.
- For the sake of completeness I observe that different wording is used in claims 4 and 29, but no question was raised as to those words in Virgin's final submissions, and since that wording was added by amendment of the patent it does not assist in determining the effect of the wording in claims 40 and 57.
- Next, submissions were addressed as to the meaning of meaning of "computer program data" in claims 4, 29, 36 and 53, which require the information signal to comprise that feature. It is common ground that the expression means executable code, and it is also common ground that that can include scripts (high level instructions). Mr Mellor submitted that it could include routines and subroutines of computer programs, and all other forms of executable code. If a "routine" is a self-contained program then of course that is within the expression. It is not clear what the technical point of sending a sub-routine via the information signal would be, or that that is of any relevance to this case, or even what it would mean in the context of the patent if it were a synonym for "computer program data". Nor is it apparent that "other forms of executable code" would bring in anything other than scripts, which it is accepted are within the expression. It is therefore not apparent that there is any meaningful dispute in the context of this case for me to decide on this point.
- Third there is a question as to whether claim 2 requires "duration data" to be broadcast. Mr Meade described this as an "arid" debate, saying it was implicit in claim 2 and, if not implicit there, it was explicit in claim 3, so it made no difference to the case. It might not make a difference to infringement, but it might go to the validity of claim 2, so I shall consider it.
- The point here is whether the timing integer ("at preselected times and for predetermined time increments") is limited to timings which are comprised within the first identification signal, or whether it includes that method of producing timings whilst leaving other methods open as well - for example, coding in the STB. Mr Turner's evidence is that both concepts were known to engineers, and I accept that evidence. I consider that the broader view propounded by Mr Mellor is the correct view. The wording does not say how the information is to be conveyed. To the skilled man there would appear to be two possibilities. Both have some practical sense, even if sending the timing in the identification signal is the more practical. Even if the signal carried the time of display, the box might contain the time increment for display and thus predetermine it there. If the skilled person had any doubt as to what might be meant, he would then be entitled to look to claim 3, which explicitly claims one of those possibilities in paragraph (iii). Even though it was added by amendment, the effect of the amendment is to amend the patent ab initio, or at least the patent should be read as a whole with the new limitation appearing in it. Where there is such an express limitation the skilled person would be likely to think that the expression in claim 2 had a wider meaning than the limitation in claim 3 - otherwise what is the point of the formulation in the dependent claim? I therefore find that the limitations in claim 3 are not implicit as limitations in claim 2 (that is to say, that claim 2 is broader than claim 3).
- I heard evidence from one expert witness from each side. A further witness produced a product and process description in relation to the allegedly infringing product, but since there was no issue about that he was not called and I can ignore him.
Mr Stephen Johnson
- Mr Johnson gave expert evidence for Rovi. He has qualifications in electrical engineering (and mathematics) and has been closely concerned with interactive applications in CD-ROMs and television commerce applications. His particular leaning was towards the development of the interface between applications and the user. He did not have an in depth understanding of all the technologies behind the products he was developing (not surprisingly). As at the priority date he was investigating the features and the operation of analogue (not digital) STBs, and had discussions with cable and satellite companies, analysing their hardware and software with particular emphasis on Electronic Programme Guides ("EPGs"), though he did not seem to have investigated the products of the satellite (as opposed to cable) broadcasters much. Subsequently he worked on designing the technical architecture and user interfaces for an interactive television programme guide and user interfaces, for deployment on various STB platforms.
- In the main Mr Johnson was doing his best to assist the court within the limitations of his experience, but I felt his evidence was adversely affected by two things. First, while it is understandable that he would not necessarily personally have knowledge as at the priority date of everything that would be known to every member of the skilled team in this case, he does not seem to have done much by way of research to acquaint himself, or remind himself, of a number of matters that would have been relevant to and known to at least some of the skilled team. From time to time his evidence proceeded on the basis of his own views and experience without an appropriate attempt to consider matters more broadly. Second, he also, in my view, demonstrated a tendency to hold on to a view or position because it suited his client's case (or at least did not damage it) rather than being prepared to accept a position I consider he might otherwise have agreed with. He was trying to hold a position rather than consider the matter objectively as an expert. That is a little more troubling in the case of an expert.
- I therefore approach his evidence with a little more caution than I would have expected or wished.
Mr Graham Turner
- Mr Turner gave expert evidence for Virgin. He has a degree in electrical and electronic engineering and for many years worked in consumer electronics and then in defence. He also did consultancy work, and in 1988 was invited to head up British Satellite Broadcasting's conditional access activities - I explain what is meant by that below. With BSB Mr Turner was responsible for both technical and commercial matters. When BSB merged with Sky in 1990 he left and set up his own consultancy. He continued to provide services relating to the encryption side of broadcasting but also had a hand in the Service Information Committee responsible for contributing to the standards for the newly emerging DVB digital television standard. "Service information" is metadata relating to the television broadcast "so that the user can be provided with information to assist in selection of services and/or events within the bitstream". He is therefore more of a technician than Mr Johnson, and in some areas his expertise went beyond that which would be the expertise of the skilled addressee. I accept that it also gave him a slightly CA-centric view of the matter. Notwithstanding that, I found his evidence helpful even though I do not accept all of it. His evidence was criticised because he was shown the patent before he was shown the prior art, and it was suggested that these items ought to have been presented to him the other way round. On the facts of this case I do not consider that to be a relevant criticism. I am not sure what he would have been able to contribute had he been shown the prior art with no real context against which to consider it. I do, however, find that his report was not as clear as it ought to have been as to the make-up of the skilled addressee (skilled team) in this case.
The skilled addressee
- The patent is taken to be addressed to a skilled addressee. The experts agreed that the skilled person would actually be a team, but there was a dispute as to the make-up of that team.
- The team of Mr Johnson (Rovi's expert) was someone working in the consumer electronics or satellite/cable TV fields, being a systems architect interested in designing interactive systems for televisions and/or satellite/cable STBs and writing software for them, probably supported by an interface design engineer.
- According to his report, Mr Turner's skilled addressee would be "an individual or team with experience in EPG's conditional access systems and associated provision of data signals including service information". That requires a little explanation. EPGs are Electronic Programme Guides - listings of programmes presented on the screen from which a viewer could see the times and dates of current and future programmes and some further details about the programmes themselves. A conditional access (CA) system was a system made up of hardware and software which allowed the broadcaster to control what programmes a viewer could and could not see. One can take as an example a programme which has to be paid for. All programmes are broadcast generally and the signal can be detected by everyone. However, the broadcaster will want only those who have paid for the programme (or the service) to be able to watch it. This process is carried out by encryption and decryption, and only those with the relevant (electronic) permissions can decrypt the signal. This selecting activity is done by part of the STB system known as conditional access (CA). Another aspect of a CA system is the ability and need to interact with the viewer by way of messages on-screen to alert the viewer to things he needs to know - the need for payment, for example, or refusals of permission to view.
- Mr Turner's team would also include a person with skills in and knowledge of the sort of signalling required from the "head end" (the broadcasting end of the system) in order to make the system work. Such a person would not need to deploy large areas of his knowledge, such as signal compression and the complexities of multiplexing (broadcasting groups of stations signals together), but he would have relevant knowledge as to how to inject "service information", that is to say information about the material that was being broadcast and which a receiving system would need to receive to be able to interact with a viewer.
- Mr Johnson did not accept that those persons, with their level of expertise, should be part of the team. He accepted that his team would know of the existence of CA, and its outward manifestation to the viewer, but would have no knowledge beyond that. Deeper knowledge was not relevant, and he partially relied on the fact that the necessary security surrounding the CA system meant that details were not disclosed as between designers. By the priority date interaction between viewer and CA was being handled by teams more dedicated to that task, so CA-related interactivity was of less importance to the team. He did not see the patent as being directed to CA at all, partly because the teaching of the patent could be deployed in a TV which would not (at least at the priority date) have any CA technology in it (that technology would be in the STB).
- All this is capable of mattering, of course, because of course it affects the common general knowledge ("CGK") which has to inform a consideration of the patent and the prior art. There was a suggestion in final submissions that the materiality of the difference between the parties did not matter because of concessions by the experts (particularly Mr Johnson) as to what other members of the skilled team would know anyway, but I do not think that this is an answer to the point. I think that the point still needs to be addressed.
- I prefer Mr Turner's evidence, and Virgin's analysis, on the point. So far as the CA expert is involved, it can be seen that the patent is addressed to an area which concerns CA. One of the objects is to allow a viewer to view the cost of an event, and another is to permit the viewer to authorise payment for it. The first brings in CA matters only peripherally, but the second engages CA considerations more directly. Later, the patent gives an example of a pay-per-view application, and the steps which would be needed to achieve notification of the need to pay and the process of authorisation. It is true that a large area of CA expertise would not be needed to implement this patent (as Mr Turner agreed) but part of that expertise would. Mr Meade submitted that Mr Turner's concessions had the effect of a concession that a CA expert would not be part of the team, but I do not think that that was the effect of them. The CA expert would be part of the team because of the apparent interface between his area and the area of the patent. He just would not need to deploy some areas of his expertise for the purposes of his notional consideration of the patent. I do not see why that should disqualify him.
- The same applies to the signals engineer (the "head end" specialist). The patent involves the delivery of two types of signal. It is not just about the interface between the user and the STB in some sort of abstract way. It involves the practicalities of signalling. Subject to the point that I am about to make, that is the concern of a head-end engineer, and he would in my view be part of the skilled team. As with the CA expert, there will be large areas of his expertise which are not necessary for the purposes of this patent, but there is an intersection with part of his expertise and he would be needed to deal with that.
- One of the problems with Mr Johnson's skilled team is that it did not seem to have anyone who was capable of dealing with the signals side of the patent at all. It was not apparent from his description that either member of his team would be capable of dealing with that. As he put it in his cross-examination, the team would take the signals for granted, though the team would need to know what was in them. To a degree, his team would be aware of the sort of things that CA systems did by way of putting messages on screen, either directed to all users or to a sub-class of users. However, even then he did not seem to have anyone within his team who was capable of dealing with the technicalities of the sort of signal which the patent would seem to require, and that is a lacuna. In my view it would be filled by the CA engineer and the head end engineer.
Common general knowledge
- I approach the question of common general knowledge ("CGK") with the following well-established principles in mind:
(i) Not everything that is published is CGK, even if all publications are available to the skilled addressee and even if widely circulated. To be CGK it has to be generally known and accepted without question by the bulk of those engaged in the particular art and becomes part of their common stock of knowledge - British Acoustic Films 53 RPC 221 at 250, cited in Beloit Technologies Inc v Valmet Paper Machinery Inc RPC 489, with the qualification that "accepted without question" might be putting it a bit high.
(ii) Notwithstanding that, a detailed work does not have to be fully present, in detail, in the mind of the skilled addressee to qualify as CGK. It is sufficient if is the sort of stock work that would be (notionally) present on his bookshelf and appropriately consulted from time to time - Raychem Corp's Patents  RPC 23.
- There were potentially significant disputes as to the scope of CGK in this case, the resolution of which was not assisted (as Mr Meade correctly pointed out) by a lack of focus on the detail of what was said to be CGK on the part of Virgin. It is unnecessary to set out a lot of the detail of what was CGK. I can take as a starting point the areas which were pointed up as being undisputed in Mr Meade's closing submissions:
(i) EPGs and their provision to and storage of information of the data by an STB.
(ii) Data transmission by cable and terrestrial and satellite.
(iii) The use of the vertical blanking interval - "VBI" - for the transmission of data. The VBI is the time taken in a cathode ray tube television for the electron beam (which paints the picture in a series of lines down the screen) to move from the bottom of the screen to the top of the screen to start a further paint. In that interval no "picture" data is sent and the time is available for the transmission of what are, by modern standards, small amounts of data. That is how data for teletext services was transmitted.
(iv) The combined use of a TV, STB, a processor for running STB software, a signal combiner and generator locking device to allow the superimposition of a display over a video image, and volatile and non-volatile memory for the storage of EPG software and data respectively.
(v) At the priority date, TVs themselves had limited memory and processing power.
- This is by no means an exclusive listing. Rather than continuing the catalogue of what was common ground, it will be more useful to focus on the areas of dispute, which inevitably means elements that are said to be CGK by Virgin which are not accepted by Rovi.
(i) The use of superimposed on-screen icons or prompts, in which the prompt was either "burnt in" (i.e. transmitted with the video data, essentially as part of the video signal) or generated locally (in the STB) from a signal or material which was not an integral part of the video signal. Mr Meade criticised the generality of this formulation, but it is really a contraction of paragraphs in Mr Turner's report which were put to Mr Johnson and accepted by him, to the effect that by the priority date it was known to broadcasters (and therefore to the skilled team) that there were those two ways of rendering prompts. I find that this was CGK.
(ii) The use of a broadcast signal or flag to instruct an STB to render an on-screen prompt (or other image) stored locally on the STB. In support of this element of CGK Mr Mellor points not to the section of Mr Turner's evidence where he deals in terms with CGK, but to parts of it in which he deals with one of the pieces of prior art (Florin). In that context he first explains the synchronisation of prompts to screens and then explains some obvious alternatives to the way in which that piece of prior art approaches "using a data signal to provide [a] prompt". In the first instance he refers to something which was not "technically .. unusual or exciting" and identifies it as having been achieved in a previous encrypted system. In the latter case he refers to instances in which "a specified field in the data signal [is used] to indicate that further information is available in an "information signal". There are a wide range of cases where data fields are used in this way. For instance, in the MPEG-2 and DBV SI standards there are tables and descriptors (data fields) which explicitly confirm the existence of other tables and descriptors and point to them"; and he gives some examples in technical documents. This is not a wholly satisfactory way of trying to establish CGK but I find that Mr Turner is right about the point forming part of the CGK.
(iii) The use of IPPV (impulse pay per view) in the US on various systems (which are identified). IPPV is what its de-construction suggests - the ability of a viewer to buy a programme at short notice, usually (but not necessarily) very shortly before it is broadcast. Mr Turner said that the normal way of inducing a purchase is to put up "something" (presumably a message) which invited the viewer to purchase it. I find that that IPPV was indeed being deployed in the US and had been for some time (at least 18 months) before the priority date, and based on Mr Turner's evidence, with a degree of confirmation by Mr Johnson, I find that that was CGK. The relevance of this is questionable, not least because of the generality of the assertion as to CGK.
(iv) The availability of digital (as opposed to analogue) broadcasting systems, and the fact that they offered more scope for data transfer than an analogue system. For about 2 years before the priority date the DVB specification for digital broadcasting was being developed in Europe, and digital services were launched in the US, Thailand and Australia. It was said by Mr Mellor that a signals engineer would keep abreast of the developing standards in the DVB proposals, including the versions of the MPEG protocol (MPEG-2) that it was proposed would be used. One relevant aspect of digital transmissions was that some of the CA messaging formerly being delivered as part of the CA system was now being sent as part of the more general (and expanded) SI (system information) packages. I accept that this would be CGK since it would be known to the signals engineer.
(v) The download of software to STBs. This was said by Virgin to be CGK. Mr Johnson's position in his report was that EPG software was programmed into non-volatile memory as at the priority date and that it was only after the priority date that it became "more commonplace" to have volatile memory in STBs which could be used to receive downloaded improved EPG software, though even then this was a risky enterprise because of the risks to the STB posed by corruption of the transmitted data. His cross-examination clarified this by saying that the possibility of downloading software was known, but perceived as risky, as at the priority date. Mr Turner's evidence was that there were risks, and they were appreciated, but despite that there were well-advanced plans to introduce that feature into the industry. During the cross-examination of Mr Johnson Virgin introduced evidence that existing systems (including an STB box with which Mr Johnson had claimed some acquaintance) offered the feature, and he admitted that other manufacturers made similar claims for their products, though he suggested that they might merely be at the level of claims. Mr Johnson had not carried out any investigation as to what was actually being done on any of those products as at the priority date. I find that Mr Johnson's cautionary words about the risks, while true so for as they reflect an acknowledged risk, overstate the degree of caution in the industry. I find that even Mr Johnson's skilled team would have been aware of this feature of products and systems which were being sold or otherwise touted in the industry, because they had become a feature of general industry products which one would expect his skilled team to be aware of, and that the perception of risk was not a significant obstacle to its implementation. I therefore find that the possibility of downloading software to an STB was CGK as at the priority date.
- Other areas of dispute as to CGK are best dealt with in the prior art context in which they arise.
The attacks on validity
- By the time of final submissions 3 pieces of prior art were in play, one having been abandoned after evidence. I will take each of them in turn. Only one of them is relied on in respect of all the claims; the others are used to attack some of the relevant claims only. The position in relation to each of the claims is set out in the context of each piece of prior art.
- This piece of prior art is the basis of an attack on the grounds of obviousness.
- Florin is a WIPO application published on 5th January 1995 on an application by Apple Computer Inc (which did not then occupy the same position in the personal computing world as it now occupies). The Field of the Invention is described thus:
"The present invention relates to the field of audio-visual systems. More specifically, the present invention relates to a system for selectively viewing and interacting with programs and services from a number of program/service sources, a control device for controlling the system, and the methods and apparatus incorporated in the system for managing selection, viewing and interacting with the program/service offerings."
It is a long document with many embodiments. The one which is of interest in this action is presumably intended to fall within one of the generalised descriptions as follows:
"The marriage of video and television technology with computer interface technology provides customers with maximum flexibility in storing, retrieving and viewing television and other audio-visual programming.... Furthermore, information such as TV program listings and additional information related to programs as well as selecting and controlling categories of interactive programs and services may be provided through the user interface of the present invention.
- Pages 29-30 sets out provisions as to system memory which includes (omitting cross-references to figures):
"…both volatile and non-volatile memory components. The non-volatile part of system memory includes read-only memory (such as ROM), which is used to store an operating system and playback software, fonts, sounds and the like used in the present invention. The non-volatile part of system memory also includes rewritable memory (such as SRAM), which is used for persistent storage of mark or record indicators, listings of programs viewed or taped, and other user preferences. Additionally, the volatile part of system memory includes a sufficient random access memory (such as RAM or DRAM) for the temporary storage of data received over the T/T cable or from other devices connected to the transceiver."
- Page 32 refers to data provided for program listing, which is provided in "the digital program listing channel". Figure 3a (which I have not reproduced in this judgment) shows how the electronic spectrum is to be used. It does so in terms of channels. Channel 100 carries the EPG listing, Channel 103 carries existing analogue TV channels, pay-per-view digital channels are shown carried by channel 106 and "Additional digital channels - for other interactive services" are carried on channel 108. Mr Mellor submits that this teaches that the additional services to which I am about to come were to be digital, not analogue, services. I accept that submission.
- The part of this long document (running to 77 pages and 50 pages of diagrams) which was relied on as founding an obviousness attack is within that part entitled "Home Shopping Interface". In order to understand the case made it is necessary to refer both to text and to the associated figures. The relevant figures appear in Annex B to this judgment.
- The relevant text reads (omitting some of the cross references to the figures):
"The present invention includes one possible home shopping interface, which will now be described. In Figure 43, a channel identified as "SHP" or "TV Shop" represents a dedicated home shopping service within the electronic spectrum illustrated in figure 3a. … In the current embodiment, the selection of the TV Shop service results in a continuous full-motion video display of various paid-for commercials or advertising messages. As shown in the representative screen of Figure 44, a TV Shop icon is displayed along with a blinking select icon (or right arrow icon) in a graphic overlay panel 400, at the start and end of each advertisement, prompting users to press the select button [this reference to a select button feature on a TV remote control).
Referring now to Figure 45, by pressing the select button… the user can view additional information relating to the advertisements shown in the full motion video section of the TV Shop. In the example of Figure 45, it will be noted that a listing of the advertisers is provided, in which the advertisers whose commercial was playing last is highlighted. Through the use of the up arrow button or the down arrow button [on the TV remote control], the user may scroll through the advertisers, sequentially highlighting each of their advertising identifiers, along with a select icon 403 prompting a user to select that advertiser ... As illustrated in Figure 46 the selection of "Attic at Casey's" advertiser with the select button… results in a display of the various specials available from that advertiser. In Figure 46, there is shown cookware highlighted in conjunction with a still image of the cookware for sale. By depressing the down arrow button on [the] remote control device, the highlighted section on the display can be moved down such that the mini-espresso (see Figure 47) is selected. Simultaneously with the highlighting of the mini espresso selection, a still image of the mini-espresso machine for sale is shown.
"Referring to Figure 48, to obtain further information regarding the mini-espresso machine illustrated in Figure 47, the user depresses the select button… on the remote control device to highlight the info icon 408. As illustrated in Figure 48, information relating to the particular product (i.e. espresso machine) is shown on the screen 180. To order the mini-espresso machine illustrated in Figure is 47 and 48, the user depresses the select button... to highlight the order icon 409. "
Florin - obviousness of claims 2, 3, 27 and 28
- This debate concerns the identification signal in the patent and the bullseye prompt in Florin. The debate about Florin revolved around the question of what it was that informed the STB to show the bulls-eye icon, whether it was obvious to do to implement that feature and whether it was obvious to add timing and duration data to whatever that prompt was. Virgin's case was that Florin discloses superimposing a prompt on screen at the start and end of each advertisement, but it does not state how it is that the STB "knows" that an advertisement is starting or ending, or how it knows how long to display the prompt. That is the difference between Florin and the above 4 claims.
- It was common ground that the relevant icon would not be thought to be broadcast as a piece of data every time it was needed but that it would be called up from the STB when needed, and the experts agreed that a skilled person would know of the ability to send flags or fields in a data stream to alert the STB to the need to render a prompt, and would know or assume that the prompt would or could be stored internally in the STB.
- At this point the cases of the parties diverged. Florin discloses an invitation to interact, and see an advertiser's name highlighted, by displaying a prompt (the bullseye feature). Virgin's case was that a signal sent with or before the advertisement was an obvious way to tell the STB which advertiser's name to highlight on the screen once the prompt was responded to via the remote control. So far as the prompt is concerned it would be obvious to inform the STB when to put the prompt on-screen by sending a signal, and one way of doing that was by a digital signal. It is true that there were two other ways known at the time and which could achieve the same result, namely via the EPG system and via the VBI. However, sending a signal via the digital channel made more sense, and was more obvious as a way of achieving the end result described in Florin. Florin itself had a digital channel. It was equally obvious to have the signal convey information about the timing and duration of the appearance of the prompt. The data relating to this on-screen prompt could be sent via the EPG system of the VBI, and each had its pros and cons, but each was obvious. Using a digital signal was likely to produce more flexibility than using the EPG system, and flexibility would be desired by advertisers.
- That was Mr Turner's evidence (in summary). Rovi criticised this evidence on a number of bases. This included the fact that his attention was drawn to the shopping channel by Virgin's solicitors at a meeting at which he was given the art, and before he had embarked on his own consideration of Florin. It says that this, and the fact that Virgin's ("illegitimate") focus was on only a small part of the disclosure meant that Mr Turner's evidence was an exercise in hindsight.
- Rovi's case was that the display of the prompt was triggered by the EPG. Although this was not expressly disclosed, that is what the skilled addressee would consider was the trigger. Similarly the EPG data would define the end time of the prompt display or the display time. The receiver would display the prompt for a fixed duration, and the data for that would be part of the EPG data in the same signal as the data which populates the TV Shop screen. There would be no distinct Identification Signal. Any signal would not transmit an icon or a pointer to it, and there would be no transmitted data which itself prescribed the start, finish and duration of the display of the icon. Another way of achieving the same result was the use of the VBI. The VBI would identify the advertisement, and the STB would be able to detect a change of advertisement and respond by displaying a prompt. There would be no pointer or duration information transmitted in the VBI. Mr Johnson did not accept that a digital signal with pointer and timing information within it would have been obvious to the skilled addressee.
- Approaching the matter within the structured approach identified in Pozzoli v BDMO  FSR 37, it is necessary to identify the differences which exist between the invention and the prior art (step iii) and assess whether, without knowledge of the alleged invention, those differences constitute steps which would have been obvious to the skilled addressee. That is not quite the exercise which is relevant in the present case, because the question is not so much whether what is said to be the extra step would be an obvious thing to do over the prior art, but what are the obvious ways of implementing the prior art. In relation to part of the reliance on Florin the point boils down to one question. It was accepted that Florin disclosed the display of a prompt which, when responded to, led to the display of the fruits of a second signal, which was capable of being the equivalent of the information signal. The bullseye prompt (together with the shopping channel logo, though that does not matter) is the equivalent of an on-screen prompt, and it was accepted by both experts that the natural reading of Florin is that that prompt was stored on the STB. The question is: How was the display of that prompt triggered? That is not disclosed by Florin (as the experts agreed). To summarise, Rovi submitted that the skilled addressee would take it to be triggered by the EPG program and data, or the VBI, and by nothing else. Nothing else was obvious. Virgin's case is that it could be triggered by a separate digital signal, which was an equally obvious solution to the EPG and VBI solutions. Furthermore, if one is sending a digital signal for that purpose, it is obvious to consider using that to control when the prompt should appear, and for how long. So the question is whether or not the digital signal route, with timing data, is obvious. In this context it must be borne in mind that the fact that certain solutions are obvious does not rule out the obviousness of others - Brugger v Medicaid  RPC 635.
- On the question of the obviousness of sending a separate digital signal to trigger the display of the prompt I prefer the case of Virgin and the evidence of Mr Turner. The latter accepted the obviousness of the EPG route and the VBI route when they were put to him, and said that the digital route was an equally obvious alternative. I accept his reasoning. It will be remembered that the Florin system had a digital channel which is available for "other interactive services". The skilled addressee would be aware of that, and Mr Johnson accepted that the interactive shopping channel would be one of the digital channels provided. This would provide an additional stimulus to considering a digital identification signal. I would have thought that that route would, if anything, be more obvious, or at least more natural, by 1995 than the VBI route. (The possibility of VBI code as being the identification signal was not really canvassed at the trial. At one stage in his final submissions Mr Mellor suggested that it could be, and for my part I wondered why it would not be, but since the possibility was not really dealt with by the experts I do not deal with that possibility further.) Mr Johnson accepted that Florin taught the transmission of data (though not specifically the request for the prompt) by a digital channel (Day 2 p 245). This possibility would in my view be particularly obvious to the CA engineer whom I have held to be a member of the skilled team, because such an engineer would be more accustomed to thinking in terms of digital signals. Mr Johnson was resistant to the idea, but I think that that was because he was wedded to his own EPG-based way of thinking and because he was constantly going back to what he thought Florin actually taught rather than obvious ways of implementing that thinking. This digital signal would be the identification signal of the patent.
- That deals with the identification signal triggering the prompt. Florin also teaches that the prompt is displayed for a period at the start and end of each advertisement. Something has to provide the STB with those parameters. They are timing parameters. Mr Johnson's case was that that would be provided by the EPG system, which dealt with the start and finish of programmes - providing information about that was its function and so it could provide the prompt with information around the start and finish which would could be used to generate, and remove, the prompt at the desired times. The duration for the display of the prompt would be hard-wired into the box, in the sense of the duration being fixed by the internal software to the box. At one stage in his evidence he suggested that this would be the EPG software (though his report said "STB software", which smacked to me more of the box's firmware). Thus the EPG system would know that the time had come to display the prompt (because an advertisement was about to start, or was about to end), would demand the display of the prompt and the software in the box would itself control the duration of the display. I accept that this would be one way of dealing with the display start and finish times. Mr Johnson accepted that it would be technically possible to deal with these features via a digital signal but did not think that that would occur to the team. For my part I do not see why it would not. I prefer Mr Turner's evidence about this. The team has (on my finding) already alighted upon the idea of sending a separate identification signal to trigger the prompt in Florin. It would in my view equally naturally occur to them to control the prompt by the same signal. In support of this idea Mr Mellor prayed in aid a perceived need for flexibility - varying length adverts, and perhaps short adverts of a few minutes rather than the hour (or half hour) suggested by Florin. He put, and later submitted (backed by some of the evidence of Mr Turner) that the need to cope with these factors, which would (he said) be apparent to the skilled team, would make it even more obvious to use the digital signal for this purpose, because it would get more complicated for the EPG system to deal with it. I do not necessarily accept that the thinking of the team would extend into this area of flexible programming, but even without it I consider that it would be obvious to use the already obviously available digital signal to control the timing as well as the initial appearance of the prompt.
- These findings mean that claims 2 and 27 of the patent are obvious over Florin. The obvious digital signal is the way in which the viewer is alerted to the presence of an information signal (shown on the screens which result from responding to the identification signal), and it is obvious to have the digital signal deal with the timing of its appearance and disappearance on screen ("at preselected times and for predetermined time increments, without viewer input"). Furthermore, if it were necessary I would accept Mr Mellor's submission that if the timing were not within the signal but were built into the STB (which was Mr Johnson's source of this information) then claim 2 would still be invalid, because the words just quoted would still be demonstrated by the system or method.
- That deals with claim 2 (and 27). It also deals with claim 3, apart from paragraph (ii). The digital signal is an obvious way of achieving paragraph (i), and it is obvious to have it contain the "data defining said preselected times and said predetermined time increments" (paragraph (iii), for the reasons appearing above). However, there was a shifting debate about paragraph (ii) (the identification comprises "data defining an icon or textual matter alerting a viewer to the availability of information or data enabling access to such an icon or textual matter as may be already stored on the computer"). It was not always clear where the debate about paragraph (ii) lay, and in final speeches it was at one point suggested that it was somehow relevant to consider the need for multiple icons. However, that point fell away, and in the end Mr Meade took the point that nothing in the obviousness case developed at the trial made it obvious to have the digital information signal which was implementing Florin contain the icon (or text) to be displayed, so the first alternative of paragraph (ii) was not obvious. That did not seem to be disputed. However, since (ii) contains alternatives, if the other alternative is obvious then the whole claim is then obvious, and there was a dispute as to the second alternative - "data enabling access to such an icon or textual matter as may be already stored on the computer." Mr Meade in effect submitted that as a matter of construction this meant a pointer - the data in the identification signal had to contain the equivalent of a pointer to the icon which was to shown on screen, and he submitted Florin did not in terms disclose that pointer (which was again not disputed) and he said that Mr Turner accepted that his identification signal would not contain such a pointer. Therefore, Mr Meade submitted, this particular integer was not covered by an obvious implementation or development of Florin.
- There was not a lot of focus on this particular point before final speeches, or even in those speeches. In one sentence in his written submissions Mr Meade pointed to a particular answer of Mr Turner given in cross-examination in which he is said to have disclaimed that his obvious implementation of Florin would contain such a pointer:
"Q. I think it follows from that last answer, Mr. Turner,
that none of the proposals that you have made for what to do
with Florin involve a signal sent at the time of the advert
change-over which includes either an icon or a pointer to an
icon. That just does not happen in your suggestions about
what to do with Florin.
A. Well, it is an indicator of the use of an icon. There may or
may not be more icons in Florin, in this situation of multiple
shopping channels, but it is the existence of that message
which effectively says put up an icon that causes the context
of the box knowing that it needs to put up either the icon or
the relevant icon. But there is not something normally in
that sort of message that would say, "Go to address this and
this". Usually it is a situation of the message itself has
its own context." (p573)
- I start by observing that the point is in part a construction point. It was not foreshadowed in any way before Mr Meade's single sentence about it in his skeleton argument (at paragraph 129), and this sentence was not preceded by a suggestion that there might be a construction point about it in his section dealing with construction. That does not of itself mean that it is a bad point, but for such an important point one might have expected a little more emphasis to be given to it.
- Since it is a construction point, before turning to what Mr Turner was actually saying, it is necessary to bear in mind what the wording of the claim is. Although Mr Meade paraphrased it as referring to a "pointer", it does not actually say that. Paragraph (ii) refers to data which "enable[s] access to an icon". That no doubt includes what Mr Meade is referring to as a pointer, but it does not actually talk in terms of pointers. And it probably depends on what one means when saying "pointer". If one talks in terms of "enabling" (which is the word used) then Mr Turner's formulation in the passage of his cross-examination referred to above is capable of falling within the wording of paragraph (ii).
- In considering what Mr Turner meant in his answer (which involves considering the preceding couple of pages of his cross-examination) it is relevant to look at the evidence in his reports. In his first report at paragraph 274 he is considering the new claim 3, having considered the identification signal in the context of claim 2. He says:
"The Skilled Person would understand the TV Shop icon is likely to be frequently used for multiple advertisements. It is therefore most efficient to store the icon locally and simply provide data (in an "identification signal") directing the STB to place the locally stored icon on screen."
- That does not talk of pointers, but it does talk of a form of instruction to produce the icon. That, in my view, is "enabling" within paragraph (ii). He returns to the point at paragraph 38 of his second report, where he says:
"(d) So, in order to create the described prompt at the start and end of each advertisement, signals must be communicated from the head-end to the STB so that the software has the necessary input(s) for scheduling the prompt to coincide with the play-out of the advertisements. This is the solution the Skilled Person would immediately understand and it would be unsurprising and predictable because it was well known to pass control messages from head-end to STB.
(e) The use of such signals is also suggested when considering the nature of the further information in Florin itself. ... As the content sent from the head-end is changing, in real-time, at the start and end of each advertisement, and as the prompt is being shown at the start and end of each advertisement, it would be very odd indeed not to send a message from the head-and at the start and end of each advertisement to initialise the display of the prompt."
- These paragraphs are not directly addressing the question of the form of instruction given to or perceived by the STB, but they do strongly suggest that there is something in the signal which actually triggers the production of the prompt rather than somehow leaving the STB to infer that the prompt is needed. Again, in my view, that is "enabling".
- Having done that, it is now necessary to re-visit the short passage in his cross-examination which I have set out above to see whether he is saying something inconsistent with that. Mr Meade seized on this cross-examination, and in truth it is all that he has got. It must be placed in its context.
- In the 2 pages which precede the passage of the cross-examination just referred to, Mr Meade cross examined Mr Turner about part of his evidence when he was dealing with how a TV channel might wish to update the channel identifier which is stored in an STB and reproduced on screen during programmes. In that context the following exchange took place:
"Q. If we can just think about this in terms of the data that will
be on the box, on the set-top box, I am not saying this is
exactly how it would be done or at the bit level how it would
be done, but if we think of it slightly conceptually.
Somewhere on the set-top box there has to be a two-column
table with the channel number and the graphic appearance of
the icon. That is what you need?
A. Yes. In practice it would normally be a pointer to an area of
storage of graphic images, but, yes.
Q. I understand your point. The actual table would be the
channel number plus a pointer to an address?
Q. And the artwork would be at the address?
Q. But conceptually you agree it is just one ----
A. Yes, it is a one-to-one mapping of channel number. It might
not be channel number but logically it is the same as channel
number. It is something that says this is this virtual
channel or this real channel because channel numbering becomes
a bit -- it is not so -- it is easy in the analog world
usually because there are physical channels. When you get
into digital, then channel numbering is a bit more of an
Q. Yes. So, if there was a desire to update that, that
information, all that needs to be sent is the channel
A. And the image.
Q. ---- and the image. That is the only signalling you need to
A. Yes and that is indeed all I was trying to say here. For
flexibility you would need to be able to download a new image.
Q. That need not be done live or at an advert break.
A. No, I was not trying to suggest that."
- What is apparent from that is that Mr Turner was being very specific about what he was referring to as a pointer – he was referring to a table in software on the machine which would actually contain a reference to a particular address in computer memory at which the information necessary to assemble the icon would be found. It is that sort of "pointer" that he was disclaiming as being part of his identification signal in the passage on which Mr Meade put such weight. He is doing no more than that. In my view there is no inconsistency between the passage relied on by Mr Meade and the passages from his reports which I have quoted and which do not deploy the same detailed concept. In my view his evidence demonstrates that he considered that the process which he had in mind fell within the concept of "enabling", and that that "enabling" arose from data in the identification signal. The contrast (if any) between what he said in cross-examination and what he had said in his reports was not put to him at this point in his cross-examination. In my view Mr Meade can only extract what he seeks from the passage relied on first by misconstruing paragraph (ii) by using the world of pointers rather than focusing on the actual wording ("enabling") and then taking Mr Turner's oral evidence out of context. In my view Mr Turner was not being inconsistent and overall his evidence demonstrates that in his view of the obvious way of implementing Florin there is some of the data in the identification signal which enables the production of the icon (in the sense that it invokes it) within paragraph (ii). I accept that evidence.
- It therefore follows that claim 3 is obvious over Florin, as is claim 2. The same is true, for the same reasons, of the corresponding method claims 27 and 28.
Florin - obviousness of claims 4, 29, 36 and 53
- Next come this range of claims, which introduce the concept of "computer program data" into the mix. Claims 4 and 29 are system and apparatus claims respectively which introduce the concept of the information signal being "computer program data" to the preceding claims which I have considered above. (They also introduce data initiating operation of a program resident on the system, but I deal with that below.) They are as the previous systems/methods, but add that:
"the first information signal comprises computer program data".
- Claim 36 describes apparatus similar to that described in preceding claims, and it is addressed by identification and information signals, but this time with:
"the broadcast information signal comprising computer program data"
- There is "means for storing … the broadcast information signal" and "a processor that …executes the computer program data and superimposes at least a portion of" that signal when the user presses the button on the remote control. Claim 56 is a method equivalent.
- Florin does not seem to disclose broadcasting a computer program within the information signal, and the question arises as to whether this difference between Florin and these claims is bridged by obviousness.
- There was no real dispute as to whether Florin disclosed the equivalent of an information signal. This element of the dispute is about the possible content of it. Mr Turner's view was that the industry was aware of downloading code over the air via cable for use in STBs in the 1990s. He gave some examples of references to the activity - a reference to it in a paper given at a conference in 1993, a joint venture known as OpenTV (referred to below), and a paper published in 1995. Other examples were given. He considered that the skilled person implementing the Florin system had a choice when it came to implementing the screens that Florin put up when the user reacted to the identification signal. He could either have hard-coded screens, which would then appear, or he could provide dynamic screens, which would be provided to the STB by downloadable software. He considered that the latter option was the more likely because it provided greater flexibility. He also pointed out that Florin was deploying computer technology, so thinking in terms of programs would be more natural. The program would interact with, or trigger, a number of standard "building blocks" already stored in the STB. He made it clear in his cross-examination that he did not envisage this download taking place every time a user interacted with the identification signal. It would take place every so often, with the program being stored in the STB ready to be called when needed. This way of dealing with the display meant that the display of one advertiser could be differentiated from that of another, and any advertiser could, from time to time, change that element of the presentation. This was likely to be desirable, and was understood to be technically feasible. The flexibility that this prospect afforded was obvious and desirable, and the skilled team would want to achieve it.
- Rovi accepted that as at the priority date it was known that one could download software but said that it was undesirably risky. Mr Johnson said that the concept was well known but it was also known to be risky. A faulty download could lead to a malfunctioning STB which would then need to be replaced or reprogrammed by an engineer. Accordingly it would not be obvious to the skilled person to include computer program data (executable code) in the information signal. When it was put to him that vendors were selling STBs which claimed to be able to be able to revise their internal software with downloadable software Mr Johnson acknowledged that many of them claimed that they could do that but "To the extent that they did, that is another question I suppose." Mr Meade relied on an answer of Mr Turner given in cross-examination in which Mr Turner is said to have acknowledged that he could not identify an instance of specific applications being downloaded (via downloaded code) for different shops so as to provide different screens before the priority date. If the code was not downloaded and executed "on the fly" (which Mr Johnson characterised as being very difficult) it would require storage; but Florin was said to acknowledge that storage was limited, so that sort of activity would have to be minimised.
- The thrust of Mr Johnson's evidence was indeed that he accepted that downloading program code was common general knowledge, but with the caveat that it was known to be unreliable. He did not say so in terms, and at one stage he seemed to be suggesting that whether or not it was common general knowledge depended on whether his version of the skilled team was correct or whether Mr Turner's was. However, looking at the evidence as a whole I do not think that deciding who is right about the identity of the skilled team is where the answer lies. The thrust of Mr Johnson's evidence was to the effect that I have just recited, with its qualification. The real question is whether the qualification is justifiable to the degree suggested by Mr Johnson (that is to say, a degree which would lead people not to do it). I do not think that it is. I think that Mr Johnson's evidence is from over-caution, and is more reflective of his own views, if anything at all. He expressed scepticism of the claims of STB manufacturers, but without anything to back up that scepticism. His concerns about the risks of the process were expressed in the context of downloading different versions of the operating systems, which I suppose might have been thought to be more perilous than the more limited downloads which are under discussion in this part of the case, but in any event I do not think that the many claims of manufacturers can be easily ignored. Rovi accepted that scripts (simple high level programming instructions) would amount to executable code for these purposes, and I find that they would present even less risk than more complex programming at operating system level. I prefer the evidence of Mr Turner. I find that by the priority date the possibility of sending downloadable code to an STB was common general knowledge, and that the risks of doing so were not perceived at the level suggested by Mr Johnson and were not such as to provide a "lion in the path" of the prospect. As Mr Mellor pointed out, the patent itself suggests the process of downloading code without any suggestion that it was providing the means of avoiding putative lions or demonstrating their absence. The fact that there may not have been any manifestations of this being done in a shopping channel context at the priority date (as Mr Turner accepted, at least as far as he knew) is relevant but not determinative, and is not sufficiently weighty to outbalance the rest of the evidence which suggested it was known technology without the level of anxious risk suggested by Mr Johnson. And reliance on the suggestion that Florin contains a reference to storage being limited is not compelling. The relevant passage in Florin does not actually say that it was limited. It merely contains a reference to the merits of its particular proposals for storing relevant EPG dateaon the footing that the amount of system memory, and therefore cost, would be reduced (see page 33). That is not the same point as that which Mr Meade sought to make about storage.
- Once that conclusion is reached it is in my view easy to conclude, and I do conclude, that it would have been obvious to the skilled addressee to send computer program data (executable code) in the information signal so as to provide differing versions of the screen in Florin which appeared when the user interacted with the prompt and pressed the button on the handset. It follows from that that these 4 claims are obvious over Florin.
Obviousness of claims 4, 29, 40 and 57 over Florin
- This point presents similar issues to those arising in relation to computer program data, save that what is transmitted in the information signal is program initiation data (or wording in claims 4 and 29 which is similar, so far as still relevant). Claim 40 is an apparatus claim which is an alternative to claim 36 (using new claim numbering) and claim 57 is a method claim parallel to 53, in each case introducing program initiation data in place of the computer program data in the information signal. In each case the information signal:
"instead comprises program initiation data for initiating execution of computer program data, and the processor executes the computer program data as directed by the program initiation data when [the user presses the button.]"
- It is important to bear in mind the whole phrase - "program initiation data for initiating execution of computer program data", and I do so at all times even if using the first three words as shorthand. I have dealt with a point of construction on this expression above.
- Virgin's case acknowledges that there is no explicit disclosure of this element in Florin, but it says it would be obvious to implement it by sending program initiation data. Mr Turner's evidence on the point in his first report is very short. He provides a claim chart in which (against the relevant integer) he says it is present in Florin and adds that:
"The instruction to the STB to highlight the correct part of the screen when the user selects the "additional information relating to the advertisements". This must be synchronised to the video."
- He does not identify what the "computer program data" is, or what the computer program is. Furthermore, in his claim chart he does not deal with the expression "executes the computer program data as directed by the program initiation data". In the body of his report he merely says:
"Further Claim 36 requires a "broadcast information signal" which comprises 'program initiation data': this is the instruction to the STB to highlight the correct part of the screen when the user selects the 'additional information relating to the advertisements'".
He does not say any more than that.
- Mr Johnson does not really consider the point in his report; he focuses on whether or not Florin actually discloses this integer, when it is common ground that it does not. In his cross-examination he was taken to figure 45 of Florin and agreed that (on the hypothesis that a separate digital signal was sent, which I have held to be an obvious implementation of Florin) the "signal is initiating the operation of the highlight element on that screen". The cross-examination then went on:
"A. It is a display of the highlight. When you say program, you mean just the display, is anything else happening.
Q. The piece of code that does the highlight?
A. That does the highlight, the display of it?
This is some evidence that the signal was triggering a piece of code. It does not indicate that it was triggering the initiation of anything which might be regarded as a complete program.
- In his cross-examination Mr Turner in substance amplified his short reference in his report (though not by express reference to that report). Mr Meade asked him to amplify what he considered to be "program initiation data". He then embarked on an explanation to the effect that what he considered to be an implementation of Florin involved the triggering of part of some code already in the computer and supplemented with some extra data appropriate to a particular advertiser. Towards the end of his evidence on Florin he said:
"So one would initiate the program and you would pass data across to it which was relevant to that advertiser, or one might have a little bit of script, it would be quite a tiny script, which said, "Build this advertisers screen in this way".
A. And that would not be downloaded, it would be downloaded dynamically in the sense that it might change from month to month or year-to-year but it would not be downloaded at the time that you wanted to show an advertisement.
Q. So saying an advertiser knows it is going to be advertising for a while, so the script is written for that advertiser and that is put in a box.
Q. And then as another advertiser comes on, a short script is written and that is put in the box and then called when necessary. Is that the scenario?
A. That is the scenario, yes." (Day 4 page 582 – 583)
- This approaches the matter in a somewhat oblique way, but I find that Mr Turner was saying that these were obvious ways of implementing Florin, and that what he was saying was that there was an information signal which required some software to run in the machine in a particular way, perhaps with the addition of some data as well. I accept that what Mr Turner was describing was obvious, and that it fell within the meaning of the disputed integer even if it was not apparent that a whole "program" was being run. Something in the nature of a program (or a routine) was being run, and that is within the meaning of "computer program data". Thus this disputed integer, not disclosed in Florin, is nonetheless obvious over it.
- Mr Mellor sought to make a case that other programs could obviously be run, and in cross-examination put to Mr Johnson the running of a calculator program to calculate the amount of sales tax and other elements necessary to make up a final price of there was an on-line ordering of an advertised product. Mr Johnson did not accept that a calculator program would obviously be run, and I agree with that. This line was in my view somewhat removed from the obvious; but Mr Mellor does not need it because he wins on the point in the manner referred to above.
- The wording that I have been considering hitherto is the wording in claims 40 and 57. As I have pointed out, the wording in claims 4 and 29 is different ("data initiating operation of an interactive operating program resident on the system"). Since I have decided that the first alternative in these claims ("computer program data") is obvious, then this claim fails anyway. However, for the sake of completeness, I record that the differing terminology of the 2nd half of the relevant integer in these claims, dealing with initiation, is not materially different from the wording in claims 40 and 57, so if it is relevant this part of claims 4 and 29 suffers the same fate of obviousness.
- Bennett is a paper from 1990 entitled "The Architecture and Security Design Goals of the Eurocypher System". Bennett is the first of the two authors, the second being Paul Moroney. It outlines the architecture of the Eurocypher conditional access system, which was described as an enhancement of a system already in use in the United States, with Eurocypher being intended for the UK market and in particular for a broadcaster known as BSB. It sets out the architecture in a high level manner with a reasonable degree of technicality involved. It is littered with acronyms, the most important of which for present purposes is ACM which stands for Access Control Module. This is a component in a TV signal receiver which processes and descrambles signals sent to it.
- The basic functions of an ACM include "access control for the service selected by the consumer" and "text display capabilities for the receiver". Part of its activities involve making sure that a viewer can only view TV programmes which he or she is entitled to view and to provide messages from time to time to the viewers. Messages can be "personal" or "tiered". Personal messages are what the name suggests – messages intended for a specific user. Tiered messages are messages intended for certain categories of users, the "tier" being that category. Programmes are also distributed according to tiers:
"If the ACM is authorised for the required tier, the receiver is permitted to access the service, either directly, or through impulse purchase."
- Section 3.3 of Bennett describes Impulse Pay-Per-View ("IPPV"). It describes how the ACM provides control features for this, authorising and authenticating purchases and reporting to an external unit via a serial interface provided by the receiver. There are then sections entitled "Audience Survey, Regional Blackout, Circular Blackout and Parental Lockout Control" before one arrives at the most significant part of the article for these purposes which is section 3.8 – "Text Features". The relevant part of that paragraph reads as follows:
"The ACM provides support for menu-driven on-screen displays for the receiver in a 16 by 31 character format, through interaction with the On-Screen Display (OSD) circuitry included in the ACM"
- The OSD is a display which supports a number of text characters. It has a number of capabilities, including the ability to set or change passwords and to provide diagnostic facilities. It will also display the name of the current or next programme. It is the next section which is the most important for the purposes of this action:
"Using the ACM text facility, it is possible to send a personal message from the SAS [Subscriber Authorisation System] to individual ACMs for display by the OSD chip in the ACM. A personal message is limited in length to one screen. Display of the message can be forced or it can be stored for later viewing. A "message arrival" indicator signals the presence of a stored message which has not yet been read. It is also possible to send "tier-addressed" messages to all ACMs viewing a channel which fall into specified classes of access control. These messages are sent from the PCS for display by the OSD chip in the ACM. As with personal messages, display may be queued. Display may also be constrained to lie within a time window, or be of indefinite duration."
- Virgin seeks to mount an anticipation claim, based on this document, in relation to claims 2 and 27. It also uses it to mount an obviousness attack on claims 2, 3, 27, 28, 40 and 57.
Bennett - anticipation of claims 2 and 27.
- Bennett is said to disclose all integers of the claim. It carries normal programming and superimposes text. The message arrival character is the message arrival indicator, and the information characters are the text message. The first identification signal is the message control code of the tier addressed message informing the ACM whether or not the message is for forced display or not, and the first information signal is the content or coding within the text field of the tier-addressed message to be used as the message itself. Figure 2 of Bennett shows a remote control, which provides the control signal from the remote control required by claims 2 and 27. If the message is not for forced display it will be stored, so "storing the first identification signal and the first information signal" is matched. The message is "processed" by the text facility in the ACM (paragraph 3.8 of Bennett) and the message control code must be processed to render the message arrival indicator on screen. That amounts to processing the first identification signal. The coding for the text within the message is also processed by the ACM in order to create the text message – that is "processing the first information signal to provide a second information signal". The message arrival indicator superimposed on screen is the second identification signal and the message text superimposed on screen is the second information signal.
- So far as the superimposition "at preselected times and for predetermined time increments" is concerned, Mr Mellor points to the fact that a stored message may be constrained to lie within a time window or to be of indefinite duration. If the message itself has that quality, then the display of the message arrival indicator (the second identification signal) must be similarly and appropriately time constrained. Therefore this integer is matched. The message (the second information signal) is superimposed when a signal is received from the remote control in response to the message indicator. It will, and can, only happen within the time window which applies to a time-constrained message.
- Mr Mellor's analysis seeks to achieve what he must achieve to make good his anticipation claim, namely that all the elements (integers) of the patent appear, either expressly or implicitly, in Bennett. I find that in relation to these two claims (claims 2 and 27) one key integer is not matched, namely the need for two initial signals - a first identification signal separate from an information signal (which are then respectively turned into second signals of each kind respectively). Both these claims refer to, and in my view require, two signals - an identification signal and an information signal. There is no suggestion in Bennett that there is more than one signal as such.
- Mr Mellor does not seek to assert otherwise. He submits that the patent allows for the two signals to be sent together, and points to a line at which it says:
"The identification and information signals to the tuner and corresponding signals to the computer may be transmitted in single or separate packages."
- There was no real expert evidence as to what that might mean; indeed the sentence does not seem to have been put to either expert. However, the constant reference in the patent in suit to separate signals cannot be ignored. A package is presumably some sort of electronic wrapper in which two things that can fairly be regarded as separate signals can be sent within that wrapper. That is not identifiable in Bennett. There may be some part of the "message" (to use Bennett's terminology) which triggers an indication that it has been received, so that the viewer can call up the message (if it is a queued message), but there is no indication that it is in any sense a separate signal. Even if it is some sort of line of code accompanying the text of the message that does not, in my view, make it a separate signal for these purposes. The same is true if it is some other sort of flag requiring a prompt to be put on the screen. It follows that every integer which is consequential upon the presence of a first identification signal (such as the reference to the second identification signal) is not present in Bennett either.
- The anticipation claim therefore fails for that reason.
- Mr Meade also takes two further points, first that Bennett does not disclose an on-screen flag that a message has arrived (necessary to the patent) and second that Bennett does not disclose the timing elements of claims 2 and 27 (the second identification signal is to appear at preselected times and for predetermined time increments).
- So far as the first of those is concerned (on-screen flag) Mr Meade relies on the fact that Bennett does not disclose in terms that the "message arrival indicator" is something that appears on the screen. It just refers to a "message arrival' indicator" which signals the presence of a stored message. The suggestion was made in the proceedings that this might be an indicator on the face of the STB itself, and an LED was suggested as a candidate. Accordingly, Mr Meade suggested that Bennett did not disclose an on-screen indicator sufficiently unequivocally.
- I do not think that Mr Meade is right about this. The relevant paragraph in Bennett is talking about on-screen displays, and the more natural reading of the reference to the message arrival indicator is that it appears on-screen. Furthermore, while Mr Johnson (Rovi's own expert) suggested that a hardware indicator such as an LED would have been in mind in 1995, because by then it was thought to be a better way of getting a viewer's attention without switching on the TV, he did not suggest that that way of thinking was prevalent at the relevant date for anticipation purposes (1990, the date of Bennett). He said that in 1990 he did not think that any hardware boxes had an indicator light, and the context of his answer was such that he really meant that the reader of Bennett in 1990 would not think in terms of a hardware indicator as opposed to an on-screen indicator. He said that by 1995 one would be thinking in terms of both an LED hardware indicator and an on-screen indicator. In the light of the wording of Bennett and the evidence of Mr Johnson I find that Bennett does disclose a message indicator which appears on-screen.
- So far as the second point is concerned (timing information), Mr Meade submitted that the wording of Bennett about the display of indicators and messages was too unclear to disclose the timing elements to which I have referred. I do not agree with that point. As Mr Turner pointed out, and as I accept, the concept of a time window implies a start time and a stop time, or duration, which is to the same effect as "at preselected times and for predetermined increments" (integer J). It is no vaguer than the wording of the patents. However, the integers require that the second identification signal be passed and displayed in that manner, and since I have found there to be no separate first identification signal there can be no second identification signal either, so this feature of the integers is not matched for that reason.
- I therefore find that claims 2 and 27 are not anticipated by Bennett.
Obviousness of claims 2-3 and 27-28 over Bennett
- In the event (which is the case) that claims 2 and 27 are not anticipated by Bennett, Virgin submits that the differences are bridged by obviousness. They also say that claims 3 and 28, with their more broken down elements, are also obvious over Bennett.
- I deal first with claims 2 and 27. The key differences between claim 2 and Bennett are, as I have held, the absence of a separate identification signal (first and second) within the meaning of the patent, and (it then follows) the automatic display of the second identification signal at a predetermined time and for predetermined time increments. Mr Mellor submitted that the creation of a second, separate, signal containing the material which has to be within an identification signal was an obvious thing to do. He did not point to any evidence of Mr Turner to this effect, and I have not been able to identify any. What he did point to was part of Mr Johnson's cross-examination in which, according to Mr Mellor, Mr Johnson agreed that the difference was "trivial".
- I have studied this passage in Mr Johnson's evidence carefully. He does indeed characterise something as trivial, but what he characterises is not the difference which Mr Mellor relies on. He had indicated that he considered the way that the STB in Bennett would give an indication to put up a message indicator was for the box to interrogate its memory periodically and if and when it found a message it would put up an indicator. It was put to him that sending "data" which instructed the machine to put the message on the screen would be "trivial". Having first expressed himself as having "no comment" on that suggestion, Mr Johnson went on to say that the suggestion and his view of what was happening in Bennett were "equally trivial". What he did not say was that the difference between his view of Bennett and sending a separate signal was trivial. Furthermore, he expressly rejected the idea that this alternative was obvious. He also expressed the view that there was "no particular need" to do it that way. The thrust of his evidence is that what was proposed was not obvious.
- Mr Turner did not seem to express a view on this aspect of obviousness. I was not referred to any of his evidence, and having re-read it I could not detect his expression of such a view, let alone a reasoned view.
- In the light of this I find that this difference between Bennett and claims 2 and 27 is not bridged by obviousness. I heard nothing to indicate why the skilled team, considering Bennett, would bother to consider sending a separate signal informing the STB about prompts. If there was something built into the system which provided for the signal to be put up at a given time, it is not at all apparent why it would occur to anyone to do that in a separate signal; and if there is a piece of the message data which provides for a message indicator to appear on the screen it is equally not apparent why the team would consider it would be useful to strip that out and put it in a separate signal.
- I therefore find that claims 2 and 27 are not obvious over Bennett.
- So far as claims 3 and 28 are concerned, the obviousness attack again requires the sending of a separate identification signal, and since this is not obvious for the purposes of claim 2 it is not obvious for the purposes of claims 3 and 28 either. If he had got over that hurdle Mr Mellor submitted that the other two limbs of claim 3 (data defining an icon or enabling access to a system-resident icon, and data defining the pre-selected times of display of the information signal) were also obvious extensions of Bennett. I think that he is correct about the first. If it is obvious to send a separate signal to signify the availability of the information signal (the text message) then it must be equally obvious that it should do so by the way in which Bennett indicates the arrival of the message, namely an on-screen indication, and to do so by triggering ("enabling") access to the icon that Bennett requires to be triggered. I accept Mr Turner's view that Bennett requires that an existing icon be "pulled up" (Day 4 p629).
- So far as the second is concerned, the point really reinforces the lack of obviousness in sending the separate signal. The obviousness argument requires two steps - first the idea of a second signal, which, even if "trivial" technologically speaking, would not be an obvious thing to do (why would you?) but it then becomes a separate signal with the additional burden of carrying the timing details. It should be remembered that in claims 3 and 28 it is the identification signal that has to contain the "data defining said preselected times and said predetermined time increments". Mr Mellor's submissions focused on the fact that the timing elements of the message ("constrained to lie within a time window") would be likely to be conveyed by the message itself, and that may well be right, but more than that is required for the obviousness case he seeks to make. He needs to establish that it is obvious to send a prior signal (the first identification signal) and to move the timing data from the "message" part into that signal. I heard no evidence which would justify the conclusion that that even more elaborate step would be obvious.
- I therefore find that claims 3 and 28 are not obvious over Bennett
Bennett - obviousness of claims 40 and 57
- This aspect of Virgin's case assumes the obviousness over Bennett of separating out an identification signal (with timing data) and then having the information signal contain "program initiation data for initiating execution of computer program data". This argument fails in the light of my earlier finding that there is no separate signal.
- Had I decided otherwise, Mr Mellor submitted that Bennett teaches a signal which fell within that description because it taught support for a number of text features such as double sized characters, underlining and blinking. These required something in the message which triggered the necessary sub-routine and that was the program initiation data for the purposes of the integer I have just referred to. For the reasons given above when I construed the phrase, I do not consider that this process clearly falls within the description. Mr Johnson explained that the effect would be achieved by something like a control tag with the text, and for the reasons that I have given I do not think that this is sufficient to amount to the "execution of computer program data". It no more falls within that description than requiring the computer to display the character itself, and it was not argued that that fell within it.
- Accordingly, this obviousness claim fails.
- This piece of prior art is part of newspaper article posted on 23rd January 1995 in the Philadelphia Daily News. It takes its name from its author, Jonathan Takiff. The article describes three new products said to have been "introduced" at the recent Consumer Electronics Show, "which merge television and telephone-based technology to produce exciting new communications applications". The third appears under the heading "Interactive video". The section of the article reads as follows:
While several companies have announced interactive operating systems that will let you request information, order products or get movies on-demand through your television set, only the new Open TV system is "affordable and ready for deployment now," announced Dr Norman Koo of Sun Microsystems. Developed in association with Thomson Consumer Electronics, the Open TV box is built around the guts of Thomson's RCA Digital Satellite System receiver. So presumably DSS subscribers will be the first to enjoy the services as soon as next year.
In its set-top decoder box version, the Open TV environment also allows cable, telephone and over-the-air broadcasters to download interactive applications. Whenever a user wants to jump in and feed back, the set-top box sends a message to the phone line via a built-in 1200 baud modem.
At the CES demonstration, viewers were alerted to Open TV interaction with a program by the presence of an on-screen icon. Demonstration of a ticket channel application began with MTV-style music video clips and then overlaid interactive graphics, which allowed you to choose and order tickets to concert events.
An interactive advertising demo showed current commercials and then invited the viewer to request more information, browse through a product catalog or place an order.
A video-on-demand application showed a list of available films and concerts. On making a choice, the selection quickly downloaded to the television through a digital interactive decoder. Wilder still, the viewer maintains full, VCR-style control over the program, including fast forward, rewind and pause."
- That is the sum total of this piece of prior art. Its context has to be considered. It is a short press piece, not particularly technical, about something demonstrated at a consumer electronics show at which (according to the evidence before me) some of the products were often "vapourware" - a product announced, with promised features, but not yet made available. No technical details are given. Mr Meade made much of its lowly status, though he did not go quite so far as to dismiss it on that ground alone. He did, however, submit that bearing in mind its nature, one had to be careful in determining what it disclosed. He said that the skilled team could not determine what some of the functionality was. I agree that this piece of prior art has to be approached with care, though (if it matters) I find that the status of OpenTV as a project was enhanced by the fact, referred to in evidence by Mr Johnson, that a major player called Nethold had announced that it would be building a system on the basis of OpenTV. I find that this would have been known to the skilled team, and would have led to the interest of the team being "piqued" (which was Mr Johnson's word) to a greater degree. The vapours were beginning to condense.
Takiff - obviousness and claims 36, 40, 53 and 57
- Takiff is used as the foundation of an obviousness attack on those 4 claims. Claims 36 and 40 disclose the apparatus with which we are now familiar, and identification and information signals, the means of storing those signals (in each case there is one of each signal, not two, as in earlier claims) and in which the information signals are computer program data and program initiation data respectively. Claims 53 and 57 can be regarded as corresponding method claims. Virgin's case in its final submissions is that Takiff discloses all the integers in these claims save that it does not disclose sending a broadcast signal to instruct the STB to superimpose an icon. There are two ways in which the icon could appear - it could be "burnt in" (that is to say, transmitted as part of the TV signal in the same way that channel identifiers are often made to appear in modern TV broadcasts) or it could be sent in a separate signal. Both are possible, and the second is obvious. Thus the difference between the disclosure in Takiff and the claims it is bridged by that obvious step. Thus these claims are obvious over Takiff. The information signal is the "interactive services".
- Rovi does not accept that Takiff discloses as much as that. It accepts that the document discloses downloadable applications, a back channel (the modem, used by the user to communicate back in relation to the broadcasts), an icon relating to interaction and some overlaid graphics, but there is said to be nothing to show how those elements inter-relate, or how any inter-relation is achieved. In particular:
(a) There is nothing to say how the downloaded applications are made to arrive - nothing to show whether they are downloaded in advance, downloaded on demand, downloaded through consumer choice or instigated by the broadcaster or downloaded following a prompt. If the applications are to be taken to be the equivalent of the information signal the patent requires a download in advance; nothing else is within the claims.
(b) While there is an on-screen icon, one cannot tell from the disclosure what the icon actually signals other than some unspecified interaction, and it is not apparent whether it appears over programming or in a menu of the user interface, whether there is one icon or more than one, whether it is burnt-in or locally rendered, whether it is displayed for a particular period of time and whether it appears on all channels or only on a single channel.
(c) What is meant by a ticket channel is unclear.
(d) It is not clear how the overlaid graphics on the ticket channel are rendered.
- There is no doubt that the article is vague in a number of areas, but that does not mean that it is vague in all areas which are relevant to this case. It does not make Rovi's case for it to point out difficulties and vagueness in the prior art if other parts are clear enough to provide relevant disclosure and to which the skilled team can apply some obvious additions. The real question is what is the disclosure and whether it is clear enough for the purposes of informing the skilled team about the prior art and to establish a base for the obviousness claim alleged.
- I agree with Mr Meade that there is a significant lack of clarity in Takiff. It takes a consumer/layman's view of what was demonstrated, and any analysis of what is going on technically is completely lacking. What is reasonably clear is that a process starts with an icon on the screen, but it is not clear whether the icon is transmitted, whether it is a pointer to a stored icon or whether it is "burnt in" in the TV transmissions. It obviously invites interaction, and presumably something else then happens. It does not make clear whether an "application" is then downloaded, or whether it is immediately made available because it has been previously downloaded. Either is possible, and Takiff introduces both possibilities because the back channel (via the modem) would allow an on-demand downloading
- Furthermore, and crucially for the purposes of a consideration of Takiff, it is not clear what the "applications" are, or at what stage they are invoked. They can be assumed to be computer programs (and therefore "computer program data") but their contents are unknown. Do they control the process once the prompt is responded to, or is there some intervening screen (not amounting to computer program data) before one gets to the interactive application? There is likely to be more than one (and possibly many more than one) way of getting to that point.
- That is an important element of lack of clarity in the disclosure. Mr Mellor's case in his final submissions was that the screens that the user gets when he presses his remote control button (in response to the on-screen prompt) are the second information signal. Those screens may not have involved the execution of computer code - they may simply be the equivalent of text prompts. If that is right then an integer of claims 40 and 53 is missing, and there is another obviousness bridge to cross in choosing to execute an already-resident piece of computer code in lieu of those prompts.
- There are therefore, in my view, at least three links which have to be established via obviousness for claims 40 and 53 to be treated as obvious over Takiff - a separately broadcast prompt (or signal invoking a resident prompt), the prior download of an application (as opposed to its being downloaded on demand) and the invocation of that application when responding to the prompt. As to each of these I find as follows:
(i) The first, by itself, is probably obvious. The alternative is a burnt-in prompt, and a separately broadcast prompt (or trigger) is, I find, equally obvious. Mr Johnson agreed with Mr Turner that that was the case. For the sake of completeness I record that, so far as may be necessary, the information signal integer in these four claims is not matched by a burnt-in prompt. These claims anticipate a separate signal - separate from the TV broadcast. Claims 1 and 27 (which are not set out in this judgment because they are no longer defended) contain different terminology which is appropriate to a burnt-in prompt, and that wording is not reproduced in any of the claims with which I have to deal in this judgment.
(ii) I do not regard a prior download as necessarily obvious. Takiff itself envisages some interaction via the modem which might lead to the downloading of applications - it is not clear. The extent to which it would be obvious to download an application in the form of executable computer code in advance would depend on various factors, such as the size of the code, the number of applications that would require to be downloaded, the memory of the STB, and the availability of routines for removing unwanted applications from the STB. There would or may be conditions in the implementation of Takiff which would make prior download more obvious than others, but bearing in mind the lack of knowledge about details of OpenTV one cannot predicate that they would exist. There may be circumstances which would make download on demand the better course and more obvious to take. Mr Turner did express the view in cross-examination (Day 4 p 594) that he would expect an application to be downloaded in advance of the display of the icon, because otherwise there would not be a sufficiently immediate interaction, but that must depend on a number of other factors such as those I have mentioned in this paragraph. They were not canvassed with him (or with Mr Johnson). In my view obviousness has not been established here.
(iii) The third gap is equally difficult for Virgin to bridge with obviousness. It may be technically possible for a response to an icon to cause the immediate triggering of a previously downloaded program, but it is not necessarily obvious that one should do so. It appears that at some stage an application would be run, but it is not obvious that it would be run on the first interaction. All the viewer sees first is an icon - that shows the possibility of interaction. But the actuality of interaction could sensibly be made to depend on a further choice, which could be presented to the viewer in the form of screens which do not, at that stage, involve the running of a computer program. This was not a point canvassed in evidence, but in the course of final submissions various possibilities of the ordering of events were suggested, and this point flows from them.
- On that analysis, therefore, claims 36 and 53 are not obvious over Takiff. What this analysis also does is to demonstrate a point that lay at the heart of Mr Meade's submissions on Takiff, which is that it is too vague to allow one really to see what the products and processes involved were, and one can only get to the invention by an inadmissible application of hindsight to that vague disclosure (for parallel reasoning, see Technograph Printed Circuits Ltd v Mills & Rockley (Electronics) Ltd  RPC 346).
- So far I have been dealing with claims 36 and 53, which require computer program data to be in a prior information signal and for it to be run as a result of responding to the prompt. The same reasoning applies to program initiation data for the purposes of claims 40 and 57. There was virtually no focus on these claims and that different aspect at the trial, and I find that the parallel obviousness attack on these claims fails for largely the same reason. It is not obvious to add a separate signal of this nature into Takiff and it is not obvious that such a signal would be triggered on the response to the prompt. In my view the lack of attention to these claims in this context was justified, because there is nothing in the attack.
- It follows that these claims are not obvious over Takiff.
A particular point on the apparatus claims - "suitable for"
- Having dealt with the particular attacks mounted on the particular claims it is now necessary to return to a particular point taken by Mr Mellor in relation to all the apparatus claims and foreshadowed above in the section on construction. He says that if the apparatus (system) claims were otherwise valid, then they are all anticipated by Florin because Florin discloses apparatus which is suitable for undertaking all the activities described in all the apparatus (system) claims. He took Mr Johnson through the various essential elements in the disputed claims and Mr Johnson agreed that the hardware of Florin, as hardware, would be capable of carrying out all those functions. The Florin system was said therefore to be "suitable for" all the functions described in the apparatus claims. He submitted that the word "for" in those claims meant "suitable for", and therefore Florin anticipated. He provided a written version of claim 2 in which the expression "suitable for" was substituted for the word "for" throughout to demonstrate how this worked, and how it could be said that Florin anticipated all the integers because its system was "suitable for" all of them in the sense that it would be capable of performing them.
- Not surprisingly, this was not accepted by Mr Meade. He accepted that apparatus "for" a function would be taken to mean "suitable for" that function (see e.g. Virgin v Delta  EWCA Civ 162), but said that that did not mean that Florin's apparatus was suitable in that sense. Were it otherwise then every TV with an STB with some memory in it would be likely to infringe (or anticipate if it came first) even if, as configured in software, it did not even begin to reproduce the key functions of the patent. The answer lay in the fact that the hardware without relevant software which enabled the functions is not "suitable for" the functions in question. A bare computer would not be "suitable for" the activities in the claims because it simply could not achieve them. To Mr Mellor's riposte that if it was software that made the difference then Rovi were in difficulties because the patent would then rely on excluded matter ("a program for a computer"), Mr Meade responded that that would not invalidate the patent because the mere fact that software was a distinguishing factor was not sufficient to make it fall into the excluded matter category - HTC Europe Co Ltd v Apple Inc  EWCA Civ 451;  RPC 30 at para 56.
- This point is irrelevant in the light of the fact that I have found the relevant claims to be invalid for other reasons, but I will express my views on it shortly despite the fact that the position of the point in the constellation of the arguments, and the attention given to it, gave me the strong impression that it was something of a sideshow which Virgin did not rate highly. I consider that Mr Meade's arguments about this are correct. Mr Mellor's submissions would lead to the striking results to which Mr Meade referred. A large number of computer-based patents would be vulnerable to this challenge if it were a good one; or a much larger number of computers would infringe other patents even if not configured in any way so as to reproduce the patented activities. They would infringe merely because they could, in theory, be made to reproduce them. It would be a very striking result.
- In Qualcom Inc v Nokia Corp  EWHC 329 (Pat) Floyd J considered the expression in the context of apparatus which might or might not require physical modification to make it "suitable for" the purposes of the patent. He said:
"73. Nevertheless, one has to be very cautious of any principle of construction which is said to codify the meaning of particular words. Perhaps more importantly in this particular case, it is important not to take the meaning of "suitable for" too far. Mr Antony Watson QC, who argued the case on the 324 Patent for Qualcomm with Mr Thomas Hinchliffe, started from the premise that an apparatus did not cease to infringe merely because it was switched off. So an apparatus for toasting bread infringes whether connected to the mains or not. He says this is just one example, and there is a general principle that an apparatus is still suitable for performing a particular function if it can be readily modified so as to perform that function. Mr Silverleaf accepts that a claim will be infringed if all that is required is to supply power. But he contends that modifications to the apparatus are not what is contemplated by "suitable for".
"74. I think Mr Silverleaf is right. Supplying power to a toaster does not change the apparatus: it simply puts into use the apparatus which is there already. The question in each case is whether the apparatus, as it stands, is suitable for use in that way. If the apparatus has to undergo physical modification before it can be used, then prima facie it is not suitable for use and does not infringe."
- The same reasoning can and should be applied in the present case. A TV with an intelligent STB, as in Florin, may be capable of the processes described in the apparatus claims in the patent but unless programmed in a particular way they will not do so. The programming (whether in terms of firmware, operating system or separate software programs) is the equivalent of the physical modification in Floyd J's example. The mere fact that it is software which contributes to the invention does not void the patent for excluded matter - see the paragraph in HTC v Apple referred to above. Without appropriate programming the apparatus cannot be regarded as "suitable for" the processes of the patent. Accordingly, Florin does not anticipate merely by virtue of its appropriate hardware.
- The alleged infringing product is Virgin's VHD STB. A Product and Process Description was provided and agreed. Since no claim survives there can be no question of infringement. Since there are no material issues on infringement it is unnecessary for me to dwell on it. I merely record again that there would not have been a relevant issue in any event, since it was conceded that if any claim was valid it was infringed, apart from claim 40 in respect of which there was no infringement claim.
- In the light of my findings all the claims that I have considered fall to be revoked. Independent validity is not claimed for the other claims, so all those other claims fail too. It follows that this action fails, it should be dismissed and the counterclaim for revocation succeeds.
Claims of EP '833 as amended
This is an agreed list of the amended claims of EP '833. For convenience, claims have been divided into integers and indented accordingly. Claims are numbered in the format new/old.
A system interactively controlled by a TV viewer remote (35) for superimposing identification (51, 53) and information (55, 57, 59, 61) characters over normal programming (50) on a viewer's display screen (33) comprising:
a tuner (17) having an input for receiving TV signals in a plurality of channels (25) and first identification (16, 26) and information (18, 28) signals,
a first output for passing a signal of any selected one of said channels (23) and
a second output for passing said first identification (16, 26) and information (18, 28) signals;
a computer (11) having input means (13) for receiving any of a plurality of control signals from the TV viewer remote (35),
a first output means for controlling said tuner (17) to pass said signal of said any selected one of said channels (23) in response to a first of said plurality of control signals (15) from the TV viewer remote (35),
means for receiving said first identification (16, 26) and information (18, 28) signals passed by said second output of said tuner (17),
means for storing said first identification (16, 26) and information (18, 28) signals, and
a second output means for passing a second identification signal (19) and a second information signal (20), each corresponding to at least a portion of said first identification (16, 26) and information (18, 28) signals; and
means for superimposing (21, 29) said signals (19, 20) passed by said second output means of said computer (11) over said signal of any selected one of said channels (23) passed by said first output of said tuner (17) to provide a display signal (31) for input the viewer's display screen (33),
said computer (11) automatically passing said second identification signal (19) at preselected times and for predetermined time increments, without viewer input, and
passing said second information signal (20) in response to a second of said plurality of control signals received from the TV viewer remote (35) during said predetermined time increments.
A system according to claim 2, wherein said first identification signal (16, 26) comprises:
i. data indicating the availability of said first information signal (18, 28);
ii. data defining an icon or textual matter alerting a viewer to the availability of information or data enabling access to such an icon or textual matter as may be already stored on said computer (11); and
iii. data defining said preselected times and said predetermined time increments.
A system according to claims 1, 2 or 3,
wherein said first information signal (18, 28) comprises computer program data or data initiating operation of an interactive operating program resident on the system.
A method for superimposing identification (51, 53) and information (55, 57, 59, 61) characters over normal programming (50) on a viewer's display screen (33) comprising the steps of:
receiving TV signals in a plurality of channels (25), a first identification signal (16, 26) and a first information signal (18, 28);
selecting a signal of said TV signals (25) by selecting a channel of said plurality of channels (23) in response to a first control signal (15) from a viewer remote;
storing said first identification signal (16, 26) and said first information signal (18, 28);
processing said first identification signal (16, 26) to provide a second identification signal (19) corresponding to at least a portion of said first identification signal (16, 26);
processing said first information signal (18,28) to provide a second information signal (20) corresponding to at least a portion of said first information signal (18, 28); and
superimposing said second identification signal (19) and said second information signal (20) over said signal of said selected channel (23) to provide a display signal (31) for input to the viewer's display screen (33),
wherein said second identification signal (19) is superimposed automatically at preselected times and for predetermined time increments, without viewer input,
and said second information signal (20) is superimposed in response to a second control signal received from said viewer remote (35) during said predetermined time increments.
A method according to claim 27, wherein said first identification signal (16, 26) comprises:
i. data indicating the availability of said first information signal (18, 28);
ii. data defining an icon or textual matter alerting a viewer to the availability of information or data enabling access to such an icon or textual matter as may be already stored on said computer (11); and
iii. data defining said preselected times and said predetermined time increments.
A method according to claims 20, 27 or 28,
wherein said first information signal (18, 28) comprises computer program data or data initiating operation of an interactive operating program resident on the system.
An apparatus that is controlled by a TV viewer remote (35)
for accessing a broadcast information signal (18, 20, 28)
while an associated broadcast identification signal is being automatically displayed,
without initiation by user input,
with a normal television program (50) on a television display (33),
the broadcast information signal (18, 20, 28) comprising computer program data,
the broadcast information signal (18. 20, 28) and the broadcast identification signal being transmitted to the apparatus,
the apparatus comprising: a tuner (17) having
an input for receiving TV signals in a plurality of channels (25) and the broadcast identification signal and the information signal (18, 20, 28),
a first output for passing a signal of any selected one of the channels (23) and
a second output for passing the broadcast identification signal and the broadcast information signal (18, 20, 28);
means for storing the broadcast identification signal and the broadcast information signal (18, 20, 28) and
a processor (11) that:
receives a user input signal from the TV viewer remote (35); and
executes the computer program data and superimposes at least a portion of the broadcast information signal when said user input signal is received from the TV viewer remote (35)
while the broadcast identification signal is being automatically displayed,
without initiation by user input,
with the normal television program (50) on the television display (33).
The apparatus of claim 36, wherein the broadcast information signal (18, 20, 28) instead comprises program initiation data for initiating execution of computer program data, and
the processor (11) executes the computer program data as directed by the program initiation data when said user input signal is received from the TV viewer remote (35)
while the broadcast identification signal is being automatically displayed,
without initiation by user input,
with the normal television program (50) on the television display (33).
A method for accessing a broadcast information signal (18, 20, 28) while an associated broadcast identification signal is being automatically displayed, without initiation by user input, with a normal television program (50) on a television display (33),
the broadcast information signal (18, 20, 28) comprising computer program data,
the method comprising:
receiving TV signals in a plurality of channels (25), the broadcast identification signal and the broadcast information signal (18, 20, 28);
storing the broadcast identification signal and the broadcast information signal (18, 20, 28);
receiving a user input signal from a TV viewer remote (35); and
executing the computer program data and superimposing at least a portion of the broadcast information signal when said user input signal is received from said TV viewer remote (35) while the broadcast identification signal is being automatically displayed, without initiation by user input, with the normal television program (50) on the television display (33).
The method of claim 53,
wherein the broadcast information signal (18, 20, 28) instead comprises program initiation data for initiating execution of computer program data,
and the computer program data is executed as directed by the program initiation data when said user input signal is received from said TV viewer remote (35)
while the broadcast identification signal is being automatically displayed, without initiation by user input, with the normal television program (50) on the television display (33).
Excerpts from Florin
WO 95/01058 pcT/US94106993