[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales High Court (Chancery Division) Decisions
You are here: BAILII >> Databases >> England and Wales High Court (Chancery Division) Decisions >> Secretary of State for Education and Skills v Frontline Technology Ltd [2004] EWHC 1487 (Ch) (17 June 2004) URL: http://www.bailii.org/ew/cases/EWHC/Ch/2004/1487.html Cite as: [2004] EWHC 1487 (Ch)

[New search] [Printable RTF version] [Help]

IN THE HIGH COURT OF JUSTICE
CHANCERY DIVISION
PATENTS COURT

		Royal Courts of Justice Strand, London WC2A 2LL
		17 June 2004

INTRODUCTION

1. This is an action brought by the claimant for the revocation of European patent (UK) No. 0.664.061 registered in the name of the defendant (the patent in suit). The patent in suit is entitled "Attendance Registration System by Radio Link" and was filed on 7 October 1993 claiming a priority date of 7 October 1992. Independent validity is asserted in respect of claims 1, 2 and 7 and there is a challenge as to the earlier priority date claimed in respect of claims 1 and 2 although it is accepted by the defendant that claim 7 is not entitled to claim such earlier priority. The question of priority date is relevant to the alleged use by an associated company of the defendant, Bromcom Computers Limited (Bromcom) prior to the filing date of 7 October 1993. Hence the attack on the validity of the patent in suit is founded both on the state of the art (including common general knowledge) prior to 7 October 1992 on the assumption that claims 1 and 2 are entitled to such earlier date and on the intervening state of the art on the assumption that none of the claims are entitled to such date.

BACKGROUND TO THE PATENT IN SUIT

2. For many years prior to October 1992 there has been a statutory requirement for schools in the UK to maintain a register recording the presence or absence of pupils both at the start of the morning and the afternoon sessions. Traditionally, paper registers were maintained in the school office and every day the teachers would collect the relevant register for its group and conduct a roll-call, marking the presence or absence of each pupil in its group in the register in indelible ink before returning to the same to the school office. On 1 August 1991, the Pupils' Attendance Regulations 1956 was amended^[1] to allow attendance records to be kept by means of a computer — as opposed to use of paper and indelible ink. This change may have been prompted by the earlier education Reform Act 1988 whereby schools were permitted to manage their own budgets under the local management scheme ("LMS") for schools. In order effectively to carry out such management duties, it was becoming necessary for schools to acquire computers and operate their own IT systems. By way of further impetus, on 20 July 1992, a new category of funding was created by the Government which allowed local education authorities ("LEAs") to apply for, and obtain, funding directed to tackling truancy.
3. As alternatives to the traditional paper/indelible ink registration system, two attendance record-keeping systems had been trialled and are referred to in the patent in suit. First, a swipe-card system which allowed individual pupils to log in. This system had a number of draw-backs including the fact that the teacher did not perform a roll calling operation and hence was open to abuse in that the card holder's presence was not a prerequisite for logging-in. Also the swipe machines/cards were open to vandalism. Secondly, the use of special forms which were completed by the teacher in the usual way and were then read using an optical mark reader ("OMR") which automatically entered the attendance data into the school's computerised database. An example of such a system is relied upon by the claimant which is called SIMS Attendance 3.1 and is discussed hereafter. Such a system suffered from the same defect as the traditional paper based registration system in that the forms for each teacher or monitor had to be collected from and delivered to the school office.

THE PATENT IN SUIT

4. The subject matter of the alleged invention as claimed relates to a student data collection system, more specifically, a school registration system for classroom roll-taking. As such, it relates to a data logging system. It is to be noted at the outset that whilst claim 1 includes classroom roll taking, it also covers the collection of any student data such as exam results. Hence claim 1 is not limited to the requirement of classroom roll taking which, in practice, requires a number of teachers to have ready access to their class lists and be able to take a roll-call, record the attendance data and send it back to the school office all within a short space of time typically within a 10-minute window at the beginning of both morning and afternoon sessions. The patent in suit proposes that instead of a paper-based system, teachers are provided with portable data collection units described in the embodiment as palmtop computers or C-folders which are equipped with RF transmitter/receiver modules alias transceiver means which communicate with a central computer via a wireless network. At paragraph 0013 of the specification, it is stated that in use the C-folder communicates with the central computer system which at the request of the teacher downloads the class list. The teacher then enters student attendance data which, once completed, is uploaded to the central system. Such a system mimics the traditional registration system in terms of the transfer of a class list from the central school office to the teacher, the recordal of attendance information and the transfer of the edited list back to the school office.
5. To allow the C-folder unit to communicate with the central computer as a wireless system, it is necessary that both the C-folder unit and the central computer are equipped with wireless transceiver means, namely means that allow both the transmitting and receiving of RF signals both to and from the portable units to the computer. Where more than one C-folder unit is involved (which is a requirement for any student registration system involving more than one group or class of students), a further requirement of any wireless system must be that there is some form of RF transmission protocol to allow each of the portable units to communicate in turn with the central computer. At paragraph 0018 et seq, one such protocol is described and forms the subject of claim 5. The defendant's expert Mr Copeland^[2] accepts that the components to such a protocol are not novel per se and, indeed, the defendant does not contend that the combined components of claim 5 are inventive if claims 1 and 2 lack inventive step.
6. A further feature described in the embodiment of the patent in suit and the subject of claims 3 and 4 is that the central computer system comprises a network of a number of RF transceiver units ("RTUs") which are wired to the central computer in one of a number of different configurations (eg by a RS-485 serial link). In practice, such RTUs are strategically located around the school/college campus so that all of the C-folders engaged in roll-call taking will be in communication with a RTU.
7. In order to control the data traffic of such a system, the RTUs are provided with a data buffer memory. As Dr Jones, the claimants' expert stated, buffering of some kind is necessary where the radio transmission between the C-folders and the RTUs operate at a different speed to the RTU data transmission to the central computer. However, as described in paragraph 0031, the buffering feature allows the RTU to hold or queue a number of requests from the C-folders whilst the central computer ("server") is acting on a request. Furthermore, at paragraph 0052, a polling feature is described whereby the server polls each RTU in turn asking it if it has any data to transmit to the server or requests for data from the C-folder. In other words, the RTUs are acting as a go-between which allows efficient transmission of data, not just between a single C-folder and the server, but between a number of C-folders interacting with a number of RTUs and the server. Mr Howe QC, for the defendant, relied particularly on para 0026 which states:

"only one active C-folder transmission per RTU is allowed at one time. However, it is possible for C-folders in different zones to initiate RF transmission simultaneously, provided there is an adequate distance between them so they do not detect each other's data carrier."

Mr Howe contends that it is these features the subject of claim 7 of buffer memory and polling (as described above) that permit such parallel processing of data transmission when the folders are sufficiently far apart. A dispute arose as to how far apart the folders had to be located. Whilst I do not consider it relevant to the issues I have to decide, I prefer the evidence of Mr Ali Guryel, the inventor, to that of Dr Jones (the claimant's expert) that, in practice, it would be a simple matter to adjust the sensitivity of the carrier detect in order to implement such a system.

8. The claims which require separate consideration are as follows:

Claim 1: A student data collection system comprising:

(i) a central computer including a data collation program and wireless transceiver means

(ii) a plurality of portable data collection units for use by teachers/supervisors each of which includes wireless transceiver means for communication with the central computer

characterised in that the central computer includes means for downloading respective sets of student data to each of the portable units and in that each portable unit includes manual data input means and a data logging program so that the user can enter student attendance and/or other data to produce an output for uploading to the central computer which is a combination of the downloaded data and the manual input by the user.

9. The following terms are in dispute, namely:

(a) A data collation program;

(b) Means for downloading respective sets of student data;

(c) Portable unit.

In construing such terms, I have in mind that in accordance with Article 69 and the Protocol, the claims should be construed purposively and that the purpose should be ascertained from the description and drawings. In this regard, I also have in mind the judgment of Lord Justice Jacob in Rockwater Ltd v Technip Finance SA (2004) EWCA ((iv) 381 paragraph 41.

(a) Data collation program

10. It was agreed that such a term is not a term of art. Furthermore the description and drawings do not refer to it either expressly or by inference Mr Arnold QC for the claimant states that it simply means a program which collates data and points to the application as filed which states that the data collation program is "for assembling data transmitted from the individual collection stations" This is indeed consistent with the dictionary meaning of collation/collate, namely "assemble information from different sources".^[3] The data being assembled in accordance with claim 1 is the data uploaded to the central computer from the plurality of portable units. Mr Howe, whilst accepting that that is one of its functions, contends that the collation program also manages the assembling of data for downloading to each of the portable units. There is nothing in the description which supports such a contention and whilst a "collation program" may include a program which so functions, it is not a requirement of such a program. In other words, claim 1 includes a system having a collation program which assembles data uploaded from the portable units and which may or may not manage data for downloading to each of the portable units.
11. A further question is whether the collation program is one which operates at a high level such as an application program as opposed to one which operates at a lower level such as a networking program (eg Novell Netware) or an operating system program (eg DOS). Applying a purposive construction, the function of the collation program is to assemble data uploaded from the portable units and it matters not how the central computer is programmed to achieve such a result — particularly as there is no mention in the description of how the collation program is intended to be implemented.

(b) Means for downloading respective sets of student data

12. The dictionary meaning of the term "downloading" is to transfer data from one computer to another.^[4] However, Mr Howe contends that in the context of the patent, the term downloading has a more restricted meaning, namely that the sets of student data are transferred from the central computer to the portable unit in such a manner that the portable unit is able to process the data independently of the central computer. This meaning of downloading, Mr Howe contends, is derived from the meaning of the words "sets of student data". A set of student data is clearly broader than a class list (which is the subject of claim 2) and includes any student data such as examination results, etc. Mr Howe contends that to constitute "a set" of data, the data must be in batch-mode such as a class list rather than pupil by pupil. The analogy he used being the difference between a data logging system "can by can" on a shelf which he classified as immediate transaction mode as opposed to assembling the data for a discrete batch of cans and transmitting data for such batch in one transaction (batch mode). As described in the patent a set of student data (eg a class list) is downloaded (transferred) from the central computer database to each portable unit as a separate user transaction. The downloading of the class list, marking of the list by the teacher and the data transmission to the central computer when all the attendance data has been collected are sequential steps which operate in batch mode as opposed to immediate transaction mode. Dr Jones accepted that both batch and immediate transaction modes would be known to those skilled in data logging systems.^[5] Applying a purposive construction, I consider the words "downloading respective sets of student data" require a transfer from the central computer database to the portable units of discrete batches or lumps of student data to which as the claim goes on to stipulate, the user then enters data to produce an output for uploading to the central computer. What constitutes a "set" or batch of data is not readily definable. It may constitute a small quantity of data or a large quantity which itself may be transferred in smaller quantities. A set of student data is nevertheless in the context of the patent student data such as a class list which constitutes a group of items of whatever number which give rise to some identifiable entity.
13. Mr Arnold contended that the tailpiece of claim 1 which stipulates the user function of entering data and uploading the output qualify the whole system rather than the features of the portable unit. I do not agree. Such user functions qualify the features of the portable unit, namely, its manual data input means and its data logging program. The means for downloading respective sets of student data are provided by the central computer. This is to be contrasted with the data logging program which is part of the portable unit and which allows the user to enter data to edit the downloaded set of student data and hence to produce the output to be uploaded to the central computer where the collation program allows the output(s) from the portable units to be assembled. This division of processing activity between the portable units on the one hand, and the central computer on the other is not only consistent with the description of the embodiment but also with the terms of both claims 1 and 2. In summary, I do not consider it helpful to construe the word "downloading" in isolation. The context of providing means for downloading sets of student data which are to be included as a part of the central computer are to be read together with the data logging program which is a part of the portable unit, the function of which is to allow the user to enter data (such as attendance data) and upload the combined downloaded data and the manual input to the central computer. The processing of downloaded student data by the user entering data to create an output for uploading to the central computer can only relate to the use by the user of the manual data input means, together with the proper functioning of the data logging program provided on the portable unit. Any other meaning would render nugatory the requirement of the data logging program being a part of the portable unit.
14. Thus claim 1 requires the central computer to have means to download sets of student data to the portable unit(s) — in other words, to transfer student data in batch mode, rather than bit/byte by bit/byte. Such downloaded data is then processed independently of the central computer by the users' portable unit using its data logging program (eg by recording attendance data) to provide an output for uploading back to the central computer.
15. Mr Arnold further contends that whether the system operates in batch mode or immediate transaction mode, there is no material difference to the user and therefore the claim covers both modes of operation. I agree with Mr Howe that the working of the invention claimed is concerned not only with the user interface but also the manner in which data is transmitted between the portable unit and the central computer and clearly there is a material difference in the way the data traffic is transmitted between batch mode and immediate transaction mode.

(c) Portable unit

16. A dispute arose as to whether portable units included computers on wheels or trolleys. Portable in the sense used in the patent bears its ordinary dictionary meaning, namely, "able to be carried by hand or on the person; easily moved, carried or conveyed" (New Shorter OED). Illustrative of such units are hand held units such as palmtops and laptops (see claim 8 and Figure 1). Desktop PCs or PCs on trolleys, whilst movable are generally not portable, not being easily carried or moved. The expression "completely portable" used in the patent (paragraphs 0007 and 0016) is used to distinguish between those units which are wireless (radio links) as opposed to wired units, also referred to in paragraph 0007.
17. Claim 2: claims a student data collection system according to claim 1 comprising:

(i) a school registration system in which each set of data comprises a class list

(ii) which is downloaded in response to a request entered by a user into a portable unit whereby the user can then enter corresponding attendance information which is subsequently uploaded to the central computer.

It is to be noted that this claim clearly relates to batch mode transmission, Thus, a class list is to be downloaded from the central computer to the portable unit and only uploaded when the attendance data has been entered — see also paragraph 0009 of the patent.

18. Claim 7 :

claims a student data collection system according to claim 5 or

claim 6 in which each transceiver of the central computer includes a data buffer memory and in which the central computer is arranged to periodically poll the transceivers to determine whether they have data for transmission, or requests for data, from the portable units.

Claim 6 is appendant to claim 5 and relates to the RF protocol between the RTUs and the portable units. claim 5 is appendant to claims 3 or 4 which, in turn, are appendant to claims 1 and 2. Claims 3 and 4 relate to the use of a plurality of RTUs and the various architectures that may be employed relative to the central computer.

19. Whilst claims 1 and 2 are both broadly conceptual, claim 7 (as appendant to claims 1 or 2, 3 or 4 and 5 or 6) relates to the manner in which the concept of the wireless school registration system is implemented. Furthermore as noted hereafter when considering Bromcom's development of their Electronic Attendance Registration System (EARS) whilst claims 1 and 2 include the use of dumb RTUs, claim 7 relates to RTUs which are intelligent namely having processing capability to control the traffic data flowing between the portable units and the central computer.

Skilled addressee

20. In considering the cited state of the art, it is necessary to consider firstly those persons skilled in the art to whom it is addressed and once identified, the common general knowledge which such persons are deemed to possess. There was a dispute as to whether those persons skilled in the art would include someone skilled in wireless technology. Mr Arnold, relying on sections 3 and 14(3) of the Patents Act 1977, contends that because the expression "a person skilled in the art" appears in both the provision relating to inventive step and insufficiency, those persons who are the addressee of the patent when considering whether there is a sufficient disclosure for the invention to be performed are the same skilled persons in the art to whom questions of inventive step are to be determined. Mr Arnold relies in particular on 3M v ATI Atlas (2001) FSR 514 at page 528 where Pumfrey J states:

"It seems to me that as a matter of principle, invention cannot lie in bringing into a notional team working on a particular problem a new notional member with different skills from that of the existing notional team. The specification necessarily describes the attributes of the team to which it is addressed … The addressee is the person likely to have practical interest in an invention ...."

I do not believe that Pumfrey J was laying down a hard and fast rule regarding the bringing in of persons outside the notional team. Inventions may lie in the application of old concepts to a new art and to apply such a rule would in many cases be in effect an ex post facto rationalisation of the invention. Whilst the words "skilled in the art" used in sections 3 and 14(3) are the same, their context is quite different in that the "art" referred to in section 3 must relate to the pre-existing art, ie prior to the patent disclosure whilst the "art" in section 14(3) clearly comprehends the matters disclosed by the patent specification.

21. In Tetra Molectric Ltd v Japan Imports (1976) RPC 547 at 583, Buckley LJ in considering whether it was obvious to combine two different arts (the manufacture of smokers lighters and piezo-electricity) posited the question whether in the state of knowledge to be attributed to the skilled person in the smokers lighters art such a person should have been put on the track of using a piezo-electrical ignition device and having been put on the track, would such a person reasonably be expected to seek advice and assistance from an expert or experts in the piezo-electricity field.
22. To ascertain the relevant art, one clearly has to have reference to the subject matter of the patent. The contribution to the art (eg the problem and the proposed solution) is indeed so determined. Once so ascertained, the European approach (at the EPO) is to determine what is the closest prior art to that problem/solution and the inventive step is then assessed applying a level of skill of the hypothetical non inventive addressee (the sort of person referred to by Lord Reid in Technograph v Mills & Rockley (1972) RPC 346 at 355).
23. The technical field or art to which the patent in suit is directed is student data collection systems being an aspect of data logging systems; in particular, it is directed to school attendance registration systems. At the priority date, the art was one which was in its infancy due to the changes in school management and the use of IT systems therefor. There is no real dispute that those skilled in this art would include persons with expertise in school administration systems who would generally be computer literate. Nor is there any dispute that the notional "team" of those skilled in the art would include a systems analyst, namely an IT person having both knowledge of programming and the design of new systems. The difficulty is that there are a number of specialised aspects in the IT field requiring specialist expertise so that the term systems analyst is an all embracing one covering a number of specialist disciplines. In this case, radio communication is a specialist aspect which an ordinary IT programmer or "systems analyst" may have little or no knowledge of (see Dr Jones XX Day 4/164-165; Day 6/133-134; Wright XX Day 9/76-77). Clearly, student data collection will involve database information technology, namely, the management and processing of data. It will also involve the linking together of different sources of data, eg attendance data classroom by classroom — in other words, the use of networking software.
24. Mr Arnold contends that in implementing the invention it will also involve expertise in radio communications both with regard to the hardware and the protocols to allow radio communications between the portable units and the central computer. However, I do not accept that such expertise in radio communications would have been possessed by the notional team grappling with the problems of student data collection and/or student attendance registration systems prior to 7 October 1992.^[6]
25. The cited prior art fails into two distinct parts, namely wired network systems and the radio communication]wireless network systems. On the assumption that those grappling with the above problems would be presented with both the wired and the wireless systems, I consider that the question boils down to whether the notional team of educational IT persons and general systems analysts having knowledge of database and network systems would reject out of hand those prior art cited wireless systems then available or would consider them worth following up, no doubt with the assistance of a radio communications specialist.

Common general knowledge

A. Excluding radio/network communication specialists

26. The defendant's admissions relating to the common general knowledge as of 7 October 1992 of those skilled in IT educational administration systems and genera! systems analysts can be summarised as follows:

(1) There was a statutory requirement to take and keep registration records of pupil attendance.

(2) Conventionally the pupil registration systems was paper-based where centralised records were kept in the school office, lists of pupils were given to teachers who annotated them with pupil attendance data (the daily taking of the roll-call) and returned them to the school office from which the centralised records were updated. The attendance data was marked with an indelible pencil.

(3) From 1991 the relevant Government regulations permitted the statutory registration record to be kept on a computer.

(4) Desktop personal computers were being used in schools for administrative purposes. A network of up to six of such personal computers could be used to permit data transfer between them.

(5) The SIMS software modules for school administrative purposes were the most widely used modules.

(6) One of such SIMS modules utilising an optical mark reader ("OMR") allowed pupil registration information from the OMR sheets annotated by the teachers to be entered into the computer.

(7) Portable laptop computers with a built-in keyboard and power source were in existence. Such laptop computers had some of the features of desktop personal computers.

(8) Palmtop computers with a built-in keyboard and power source were in existence. Such palmtop computers were able to transfer data to and from a PC by wired means. Although there were palmtops with infra red links, I do not believe this knowledge formed part of the common general knowledge.

(9) Wired networking operating systems having LAN hardware and software (eg Novell) for networking personal computers were in existence. Such systems permitted the user of a computer to access files of one computer (the server) on the network.

Knowledge of wireless LAN systems was not a part of the common general knowledge of the non-specialist systems analyst at that time. It is true that prior to October 1992, articles were being written about such systems (eg NCR's WaveLAN) in magazines such as Byte; June, July 1991 and April 1992.^[7] Byte is a US magazine which was also published in Europe. It was widely read in the UK by those interested in computers (Wright XX Day 9/74). Also there was an article in Computer Weekly (13 February 1992) which was a widely read UK magazine about wireless LANs, headed "Freeing the laptop from the constraints of wire". It is to be noted that in the April 1992 Byte article the author states:

"while there is much excitement and enthusiasm about wireless networks, the technology is still in its infancy and it must overcome a variety of technical and regulatory hurdles before it becomes a major component in today's computer systems."

Mr Wright, the defendant's IT school administration systems expert, fairly states that: "if it was in the general press, then you may expect that people may be starting to be aware of it" (XX Day 9/76).

B. Including a radio communication specialist having radio data network skills

27. Such persons would have knowledge of wireless LANS as part of their common general knowledge. They would have an understanding of both the available hardware and the types of software involving such systems including a general understanding of the RF protocols described in Tanenbaum's Computer Networks textbook which was referred to by Dr Jones as the standard textbook on computer networking in computer science courses (see Dr Jones First Report paragraph 61; XX Day 4/170-172). RF protocols described include the slotted ALOHA Protocol and non-persistent Carrier Sense Multiple Access ("CSMA") which are referred to in the patent in suit (paragraph 0039), and when combined form the subject of claim 5.

GROUNDS OF INVALIDITY

The claimant has raised a plethora of grounds of invalidity set out in its re-re-re-amended grounds of invalidity which I deal with now:

NOVELTY

28. (1) Lack of novelty due to the prior disclosure by Symbol Technologies

Limited and/or Endata of systems comprising portable radio terminals designated PRT 1520 portable units and base stations, by the use of such units

(a) at exhibitions to register delegates' data, including the opening of the Birmingham International Convention Centre in June 1991

(b) at supermarket and cash and carry outlets for retail price checking, eg Anglia Co-op food shops, Safeway and Somerfield supermarkets.

(2) Lack of novelty due to the prior disclosure by Symbol-MSI in the USA by the manufacture and sale of Spectrum One systems comprising portable radio terminals designated LRT 3800 and a central computer. Such units were used in particular for warehouse picking and stock-taking.

29. The Symbol PRT 1520 terminal is linked by means of a short cable to a bar-code scanner (Symbol LS 2000 laser scanner) both of which were hand-held in operation. The PRT unit had a keyboard and a small display. It had an internal micro-processor which was fully programmable and was powered by a battery and it had a RF transceiver broadcasting at about 450MHz using proprietary protocols for communication with an IBM personal computer equipped with its own transceiver. The reason for the particular broadcasting band of 450MHz described as narrow band was that at that time in the UK the regulations restricted the use of unlicensed wireless network systems to such frequencies.
30. The PRT unit could be operated in two modes; first, the interactive mode which allows it to transmit data to the host computer and receive an immediate host response, eg a real time price checking exercise. Secondly, the PRT unit could be operated in asynchronous mode which allows either it or the host computer to transmit one-way data transactions with merely an acknowledgment of receipt. This latter mode only applies where no immediate response is required and is stated to be useful for the transmission of large amounts of data such as data files - see Symbol Programmer's Guide.^[8]
31. The LRT 3800 unit differed from the PRT 1520 unit in two ways. First, the bar-code scanner instead of being linked by a short cable to the terminal was built into the unit as an integral part of it. Secondly, the unit broadcast at about 900MHz, namely broad-band, which was permitted in the USA. This resulted in faster data transaction rates and allowed multiple transceivers to be added to the host computer.
32. Whilst the ambit of claim 1 of the patent in suit does not exclude the use of portable units which are linked to a bar-code scanner, I do not consider the prior use of either the PRT 1520 unit or the LRT 3800 unit renders such claim lacking in novelty. First, claim 1 is for "a student data collection system". The Symbol portable units relate to a bar-code system for price checking, stock-taking and delegate registration, etc. depending on how they are specifically programmed. Secondly, claim 1 requires the portable unit to have "a data logging program so that the user can enter student attendance and/or other data to produce an output for uploading to the central computer which is a combination of the downloaded data and the manual input by the user" The Symbol portable units are not so programmed. To enable them to be so used it would be necessary to reprogram them to function in the manner claimed. Thirdly, it does not appear that the central computer used with the portable Symbol units has a collation program which assembles data transmitted from the units, certainly there is no disclosure that it can collate student data so transmitted.
33. Claims 2 and 7 which are appendant to claim 1 are also novel for the above reasons. Furthermore, such claims are more restricted as to how the student data collection system is to function. None of the features so claimed are disclosed by the pleaded use of the Symbol units.

INVENTIVE STEP

34. In approaching this issue, I have borne in mind the role of the expert as recently expounded by Jacob LJ in Rockwater Ltd v Technip France SA (supra) at paragraphs 11 – 15 regarding the specific attributes of expert witnesses.
35. In this case, Dr Jones, the claimant's main IT expert, is exceptionally well qualified in the field of RF network communications technology in addition to his academic and research expertise in general IT systems. Such expertise is clearly beyond that of the "notional skilled unimaginative addressee" grappling with the problems of student data collection and registration systems relating to student attendance data. The expertise of Mr Copeland, the defendant's main expert was not academic or research orientated, but more of a practical nature. He had experience of the use of Symbol PRT terminals as data collection units as part of a radio data network whilst working for Amtrak in 1992 (see XX Day 9/138-150).
36. Criticism was made by Mr Arnold of some of Mr Copeland's evidence. In particular, it is said that he sought to argue the case against the weight of evidence to the contrary. A good example was the question of the availability by October 1992 of wireless LANS. Mr Copeland distinguishes wireless LANS from radio data networks (see XX Day 9/150) and his evidence was that NCR Wave LAN technology was untried and untested and was not available to him (in the UK) (XX Day 9/153-154). There seemed to be some confusion between knowledge about wireless LANS, something Mr Copeland accepted, and their availability prior to October 1992 in the UK which he did not accept. I do not think Mr Copeland was seeking to argue the case anymore or less than Dr Jones was putting forward possibilities which on occasions best suited the claimant's case, eg his evidence regarding the Priority Document disclosure, dealt with hereafter, and his evidence regarding the volume of traffic on SIMS Attendance Module (XX Day 5/154-178). A further criticism of Mr Copeland's evidence was that he had changed the wording of an article (the Carnegie Mellon article) regarding file sharing architecture. As he rightly explained, this change was because SIMS/Kingshurst systems did not download files to a client's own disk, but rather accessed files in the server. Making due allowances for the above, I have found both Dr Jones and Mr Copeland, whilst seeking to put forward their respective parties' best case, to have been fair and provided clear and helpful evidence to the court. No criticism was made of the parties other "educational" IT experts, namely Mr Topham (for the claimant) and Mr Wright (for the defendant).
37. As regards the principles to be applied in considering the question of lack of inventive step, the defendant relies on the well known warnings against ex post facto analysis in Fletcher-Mou1ton LJ's judgment in British Westinghouse v Braulik (1910) 27 RPC 209 at 230, namely:

"I confess that I view with suspicion arguments to the effect that a new combination, bringing with it new and important consequences in the shape of practical machinery is not an invention, because, when it has once been established, it is easy to show how it might be arrived at by starting from something known and taking a series of apparently easy steps. This ex post facto analysis of invention is unfair to the inventors and in my opinion is not countenanced by English patent law".

I consider such warnings are particularly apposite when considering claim 7.

38. Bearing the above warnings in mind, I shall consider the question of obviousness, following the structured approach of Windsurfing v Tabur Marine (1985) RPC 59 at 73-74.

Step (1): Inventive concept

39. In its broadest aspect (claim 1) the inventive concept is for a student data logging/collection system which comprises a central computer linked by wireless communication (eg radio) to a number of portable units which allows sets of data to be downloaded to the portable units, where the data can be manually edited (eg student attendance data can be entered) and which allows the user(s) operating the portable unit(s) to upload the combined downloaded and added data back to the central computer where the same is collated. This inventive concept is particularly applicable for use as a classroom roll taking system as part of a school registration system (claim 2). In such cases, class lists are downloaded in response to requests entered by the teachers/supervisors into their respective portable units.
40. In its narrower form (as claimed in claim 7), the inventive concept relates to the implementation of the above concept and comprises a combination of the following features:

(i) the requirement of a plurality of wireless transceivers connected to the central computer (claims 3 or 4);

(ii) a specific RF protocol allowing each of the portable units' transceivers to communicate with the central computer's transceivers (claim 5); and

(iii) the requirement that each central computer's transceivers should have a data buffer memory and that the central computer must periodically poll the computer's transceivers to control the incoming traffic from the portable units (the intelligent RTU).

Step (2): The common general knowledge

41. Already identified.

Step (3): The differences between the prior art and the alleged invention

Step (4): Whether those differences would have been obvious steps

42. I shall consider steps (3) and (4) in the light of the following cited prior art:

(1) Symbol

(2) (a) Kingshurst

(b) SIMS

(3) Common general knowledge

(4) Gombrich

(5) Nobles.

SYMBOL

43. As already described, the Symbol portable units PRT 1520 and LRT 3800 were special purpose networks which as Dr Jones explains (Final Report paragraphs 64 - 67) were packet based networks which had been optimised for specific data collection purposes using proprietary software and protocols. The Symbol units were designed to operate in conjunction with a bar-code scanner although data could be entered via the keypad as a fall back if the bar-code reader malfunctions (Fryett XX Day 2/119-120).

(i) PRT 1520 — Differences (Step 3)

44. Because this portable unit operated on a narrow band transmission, only one unit could communicate with the base station at any one time. There was also a problem of loss of data due to outside interference (eg from other PRT units). As Mr Fryett (First W/S paragraphs 15 to 18; XX Day 2/107-109) explains, Symbol developed their own proprietary protocols to deal with these problems. As regards the architecture, the base station capacity was stated to be up to 8 PRT units — see The Programmers Guide Manual.^[9] As Mr Fryett stated (Third W/S paragraph 2) since the RF link was narrow band, a single base station would generally be used. Thus, although more than one base station connected to the central computer with separate wires was possible, in that event each base station would require different radio channels to avoid interference. Mr Fryett admitted that the suggested manner in which each base station may be connected to the central computer using standard RS-232 linkages (a serial link) managed by conventional "hand shaking" to control the flow of traffic was one he would not recommend (see Fryett XX Day 2/139-141).

[INSERT DIAGRAM HERE]

Architecture of PRT Units
45. None of the pleaded prior uses relating to the PRT unit used more than a single base station. In operation although the PRT unit had a RAM programmable memory, as regards the pleaded use, the manner in which such units were programmed was to operate in what Mr Howe referred to as "immediate transactions" or immediate transaction mode as opposed to batch mode (see Duggan XX Day 3/35). Thus when used for retail price look-up, the scanner read the bar-code on the can which the unit immediately transmitted to the central or host computer can by can rather than storing up data for a series of cans before transmitting the data to the host.
46. As explained above the PRT unit whether operating in the immediate transaction mode or the batch mode could function interactively or asynchronously. Interactively the host computer would process the request and provide the correct price to the unit. The only difference when functioning asynchronously is that instead of being in the form of a request, the data sent by the unit is in the form of information, eg a picking list (used in warehousing operations like a shopping list). In this case, the host computer sends an acknowledgement back to the unit that it has received the data (See Fryett XX Day 2/121-129).
47. The Programmers Guide Manual ^[10] specifically directs anyone who wished to program the PRT unit to operate in batch mode to do so. Reference is made to the PRT's use in asynchronous transaction mode where by way of example it is stated:

"the host may first send the PRT a data file containing a list of each piece of equipment to be measured The PRT operator can then go to each piece of equipment and enter the production numbers into the PRT After completing the data collection process, the operator has created a production run data file in the PRT's memory. This data file can then be transmitted to the host computer as input to the company's production control system .."

This example is illustrated by simple pictorial diagrams of in effect downloading of the list and uploading of the completed production run data file.^[11] The guide manual (some 430 pages long) was pleaded as part of the state of the art in support of the user of the LRT 1520 unit, although no specific reliance was placed on this example as a disclosure per se However, it is the case that anyone wishing to reprogram a PRT unit to allow it to function in the manner suggested in the manual would have no difficulty in doing so.

48. There was some evidence (not pleaded) relating to the use from about October 1990 by the John Lewis Partnership of a bar-code system, designated Move Stock Application which was stated to operate such PRT 1520 units asynchronously in batch mode as opposed to immediate transaction made. Although the specification relating to how the system operated was confidential (Duggan XX Day 3/3 8), such use was relied upon as demonstrating that the PRT units could be reprogrammed to operate in batch as opposed to immediate transaction mode. How successful such a system was is open to doubt. Mr Stephenson's evidence was that the Symbol asynchronous transaction mode systems he came across whilst working for Symbol in 1995 were unreliable and regularly lost transactions (XX Day 8/104-106).
49. From the foregoing, the differences between the invention claimed in claim 1 and the pleaded PRT 1520 systems can be summarised as follows:

1. The PRT systems were bar-code systems which, without reprogramming the PRT units and the related host computer, were not capable of being used as a "student data collection system".

4. As programmed, the PRT units were part of a special purpose radio network operating in immediate transaction mode — as opposed to batch mode. It follows that as so programmed, the units did not have a "data logging program" that the user can enter student attendance and/or other data to produce an output for uploading to the central computer which is a combination of the downloaded data and the manual input by the user.

3. The central computer did not have a collation program for assembling data transmitted from the portable units.

4. The central computer did not have means for downloading respective sets of student data, or in case of claim 2, a class list in response to a request entered by a user into a portable unit.

50. As regards claims 3 or 4, whilst the use of a plurality of wireless transceivers connected to a central computer was a possibility, the systems as pleaded and relied upon only had a single base station. As regards claim 5, the radio protocols were proprietary although they could have been ascertained by reverse engineering. Mr Fryett (XX Day 2/107-109) states that he had not been able to get full details from the U.S. It does not appear to have operated in the way claimed in claim 5 where the transceivers of the central computer (ie more than one) are arranged to generate periodic signals indicating free slots for transmission. As regards claim 7, there was no polling of the transceivers of the central computer nor were there data buffers on the transceivers to regulate the flow of data traffic from the portable units to the central computer, As stated above, Mr Fryett disowned use of some sort of "hand shaking" management of the base stations — something that in any event was not the subject of the pleaded prior use.

(i) PRT 1520 — Inventive step (Step(4)

51. The first question is whether in October 1992 to the notional team of skilled persons in the educational IT art and a general systems analysts, the use of the Symbol PRT 1520 wireless unit as then being employed in conjunction with bar-code scanners for retail and warehousing applications as well as for registration purposes at exhibitions, would be useful in the field of student data collection and, in particular, school roll-call taking. Bar-code systems had been suggested for schools in place of the paper based OMR registration system. In operation, a register having specific bar-codes for the class and individual pupils could be scanned by a teacher using a hand-held pen and the attendance data so read and stored could be fed into a central computer in the school office (see Topham XX Day 4/101-102). An alternative bar-code system which Mr Topham considered could have been used for school attendance following on from its use at exhibitions for registration of delegates (eg the opening of the Birmingham International Conference Centre) would be if each student had a bar-code label either to be worn on their lapel or carried as an alternative to a swipe card (See Topham XX Day 4/130-131 and Jones XX Day 6/62-63). The practical problems of using such a system would be similar to that of swipe cards, namely, swapping or losing cards.
52. If the Symbol system using PRT 1520 units in the manner used as pleaded were to have been implemented in schools, transmission of data to the host would have been in immediate transaction mode, (pupil by pupil) as opposed to batch mode and would not fall with claim 1. However, as referenced in the Programmers Guide Manual (supra) and stated by Mr Copeland it was possible to program the unit to capture data, (eg a class list downloaded from the host) consolidate it (eg add attendance data to the whole list) and transmit it back to the host. The question is whether that was an obvious way to proceed.
53. It was Mr Copeland's evidence that, in his experience, the use of the PRT units at Amtrak in 1992 was not particularly reliable (First Report paragraph 6.3 XX Day 9/146 and Day XX 10/32-33). Two such units each with a hand-held bar-code scanner and a portable label printer formed a radio data network with the depot PC which allowed data to be transmitted from the portable unit to the depot PC and from the depot PC to the host computer (See Copeland XX Day 9/139-141). As Mr Copeland stated, compared with a hard wired configuration using a fixed workstation directly cabled to the host computer which provided a consistent and quick response, the response using the radio link varied tremendously (Day XX 9/146-147). Mr Copeland's experience was that whilst the radio network system with one user (ie only one PRT unit being used at a time) was okay, with two users (ie both units being used at the same time), the system deteriorated quite badly (Day XX 9/148-148).
54. Instead of PRT units being programmed in the immediate transaction mode, but in the batch mode (eg John Lewis), the evidence as to reliability was certainly no better. Indeed, at that time, if one wished to operate such units in batch mode, the more reliable and lower cost way to proceed was to use a portable unit with a cradle or docking station which was wired to the central computer. (See Stephenson W/S para 8; Christopherson W/S para 8.) An RF link was only considered to be worth the extra cost where it was important that the data transactions, eg the checking and/or collecting of data were processed in real time. The importance of reliability in terms of attendance registration for schools was referred to by Mr Topham not only because of statutory obligations to maintain accurate records but also for fire drill purposes. (See Topham XX Day 4/112-113; 138.)
55. An issue in the case which has been referred to as scaleability is that even if the then existing RF Symbol systems could only be operated with a small number of portable units, eg limited to say two to three classes, Mr Arnold contends that if it were obvious to do so in the manner claimed, then the invention is obvious as the claims only specify that there should be more than one unit. That proposition is clearly correct. But when considering the type of student collection system to be deployed, eg to replace the existing traditional paper system, it is relevant to ask first whether the paper system needs replacing at all (eg where only two or three classes), and if so, as a matter of proportionality, the changes required and the cost involved compared with the likely practical value that will result if successful. Schools and colleges vary in sizes from anything between 50 to 3,000 students and registration groups or classes between 5 to 150 (Copeland First Report 7.2.1). As Mr Copeland states (First Report 4.2.4):

"what works and is economic for 20 users may not work or be economic for 200 users ... For a system to be scaleable, it needs to be designed so that additional processing, storage and networking resources can be easily added as the throughput generated by extra users increases."

56. Taking the following into account:

1. The relative unreliability of the Symbol PRT 1520 RF Systems at that time (1992) compared with the reliable wire equivalent systems and their lower cost.

2. The limitations of the Symbol PRT 1520 RF Systems in terms of number of portable units that could be practically utilised (ie less than the specified eight maximum).

3. The fact that the Symbol PRT 1520 System was a bar-code system linked with a bar-code scanner.

4. The fact that the PRT 1520 System utilised a proprietary special purpose network, i.e. not known and not subject to any standard; this latter matter would have been a concern to the educational purchasing authorities. (Topham XX Day 4/138-139)

I do not consider that it would have been obvious to those skilled in the art (namely the IT educationalist and the general systems analyst) to reprogram the Symbol PRT 1520 unit to function as part of a student data collection system in the manner claimed in claim 1.

57. It follows that for the same reasons neither claim 2 or 7 is obvious. Furthermore, claim 2 being limited to the downloading of a class list for the purpose of roll-call taking implicitly requires those teachers/supervisors using the portable units to be able to conduct the roll-call taking exercise concurrently, typically within a 10 minute period for both morning and afternoon sessions. This in itself, will present problems in the design of the system. claim 7 (as appendant to claims 1 or 2, 3 or 4 and 5 or 6) provides an implementation of such a system using a plurality of intelligent transceivers connected to the central computer which is not in any sense foreshadowed by such use.

(ii) LRT 3800

58. In the LRT 3800 unit, the bar-code scanner is an integral part of the unit. Mr Fryett was unwilling to accept that both it and Spectrum One network (the broad-band RF protocol) were designed for bar-code data transactions (See Fryett XX Day 2/88-96). However, it is quite clear that the LRT 3800 unit was so designed. Mr Fryett in fact was only concerned with the RF technology of the PRT unit. It is accepted that the LRT 3800 unit was capable of operating both faster and with more portable units. Mr Christopherson (W/S paras 5 to 10) gives evidence as to its capabilities prior to re-engineering it. His evidence was that it could barely handle 10 portable units operated as dumb terminals. For the same reasons why it would not have been obvious for those skilled in the art to reprogram the PRT 1520 unit to operate as a student data collection system in accordance with claim 1, I hold claims 1, 2 and 7 to be inventive over such prior use.

KINGSHURST/SIMS

59. Both the Kingshurst and SIMS prior disclosures relate to wired (ie not wireless) IT school attendance registration systems which operate on a file sharing architecture. It is common ground that in terms of architecture no distinction is to be made between them.

(a) KINGSHURST —Differences (Step3)

60. Kingshurst City Technology College, North Solihull was a purpose built school which opened in September 1988. At the time it was considered to be a "high tech" college. By the second year there were 16 teaching groups with an IBM PC computer networked into every classroom. As the number of teaching rooms expanded the network was expanded up to some 30 IBM PCs for both teaching and administrative use. The network architecture was a token ring and included up to three file servers, The IBM PCs were located either close to a teacher's desk or on a trolley for ease of being moved. As Mr Halliwell-Ewen states (W/S paras 6 and 7) two software systems were used, the operational differences between the two not being material. The first system used in the first academic year was written by Greg Marshall and used Paradox 2 (a network compatible database language). This was replaced in September 1989 by a system called the Tulip system which was developed by Martyn Bryant (an employee at the time of Tulip Computers), using a language called Clipper. Both systems operated as follows:

(1) The teacher using a classroom IBM PC would log onto the system.

(2) The system would download a list of students from a central computer (server).

(3) The teacher would enter the presence or absence of students.

(4) The resulting attendance data would be uploaded to the central computer.

(5) The uploaded data would then be available to be printed or accessed elsewhere on the network.

In addition to the daily attendance registration, the system was also used for administration purposes such as the enrolment of students and accessing other student data.

61. The central computer/file server contained the students' database files which could be accessed by a series of keys or index. The software running on the server to organise and store data in the database files included Netware software (eg Novell Netware) at the network interface level and BIOS (Basic Input Output) at the index level. On the client workstation or teachers' classroom PC an executable applications program loaded into the client PC RAM allowed the teacher to control the transfer of the data (eg class lists) from the file server to the client workstation. Mr Howe described the "upload" manner in which the client's applications program worked as being analogous to each teacher "pigeon holing" the attendance data it produced into the correct "pigeon holes" of the database by means of the client's applications software and the use of the file server's keys or indexes. Mr Howe contends that an arrangement of keys or indexes does not constitute a "collation program" and it is the client's application software that is doing the collating of attendance data. Mr Arnold accepted that at the application level it is the case that the application program running on the client PC that controls the processing of the attendance data. However, as explained above at the organisation and storage level (file management processing) the network software and BIOS software is responsible for the "pigeon holing" of the data. In other words one cannot function without the other (see Jones XX Day 5/126-132). The patent in suit's only reference to a data collation program is that it is included with the central computer. The Kingshurst file server is managed by the network operating system and has file-based software controlling things at file level. Operating in such a manner the file server does include what adds up to a data collation program or at any rate it provides a data collation function of allowing data from different client PCs to be collated and stored in the database file.
62. There was a question as to whether the data transferred from the file server to the client PC was "downloaded" to the client. It was Mr Marshall's view that the student data (eg class list) requested by the client PC once pulled out from the database files is held in the Novell Network cache on the server, which may be written back to the server's hard disk if it requires updating. The data requested by the client PC is transferred from the server to it in frames and packets across the wire for processing by the client before uploading to the server (see XX Day 3/130-133). I have no doubt that such requested data is being downloaded from the server to the client PC.^[12]
63. Finally there was the question of whether the client PCs at Kingshurst were "portable" units. Whilst some of such units were on trolleys and hence once unplugged they could be moved from room to room, I do not believe them to have been portable within the context of the patent in suit, namely something which is not just moveable but also can be readily carried.
64. It follows that the differences between the data collection system the subject of claims 1 or 2 and the Kingshurst system are:

(1) the use of a plurality of portable units;

(2) that the portable units are wireless units and hence the requirement of transceiver means both for the central computer and each of the units.

Inventive step (Step 4)

65. Mr Latham (the art and design co-ordinator at Kingshurst from 1988 to 1996) refers to the 10-minute roll-call taking registration period in the morning and the fact that the whole registration process only took a few minutes (XX Day 3/55-56). I do not accept there was a delay problem with the Kingshurst system. It is true that Mr Lynch (Science and Technology teaching responsibility at Kingshurst from 1988 until mid-1990) states that a reason why Mr Bryant was asked to write the new Tulip system was due to the poorer performance of the Marshall system. This was due to the Marshall system being written in Paradox which was an interpretative language whilst the Bryant/Tulip system was written in Clipper which was a compiled language which is more efficient and therefore faster (see Lynch XX Day 3/89-93). However Mr Lynch's evidence was that the Kingshurst system was perfectly workable and logging on was no slower than waiting for 2 or 3 minutes for a pupil to fetch a paper register (XX Day 3/85-86). Mr Lynch who holds strong views as to the technical merit of the patent in suit also claims that he is something of a perfectionist and could have written the Kingshurst system in a lower level language C (BBC Basic) to make it even more efficient.
66. The claimant contends that replacement of the IBM PCs in the classroom with portable hand held units eg laptops or palmtops wired in the same way as the IBM PCs were wired would not have involved any change in the system from a technical point of view. In other words the portable laptops would be networked using the same software and same file servers and would function in an identical manner. I do not believe there is any doubt that would be the case. It is common ground that both portable laptop and palmtop computers were common general knowledge. The advantage of a portable unit as opposed to a desktop PC is that being portable it is user friendly, it occupies less permanent or semi-permanent space and can be stored away when not in use.^[13]
67. The question is whether the idea of substituting wireless portable units in place of wired portable units would have occurred to those skilled in the art (an IT educationalist and/or a general systems analyst) in 1992 and if so whether the idea would be considered worth pursuing eg by involving a radio communications IT specialist.
68. Whilst I have held that knowledge of wireless LAN systems was not part of the common general knowledge to the non-specialist systems analyst, nevertheless by October 1992 the concept of wireless networks was becoming known from articles appearing in Byte and Computer Weekly albeit it was still in its infancy. That the concept of wireless LANS was generally known can be seen from the evidence of a number of defendant's witnesses^[14] including Mr Copeland and Mr Wright, the defendant's experts.^[15] There was a dispute as to availability in the UK of wireless LAN products. Whilst Mr Copeland's evidence was that they were not available to him in the UK, Dr Jones's evidence was that they were commercially available and had been for some years prior to 1992. I believe Dr Jones to be correct, not only because of the references to NCR's wireless Wave LAN product in the Byte articles but also the defendant sought to rely on a paper^[16] about a student electronic notebook devised by Columbia University pre-1992 to show that the off-the-shelf wireless network cards referred to (the NCR WaveLAN card and the ARLAN card) were full-size ISA cards which would be too bulky to insert into an ordinary laptop. However size did not deter Columbia University using such off the shelf wireless network modules. The cards were merely housed in a separate enclosure connected to the back of the portable unit. In fact, Columbia University had developed a wireless network system using a number of Toshiba 5200 portable work stations (notebook size laptop) equipped with NCR WaveLAN cards, as well as a system using an ARLAN wireless network card. Whilst these developments at Columbia University were not cited as prior art, they do provide support that the use of wireless LAN cards with laptops that were generally available not only was possible with relatively minor alterations but had actually been done. Furthermore, Dr Jones had located a brief case size laptop (Grid case) that was available at the same time which did not require any alteration at all; something no doubt the radio communications specialist would have considered doing had he been asked to investigate the use of wireless LANS (see XX Day 5/75; Day 7/138).
69. It is true that wireless LANS were only in their infancy in 1992 and there were questions about their speed, as well as regulatory considerations, Generally it was felt that compared with a wired LAN a wireless LAN ran at about half the speed. At Kingshurst the IBM token ring was running at 4 Mbs and with up to 30 IBM PCs it was found by users to be perfectly acceptable. NCR WaveLAN ran at 1.2 to 2 Mbs which would indicate that with the same file sharing architecture a wireless system managing eg 10 to 15 portable units would result in similar performance.^[17] As Dr Jones stated, substituting a NCR WaveLAN card in place of the wired network cards in the Kingshurst system could be done without any modification to the system.^[18] It was his evidence that whilst one would expect changes in performance it was not necessarily the case that the radio network would be slower depending on the efficiency of the token ring.^[19]
70. The primary reason for a wireless system versus a wired system was the elimination of the need to install cables particularly in old buildings or buildings where it is difficult to lay cables. Another advantage is that the user of a wireless portable unit does not have to worry about wires linking it to the network, Mr Wright, the defendant's IT educationalist expert accepted that by 1992 the advantages of portable battery-powered laptops and palmtops were well known (XX Day 9/89, 94-96) It was suggested to Dr Jones^[20] that a systems analyst at that time would have shied away from a radio network solution to the problem. Dr Jones' evidence was as follows:

"some people would have shied away from a token ring solution to the problem because they would have thought that was too complex perhaps, but some people would have embraced it and thought that was the right solution. That is why you see so many different systems out there, because everyone has their own particular experience of different sorts of networks. Token rings were a harder network to manage than some other networks, for instance Radio networks tend to be very easy to manage.

(Q) It may be easy for you to manage Dr Jones.

(A) Adding a terminal to a radio network does not involve a cabling process and assigning of parts and that sort of thing."

71. I consider that the concept of a wireless network system using portable laptops or palmtops in place of the wired IBM PCs in the classrooms to be no more than a logical alternative to the Kingshurst system and that such a concept was obvious in October 1992 to a team of someone skilled in the IT education field and a general systems analyst considering the Kingshurst system. I believe that to implement such a concept such a team would have consulted an IT specialist in RF networking regarding his or her common general knowledge of LAN products and how best to design the network including what RF protocols to use. Following the manner in which the Kingshurst system operated (as set out above) using wireless laptops or palmtops instead of wired desktop IBM PCs, would inevitably result in a system falling within claims 1 and 2.
72. Whilst I consider claims 1 and 2 are conceptual claims, claim 7 relates to how a RF system may be implemented. This is indeed reflected by the fact that the defendant through its associated company Bromcom only devised its EARS system the subject of the patent in suit in the period October 1992 to October 1993. I do not consider claim 7 relating to the use of intelligent RTUs connected to a central computer as appendant to claims 1 or 2 and 3 or 4 and 5 or 6 is obvious in the light of the Kingshurst system. The requirement of a number of transceivers linked to the central computer and the manner in which they control the traffic flow of data to and from the portable units by means of the RF Protocols the subject of claim 5 combined with the use of the RTUs controlling the flow to and being polled by the central computer is I believe only with hindsight something those skilled in the art with assistance of a RF IT specialist could have arrived at.

(b) SIMS

73. SIMS was a school administration software package installed on hardware consisting of a central file server (generally located in the school's front office) and a wired network of three to four PC workstations. A number of employees of Capita Business Services Ltd (the successor in title to Sims Limited) gave evidence regarding the working of the system and in particular to its Attendance module (known as Attendance 3.13). It was pointed out by Mr Howe that Capita had an interest in the outcome of this dispute and wished to compete with Bromcom in the provision radio based attendance registration systems. That may well be the case but it was not suggested that I should disregard the evidence of such witnesses on this account.
74. Attendance 3.1 was a module run on the SIMS network which was a paper based attendance registration system which allowed the data entered on the form lists by the teacher/supervisor at the time of the roll-call taking to be read by an OMR and thereby automatically entered into the central file server database. The file server was provided with a printer which allowed (eg monthly) attendance records of students to be run off The file server and workstations had manual entry means (keyboards) which allowed attendance records to be edited, updated or corrected by the teacher for example where such a workstation was located in a staff room. It was common ground that the manner in which data was transferred to and from the central file server to the workstation was in principle the same as the Kingshurst system, other than in the SIMS system the manual entry was pupil by pupil (rather than by batch entry eg class by class). Attendance Version 3.3. released in late 1993/early 1994 did allow such batch entry.
75. In addition to the fact that the Attendance 3.1 module was a paper based OMR system an important further difference to the Kingshurst system was that due to SIMS's other administration functions, both the volume of traffic was considerably greater and the rate of transfer of data from server to workstation was considerably slower than that of the Kingshurst system.

There was a dispute as to the volume of traffic and consequently the number of PC workstations that could operate Attendance 3.1. Following the claimant's response to the defendant's request for Admissions that the volume of traffic was only 40Kb in contrast to the defendant's figure of 4Mb an experiment was performed by the defendant on a later Attendance 3.41 (dating from 1997). No reply experiment was conducted by the claimant. Instead Dr. Jones put forward a number of ways in which he considered the system could be modified to reduce the volume of traffic. In fact two of the suggested ways namely batch entry and transferring the executable to the client PC were implemented in the experiment which reduced the volume of traffic from the 4 to 2.7 M bytes. Dr Jones' carried out certain calculations based on 40Kb. It was not clear what extra traffic he was including and his figures of 70Kb minimum to 1.2 Mb maximum were purely theoretical and of no assistance. What was clear from the evidence was that in running Attendance 3.1 there was a very large amount of traffic which was not related to attendance data gathering. The precise amount of traffic is not material to the questions I have to decide. As already indicated in relation to the Kingshurst system, there was no problem in operating a registration system with up to 30 PC workstations and a way of permitting the use of more PCs on the network to operate is to minimise the traffic relating to data not connected with the registration system.

76. Doing away with the OMR paper based system which is the heart of the Attendance 3.1module is to radically depart from the essence of how the system was designed to function. It was suggested that one might replace the OMR paper based system with a wireless system limited to say two or three classrooms. I do not consider this approach to be a realistic approach. Why do away with either the existing paper registration system or the SIMS wired network system of three or four workstations if only two or three forms/classes are involved (see Copeland XX Day 10/18, 20-26). It was not suggested that a wireless system limited to two or three classes would be an improvement on a wired system particularly bearing in mind the volume of traffic being handled by the SIMS system.
77. For the above reasons I do not consider claims 1, 2 or 7 are obvious or lacking in inventive step in view of the prior use of SIMS's Attendance 3.1 module. Given the acknowledged prior use of the Kingshurst system operating an acceptable attendance system with a wired network of 30 PCs, it was not clear why the claimant relied on the use of the SIMS Attendance 3.1. If they could not succeed in view of the Kingshurst prior use, it was not suggested they could succeed in the light of Attendance 3.1.

COMMON GENERAL KNOWLEDGE

78. I have already set out my findings as to what constitutes common general knowledge to those skilled in the relevant field (see part 3A above). The Kingshurst network system was not part of such common general knowledge albeit the SIMS system using an OMR reader was. For reasons given as to why claims 1, 2 and 7 are not obvious in view of the SIMS OMR system I do not believe such claims lack inventive step over the relevant common general knowledge. Thus, even though those skilled in the art were becoming aware of wireless LANs, there was no reason to think that systems such as the SIMS OMR system could be improved by substituting a wireless system for such wired systems.
79. That such claims were not obvious in view of the common general knowledge at the time is supported by the fact that Nord Education Limited (a subsidiary of Nord Systems, distributors of ICL computers and a software house) who developed a multi-user system called SCRIPT which was widely used in schools for school administration in competition with the SIMS system, looked at a number of IT options in 1993 to deal with pupil attendance registration. They considered both paper based and non paper based systems such as swipe cards and decided upon a low cost paper based system using an electronic pen which the teacher used to capture the attendance data using a bar-coded paper form list. The system was appropriately named AWOL. Wireless systems were not considered.^[21] Likewise Key Solutions Limited an education management consultancy business which developed software to support school management activities in 1993 looked at better ways to manage attendance registration at schools. It decided upon a system allowing the schools a choice of one of three systems, namely a keyboard entry system, an OMR system and a barcode system using an electronic pen. All three systems were based on the teachers' familiarity with the existing paper markings. Wireless systems were not considered.^[22]

GOMBRICH (PCT WO 87/00659)

80. This application which was published on 29 January 1987 relates to a patient identification system where patients wear a bracelet bearing their own barcode identification. The barcode can be read by the nurse/doctor using a hand held barcode reader. Data relating to the patient can be transmitted as a RF signal to a RF data overvoice transceiver located in the patients' ward which is connected by telephone wire to a multiplexer located in the hospital. From the multiplexer there is a cable link to the hospital's computer system. There is no suggestion in the disclosure that this system can be used for student data collection. Dr Jones's evidence that it would be obvious that the system could be programmed to download/edit/upload student data (Jones's first report paras 177-181) is far from convincing. At the heart of the system is a barcode reading system which utilises the hospital's computer system via the existing telephone lines which are connected to a multiplexer. It presupposes each student would carry/wear some form of barcode ID and that there would be wired telephonic access in each of the (class) rooms where roll-call taking or some other student data collection is taking place. I do not consider this disclosure makes any of the disputed claims obvious.

NOBLES (US Patent No 4, 820, 167)

81. As the title states this patent relates to an electronic-school teaching system. The system comprises a central school computer which is connected ie, wired to a plurality of classroom computers, one for each teacher. Each student is provided with a portable unit the size of a single school textbook which has a display screen and an input/output port for interfacing with the teachers' classroom computer. The idea is to do away with conventional textbooks and allow the students to download texts or test work from the particular classroom computer that the student is working with at the time by simply plugging in his unit. It is not a wireless system and although claim 36 refers to a further refinement of reporting student attendance data to the central computer, the patent provides no details as to how the attendance data is to be collected or processed. Mr Copeland stated (XX Day 10/81) "it is not designed as an attendance system. It is a by-product". As Dr Jones states (first report paragraph 183) the emphasis is on providing information and data to the students rather than collecting data about students. Dr Jones goes no further than indicating that some downloading and uploading of attendance data would have been contemplated with the possibility of such a system being networked on a wireless basis. I consider that the school teaching system described in Nobles does not render any of the claims 1, 2 and 7 obvious all of which relate to a student data collection system and as Dr Jones acknowledges apart from the passing reference in claim 36, there is no indication as to why or how one would adopt the teaching system disclosed to a student data collection system.

PRIORITY OF CLAIMS 1 AND 2

The law

82. Section 5(2) (a) provides that a claim to priority requires the invention to be "supported by matter disclosed in the priority application" Invention is defined in section 125(1) as that specified in the claim of the specification as interpreted by the description and any drawings. This is subject to the Protocol on the Interpretation of Article 69.

Section 5(2) (a) is to have as nearly as practicable the same effects in the UK as Articles 87 (1) and 88 of the European patent Convention.

Article 87 (1) provides that:

"A person who has duly filed in ... any State party to the Paris Convention.., an application for a patent... shall enjoy for the purpose of filing a European patent application in respect of the same invention, a right of priority during a period of 12 months from the date of filing of the first application."

Article 88(4) provides that:

"If certain elements of the invention for which priority is claimed do not appear among the claims formulated in the previous application, priority may nonetheless be granted, provided that the documents of the previous application as a whole specifically disclose such elements."

Whilst the above provisions have been considered by the House of Lords^[23] as to what constitutes the necessary support, namely that the earlier priority application must be an enabling disclosure in the same sense as that required by the sufficiency provisions, the question in this case was whether or not there was a disclosure in the priority application of the same invention claimed in claims 1 and/or 2.

83. In the case of X/Same Invention G2/98 (2002) EPOR17 the Enlarged Board of Appeals has set out the principles to be applied in interpreting articles 87(1) and 88 of the EPC as to what are the necessary requirements that constitute a disclosure of the same invention in the earlier priority application. In that case it is stated that to be entitled to the earlier priority the invention the subject mater of the claim must be directly and unambiguously derivable from the earlier application using common general knowledge (see para 9 of the Reasons and the Conclusions). In other words the priority application must expressly disclose the invention claimed or one skilled in the art can unambiguously derive the invention from the earlier application.

The priority application

84. As the claimant points out, the application is a jejune document. It has no claims and only hand drawn sketches of the basic components illustrating the palm held teachers' folder and the school office central computer system. What is said is that the priority application does not disclose the following features claimed in claims 1 and 2:

(i) means for downloading respective sets of student data (from the central computer) to each of the portable units;

(ii) an output for uploading to the central computer which is a combination of the downloaded data and the manual input;

(iii) downloading data in response to a request entered by a user into a portable unit (claim 2).

Dealing with each of these in turn:

(i) Downloading from the central computer

85. As described and illustrated, the central computer has a transceiver which communicates with the teachers' folders via their transceivers. It is common ground that the manner in which classroom names of pupils is loaded into the teachers' folders is not stated in the priority application. Dr Jones postulated (Jones XX Day 7/38-58) that in addition to downloading the classroom list over the radio link the data could be entered into the teachers' folders either manually using the keyboard or by downloading over a wired link eg once per term at the school office. Dr Jones did not accept that these were merely theoretical possibilities. This evidence is to be contrasted with the evidence of Mr Copeland (Copeland XX Day 10/90-94) and Mr Wright (Wright XX Day 9/112-113 Re X Day 9/130) who approached the matter from a more practical view point namely that manual input whilst technically possible was labour intensive and would be rejected as a possibility. So far as downloading over a wired link at the school office, such a possibility also seems highly unrealistic when the obvious way to enter classroom lists into the portable unit would be via the described radio link which is stated to be the preferred way of performing the system as not requiring any temporary or permanent connection (emphasis added) to fixed apparatus installed in the premises — see page 2. I prefer the practical approach of Mr Copeland and Mr Wright to the more theoretical approach of Dr Jones and conclude that to those skilled in the art the downloading of a class list from the central computer to the portable unit(s) by means of the radio link can be derived from the Priority Application disclosure.

(ii) Output being a combination of the downloaded data and manual input

86. It is common ground that this feature is not explicitly disclosed in the Priority Application. It was also accepted^[24] that the output to be uploaded could consist of just the attendance data manually entered by the teacher, eg attendance codes with an identifier rather than such attendance data combined with the downloaded class list. Dr Jones accepted that the latter was a common choice for many systems. A third output choice was postulated by Dr Jones namely a mixture of manually entered attendance data with some of the downloaded data. Dr Jones's evidence was that the choice between the three types of output was a trivial one and dependent on the precise nature of the tasks to be performed. What is disclosed in the priority application (page 3) is:

"when the operation (roll-call taking) has been completed in respect of all pupils on the list held in the device, a signal including all the assembled attendance information preferably with an identifier characteristic of that particular 'station', will be transmitted to the central system at the school office."

One of the alternative ways of carrying out such a task that can be derived from this disclosure is to combine the downloaded class list with the manually entered attendance data. Mr Copeland refers to it as the obvious and simplest way of doing it. I accept this evidence and find that such feature is directly derivable from the priority application.

(iii) Downloading data in response to a request entered by a user into a portable unit (claim 2)

87. It is common ground that this feature is not explicitly disclosed in the priority application. Dr Jones postulates^[25] that in addition to the data being downloaded in response to a request by the user it could be downloaded automatically by the central computer. However, Dr Jones accepts that the most common and simplest way of initiating the transfer of class list data would be in response from a request entered by the teacher/user. I have no doubt that a skilled person would have inferred from the Priority Application that to load the classroom list into the portable unit the most obvious way which this was to be achieved would be at the request of the user.

Conclusion on priority

88. For the reasons set out above, I conclude that claims 1 and 2 are entitled to the earlier priority date, However it is accepted that claims 3 to 7 are only entitled of the later filing date. It is therefore necessary to consider the pleaded intervening use by Bromcom to the EARS system.

INTERVENING USE AND DISCLOSURES

89. The re-re-re-amended grounds of invalidity (para 8) refers to the sale installation and/or use of EARS software and associated hardware by Bromcom prior to the filing date (7 October 1993) in respect the following schools or colleges:

(1) St John Bosco High School, Liverpool

(2) Wilberforce College, Hull

(3) Spennymoor School, County Durham

(4) Brentford School for Girls.

90. In their closing submissions the claimant relied specifically on the use at St John Bosco, Spennymoor School and to a lesser extent Brentford School for Girls. I shall not deal with Wilberforce College as it is accepted that if the claimant can not succeed on St John Bosco or Spennymoor, they will not succeed on the use by Wilberforce College. In addition to such intervening use the claimant relies on certain publications with regard to claims 1 and 2. The only ones of significance that I was addressed upon were (1) a document entitled "Computerised Registration" produced by Southwark LEA and dated 6 October 1993. There was a dispute as to whether this document was made available to the public prior to 7 October 1993 and (2) a brochure entitled "Radio EARS — a revolutionary new system for class attendance registration and productivity in and around colleges". Again there is a dispute as to when such brochure was first made available to the public.
91. I shall deal with the instances of intervening use first. Mr Tigin Guryel, head of research and development and production at Bromcom in his third witness statement (paras 6 et seq) describes the development by Bromcom of the Electronic Attendance Registration System (EARS) the subject of the of the patent in suit. The software for EARS was developed in two phases.

Phase 1

Phase 1 software which began sometime after October 1992 formed the basis of the Brentford School for Girls pilot and reference site. As Mr Guryel explains the system was very simple — the computer transceivers (RTUs) were dumb RTUs with no processing capacity or memory. The central server would interrogate the folders directly via the dumb RTUs whose only function was to convert data from cable signals to RF signals and vice-versa . Phase 1 software clearly had none of the features the subject of claims 5 to 7. It was contended by the defendant that whilst such a system had the remaining features of claim 1 it did not have a collation program. This contention depends on the proper construction of a collation program which I have already decided is to be constructed broadly. It follows that phase 1 software when installed at Brentford School for Girls had all the features of claim 1.

92. As to whether such features were made available to the public, it is admitted (defendant's admissions dated 13 October 2003 para 1) that the following features were publicly disclosed by the sale and use of the EARS system at Brentford School for Girls namely: A student data collection system comprising:

(1) a central computer

(2) wireless transceiver means

(3) portable data collection units for use by teachers/supervisors each of which also includes wireless transceiver means for communication with the central computer

(4) each portable unit including manual input means.

What is contended by the defendant is that the Brentford EARS installation was a pilot reference site which at all material times remained the property of Bromcom and access to the software was subject to an obligation of confidence. Mr Graham Wright was deputy head teacher at Brentford School for Girls at the relevant time and was involved in the installation of the EARS system there. It was clear to him that as far as the school was concerned it would not have allowed third parties to have access to the system itself. In paragraphs 12 to 20 of his first witness statement he draws a clear line between allowing interested teachers and educationalists from other schools seeing a demonstration of the working of the system and third parties investigating or having access to the system itself. It was his view that only Bromcom staff had access to the system. I have no doubt that this was indeed the position.

93. What was made available to interested teachers and educationalists is to be found in the Southwark LEA document dated 6 October 1993 (pleaded as an intervening publication). The system is described as follows:

"This system is based on direct radio contact between the teacher and an administration computer. This is accomplished by installing radio transceivers out of sight around the school. These transmit information from the teacher to the central computer and vice versa.

The Computer Folder

Each teacher has her/his personal folder, which is approximately A4 sized and is fairly light. The folder has a built in miniature computer and radio transmitter-receiver. It has a keyboard with a screen at the top of the folder. The teacher keys in the tutor group which is to be registered. The list of pupils scrolls through on the screen and the teacher marks each one present or absent …"

Appendix 2 to the document refers to a visit dated 20 July 1993 by personnel from Hounslow LEA at which there was a demonstration of the registration process. There it is stated that once the form tutor had keyed in his personal code and the form's description, the girls' names were scrolled on the screen and the form tutor marked the register by keying in symbols for present or for categories of absence. He then transmitted that data which went directly to the school's computer by radio. Mr Bartlett who was responsible for promoting the EARS system at Bromcom confirmed^[26] that those witnessing the demonstrations were told that the teacher would call up a register on the folder by radio from the central computer based in the school office, the teacher would take the register and mark the attendances and absences and then transmit the information back to the central computer.

94. Whilst not explicitly stating that the central computer will assemble the data transmitted from the individual folders or that the attendance data transmitted back to the central computer is a combination of the class list and the manually entered data, I consider that such features to those skilled in the art are features which can be derived from what was being demonstrated to IT educationalists or teachers at Brentford. It follows that if claims I and 2 are not entitled to the earlier priority date, such claims would lack novelty or if novel would be obvious to those skilled in the art in view of the Brentford School for Girls use of the EARS system (Phase 1). However I do not consider that such use renders claim 7 obvious. How traffic using a number of transceivers (RTUs) is to be controlled both as regards the radio link concerning radio protocols to be used (claims 5 or 6) and as regards the data transmission from the RTUs to the central computer (claim 7) cannot be derived from such use at Brentford nor would those features of claims 5 or 6 combined with the features of claim 7 have been obvious to those skilled in the art.

Phase 2

95. Phase 2 software referred to by Bromcom as SERHAND was a continuing development of the EARS system primarily between March and October 1993. As Mr Tigin Guryel states^[27]development work on the full EARS system based on the source code used at Brentford was slower than expected and this resulted in the need to produce two further versions of the executable code to meet customer demands for installation. These were the "RF Survey" version to allow Bromcom employees to survey customer sites when installing the RF transceivers and the "Brentford Type Polling/Dumb RTU" version which allowed customers to start training and to phase in the electronic attendance registration system. This latter version in effect mimicked the Brentford pilot system by converting intelligent RTUs to work in dumb mode. In other words it is based on a simple point-to-point polling operation controlled directly by the central computer so that only one RTU and one folder were active at any given time. The fully functional version of the software for the EARS system was not released to customers until the end of October 1993. This version managed the EARS protocols and communicated with the intelligent RTUs which in turn operated asynchronously communicating with the folders. The claimant attacked the credibility of Mr Tigin Guryel. In particular a sustained attack was made as to when the various software versions were released from R&D to customers as opposed to released to QA for internal testing. I found Mr Tigin Guryel to be extremely knowledgeable about the development of the EARS software and that he gave his evidence in a careful and clear manner. I reject the attack on the accuracy of his version of the events.

Folders

96. There was an issue as to when the portable units or folders were first supplied to customers. Bromcom's first folders used at Brentford School were Casio folders. It was Mr Tigin Guryel's evidence that only 10 such folders (Casio SF9600) fitted with RF modules were ever produced although 10 further RF modules were made by M & B Electronics and used for testing purposes or as spares (XX Day 7/180-1 82).^[28] As part of the Phase 2 development in March 1993 new purpose built folders were produced and were designated AR (Attendance Register) folders and AR plus Such folders were intelligent folders having their own programmable chip (EPROM). The software sometimes referred to as firmware was designed to correlate with the software on the RTUs. However as Mr Bartlett states there was a problem of the delivery of the new folders due to a shortage of the resin for the chips from Japan. Mr Bartlett records^[29] that a number of orders were placed for EARS installations including the supply of AR folders. However precisely when such folders were supplied to the various customers is far from clear. It was put to Mr Bartlett that in the case of St. Benedict's, which included the supply of 80 AR folders, it is recorded that Bromcom were paid in full in September 1993, The end of September was Bromcom's end of the financial year and Mr Bartlett's belief was that the folders were not delivered until October 1993. This evidence is consistent with that of Wilberforce School which it is accepted did not start receiving folders until 14 October 1993.^[30] It is also consistent with that of Spennymoor who did not receive its 46 AR folders until 8 October (as noted below). Finally it is consistent with the evidence of Mr Tigin Guryel that prior to the completion of the SERHAND software to be used both by the central computer and by the RTUs (end of October) the folders that were in use were in conjunction with a Brentford Type Polling/Dumb RTU system referred to above. There was therefore no reason to supply properly programmed AR folders until the SERHAND software had been finalised.
97. Dealing specifically with the installations of the EARS system at St John Bosco and Spennymoor School:

St John Bosco

The evidence relating to this school was given by Mr Peter Smith, deputy head teacher. It appears that the school ceased using Bromcom in 1998 and had disposed of the paperwork relating to its use of the EARS system. Mr Smith's recollection of the events of 1993 understandably was poor. It appears that the school was told by Bromcom that the EARS system to be installed would be operational from September 1993. Mr Smith recalls that the cabling was installed during the summer holiday of 1993. He had no clear recollection of when the folders were delivered or the system first used by the school other than he believed it was first used in September 1993 (see XX Day 3/167-168). The only documentary evidence relating to the sale and installation was an invoice dated 16 September 1993 and a reminder dated 22 October 1993 in respect of a number of items including 60 EARS AR folders . Payment terms were specified as payment on installation with a statement - "part invoiced half due now" The defendant points out that this is an invoice and not a delivery note. Payment was not made by St John Bosco until 22 October 1993. Mr Smith was not clear why payment was not made earlier — one suggestion, not suggested by Mr Smith, was because of a typo in the first invoice regarding the VAT rate. What is clear from the express words of the invoice is that the goods the subject of the invoice remained the property of Bromcom until paid for which was not done until after the filing date. What is not clear from the documents or the evidence of Mr Smith is what software was installed at St John Bosco prior to 7 October 1993. Mr Bartlett (second W/S para 33) believes that the version would have been the phase 2 — Brentford Type Polling/Dumb- RTU version described by Mr Tigin Guryel which was used as a stop gap at this time. This evidence accords with that of Mr Tigin Guryel regarding the various stages as to how the software was installed or rolled out at the various schools with reference to Frank F Harrison School — (see third W/S paras 17-22). It was not until the end of October that the fully functional software version was released to schools. As regards the claimant's attack on the credibility of Mr Bartlett, none of the criticisms of his evidence (which in any event have no substance) affect the evidence relied upon in this judgment.

98. I do not consider the sale and installation at St John Bosco of the EARS system prior to 7 October 1993 advances the case over the Brentford School for Girls disclosure. Both were reference sites and hence available for inspection to educationalists and teachers. Both operated on the "dumb RTUs" principle and both remained the property of Bromcom so that at the very least no invasive investigation of the software could be performed. Any non-invasive tests in the manner postulated by Dr Jones^[31] could only have revealed the features of claims 1 to 4. claim 7 is not obvious in view of the St John Bosco disclosure for the same reasons that claim 7 is not obvious in view of the Brentford disclosure.

Spennymoor

99. Mr Ogden was head teacher at Spennymoor from 1989 to 2001. He accepted that his recollection of events was based on the contemporary documents which in so far as the supply of relevant items is concerned are as follows:

18 August 1993: delivery note including one PC and 12 RTUs stating "printer and folders to follow".

7 October1993: delivery note for printer and 46 folders

20 October 1993: invoice for the 46 folders and printer. In addition a further 17 RTUs, cabling and installation. Payment terms on delivery.

A condition of sale was that all goods would remain the property of Bromcom until paid for. As Mr Ogden explains Spennymoor's requirement was not to have a morning registration (eg form by form) but only an afternoon registration. The morning roll call taking was to be by classroom. The reason for this requirement was because the school was a split-site (Upper and Lower about a mile apart) and pupils went straight to their classrooms for their first lesson. In a letter dated 10 June 1993 Mr Ogden made it clear that the system was to be up and running by 6 September 1993 — training of staff on Friday 3 September. In a letter of complaint dated 9 September 1993 to Bromcom, Mr Ogden pointed out Bromcom's failure to provide a system up and running. At this stage Mr Ogden had handed over management of the installation and operation of the EARS system to a Mr Warburton. In the letter of 9 September 1993 Mr Ogden reports that it was his understanding that they then had "46 folders configured, two computer terminals which cannot speak to each other because the radio link is not installed …" Later on in the letter Mr Ogden points out that they had no other registration system in place and that although they can now manage parts of the afternoon session they have no software for the morning. Mr Ogden (XX Day 5/23 and 28-29) accepted that he might have been mistaken about the 46 folders as he did not understand enough about the technology. He also accepted that the AR folders did not come until October. It was Mr Ogden's evidence that Mr Aker replaced whatever folders they had by the AR folders. Bromcom's response to this letter was to assist Mr Warburton in the setting up and testing of the class registration program which in Bromcom's letter dated 21 September 1993 was stated to have begun on 20 September. Mr Ogden's evidence was that because of their statutory duties to conduct a pupil registration system they reverted to a paper based system whilst their problems were being sorted out (XX Day 5/25-27; Re X Day 5/36) and the class lists were printed out from the central computer. It was Mr Aker's evidence that he visited Spennymoor on 8 October 1993 to change the database software to allow class registration which would be taken at the first lesson of each day (instead of the tutor group registration system). Mr Aker had been employed by Bromcom at that time to develop the database software for EARS working with Mr Tigin Guryel who was responsible for the hardware and communication side of the software. Mr Aker states (W/S para 11)^[32] that the class registration as of 20 September would have been paper based as a stopgap. Teachers were provided with printouts of class lists and attendance data recorded by them was manually entered into the central computer by the system administrator. Mr Aker (W/S para 12) states that be was asked to deliver the EARS folders and a printer to the school and I have no doubt that this refers to the 46 folders and printer the subject of the delivery note dated 7 October 1993. The claimant attacks the credibility of Mr Aker. His evidence was given over a video-link from Cyprus. He was questioned about Mr Ogden's letter of 9 September concerning the 46 folders referred to in the letter. It was quite clear from the start that he knew nothing about the folders referred to (see XX Day 8/8). It was suggested to Mr Aker by Mr Arnold that the 46 folders delivered by him on 8 October were replacements for 46 folders that Spennymore already had (XX Day 8/17-18). Mr Aker did not agree. There is no support for the suggestion in the documents in particular there is no note on the delivery note dated 7 October to collect folders. I accept Mr Aker's evidence that the 46 folders delivered by him were not replacement folders I also accept his evidence, supported by Mr Ogden that Spennymoor had in place a paper based system for the morning class registration roll-call taking because as Mr Aker explains the class registration software was not finalised until late October. The evidence of Mr Tigin Guryel^[33] corroborates Mr Aker's evidence although Mr Guryel was not personally involved in Spennymoor until much later.

100. As regards precisely what software had been installed at Spennymoor by 7 October 1993 is far from clear. Whilst some sort of registration system was apparently available for afternoon registration, the evidence of Mr Tigin Guryel was that until the release to schools of the fully functional version which was not until the end of October, the only version available would have been the Brentford Type Polling/Dumb — RTU version. In other words no RF protocol would have been in place nor would the RTUs have been utilising any data buffer. Precisely what folders were used in conjunction with the dumb RTU version is not clear from the evidence. It is accepted that the system remained Bromcom's property prior to 7 October 1993 — in fact it was not paid for until the end of the month. Furthermore, the evidence was that it did not become a reference site until after the autumn half term break when the system was first up and running (ie late October).
101. Even if access to the Spennymoor system to analyse how it functioned in the manner postulated by Dr Jones was permissible, eg by pupils or teachers, which in itself seems highly improbable, as the software installed in the EARS system at that time was operating the dumb RTU version, it would only have revealed those features the subject of claims 1 to 4 of the patent in suit. For the reasons set out above relating to Brentford and the St John Bosco school I find that the Spennymoor use neither anticipates nor renders obvious those features of claim 7 combined with claims 5 or 6.

Intervening publications

102. As regards the Southwark LEA document relating to Brentford, I have already considered its impact. As regards the undated Bromcom brochure, relied upon,^[34] the claimant contends that it was published in March 1993 because it shares the same artwork and some of the text with an advertisement published that month. It was Mr Bartlett's evidence (XX Day 8/129-130) that it was not published until about January 1994 due to the fact that it contained information eg relating to the Teacher Organiser and the Teachers Palmtop that was not available as of March 1993 but only became available in January 1994. I do not accept that the claimant has established that the Bromcom brochure was published prior to 7 October 1993. In any event it is not alleged that it discloses features other than the features of claims 1 and 2.

CONCLUSION

103. 1. I find claims 1 and 2 to be obvious and lacking in inventive step in view of the Kingshurst prior use.
     2. I find claim7 is not obvious in view of any of the cited prior art and intervening use and disclosures.

Note 1   The Education (Pupils’ Attendance Records) Regulations 1991; regs 1 and 2.    [Back]

Note 2   Copeland first report para 7.5.2.    [Back]

Note 3   The New Shorter Oxford English Dictionary.    [Back]

Note 4   New Shorter Oxford English Dictionary (1993); Collins Dictionary (1994).    [Back]

Note 5   Jones First report paras B9-lO; XX Day 6/17-18.    [Back]

Note 6   See para 79 below regarding two such teams.    [Back]

Note 7   D3/179,181 and 183.    [Back]

Note 8   D2/128-138.    [Back]

Note 9   D2/124.    [Back]

Note 10   See also Copeland first report-para 4.5.7:1.1; XX Day 10/33-34.    [Back]

Note 11   D2/136-137.    [Back]

Note 12   See also Bryant XX Day 6/11.    [Back]

Note 13   See Wright XX Day 9/94    [Back]

Note 14   Bryant XX Day 6/11-12; Christopherson Day 6/164; McDonnell Day 6/149-150 Stephenson Day 8/107.    [Back]

Note 15   Wright Day 9/74; Copeland Day 9/150.    [Back]

Note 16   Bundle D4 / 298, 302.    [Back]

Note 17   Jones XX Day 5/79-80; Day 6/52-53 - WaveLAN will run at 1.2 to 1.5 Mbs including overheads.    [Back]

Note 18   Jones First Report para 167; XX Day 5/75-76.    [Back]

Note 19   Jones XX Day 6/50-53.    [Back]

Note 20   Jones XX Day 6/60-61.    [Back]

Note 21   See JC McDonnell W/S paras 12 to 14.    [Back]

Note 22   J Williams W/S paras 7 to 17.    [Back]

Note 23   Asahi Kasei Kogyo KK’s Appl (1991) RPC 485 followed in Biogen Inc v Medeva plc (1997) RPC at 46. (Lord Hoffmann)    [Back]

Note 24   Jones XX Day7/58-72; Wright XX Day 9/114-1-16; Copeland XX Day 10/94-98.    [Back]

Note 25   Jones XX Day 7/73-78; also Wright XX Day 9/113-114.    [Back]

Note 26   Bartlett XX Day 8/153-154.    [Back]

Note 27   Tigin Guryel .(3rd W/S para 12).    [Back]

Note 28   See also Bartlett (Ref XX Day 8/181-182).    [Back]

Note 29   SeeE2/11A; Bartlett XX Day 8/174-175.    [Back]

Note 30   Overment XX Day 4/91-92.    [Back]

Note 31   Jones First Report paras 316-319.    [Back]

Note 32   See also Aker XX Day 8/14-15.    [Back]

Note 33   Tigin Guryel third W/S para 28; XX Day 8/58-75.    [Back]

Note 34   D4/129-139.    [Back]