England and Wales High Court (Chancery Division) Decisions

IN THE HIGH COURT OF JUSTICE
CHANCERY DIVISION

		Royal Courts of Justice Strand, London, WC2A 2LL
		22nd July 2004

Mr Justice Lewison:

Introduction

1. Ultraframe (UK) Ltd is one of the market leaders in the manufacture and supply of conservatories. Two of its modular systems are called "Ultralite 250" and "Ultralite 500" respectively. Until 2002 Eurocell was one of the largest distributors of the Utralite systems. But in 2002 it began to manufacture its own system, which is called "Pinnacle 500". It was designed to compete directly with Ultralite 500.
2. Ultraframe claims that the Pinnacle 500 system infringes a patent of which it is the proprietor; and also its design right in both the Ultralite 500 system as a whole and also in many of the components. So far as the claim to patent infringement is concerned, Eurocell say:

i) Their system falls outside the scope of the monopoly claimed by the patent;

ii) And anyway the patent is invalid because:

a) The patented invention was not novel, but had been anticipated by previous publications;

b) Or if it was not anticipated, it was obvious.

3. So far as design right is concerned, Eurocell say:

i) Ultraframe's designs are not original, but were copied;

ii) Anyway, the designs were commonplace at the time that they were created;

iii) Parts of the designs in which Ultraframe say they have design right are outside the scope of design right because they fall within the so-called "must fit" and "must match" exclusions from design right;

iv) Eurocell's designs were themselves original, and were not copied from Ultraframe's designs, so that there has been no infringement.

Industry background

4. Mr Savage, Ultraframe's expert, explained the background to the industry in his initial report. He did so in non-technical language; and I adopt his summary, which was not challenged. There are three basic types of conservatory.

"Dual pitch" conservatories

5. These are the best recognised types of conservatory with a ridge and dual pitch roof. These roofs tend to look elegant and expensive, but involve substantially more components than lean-to roofs and take more time to fabricate, and so are more expensive. A common type of dual pitch roof was in 1992 (and still is) the "Victorian" style roof.

"Steep lean-to" conservatories

6. This expression describes lean-to (mono-pitch) roofs of between around 10° to 30°. These fairly common roofs can and should be distinguished from the more specialised "low-pitch" conservatory roofs. Because lean-to conservatories are rather easier to install and are cheaper than dual pitch roofs, they are often favoured by less experienced fabricators. However, the options are generally far more limited with this type of roof, and they do not tend to look as attractive. Some early lean-to conservatories were little more than glorified carports.

Low-pitch roofs of less than around 5°

7. In the early stages, 5° was really the lower limit of these roofs, since the materials used (slab polycarbonate sheets for example) were not guaranteed for use at an angle less than this. The angle was limited by a number of things, but particularly rain-water run-off. Serious problems result if rain-water starts to pool on the roof and whether it does or not will depend on a number of factors, such as the stiffness of the panels and so on.
8. The last of these three roof types began to grow in the early 1990s because the first two roof types were less than ideal for a number of applications. For example where the wall to which the roof will be affixed is of fairly low height, for example a bungalow wall, it is not possible to use a "steep lean-to" roof. This is because the height of such a conservatory at the house wall needs to be substantially higher than the height of the conservatory at its eaves. It is also difficult to use a dual pitch roof, since the conservatory will then have to have a box gutter running between it and the bungalow, which is expensive in the short term and often troublesome in the long term. A solution was the use of low-pitch lean-to roofs with pitches of 5° or even less. These could start at the highest point of the bungalow wall and yet maintain an acceptable internal height up to the eaves. In 1992/3 the market for low pitch roofs had not really been tapped.
9. Just like the dual pitch conservatories before them, low-pitch conservatory roofs were initially fabricated on site by builders who would cut and fit all the components. This would preferably be done using timber and glass for example. Gradually builders began to try and "botch" together roofs using plastic profiles from vertical window systems and the glazing bars from the systems that were being and had been developed at the time for dual pitch roofs (i.e. modular component systems for use in conjunction with plain glass sheets) with polycarbonate sheets.
10. Then manufacturers began to produce modular roofing panel systems to make putting together low-pitch roofs easier. These panels would normally be attached to each other, rather than via glazing bars or timber beams for example. However, these roofs were still the poor relations of Victorian-style systems. They tended to be fairly basic in appearance (often looking a bit like the corrugated iron or plastic sheeting you find on carports, very basic lean-to structures etc), and certainly nobody seemed to have given a great deal of thought to the aesthetics.
11. The design considerations for the different types of roof differed from each other. Designing a dual-pitch conservatory is a bit like designing a half-timbered or a steel framed building. The framework provides the structural rigidity and strength, and then the walls are filled in. In the same way in a dual pitch conservatory (particularly the later, modular component types), extruded aluminium components provide a framework into which sheets of glass etc fit. Calculating the necessary thickness of the structural components for example is a straightforward calculation of I-values etc. With modular panels as used in the later low-pitch roofs on the other hand, it is the roof panels themselves that provide the strength and stiffness in the roofs. Whilst there were often stiffening members added, the calculations of how thick the panels needed to be etc is a complicated one, particularly because of the hollow ducted nature of the panels. The principle of the two types of roof from a structural point of view is therefore rather different, although naturally most of the factors that must be taken into consideration (e.g. wind loads, snow loads etc) remain the same.

Everlite

12. Everlite is a system that was invented and patented by Mr Hans Sokoler. His patent was referred to in the course of the trial as "Sokoler 83". The Everlite system consists of hollow plastic panels with convex top and bottom walls. Each panel has a horizontal intermediate wall between the top and bottom walls, and vertical end and intermediate walls. The intersections between the vertical and horizontal walls form a series of chambers or ducts in two layers. Each panel has integral coupling members on opposite sides, which fit together so as to join the panels together. Each panel is 265 mm wide and 60 mm deep. Each panel also has a pre-cut eaves drip to direct rainwater into the gutter.
13. The Everlite panels had first to be screwed into position. The panels are then stiffened by inserting an aluminium bar into the duct which is formed by the join between two adjacent panels. The bar is a solid, tall, rectangular extrusion and when it is inserted, after the panels have been linked together, it has the additional effect (apart from stiffening) that it prevents the panels from moving in a vertical direction. This means that the coupling elements can no longer be unhooked; because the bar wedges them together. The bar does not, however, hold the panels together laterally. That is done by the coupling members that are integral to each plastic panel. I reproduce the schematic diagram of the stiffening member inserted into the panels prepared by Dr Kent, Eurocell's expert. (Dr Kent was not called to give evidence, so I have not taken his written evidence into account, but I have found his simplified diagrams of the various roofing systems helpful).

14. Eurocell relies on this patent as showing that Ultralite 500 is either not novel or is obvious.

Ultralite 250

15. Ultralite 250 went onto the market in about 1989. The system consisted of panels which were flat on their underside, with a contoured top surface which created a series of curves and troughs. As its name suggests, each Ultralite 250 panel was 250mm wide. The panels connected to each other by means of a "hooking" mechanism at the ends of the panels. This mechanism requires the installer to lift and tilt one of the panels and lever it into position along the whole length of the joint. The panels were strengthened by the use of an aluminium stiffening bar. The bar fitted into a duct formed by the plastic roofing panels, when joined together. But it did not fit snugly, and did not act as a wedge; although it does prevent the adjoining panels from being unhooked by vertical displacement. Ultraframe patented this system by patent GB 2 243 170. I refer to this as "Ultralite 250". Eurocell also relies on this patent as showing that Ultralite 500 is either not novel or is obvious.
16. The patent shows two versions of the system. In one version, the bar is wholly contained by the plastic roofing panel. In the other, it is indented at the top. This enables a cap to be inserted into the indentation in the bar.
17. The two systems I have described both had convex top surfaces to the panels. This was common in modular roof systems at the time. There are two reasons for this. First, a curved top and bottom surface made the panel stiffer and stronger. Second, a curved profile was also more forgiving to extrusion flaws.

Other products and designs current in the early 1990s

18. Mr John Lancaster (who was the founder and managing director of Ultraframe at the time, and who is still a non-executive director) was well versed in the conservatory business. He regularly read a number of trade magazines, including Conservatory Industry, Glass Age, Window Industries and Window Trade News. He was also a regular attender of trade exhibitions, including Glassex in Birmingham, and Fensterbau in Germany. At some time in the early 1990s he met Mr Schneider, who was then the marketing director of Rodeca, a German extruding company.
19. Products available in the market in the early 1990s included triple-wall polycarbonate sheeting. This kind of sheeting was usually 16 mm deep, and each sheet was divided into ducts. Sometimes the ducts were rectangular, and sometimes triangular. The available products also included top caps, which were used to cover joins between plastic panels; and end caps, which were used to seal off the open section left once the top cap was put in place.
20. There were also panels designed more specifically for conservatory roofs (such as Eurolyte), but these had to be clipped or screwed together.

Ultralite 500: the design process

21. Mr Lancaster says that following the Glassex exhibition in 1992 he realised that the current low pitch roof Ultraframe were selling (Ultralite 250) could be significantly improved. Having spoken to customers who were visiting the exhibition, he understood that there was difficulty in fitting Ultralite 250 to low-eaved buildings, such as bungalows. In particular he wanted a complete low pitch roofing system which could be more easily assembled and which was more pleasing to the eye. He therefore thought that "flat panels" would be better to use and he then started to think about ways these panels could be connected without having to hook the panels together physically as they did with Ultralite 250. He also wanted wider panels so that they could reduce the number of aluminium bars and so that they could also incorporate roof vents. The width he chose was 500mm, which effectively doubled the width of Ultralite 250 and therefore halved the number of bars. This was also very cost effective. He goes on to say that he changed the joining method for the panels from the arrangement used in Ultralite 250 by using the aluminium strengthening bar as the means of connection. In Ultralite 250 the panels were connected together before they inserted an aluminium strengthening bar whereas in Ultralite 500 the strengthening bar is inserted between two unconnected panels laid side by side to draw/connect the panels together. This mechanism therefore connected the panels and strengthened them at the same time with minimum effort.
22. Once Mr Lancaster had formulated the concept, he showed a sketch of his ideas to Mr Chris Richardson and Mr Lee Calvert (who were all employees of Ultraframe). Mr Richardson recalled that the "driver" of the idea was that Mr Lancaster wanted to be able to lay panels side by side without connecting them and then to slide a bar between them to connect them, in order to make installation easier. The three of them worked on the detailed design. Mr Lancaster was not satisfied that extruders in the UK would be able to extrude a panel as wide as 500 mm and he recalled his meeting with Mr Schneider. After speaking to him on the telephone, he decided to go to Rodeca's factory to assess its ability to extrude the panels. The visit took place in April 1992. Mr Richardson went too. Mr Lancaster said that he took "a basic drawing" of the panel with him. While he was there he was given a technical leaflet showing some of Rodeca's products. These included a three-walled extrusion (X 5603) that was 500 mm long and 60 mm deep, with light diffusing pips on the top and bottom inner surfaces of the extrusion. The visit was a success, and Mr Lancaster came back confident that Rodeca could extrude the panel that he wanted. Rodeca's principal concern was the smallness of the tolerances that Ultraframe were prepared to countenance. Mr Richardson then took up the story of the detailed design process.
23. The panel. They wanted a panel that when built up would make up most conservatory roof widths, without having to be cut to size and wasting material on site. Mr Richardson was also concerned to ensure that the panel did not sag between the bars in hot weather and he thought a 500mm wide panel might well be the limit. Also, as the panels are made wider, the height of the aluminium stiffening bars must increase to take the extra loading, which increases the overall system depth. Because this was a system for a low pitch roof they wanted to minimise this depth and also wanted to minimise material cost, all of which meant they could not make the panels too wide. Eventually, in about April/May 1992, after some deliberation between Messrs Lancaster and Richardson, they settled on a 500mm panel as the optimum width. The height/thickness of the main body of the panel was designed at 60mm. Mr Lancaster wanted a thick panel of "substance" which would impress customers and have a good thermal performance and sound insulation. The height/thickness of the panel also took into account the system height with the aluminium stiffening bar (of 75mm). (75mm was structurally the ideal depth in the light of the maximum 4m wide span of the roofing system.) For aesthetic reasons they did not want to design the panel much thinner than the overall system height as this would make the stiffening bar top caps look too big. They also did not want too thin a panel, as this would have less structural integrity and the panel could be more prone to sagging when warm. It was a balance between the depth of overall system, cost of material and structural performance. In April/May 1992 they decided upon a panel thickness of 60mm.
24. Each panel had ducts within it. The proportions of the ducts and therefore the internal webs were chosen to provide a balanced solution ensuring a panel that felt substantial, that retained its integrity under stress and yet was not wasteful of material. One idea was to have the ducts of equal width. However Mr Richardson was very concerned about panel sagging when warm. In May 1992 they therefore designed a narrow central duct to hold an additional bar to strengthen the panel in case it started to sag. This smaller duct was 30mm wide. Part of the reason for this was that because the depth of the panel was 60 mm, and was divided in two by the inner horizontal wall, a central duct 30 mm wide gave a section 30 mm square; which was the profile of a standard aluminium extrusion. As the 60mm depth and the 500mm width of the panel had already been determined they used 60mm as a starting point for the width of the ducts and divided this into 500mm. Then, taking into account the narrower middle duct (of 30mm), they were left with nine ducts. In fact the narrow central duct has never been needed for extra stiffening. The smaller central duct was shown on a drawing that Ultraframe supplied to Rodeca. Mr Conterno (who is now the CEO of Rodeca and was involved in 1992 in the extrusion of the plastic panel) was clear that the design of the panel was Ultraframe's, and not Rodeca's or that of its tool-making subsidiary, Cos-Mac. I accept his evidence. By the end of the trial Eurocell accepted that the design of the Ultrafame panel had not been copied from or made by Rodeca. However, the question still remained whether the design of the Ultralite 500 panel had been copied from the previous Ultralite 250 panel. I think it fair to say that the depth of 60 mm was inspired by Ultralite 250, but all the other features of the Ultralite 500 panel (width of panel, flat top, ribs, narrow central duct, shape of other ducts and shape of upstands) were not. In my judgment they were original.
25. The drip overhang. One design consideration when designing the panel was that the Ultralite 500 was intended to be used at low pitches, for example on bungalows where the conservatory roof would need to be almost flat (with the lowest pitch being 2.5^o) due to the low level of the adjoining bungalow roof. The pitch had to be that low, because otherwise it would be impracticable to fit a normal size door at the front of the conservatory (without an expensive load bearing additional box gutter at the bungalow) because there would not be enough height. In addition they needed to position a gutter under the edge of the roof to collect the water running off the roof; but also needed to ensure that the gutter could be positioned over the top of doors and windows so that it would not be an obstacle for people entering the conservatory through the front. They therefore decided to incorporate a "drip overhang" in the panels to be used in the Ultralite 500 system. Ultraframe had used a pre-cut drip overhang before, in Ultralite 250. However because the top surface of the panel in Ultralite 250 had a curved outer surface, the pre cut drip overhang had a different appearance to the overhang in Ultralite 500 where the top surface of the panel is flat.
26. The upstands. Mr Lee Calvert worked with Messrs Lancaster and Richardson on the upper parts of the protruding upstands at either edge of the upper surface of the roof panels. These upstands bend in slightly when the intermediate bar top cap is fixed. This action helps keep the panels together at the top and the upstands also help prevent water ingress.
27. Connecting the panels. As I have said, in the Ultralite 250 system, the panels connect to each other by hooking over each other in a way in which they hold themselves together. Mr Lancaster had the idea of changing the connection of the panels in Ultralite 500, of laying the panels side by side and using the aluminium stiffening bar to join the panels, pulling them together as the bar is put into place. Messrs Lancaster, Richardson and Calvert worked on the design and dimensions of the connections at the sides of the panel, including the channel sections at either edge of the base of the roof panels. These channel sections help to ensure any rainwater or any water that might condense on the sides of the stiffening bar does not leak into the room below the conservatory. The channel sections were designed between May and June 1992. Their design evolved alongside the design of the stiffening bar since the ends of the panels and bar have to work together.
28. The stiffening bar. Mr Richardson was principally responsible for the design of the stiffening bar.
29. I am satisfied that Ultralite 500 and all its component parts were independently designed by the Ultrafame team and were not copied from any other source. In particular, I find that the panel was designed by the Ultraframe team and not by Rodeca.
30. I reproduce the diagram of the interface between the stiffening bar and the lower coupling members prepared by Dr Kent.

31. It will be seen that the left hand lower coupling member is a rotated "L-shape". The right hand lower coupling member is a rotated "U-shape". When assembled, the L-shape fits into the U-shape. Thus the two lower coupling members restrain vertical movement. They are prevented from coming apart horizontally by the internal projections in the lower limbs of the stiffening bar. These give lateral restraint, although they do not actually grip or clamp the coupling members.
32. Ultralite 500 went onto the market in March 1993.

What is clever about Ultralite 500?

33. Mr Savage explains the difficulties with roof systems that preceded Ultralite 500. Previous roof systems of this type had often made use of connections between adjacent panels. Nearly all roofs built using panels of this type would be made up of several panels, and the panels would need to be fixed somehow. Sometimes there would be some sort of bar between panels; and these bars would often be placed in the duct between the panels. These bars were there primarily to add strength and stiffness to roofs made with such panels. The failing of most of these panels was that it was time-consuming and fiddly to connect the panels. Normally you would fix the panels one at a time, so when fixing the next panel you would be working above your head. Each panel would invariably have some flexibility, as well as uneven edges produced either by the extrusion process, or their subsequent treatment on site or in transport. The extruded plastic panels were often laterally bowed (caused by thermal shrinkage in extrusion) i.e. were slightly banana shaped along their length. All of this meant that when the first panel was fixed in place, it would be very difficult to line up the next panel all the way along its length (typical panel lengths were 4m) particularly since you would usually have to engage the second panel at an angle to the first. If you did not get the panels fitted together perfectly along their length then the stiffening member would not slide in. This all made the job of fixing panels extremely unwieldy, and the larger the panels one was dealing with the more difficult the job became. It was very much a two-person job to fit these panels together. Tilting one panel to connect it to another, also made it difficult to work where there were low height clearances, such as on bungalows.
34. In the Ultralite 500 system, the clever aspect is that the panels are "zipped" together when the stiffening member is inserted. The zipping together of the panels by the stiffening bar is something that is of real benefit to the installer of roofing panels. Instead of first having to ensure a perfect fit along the whole length of the panels, all the installer has to do with the two Ultralite 500 is ensure that the two coupling formations meet at one end of the panels, which is a simple task. The stiffening beam then actively pulls the two coupling members together, lining the panels up as it slides in. This also overcomes the problems of lateral bowing - the stiffening bar smoothes out the bowing as it pulls the panels together. It becomes a one-person job rather than a two person job to fit these panels together.
35. The zipping together of the panels also makes it easier to work where there are low height clearances and means the panel widths can be greater. When one panel has to be tilted to connect it to the other panel, this restricts the width of the panel where there are low height clearances between the top of the low-pitch roof and the gutter/soffit, which is typically the case for bungalows, the main market for low-pitch roofing systems. Zipping together the panels removes this restriction since the panel being installed does not need to be tilted but can be engaged horizontally from the side.
36. I did not understand Mr Thorley QC to dispute that when Ultralite 500 is assembled on site, something like this zipping process does or can take place. However, he said that this was not something that the patent itself disclosed, and it was not, therefore, part of the invention.

Design changes in 1995

37. The design of Ultralite 500 was changed in some respects in 1995. The change arose out of Mr Lancaster's decision to attempt to reduce production costs. In part this was achieved by transferring the extrusion process from Germany to the UK. The two principal design changes were the change of the shape of the light-diffusing shapes on the upper face of the lower horizontal wall and the lower face of the upper horizontal wall of the panel from prisms to "pips"; and the removal of some of the thickness of the plastic. In addition, the starter bar top cap was redesigned at about the same time. Originally the starter bar top cap was designed with an overhang designed to hold the side wall of the conservatory in place. In about 1995 a new piece, called a "firring", was introduced. This obviated the need for the starter bar top cap to hold the wall in place. In consequence the starter bar was redesigned. The top of the bar became longer and the vertical straight bar became shorter.

Pinnacle 500

38. Mr Adrian Redshaw is the technical design manager of Eurocell. In mid 2001 Mr Hartshorn, the managing director of Eurocell, decided that Eurocell should develop its own modular roofing system for lean-to roofs. Mr Redshaw was in charge of designing it. At the time, Eurocell was still the largest distributor of Ultralite 500, doing an annual turnover of about £1.6 million. Ultralite 500 was the biggest seller in the market. I have no doubt that Pinnacle 500 was intended to compete directly with Ultralite 500.
39. Mr Redshaw had at least a sample of the Ultralite 500 panel, together with a sales brochure and the installation guide. He said in his witness statement that:

"Once we had decided that that was what we were going to do, it was a question of looking at competitor's systems, making sure that we were aware of exactly what the competition could do and what it could not do and any strengths and weaknesses of their systems."

40. Of Ultralite 500 he said:

"In terms of strengths and weaknesses, we were aware of many of the features within the Ultralite 500 system which we felt could be improved on. We noticed that the rafter top caps … on Ultralite 500 actually pass through the wall [plate] top cap, which is notched to suit the contour of those top caps, adding unnecessary costs. The wall plate top cap is fitted last, meaning the fitter has to climb over the roof to fit it. The eaves closure trim is very difficult to assemble. Having spoken with our Technical Department, they have found this very difficult to assemble on site. The drip trim is supplied in short lengths, and one for every panel is needed, meaning eight trims are needed for a 4m wide conservatory. The non modular cutting of the panel specifies a block of wood to support the panel along its cut length. All of the above were perceived weaknesses with the Ultralite 500 system and we spent a considerable time solving all of the above."

41. I have no doubt that Ultralite 500 was the primary object of study.
42. Although Mr Redshaw's witness statement was careful to suggest independent decisions and reasons for the various features of Pinnacle 500, I do not consider that this evidence withstood Mr Speck's cross-examination. There were a number of significant questions and answers:

"Q. You basically based it on the Ultralite 500, did you not?

A. I suppose you could say that. It was aimed to compete with the Ultralite 500. We can argue about that for some time but it was there to compete with Ultralite 500.

Q. You took the Ultralite 500 and thought about various changes, some of which you discarded, a couple you incorporated, but the rest of it was basically you took the Ultralite design?

A. Yes, I suppose so; yes."

43. Mr Redshaw's subsequent denials of copying were less cogent in the light of these answers. In addition I was not convinced by some of the explanations given in his witness statement for design decisions. For example:

i) "The size of the rafter itself had to be sufficiently strong to bear the loads so the height of it almost determines itself." Mr Redshaw had to accept in cross-examination that the strength of an I-beam can be significantly increased by increasing the width of the horizontal planes, rather than by increasing its height; so that the height did not determine itself;

ii) "The choice of 500mm wide was automatic, given the modular nature of the products intended market – eg 3m, 3.5m, 4m, etc. For example, a 4m wide roof needs 8 panels. There were other panels such as Ultraframe's Ultralite 250 on the market in the early 1990's, which were approximately 250mm wide and so 500mm was a natural progression." This seems to suggest that the decision to specify a width of 500 mm was arrived at without reference to Ultralite 500. But not only was Ultralite 500 the obvious comparator (since Pinnacle 500 was specifically intended to compete with it), but as Mr Redshaw said, he had a sample of the Ultralite 500 panels and "it would have been daft not to refer to it, not to play around with it or twist or bend some of the components". As Mr Rick (who was also part of the Eurocell design team) put it; "it was a 500 panel, as per the Ultralite 500, yes."

iii) At the time that the Eurocell design team were designing Pinnacle 500, Eurocell was the biggest distributor of Ultralite 500. Given that Pinnacle 500 was specifically designed as a competitor to Ultralite 500, and given also that Eurocell examined its competitors' products, I found unconvincing the evidence that the Eurocell design team did not have a complete Ultralite 500 system to examine.

44. Mr Rick was primarily responsible for the design of the top caps. In his witness statement he said:

"There are only certain ways in which top caps and end caps can be designed so as to cloak the aluminium. To go around a rafter for example, the top cap has to be roughly the same profile as the rafter. The end caps then follow the outline of the top caps. There is nothing innovative about top caps and end caps which have been used on conservatories for years. Top caps generally have a slight flat top section and angled legs towards the panel."

45. However, in cross-examination he accepted that although the top caps had to span the bar assembly there are lots of different styles that are possible. There was, therefore no need to have a top cap with a flat top and simple splayed legs. Eurocell had another system already in production, simply called "Pinnacle". It was used in assembling more traditional styles of conservatory. It had top caps and starter bar top caps among its components. Mr Redshaw said that the design of the Pinnacle 500 top caps and starter bar top caps was taken from the earlier Pinnacle system, rather than from Ultralite 500. He illustrated the similarities with an overlay. I accept his evidence on this point.
46. I find that the Eurocell design team were instructed to design a modular roofing system to compete directly with Ultralite 500; that with the exception of the top caps, they took Ultralite 500 as their starting point and made certain changes to it, partly to address perceived deficiencies in Ultralite 500 and partly to "design round" the patent and Ultraframe's design right. The real question is whether they succeeded.
47. I reproduce the diagram that Dr Kent produced of the lower coupling members and the lower part of the stiffening bar in Pinnacle 500. It is slightly inaccurate, but the inaccuracy does not matter for present purposes.

48. There are two features of this I must notice. First, the lower ends of the panels do not fit inside each other. They do not, therefore, restrain movement (either vertical or horizontal) in the panels themselves. Second, although the stiffening bar restrains the panels laterally, it does not actually exert any constant pressure on them.
49. Pinnacle 500 went onto the market in the spring of 2002. Mr Allen, Ultraframe's sales director, saw Pinnacle 500 on Eurocell's stand at the Glassex exhibition in March 2002. At first sight he thought it was Ultralite 500. It was only on closer inspection that he realised it was a Eurocell product. He said that "the thing that gave it away as a copy and not the Ultralite 500 was the guttering was slightly different in shape and appearance". Mr Wharf, who is responsible for technical training at Ultraframe, also saw Pinnacle 500 at the 2002 Glassex exhibition. He described his reaction as follows:

"I then went over to the stand to have a closer look at the mock conservatory with what I thought was an Ultralite 500 roof installed on top of it. It was only on closer inspection of the product that I realised that it was slightly different from our own low-pitch roof and that Eurocell had obviously manufactured it themselves. I was quite shocked that despite training people and working with the Ultralite 500 on a daily basis I couldn't actually tell that this was a different product to our Ultralite 500 until I started examining the product at close range. I had never ever seen anything as close to our system before. I recall saying to myself at the time that it was as close a match as you would ever get."

50. There was no challenge to the evidence of Messrs Allen or Wharf.

The patent

51. Both Mr Speck and Mr Thorley were kind enough to introduce me to the world of patents and what they both called "patentspeak". I hope that I have understood their teaching.
52. The form of an application for a patent is regulated by section 14 of the Patents Act 1977. Every application must contain a specification containing a description of the invention, a claim or claims and any drawing referred to in any description or any claim: section 14 (2). The specification must disclose the invention in a manner which is clear enough and complete enough for the invention to be performed by a person skilled in the art: section 14 (3). The claim must define the matter for which the applicant seeks protection; it must be clear and concise, and it must be supported by the description: section 14 (5).
53. The disputed patent ("the patent in suit") is GB 2300012. It has a priority date of 18 July 1992. The invention is called "Structures"; and the name of the inventor is given as Mr Lancaster. The patent specification began by setting out what was already known about self-supporting roofs ("the prior art"). It described roofing systems consisting of a number of plastic panels connected side by side, some of which had shaped parts which, when connected to each other, formed ducts. It referred to a number of earlier patents, including European Patent Specification No. 0070930A (which described "Sokoler 83"). It then set out the problem which the invention was intended to solve ("the deficiency"):

"Each of these elements has to be secured, usually by means of screws, before the next element can be connected, which can be time-consuming and does not readily permit later adjustments of element positions.

Furthermore, there is a limit to the reasonable span of such elements for strength purposes. Even multiple duct elements have a limited span because they have main ducts based on single duct elements which themselves have a structural limit on their widths.

An object of this invention is to provide an improved building structure."

54. The specification then goes on to give a general description of the invention ("the consistory clause"). This description is repeated word for word in claim 1. The description of the invention is:

"a structure comprising hollow building elements of plastics material coupled together side by side, the elements comprising flat top and bottom walls, vertical end and intermediate walls and an intermediate wall between and parallel to the top and bottom walls to form a plurality of rectangular ducts in two layers one on top of the other, the elements further comprising integral coupling members on opposite sides, wherein the lower coupling members comprise formations that interengage with formations of neighbouring elements and are held together by a stiffening member between the elements."

55. The next part of the specification describes preferable features that may be present in a product made in accordance with the invention. A feature described as "preferable" is patentspeak for "do not try to limit the scope of my claims by referring to this feature". The preferable features in this case include a type of lower coupling member described as follows:

"Preferred lower coupling members of building elements comprise a first member that is a horizontal channel into which a second member fits, the stiffening beam having a formation which fits over the said engaged first and second members to hold them together."

56. The next part of the specification describes the invention "by way of example only", by reference to two figures or diagrams (an "embodiment"). I reproduce figure 2.

57. Numbers in the text are references to numbered elements of the diagrams. Two parts of that description are important. First, it says:

"The stiffening beam 40, preferably extruded from aluminium, has upper and lower formations 42 and 44 respectively whereby adjacent building elements 10 are held together. The lower formation 44 comprises a downwardly opening channel 46 having a pair of opposed internal projections 48 that extend towards each other but do not meet. The formation 44 is sized so that when the lower coupling elements of adjacent building elements are brought together they fit between the projections 48 thereby holding them together."

58. Second, it describes the process of constructing a roof in the following terms:

"To construct a roof using building elements 10, the elements are laid side-by-side on a structure providing support at opposite ends of the elements with their lower elements engaged. A stiffening beam 40 is then slid into the space between the elements to fix the lower coupling members and engage the upper coupling members of the elements."

59. Finally the patent sets out the claims. The claims define the scope of the monopoly which the inventor claims. Claim 1 (and in particular the part that begins with the word "wherein") is the part of the patent that tells the world what the new and clever part of the invention is ("the characterising clause"). It is a repetition of the general description of the invention:

"A structure comprising hollow building elements of plastics material coupled together side by side, the elements comprising flat top and bottom walls, vertical end and intermediate walls and an intermediate wall between and parallel to the top and bottom walls to form a plurality of rectangular ducts in two layers one on top of the other, the elements further comprising integral coupling members on opposite sides, wherein the lower coupling members comprise formations that interengage with formations of neighbouring elements and are held together by a stiffening member between the elements."

60. It can be seen that the claim is for an invention made up of a number of different components ("integers"). Claim 3, which is the only other relevant claim, is:

"A structure as claimed in claim 1 or 2, wherein a lower coupling member comprises a first member that is a horizontal channel into which a second member fits, the stiffening beam having a formation that fits over the said engaged first and second members to hold them together."

61. This repeats the description of the "preferred" lower coupling members, which I have already quoted. Coupling members of this shape are shown in the diagrams, which are the only specific embodiment of the patent.
62. What is in dispute is the part of claim 1 which describes the lower coupling members, and in particular, the parts in italics:

"the elements further comprising integral coupling members on opposite sides, wherein the lower coupling members comprise formations that interengage with formations of neighbouring elements and are held together by a stiffening member between the elements."

63. It is agreed that neither word or phrase is a technical term. In a nutshell the dispute is this. Does the claim cover only a system in which the roofing panels partially restrain each other vertically from moving before the stiffening beam is slid into the gap between them? Or does it also cover a system in which the roofing panels are simply placed side by side, and are held together by the stiffening beam alone? It is agreed that this is a question of the interpretation of the patent. It is also agreed that if the patent is to be interpreted in the former sense, then Pinnacle 500 does not infringe. However, if the patent is to be interpreted in the latter sense, then it is agreed that it does.

Interpretation of the patent

64. The patent was granted under the Patents Act 1977. So the starting point for interpreting it must be the instructions contained in the Act itself. They are found in section 125 which says:

"(1) For the purposes of this Act an invention for a patent for which an application has been made or for which a patent has been granted shall, unless the context otherwise requires, be taken to be that specified in a claim of the specification of the application or patent, as the case may be, as interpreted by the description and any drawings contained in that specification, and the extent of the protection conferred by a patent or application for a patent shall be determined accordingly.

…

(3) The Protocol on the Interpretation of Article 69 of the European Patent Convention (which Article contains a provision corresponding to subsection (1) above) shall, as for the time being in force, apply for the purposes of subsection (1) above as it applies for the purposes of that Article."

65. The Protocol, which the Act applies to section 125 (1), says:

"Article 69 should not be interpreted in the sense that the extent of the protection conferred by a European patent is to be understood as that defined by the strict, literal meaning of the wording used in the claims, the description and drawings being employed only for the purpose of resolving an ambiguity found in the claims. Neither should it be interpreted in the sense that the claims serve only as a guideline and that the actual protection conferred may extend to what, from a consideration of the description and drawings by a person skilled in the art, the patentee has contemplated. On the contrary, it is to be interpreted as defining a position between these extremes which combines a fair protection for the patentee with a reasonable degree of certainty for third parties."

66. When the court is trying to interpret a written document, it is trying to ascertain what the document would mean to its target readership. A patent is no different. The target readership of a patent is that of people "likely to have a practical interest in the subject matter of the invention" with "practical knowledge and experience of the kind of work in which the invention was intended to be used" (Catnic v. Hill & Smith [1982] R.P.C. 183 at 242-3). So in order to interpret a patent, the court must equip itself with the practical knowledge and experience of the target readership ("don the mantle of the man skilled in the art"). For this purpose, the evidence of experts is not only helpful, but often essential. For the purpose of interpreting the patent the knowledge with which the court must equip itself is called "common general knowledge". This is not "general knowledge" in the sense in which most people understand it, but:

"what would in fact be known to an appropriately skilled addressee - the sort of man, good at his job, that could be found in real life."

General Tire & Rubber Co v. Firestone Tyre & Rubber Co Ltd [1972] R.P.C. 457 at 482.

67. It is, therefore, the technical background and mental equipment which it is necessary to have in order to be competent in the field of science or technology in question. It includes not merely what is in the skilled addressee's head, but also what is contained in frequently used text-books or trade literature. Nevertheless, the interpretation of documents remains a question of law for the court and is not answered by experts.
68. The principles that the courts have developed for the interpretation of patents are, in many respects, similar to the principles that they have developed for the interpretation of contracts (although, of course, a patent is a unilateral expression of intention). Specifically:

i) A patent is interpreted objectively, through the eyes of a skilled addressee;

ii) The court must have regard to the background as it existed at the date of publication of the patent;

iii) The court should admit evidence of technical terms;

iv) The patent must be read as a whole, both the body of the specification (together with any drawings) and the claims;

v) The court must adopt an interpretation which gives effect to the purpose of the author, judging the purpose from the language of the patent and the available background.

Glaverbel SA v. British Coal Corp. [1995] R.P.C. 255

69. However, the statutory instruction to apply the Protocol means that the court must steer a middle course between fairness to the inventor and certainty for third parties. This means that the language of the claim, as interpreted in the light of the drawings and specification, cannot be disregarded: Wheatley v. Drillsafe Ltd [2001] R.P.C. 133. On the one hand fairness to the inventor means that the claim must not be interpreted too liberally; otherwise it is at risk of being invalid. On the other hand it means that the claim must not be interpreted too narrowly, otherwise it will allow third parties to avoid the monopoly: Wheatley v. Drillsafe Ltd. But respect for the language of the claim means that an integer cannot be treated as struck out if it does not appear to make any difference to the inventive concept. It may have some other purpose buried in the prior art and even if this is not discernible, the patentee may have had some reason of his own for introducing it: Société Technique de Pulverisation v. Emson Europe Ltd [1993] R.P.C. 513 at 522.
70. As a general principle it is reasonable to infer that the inventor intended to include within his monopoly immaterial variants of the invention that he has described. A variant will be immaterial if it is not material to the way in which the invention works. In order to decide whether a variant is immaterial in this sense, Hoffmann J, in Improver Corporation v. Remington Consumer Products Ltd [1990] F.S.R. 181 proposed the following questions (the "Improver questions" or, nowadays, the "Protocol questions"):

"If the issue was whether a feature embodied in an alleged infringement which fell outside the primary, literal or acontextual meaning of a descriptive word or phrase in the claim ("a variant") was nevertheless within its language as properly interpreted, the court should ask itself the following three questions:

(1) Does the variant have a material effect upon the way the invention works? If yes, the variant is outside the claim. If no--
(2) Would this (i.e. that the variant had no material effect) have been obvious at the date of publication of the patent to a reader skilled in the art. If no, the variant is outside the claim. If yes--
(3) Would the reader skilled in the art nevertheless have understood from the language of the claim that the patentee intended that strict compliance with the primary meaning was an essential requirement of the invention. If yes, the variant is outside the claim.

On the other hand, a negative answer to the last question would lead to the conclusion that the patentee was intending the word or phrase to have not a literal but a figurative meaning (the figure being a form of synecdoche or metonymy) denoting a class of things which included the variant and the literal meaning, the latter being perhaps the most perfect, best-known or striking example of the class."

71. Strictly speaking, only the third of the questions is a question of construction. The first two are essentially questions of fact, as Hoffmann J himself noted. The answers to these two questions provide the factual background against which the third question is answered. In Wheatley v. Drillsafe Ltd, Aldous LJ clearly regarded the Protocol questions as being part of the process of interpreting the patent. Mr Thorley tells me that whether answering these questions is part of the process of interpreting the patent or is part of the process of deciding whether there has been an infringement is under consideration by the House of Lords. For the moment, however, the questions remain part of the process of interpreting the patent.
72. The purpose of the specification is to describe the inventor's invention and to explain why it is new. In explaining why it is new, the specification will set out what the inventor already knows to be old. Hence a reference in the specification to an earlier patent is a factor to be taken into account in interpreting the patent. As Jacob J explained in Beloit Technologies Inc. v. Valmet Paper Machinery Inc. [1995] R.P.C. 705 at 720:

"There is normally no reason to suppose the patentee when he set the limits of his monopoly knew of a particular piece of prior art which is therefore irrelevant in deciding what those limits are. Of course the position is different if the prior art is specifically acknowledged in the patent. The purposive construction would lead to a construction of a claim which did not cover that acknowledged prior art: it can hardly have been the inventor's purpose to cover that which he expressly recognises was old."

73. Mr Thorley submitted, and I agree, that much depends on the way in which the prior art is acknowledged. A mere reference to a prior patent does not necessarily require the addressee of the patent to dig it out and study it in detail. On the other hand if the specification identifies some particular feature of the prior patent as disclosing a problem which the inventor claims to have overcome, it may be of considerable relevance in interpreting the width of the claim. It is not that the prior patent is irrelevant in interpreting the patent; it is a question of what to do with it. If I may borrow from an analogous principle in interpreting contracts:

"First, in respect of contracts and contractual notices the contextual scene is always relevant. Secondly, what is admissible as a matter of the rules of evidence under this heading is what is arguably relevant. But admissibility is not the decisive matter. The real question is what evidence of surrounding circumstances may ultimately be allowed to influence the question of interpretation. That depends on what meanings the language read against the objective contextual scene will let in."

(Mannai Investment Co Ltd v. Eagle Star Life Assurance Co Ltd [1997] AC 749 at 768 per Lord Steyn).

74. Having described his invention, the inventor will make his claim to a monopoly. This is the function of the claim. In determining the scope of the monopoly, it is the claim rather than the specification which has primacy. As Laddie J put it in Merck & Co Inc v. Generics (UK) Ltd [2004] R.P.C. 607 at 623:

"The notional addressee is not expected to find the patentee's presumed intention from the specification."

75. In addition, the claim itself is divided into two parts, separated by the word "wherein" (in other cases by "characterised in" or another such phrase). Again I quote Laddie J in Merck & Co Inc v. Generics (UK) Ltd:

"Furthermore, the patentee may choose a form of language which emphasises which features of an invention are important and which are not. For example it is common to find claims which start with general description followed by "characterised in" followed by a list of features. The addressee would appreciate that the latter features are particularly important but the features before the words "characterised in" are less so. If there is a variant to the latter which obviously does not affect the way in which the invention works, the notional reader may be reasonably confident that the inventor wanted to cover this variant as well. In these types of cases, the monopoly is likely to extend to the new variant."

76. I take from this that it is more likely that the patent requires strict adherence to the integers that follow the word "wherein", than to the integers that precede it.
77. A patent may (and usually does) make multiple claims. It is common ground that the conventional hierarchy of claims is to start with the widest, and progress through subsidiary claims of ever narrowing scope. Thus in interpreting a claim it is reasonable to infer that an earlier claim would be read as wider than a later subsidiary claim.

Interpretation of the patent in suit

78. After this somewhat lengthy introduction, I can come to the interpretation of the patent in suit. I must first describe what, in my judgment, is the relevant technical knowledge with which the addressee is equipped. He would at least know that:

i) If you were fixing roofing panels for a lean-to conservatory you would fix them one at a time;

ii) While you were fixing them you would be working above your head;

iii) To fix them at all (if you were working on your own) you would need to support one end of the panel at the wall-plate (against the wall of the main building) and the other end on the front frame of the conservatory;

iv) Each panel would invariably have some flexibility, as well as uneven edges produced either by the extrusion process, or their subsequent treatment on site or in transport. The extruded plastic panels would often be "laterally bowed" (caused by thermal shrinkage in extrusion) i.e. would be slightly banana shaped along their length;

v) When the first panel was fixed in place, it would be very difficult to 'line up' the next panel all the way along its length especially if you had to engage the second panel at an angle to the first;

vi) Because of these characteristics of extruded plastic panels, it was difficult for the manufacturer to achieve exactness, and tolerances would be expected.

79. Although, as I have said, it is agreed that the word "interengage" is not a technical term, I am not sure that it is an "ordinary" English word either, not least because, so far as counsel's researches go, it is not found in any standard dictionary (and, for what it is worth, my spellchecker does not recognise it either). Claim 1 of the patent applies it to the lower coupling members, which are to interengage. In my judgment that means that they must engage with each other. But what is meant by "engage"? One of the meanings given to the word "engage" by the Shorter Oxford English Dictionary is:

"Of part of a mechanism: come into contact with or fit into a corresponding part, so as to prevent or transmit movement, cause to do this; interlock."

80. Another, given by the New Oxford Dictionary of English, on which Mr Speck relied, is:

"(engage with) establish a meaningful contact or connection with … (of a part of a machine or engine) move into position so as to come into operation".

81. Mr Speck submits that the lower coupling members of the panels are "interengaged" if the panels are simply laid side by side, leaving it to the stiffening member to bring them together, in the manner of a zip that Mr Savage described. This is not, to my mind, the most obvious meaning of the word "interengaged", even if it is a possible meaning of the word "engaged". However, it might be possible to interpret the word in the way that Mr Speck says it should be interpreted if the specification, as it would be read by a person with the common general knowledge, clearly described the "zipping" action. The process is described as follows:

"To construct a roof using building elements 10, the elements are laid side-by-side on a structure providing support at opposite ends of the elements with their lower elements engaged."

82. At this stage in the process the reader is told that the panels must be laid side by side. He is told that the panels must be supported at opposite ends. So they will be vertically aligned. (The vertical alignment will follow the pitch of the roof). The process of laying the panels side by side will "engage" the lower elements. They are therefore engaged before the stiffening bar is inserted into the assembly. This operation will be carried out from the underside of the panels, probably on a ladder. The next step is to slide in the stiffening member. This has to be done from the front of the panel. So the installer will have to leave the panels and go to the outside of the conservatory structure in order to slide in the stiffening bar. The specification describes this as follows:

"A stiffening beam 40 is then slid into the space between the elements to fix the lower coupling members and engage the upper coupling members of the elements".

83. Thus what the stiffening member does is two things. It engages the upper coupling members, and it fixes the lower ones. It does not need to engage them, because they are already engaged. Engagement of the lower coupling members (whatever it means) is not carried out by the insertion of the stiffening member. This seems to me to be consistent with the language of claim 1 which describes the lower coupling members as being "held" together by the stiffening member rather than "pulled" or "brought" together. Thus far the "zipping" process has not been described. Mr Speck relies on that part of the specification that says:

"The formation 44 [i.e. the lower part of the stiffening bar] is sized so that when the lower coupling elements of adjacent building elements are brought together they fit between the projections 48 [i.e. the projections in the stiffening bar] thereby holding them together."

84. He says that the description of the coupling members being "brought together" describes the "zipping" process, in which the stiffening bar "zips" the coupling members together. The linguistic problem with this is that the specification does not say that the coupling elements are brought together by the stiffening bar itself; it simply says that they are brought together, without saying how. Mr Savage agreed that the words of the specification, taken on their own, did not explicitly describe the "zipping" process. But he said that, with his technical knowledge, he could see from the drawings that this is how the invention would work. He appreciated that the stiffening bar would work by actively pulling the panels together:

"… because there is no other means to keep them together when you are working with long lengths of propyl overhead with those formations. They just would not sit there waiting for the additional member."

85. He also drew attention to the tolerances that one would expect in extruded plastic; so that precise horizontal alignment would be very difficult. Mr Speck also submitted that the fact that extruded plastic necessarily has tolerances means that in a practical sense the panels are engaged, even if they are not actually touching each other along their whole length. In addition, the fact that tolerances are needed in the plastic coupling members means that the lower portion of the stiffening bar, and in particular the projections (48) "hold" the coupling members together, even though they do not actually exert clamping pressure on them.
86. Mr Speck now turns to claim 3 (which is the embodiment actually described in the parts of the specification I have quoted). It claims:

"A structure as claimed in claim 1 or 2, wherein a lower coupling member comprises a first member that is a horizontal channel into which a second member fits, the stiffening beam having a formation that fits over the said engaged first and second members to hold them together."

87. It speaks of two coupling members, one of which "fits" into the other. This integer is additional to claim 1, so that claim 1 should not be interpreted so as to require one coupling member to "fit" into the other. Even in this claim, where one member must fit into the other, the coupling members do not hold themselves together; they are held together by the stiffening beam. It follows, therefore, that in claim 1 "interengaged" must be given a broader meaning than simply one member fitting into the other.
88. Mr Speck next refers to the specification to Sokoler 83. He says that the claim cannot realistically have been intended to cover the invention disclosed in that specification. The stiffening bar in Sokoler 83 is (in section) a simple rectangle. It operates as a wedge. The adjoining panels must be pushed together and interlock before the bar is inserted into the duct that they form, otherwise it will have no effect at all. The stiffening bar impedes no lateral movement in the panels at all. It merely impedes vertical movement. By contrast, in the patent in suit, the stiffening member impedes lateral movement in the coupling members by holding them in position.
89. In the light of all this, Mr Speck now turns to the disputed part of claim 1. I repeat the relevant part for convenience:

"wherein the lower coupling members comprise formations that interengage with formations of neighbouring elements and are held together by a stiffening member between the elements."

90. The first point that Mr Speck makes is that what is held together by the stiffening member is not the panels, but the coupling members. This differentiates the claim from Sokoler 83. The second point that Mr Speck makes is that the only thing that holds the coupling members together is the stiffening member. The coupling members do not hold themselves together. This, too, differentiates the claim from Sokoler 83. Third, the concluding integer of claim 1 makes it clear that the stiffening member holds the coupling members together directly, unlike Sokoler 83 which only holds them together indirectly. Fourth, because claim 3, which is narrower than claim 1, requires one coupling member to fit into the other coupling member, it is reasonable to infer that claim 1 makes no such requirement. Fifth, in the mind of a skilled addressee (such as Mr Savage) it would be apparent that the stiffening member actually performed the function of bringing the coupling members into alignment. Alternatively, if the claim was ambiguous or unclear in this respect, it was permissible to go to the specification to resolve the uncertainty. Sixth, the fact that the stiffening member must hold the coupling members together does not mean that the stiffening member must act as a clamp. All that is needed is a restriction on lateral movement on both sides. The skilled addressee would know that the practical problems of extruding plastic means that they cannot be extruded with pinpoint accuracy such that the coupling members will fit hard up against the projections in the stiffening member while at the same time allowing the stiffening member to be slid in between the panels. Seventh, once one realises the function of the stiffening member it becomes clear that the claim does not require any degree of interlocking or fixing of the coupling members before the stiffening beam is slid into place.
91. Mr Thorley, on the other hand, said that the "ordinary" meaning of "interengage" carried with it some connotation of interlocking. It was necessary for the coupling members to be interlocked in this sense, because they would need to be vertically aligned so that the upper coupling members could be engaged by the stiffening member. The problem with this latter submission is that, although the specification refers to upper coupling members, the claims do not. So the extent of the monopoly claimed does not include any requirement for upper coupling members. Moreover, it seems to me that the necessary amount of vertical alignment would in any event be supplied by the direction to support the panels at each end as they are laid side by side. Mr Thorley also said that the only use that the skilled addressee would make of Sokoler 83 was to identify the problem (or deficiency) that the invention was designed to solve. This was itself spelled out in the specification and consisted only of the need to screw each panel in place before the next one was fitted. I think that this is too narrow a reading of the deficiency. The patent in fact describes the deficiency as follows:

"Each of these elements has to be secured, usually by means of screws, before the next element can be connected, which can be time-consuming and does not readily permit later adjustments of element positions."

92. The deficiency thus described is that the panels have to be secured. Although the usual method of securing them is by screws, the identified deficiency is that they have to be secured at all. Thus one would expect that the inventor has invented a system in which the panels do not have to be secured at all before the next panel can be connected. Moreover, the deficiency described also foreshadows an invention which enables later adjustment of the panels to take place. One means of doing that disclosed by the specification is by the insertion of the stiffening beam.
93. In my judgment Mr Thorley's strongest point is the word "interengage". In its natural meaning, I would subscribe to the essence of the meaning in the Shorter Oxford English Dictionary, namely that the function of engagement is to transmit or restrain movement. Mr Thorley submits that the informed reader of the claim and specification would understand that the lower coupling members of the panels had to be fitted together ("interengaged") before the stiffening bar was inserted, and that the stiffening bar performed a static function ("held together") rather than a dynamic one. Mr Savage's heavy reliance on what he, as a skilled addressee, would deduce from his reading of the specification and diagrams, against the background of his technical knowledge, seems to me, on the face of it, to do what the Protocol says should not be done: namely to use the claim as a guideline, and to find from the consideration by a skilled person of the description and drawings, that which the patentee has contemplated. It is here, I think, that the Protocol questions come into play.
94. In my judgment the fact that the panels in Pinnacle 500 are not restrained at all by any part of the panels themselves does not have a material effect on the way in which the invention works. I have in mind especially that the informed reader of the specification would know that plastic panels would often arrive on site laterally bowed, and that it would, in practice, be very difficult to line them up all along their length. All that was necessary to make the invention work would be to place them side by side, sufficiently close together to enable the stiffening bar to act as a "zip". I therefore answer the first Protocol question "No". In the light of this knowledge, I consider that this would have been obvious to an informed reader of the patent and specification. It would therefore have been obvious to such a reader that a variant which did not require the coupling members to restrain each other before the stiffening beam was inserted would have no material effect on the way that the invention works. I therefore answer the second Protocol question "Yes". I pass to the third question, which is the question of construction. Although I accept that claim 3 is narrower than claim 1, I do not consider that it is as broad as Mr Speck suggests. What it describes is a particular shape of coupling member, namely one with a horizontal channel. So it is not simply describing an invention in which the two lower coupling members fit into each other, but an invention in which they fit into each other in a particular way. This does not, in my judgment, support an interpretation of claim 1 in which the coupling members do not interlock at all. Moreover, it seems to me that the very description of the members as "coupling" members, suggests that they do more than merely touch each other. I agree with Mr Thorley that the word "interengage" conveys the sense that the coupling members in some way interlock or at least partially restrain movement. When added to the description of the stiffening member as holding rather than pulling the coupling members together, I agree also that the stiffening bar is described as having a static rather than a dynamic effect. It seems to me, therefore, that to ignore these two integers would be to strike out an integer (indeed both integers following the word "wherein") simply on the ground that they do not seem to make any difference to the inventive concept. I conclude, therefore, that for reasons which the patentee knows best, the skilled reader would conclude that the patentee intended that strict compliance with those integers was an essential part of the invention. I therefore answer the third Protocol question: "Yes".
95. It follows that Pinnacle 500 is outside the monopoly claimed by the patent in suit, and the claim to infringement fails.

Novelty

96. Eurocell counterclaims for an order revoking the patent. One of the grounds on which it makes that claim is that the invention is not new. An invention is new if it "does not form part of the state of the art": Patents Act 1977 s. 2 (1). Section 2 (3) says:

"The state of the art in the case of an invention shall be taken to comprise all matter (whether a product, a process, information about either, or anything else) which has at any time before the priority date of that invention been made available to the public (whether in the United Kingdom or elsewhere) by written or oral description, by use or in any other way."

97. What Eurocell relies on is the prior publication of two patents: Ultralite 250 and Sokoler 83. Where a patent is alleged to be invalid on the ground of prior publication, the legal test is that described by the Court of Appeal in General Tire & Rubber Co v. Firestone Tyre & Rubber Co Ltd [1972] R.P.C. 457 at 485:

"If the prior inventor's publication contains a clear description of, or clear instructions to do or make, something that would infringe the patentee's claim is carried out after the grant of the patentee's patent, the patentee's claim will have been shown to lack the necessary novelty, that is to say, it will have been anticipated. The prior inventor, however, and the patentee may have approached the same device from different starting points and may for this reason, or it may be for other reasons, have so described their devices that it cannot be immediately discerned from a reading of the language which they have respectively used that they have discovered in truth the same device; but if carrying out the directions contained in the prior inventor's publication will inevitably result in something being made or done which, if the patentee's patent were valid, would constitute an infringement of the patentee's claim, this circumstance demonstrates that the patentee's claim has in fact been anticipated.
If, on the other hand, the prior publication contains a direction which is capable of being carried out in a manner which would infringe the patentee's claim, but would be at least as likely to be carried out in a way which would not do so, the patentee's claim will not have been anticipated, although it may fail on the ground of obviousness. To anticipate the patentee's claim the prior publication must contain clear and unmistakable directions to do what the patentee claims to have invented."

98. In a nutshell, what is new is the way in which the lateral restraint of the coupling members of the plastic panels is performed by the stiffening beam, rather than by "hooks" in the plastic panels themselves. It is also new that the panels can be fitted together by horizontal movement alone, without any lifting or tilting. Instructions to do this are not, in my judgment, contained in any of the prior art on which Eurocell relies. The objection based on lack of novelty therefore fails.

Obviousness

99. Assuming that the invention described in the patent is new, Eurocell say that it does not involve an inventive step. Section 3 of the 1977 Act says:

"An invention shall be taken to involve an inventive step if it is not obvious to a person skilled in the art, having regard to any matter which forms part of the state of the art by virtue only of section 2(2) above …"

100. The courts are regularly enjoined to apply the structured approach of Oliver LJ in Windsurfing International Inc v. Tabur Marine Great Britain Ltd [1985] R.P.C. 59 at 73 in answering the question whether a claimed invention is obvious. This involves asking and answering four questions:

"The first is to identify the inventive concept embodied in the patent in suit. Thereafter, the court has to assume the mantle of the normally skilled but unimaginative addressee in the art at the priority date and to impute to him what was, at that date, common general knowledge in the art in question. The third step is to identify what, if any, differences exist between the matter cited as being "known or used" and the alleged invention. Finally, the court has to ask itself whether, viewed without any knowledge of the alleged invention, those differences constitute steps which would have been obvious to the skilled man or whether they require any degree of invention."

101. Mr Savage described the inventive step as follows:

"Instead of the panels being linked together, and then the beam being added … the stiffening member was given an active role which meant that the panels did not already have to be linked in place. In order to perform this active role the stiffening member is shaped to hold the formations of the lower coupling members together. Instead of the installer fabricator having to do the difficult job first before he added the stiffening member, the action of inserting the stiffening member did the difficult job for him. Using the stiffening member itself to perform this active role was something which went against the grain."

102. I accept that what Mr Savage described is an inventive step, but I have held that it was not an inventive step "embodied in the patent". Nevertheless, it seems to me that a modified version of this description of the inventive step was embodied in the patent. As I have said, what was new in the patent was that the lateral restraint of the coupling members was a function of the stiffening beam, and not any part of the panels themselves; and also that the panels could be fitted together without any lifting or tilting. That was, in my judgment, an inventive step.
103. Mr Savage described the "state of the art" as follows:

"The approach in the Ultralite 250 Patent was that the panels would be linked together directly to each other. This was the customary approach, as can be seen from the other pieces of prior art cited by the Defendants which are discussed below. The stiffening member was then added to give extra stiffness and strength to the panel. The stiffening member was an "I -section" in a simple shape to provide maximum strength at minimum cost. When considering a new design for the connections between two panels, the design team would therefore concentrate on the interlocking formations, not the stiffening bar. It is true that the stiffening member had a limited "locking" role, in that it prevented the panels from being unlinked whilst the stiffening member is in place, however, its role was a passive one which was secondary to its principal role as a stiffening beam."

104. Again, I accept this description.
105. It follows from these two descriptions that the difference between the inventive step and the state of the art was that the stiffening beam in the patent in suit itself exerted lateral (or horizontal) restraint on the coupling members.
106. Mr Savage was of the view that this change would not have been obvious to a person (or a design team) skilled in the art. There was no evidence to gainsay him. Mr Thorley suggested in cross-examination that if there was a problem with the strength of the joint between the plastic panels, it would have been obvious to resort to the stiffening member as a clamp. Not only was this, at least to my mind, not an obvious solution at all, it would have required a major redesign of the panels. In my judgment the attack on the ground of obviousness also fails.
107. I should mention by way of footnote that Ultralite 250 and Sokoler 83 both disclosed arcuate rather than flat tops to the plastic panels. But Mr Savage accepted that a change from arcuate tops to flat tops was an obvious step, and did not try to suggest the contrary.
108. It follows that the counterclaim for revocation of the patent fails.

Design right

109. Design right was created by the Copyright Designs and Patents Act 1998 ("the 1998 Act"). Section 213 of the 1998 Act says:

"(1) Design right is a property right which subsists in accordance with this part in an original design.

(2) In this Part "design" means any aspect of the shape and configuration (whether internal or external) of the whole or any part of an article.

(3) Design right does not subsist in -

(a) a method or principle of construction,

(b) features of shape or configuration of an article which -

(i) enable the article to be connected to, or placed in, around or against, another article so that either article may perform its function,

(ii) are dependent upon the appearance of another article of which the article is intended by the designer to form an integral part,

…

(4) A design is not "original" for the purposes of this Part if it is commonplace in the design field in question at the time of its creation.

…

(6) Design right does not subsist unless and until the design has been recorded in a design document or an article has been made to the design."

110. "Original" and "commonplace". Subject to the exception for "commonplace" designs, a design is "original" for the purpose of design right if it is the product of skill and labour; and it has not been copied from another pre-existing design. This type of originality is called originality "in the copyright sense". It is not necessary that the design be novel. If a design is original in that sense, the court must then consider whether it is "commonplace". If it is commonplace then it cannot be "original" within the meaning of section 213. To that extent there is a difference between a design that is original "in the copyright sense", and a design that is original "in the design right sense". It is important to note that the exclusion from originality is a design that is commonplace, and not part of a design that is commonplace. So you do not strip away those parts of a design that might be said to be commonplace and then look at the rump. If you did that, you might find that there was nothing left to protect. Thus the fact that a design is made up of features which, individually, are commonplace does not necessarily make the design as a whole commonplace, provided that the combination is not commonplace: Farmers Build Ltd v. Carier Bulk Materials Handling Ltd [1999] RPC 461 at 476. Because design right has a relatively short life and confers only limited protection against copying, the references to "commonplace" designs should be narrowly construed. Thus "commonplace" is not to be equated with "well-known": Farmers Build Ltd v. Carier Bulk Materials Handling Ltd [1999] RPC 461 at 481. In the Farmers Build case the Court of Appeal set out a five-stage approach which the court should follow in deciding whether a design is (a) original in the copyright sense and (b) not commonplace:

"(1) It should compare the design of the article in which design right is claimed with the design of other articles in the same field, including the alleged infringing article, as at the time of its creation.

(2) The court must be satisfied that the design for which protection is claimed has not simply been copied (e.g. like a photocopy) from the design of an earlier article. It must not forget that, in the field of designs of functional articles, one design may be very similar to, or even identical with, another design and yet not be a copy: it may be an original and independent shape and configuration coincidentally the same or similar. If, however, the court is satisfied that it has been slavishly copied from an earlier design, it is not an "original" design in the "copyright sense" and the "commonplace" issue does not arise.

(3) If the court is satisfied that the design has not been copied from an earlier design, then it is "original" in the "copyright sense". The court then has to decide whether it is "commonplace". For that purpose it is necessary to ascertain how similar that design is to the design of similar articles in the same field of design made by persons other than the parties or persons [connected] with the parties.

(4) This comparative exercise must be conducted objectively and in the light of the evidence, including evidence from experts in the relevant field pointing out the similarities and the differences, and explaining the significance of them. In the end, however, it is for the court and not for the witnesses, expert or otherwise, to decide whether the design is commonplace. That judgment must be one of fact and degree according to the evidence in each particular case. No amount of guidance given in this or in any other judgment can provide the court with the answer to the particular case. The closer the similarity of the various designs to each other, the more likely it is that the designs are commonplace, especially if there is no causal link, such as copying, which accounts for the resemblance of the compared designs. If a number of designers working independently of one another in the same field produce very similar designs by coincidence the most likely explanation of the similarities is that there is only one way of designing that article. In those circumstances the design in question can fairly and reasonably be described as "commonplace". It would be a good reason for withholding the exclusive right to prevent the copying in the case of a design that, whether it has been copied or not, it is bound to be substantially similar to other designs in the same field.

(5) If, however, there are aspects of the plaintiff's design of the article which are not to be found in any other design in the field in question, and those aspects are found in the defendant's design, the court would be entitled to conclude that the design in question was not "commonplace" and that there was good reason for treating it as protected from misappropriation during the limited period laid down in the 1988 Act. That would be so, even though the design in question would not begin to satisfy any requirement of novelty in the registered designs legislation."

111. I have put "connected" in paragraph (3) in brackets, because the report reads "unconnected"; but this seems to be a typographical error (although it might depend on how you punctuate the sentence).
112. Whether a design is "commonplace" must be judged by reference to "the design field in question". Identification of the design field in question is one of fact and degree: Scholes Windows Ltd v. Magnet Ltd [2002] F.S.R. 172.
113. The "must fit" exception. The exclusion from design right referred to in section 231 (3)(b) (i) is called the "must fit" exception. Its genesis lies in the policy that there should be a free market in spare parts which must, therefore, be manufactured so as to fit into the assembly of which they are parts. The purpose of the exclusion is to exclude "all interface features": Ocular Sciences Ltd v. Aspect Vision Care Ltd [1997] RPC 289 at 425. But it is only the interface features that are excluded from design right: Ultraframe UK Ltd v. Fielding [2003] R.P.C. 435 at 458. Thus if a designer wants to design, say, a light bulb to fit a standard lamp fitting, that part of the bulb consisting of the bayonet fitting is excluded from design right. But above the bayonet, the bulb may be a globe, a candle, a rod, a mushroom or any other shape (some shapes may be excluded from design right as being commonplace, but that is another matter).
114. There is one point of principle that arises under this head. It is the extent of the "must fit" exclusion. The point arises in this way. The coupling members of the Ultralite 500 plastic panels terminate in a "U-shaped" channel. The "U-shaped channel" is the channel into which the legs of the stiffening beam fit. But the fit is not a snug one, partly due to the need for tolerances in the size of the channel, attributable to the process of extruding plastic. The channel could have been made wider, or it could have been made narrower. Mr Speck submits that because the fit between the channel and the lower limb of the stiffening beam is not a snug one, there is no part of the channel which falls within the "must fit" exclusion. On the face of it that submission is a surprising one. Section 213 (3)(b)(i) is concerned with features of an article which enable the article to be connected to, or placed in, around or against, another article so that either article may perform its function. The features in question are actual features of an actual article. They are not hypothetical features. So the starting point must be the article as it is actually designed. In relation to any particular feature of the article, the question is: does that feature enable the article to be connected to etc another article? Transposed into the articles in the present case, I must ask: does the channel (the feature of the panel under consideration) enable the panel to be connected to another article (the stiffening beam)? The answer to that question must, in my judgment, be "Yes". The fact that it could have been a differently designed feature performing the same function does not seem to me to matter.
115. The "must match" exception. The exclusion from design right referred to in section 231 (3)(b) (ii) is called the "must match" exclusion. Like the "must fit" exclusion, the "must match" exclusion is a narrow one. It applies only to those features of the design that must match the remainder of the assembly of which it is a part: Ultraframe UK Ltd v. Fielding [2003] R.P.C. 435 at 458. As the sub-section makes clear, this is limited to features dependent on the appearance of another article. Thus internal (invisible) features of an article are not excluded by the "must match" exclusion.
116. What is the design? Laddie J explained in Ultraframe UK Ltd v. Fielding [2003] R.P.C. 435 at 457 that you must first exclude those aspects of the design of an article that fall within the "must fit" and "must match" exclusions. He continued:

"The combination of features left after individual features are excluded under those provisions is the design in issue. If that combination of features is commonplace, the design is not protected."

117. I have some reservations whether this is strictly correct. Whether a design is commonplace is relevant in deciding whether it is original for the purposes of section 213 (1). If it is original then design right subsists in it, with the exclusion of those features which fall within the "must fit" and "must match" exclusions. So it seems to me that the question whether a design is commonplace is a question that must be considered before subtracting from it any features that fall within those two exceptions: see the Farmers' Build case. However, I do not think that this makes any difference in the present case.
118. What is the article? In Ocular Sciences Ltd v. Aspect Vision Care Ltd [1997] RPC 289 at 422 Laddie J said:

"The proprietor can choose to assert design right in the whole or any part of his product. If the right is said to reside in the design of a teapot, this can mean that it resides in design of the whole pot, or in a part such as the spout, the handle or the lid, or, indeed, in a part of the lid. This means that the proprietor can trim his design right claim to most closely match what he believes the defendant to have taken."

119. (The reference to "trimming" the claim is not to be understood in any pejorative sense, because the designer of an article is entitled to design right not only in any aspect of the design of the article as a whole but also in any aspect of the design of each part of it: A. Fulton Co. Ltd v. Totes Isotoner (UK) Ltd [2004] R.P.C. 301). But it does mean that it becomes of great importance that the claimant identifies clearly those aspects of the design that he says have been reproduced. The claimant's pleading is therefore crucial.
120. The question "what is the article" also has a significant bearing on the extent of the "must fit" exclusion. In Electronic Techniques (Anglia) Ltd v. Critchley Components Ltd [1997] F.S.R. 410 Laddie J refused to strike out a defence based on the proposition that that the "must fit" provisions apply to two interfitting articles carrying the features which were assembled together to form the whole or part of another larger article. However, in Baby Dan AS v. Brevi SRL [1999] FSR 377 Mr David Young QC, sitting as a judge of the Chancery Division, refused to follow this approach. He held that where the claimant claims design right in his product as a whole, he is also entitled to rely on design right in the individual components of the larger article as part of the larger product; and the "must fit" exclusion does not exclude from design right the shape or configuration of the various parts necessary to enable those parts to be assembled into the larger product.
121. I am therefore faced with two conflicting decisions of judges of co-ordinate jurisdiction. In those circumstances, the ordinary rules of precedent require me to follow the second decision, in preference to the first, unless I am convinced that the second judge was wrong not to follow the first: Colchester Estates (Cardiff) v. Carlton Industries plc [1986] Ch. 80. Mr Thorley accepted that at first instance Mr Young's decision represents the law; but he reserved the right to argue the contrary in the Court of Appeal, if need be. This considerably curtails the scope of the "must fit" exclusion, to the extent that Ultraframe relies on the assembly as a whole.
122. Mr Speck concentrated his final submissions on design right on the assembly as a whole. For this purpose the assembly as a whole is the assembly as pleaded in paragraph 4(c) of the Particulars of Claim.
123. Is the design of the assembly of Ultralite 500 (considered as a whole) original in the copyright sense? In my judgment it is. It was the product of skill and labour, and was not copied from anything else.
124. Is the design of the assembly of Ultralite 500 (considered as a whole) commonplace in the design field in question? The first thing to do is to identify the design field in question. There appeared at one stage to be an issue whether a design that was commonplace in, say, Germany was a design that was common place in the design field in question; or whether the design field was restricted to the United Kingdom. In A. Fulton & Co Ltd v. Totes Isotoner (UK) Ltd [2003] R.P.C. 499 at 522 HH Judge Fysh QC held that the inquiry whether a design was commonplace should be restricted to the United Kingdom. Although Mr Thorley at one stage submitted that this was wrong, in the end, however his submission did not seem to me to conflict with this at all. His submission, as it developed, was that a design that originated in and was commonplace in, say, Germany, could be commonplace for the purposes of section 213 if it was well-known to designers in the field in the United Kingdom. Subject to the gloss that "well-known" is not the same as "commonplace", I do not consider that there is any conflict between Judge Fysh's decision and Mr Thorley's submission. What is important is not whether articles made to a design are commonplace, but whether the design is commonplace. If it is, then it does not seem to me to matter where it comes from. The only other point that I need to make at this stage is that the design field is, I think, limited to the range of products (or designs) that a particular designer (or design team) could be expected to design, and does not (or does not necessarily) extend to the whole range of sources from which they might derive inspiration. Mr Thorley said that the design field in question in the present case is not limited to conservatory roofs, let alone low pitch conservatory roofs. On the contrary it extends to any roof structure involving the use of hollow plastic extruded elements. I agree with him that the design field is not limited to low pitch conservatory roofs. But Mr Savage said that the design field in question was limited to conservatory roofs; and I do not think that this part of his evidence was seriously challenged. I do not regard the evidence that people read trade magazines in which windows were also advertised as enlarging the design field.
125. Even if individual features of the design were unoriginal or commonplace (which I return to later), the overall assembly was neither. Mr Savage put it as follows:

"I believe that the Ultralite 500 system was groundbreaking because it was really the first time that a designer had designed a system specifically for low-pitch conservatory roofs and had given any great thought to the whole package of the roof, including the panels, the top caps and end caps and so forth. Those roofs made with 'modular' panels before this date had not been designed as a whole package in which each component complemented the others. The assembly of the Ultralite 500 system as a whole has an expensive feel – it was a substantial step forward from what was previously available."

"To me in 1992 (and even today) the whole design of the Ultralite 500 system and the individual elements of it, stood out immediately (and still stand out) from the alternatives that were available."

126. Based on the examples of contemporary designs that I was shown, I agree with this.
127. Must fit/must match. It is conceded that (at least at this level in the judicial hierarchy) neither of these exclusions applies to Ultraframe's reliance on design right in the assembly as a whole. The assembly as a whole does not have to fit or match anything else (apart of course from the wall of the house).
128. Does Pinnacle 500 infringe design right in the assembly of Ultralite 500 as a whole? Design right is infringed by a person who, without the licence of the design right owner, does anything which is the exclusive right of the design right owner: 1988 Act s. 226 (3). Design right gives the design right owner the exclusive right to reproduce the design for commercial purposes by making articles to that design: 1988 Act s. 226 (1). This means copying the design so as to produce articles exactly or substantially to that design: 1988 Act s. 226 (2). This involves two questions: was Pinnacle 500 copied from Ultralite 500? If so, did it produce articles exactly or substantially to the copied design? I have already held that the Eurocell design team took Ultralite 500 as their starting point and made certain changes to it, partly to address perceived deficiencies in Ultralite 500 and partly to "design round" the patent and Ultraframe's design right. I find, therefore, that Pinnacle 500 was copied from Ultralite 500 (apart from the top caps). I do not consider that Eurocell produced articles "exactly" to the Ultralite 500 design. As I have said, some changes to it were made. Did it produce articles "substantially" to that design? In evaluating this question I must look at the design through the eyes of the person to whom the design was directed. In this case that means the householder or possibly the builder. The question is whether any differences in design, seen through the eyes of such a person would attract the eye and would be seen by him as functionally significant: C & H Engineering Ltd v. F Klucznik & Sons Ltd [1992] FSR 421. Bearing in mind the evidence of Messrs Allen and Wharf, who were intimately acquainted with Ultralite 500, reinforced by the evidence of Mr Savage, I consider that the differences in design would have been imperceptible to the eyes of the person for whom the design was directed, whether that person is a householder or a builder. I hold, therefore, that Eurocell has infringed Ultraframe's design right in the assembly of Ultralite 500 as a whole.
129. The impact of the 1995 redesign. Ultralite 500 was designed in April or May 1992 and went onto the market in March 1993. Design right in the system as first marketed therefore expired in March 2003: 1988 Act s. 216 (1)(b). Did the 1995 redesign create a new design right in the system as a whole? Mr Thorley says that it did not. He says that the 1995 redesign either created no new design right at all; or, if it did, it created a new design right only in the changes that were made in 1995.
130. The starting point must be to consider whether the 1995 redesign produced an assembly which was "original in the copyright sense". The leading case on this question is Interlego AG v. Tyco Industries Inc [1989] A.C. 219. In that case Lord Oliver said:

"Take the simplest case of artistic copyright, a painting or a photograph. It takes great skill, judgment and labour to produce a good copy by painting or to produce an enlarged photograph from a positive print, but no one would reasonably contend that the copy painting or enlargement was an "original" artistic work in which the copier is entitled to claim copyright. Skill, labour or judgment merely in the process of copying cannot confer originality. In this connection some reliance was placed on a passage from the judgment of Whitford J. in L.B. (Plastics) Ltd. v. Swish Products Ltd. [1979] R.P.C. 551, 568-569, where he expressed the opinion that a drawing of a three-dimensional prototype, not itself produced from the drawing and not being a work of artistic craftsmanship, would qualify as an original work. That may well be right, for there is no more reason for denying originality to the depiction of a three-dimensional prototype than there is for denying originality to the depiction in two-dimensional form of any other physical object. It by no means follows, however, that that which is an exact and literal reproduction in two-dimensional form of an existing two-dimensional work becomes an original work simply because the process of copying it involves the application of skill and labour. There must in addition be some element of material alteration or embellishment which suffices to make the totality of the work an original work. Of course, even a relatively small alteration or addition quantitatively may, if material, suffice to convert that which is substantially copied from an earlier work into an original work. Whether it does so or not is a question of degree having regard to the quality rather than the quantity of the addition. But copying, per se, however much skill or labour may be devoted to the process, cannot make an original work. A well executed tracing is the result of much labour and skill but remains what it is, a tracing. Moreover it must be borne in mind that the Copyright Act 1956 confers protection on an original work for a generous period. The prolongation of the period of statutory protection by periodic reproduction of the original work with minor alterations is an operation which requires to be scrutinised with some caution to ensure that that for which protection is claimed really is an original artistic work."

131. The 1995 changes were relatively minor. They amounted to the redesign of two subsidiary components of the system and a small alteration to the plastic panel itself. They are, to my mind, closely comparable to the changes to the "Lego" brick that were held not to be sufficient in Interlego to give rise to a new copyright work. I therefore accept Mr Thorley's submission that the 1995 redesign did not give rise to a new design right in the assembly as a whole.
132. The panels themselves. The Particulars of Claim set out in paragraph 4 a number of features of the panels. These include:

i) The overall shape and configuration of the panel;

ii) The width, height and relative configuration of the panel;

iii) The size, shape and positioning of the individual ducts;

iv) The shape and configuration of the pre-cut drip overhang;

v) The upstands at the top of each panel;

vi) The light diffusing ribs on the upper face of the lower wall of each panel.

133. These features are relied upon both individually and in combination. Mr Thorley's strategy was to concentrate on individual features of the panel, picking them off one by one as being commonplace. In my judgment the evidence does establish that, taken in isolation, some of the features of the panel were (and are) commonplace. In particular:

i) Triple-wall polycarbonate and uPVC sheeting was well-known in the early 1990s and can fairly be said to have been commonplace. But it was not commonly used in conservatory roofs (as opposed to industrial roofs), and consequently it was commonplace in a different design field. Moreover, triple-walled polycarbonate and uPVC sheeting was not usually more than 16 mm deep, and was sold in large sheets;

ii) Plastic panels of a depth between 60 mm and 70 mm were (and are) commonplace in the design of conservatory roofs;

iii) The stock lengths were (and are) commonplace;

iv) Rectangular ducts in plastic panels were (and are) commonplace in the design of conservatory roofs;

v) The dimensions of the pre-cut drip overhang were (and are) commonplace in the design of conservatory roofs;

vi) Light diffusing ribs or "pips" were (and are) commonplace in the design of conservatory roofs.

134. The width of the panels (500 mm) was not commonplace in conservatory roofs. Indeed it was unique. Nor was it commonplace for the ducts to be differently sized (as was the case with the central duct in Ultralite 500). In addition, the flat (as opposed to arcuate) top was not commonplace in conservatory roofs.
135. However, since the panels are extruded as a whole, and the features are also relied on in combination, it seems to me to be sensible to consider them in combination. As I have said, the fact that a design is made up of features which, individually, are commonplace does not necessarily make the design as a whole commonplace, provided that the combination is not commonplace. Mr Savage said:

"I do not think that the shape of the combination of the panel and the pre-cut drip cutaway was commonplace. Because the shape of the Ultralite 500 panel was very different from those panels that had gone before it (they had curved upper surfaces for example), the shape that remained after the 'pre-cut' section was cut away was very different from the shape of, for example, the Ultralite 250 with the pre-cut drip overhang. The section of the panel that remains after the pre-cut section has been removed is highly visible to the consumer and is a distinctive aspect of the design of the panel and of the assembly."

136. I accept this evidence. I have already found that the design of the Pinnacle 500 panel was copied from the Ultralite 500 panel, with some changes. Some of the details of the Pinnacle 500 panel differ from those of the Ultralite 500 panel. For example: the Pinnacle 500 panel is 65mm deep, as opposed to Ultralite 500's 60 mm. The ducts in Pinnacle 500 are equispaced, without the narrow central duct. There are changes in the upstands. There are also changes in the coupling members, but in my judgment these are excluded from design right because of the "must fit" exclusion. So are the "U-shaped" channels, to which changes have also been made. The Pinnacle 500 panel also has a "U-shaped" indentation at the level of the central horizontal internal wall. Eurocell has not, therefore, produced panels "exactly" to the design of the Ultralite 500 panel. So I pass to the question whether Eurocell has made articles "substantially" to that design.
137. This is a question of fact and degree. Although the evidence of experts may be helpful, in the end it is a matter for me. Models of each system were assembled and brought to court for the five days of the trial. I have retained samples of the panels in my room while writing this judgment. I still have to make a conscious and determined effort to tell them apart. I have no doubt that the similarities between the Ultralite 500 panel and the Pinnacle 500 panel are far greater than the similarities between those two panels, on the one hand, and all the other samples I was shown. Left only to my own visual impression, I conclude that the Pinnacle 500 panel is made substantially to the design of the Ultralite 500 panel. I am reinforced in this conclusion by the evidence of Mr Savage, which I agree with. He said:

"Most of the other panels available have "corrugated" top surfaces which give the roofs made from those panels a "lumpy" profile when viewed head on or from above. This obviously creates a very different impression to the flat surfaces of the Ultralite 500 and Pinnacle 500 panels. This feature is even more pronounced when the panels are all assembled together, because apart from the very small lip which covers the join between panels, the whole bottom inside surface of the roof is flat, which was very unusual and gives a very pleasing aesthetic effect."

"I do not believe that the very minor differences in dimensions are significant. The proportions appear identical to the eye. The proportional ratios of width to depth to length in relation to the Ultraframe 500 panel (for each stock length) are almost identical to the Pinnacle 500 ratios (for each stock length)."

"Other than the central duct, which is slightly narrower in the Ultralite 500 panel, the number of ducts is the same, and their size, shape and positioning is overall very similar. Again, without looking closely it is very difficult to tell the two apart."

"From a design point of view, I believe that the shape and configuration of the combination of the pre-cut drip overhang and the panel in the Pinnacle 500 system is the same or substantially the same as the combination of the pre-cut drip overhang and the panel in the Ultralite 500 system."

138. I also bear in mind that neither Mr Wharf nor Mr Allen could easily spot the difference. I conclude that Eurocell have infringed Ultraframe's design right in the Ultralite 500 panel.
139. Top caps and end caps. I have already held that the Eurocell design team used the Pinnacle top caps as their starting point for the Pinnacle 500 top caps. It follows that there has been no reproduction of Ultraframe's design in the sense of copying. However, I should deal with the other points raised under this head. The use of top caps as a means of protecting gaps between panels from the elements was (and is) commonplace. The particular design of the top caps was a simple flat top, with splayed legs. Mr Calvert accepted in cross-examination that there were other similar designs on the market, although the precise dimensions were different. Mr Savage said with, I think, a certain degree of understatement that the Ultralite top cap and the Pinnacle 500 top cap did not "differ hugely from other products on the market in the same way as the other articles in question." In my judgment, this too was (and is) commonplace. The co-extruded gasket, which acts as the weather seal between the top cap and the panel is, in my judgment, excluded from design right by the "must fit" exclusion. I agree with Mr Thorley that if design right exists in the top caps, it can only exist in the precise dimensions of the top cap. The precise dimensions of the Pinnacle 500 top cap are different from those of the Ultralite 500 top cap. The splayed legs are of a different length, protrude from the top at a different angle, and the co-extruded gasket is a continuation of the line of the splay, rather than projecting from the splay at a different angle. I do not consider that, when considered in isolation, the Pinnacle 500 top cap is an article made substantially to the design of the Ultralite 500 top cap. I conclude therefore that Eurocell has not infringed Ultraframe's design right in the top cap, considered in isolation.
140. The same reasoning applies to the starter bar top cap. The shape of the starter bar top cap is relatively simple. The precise dimensions of the Pinnacle 500 starter bar top cap differ from those of the Ultralite starter bar top cap. The flat top of the former is longer than the flat top of the latter; the splay is at a different angle and the co-extruded gasket is gasket is a continuation of the line of the splay, rather than projecting from the splay at a different angle. I do not consider that, when considered in isolation, the Pinnacle 500 starter bar top cap is an article made substantially to the design of the Ultralite 500 starter bar top cap. I conclude therefore that Eurocell has not infringed Ultraframe's design right in the starter bar top cap, considered in isolation.
141. The end cap designed in 1993 was a trapezium. The redesign in 1995 added a rectangle to the bottom edge of the trapezium. The overall 1995 design was not, in my judgment, commonplace. Mr Thorley's main attack on the end cap was that it was, in essence, a copy of the 1993 design, and that the addition of the rectangle required insufficient skill and labour to attract fresh design right. Mr Calvert agreed in cross-examination that "not very much" work was involved in designing the rectangular extension to the original trapezium. I took the question and answer to be directed to the amount of labour that went into the redesign. But I do not consider that originality depends on the amount of labour that went into producing the work (or design) in question. The amount of labour needed to produce a work (or design) may be inversely proportional to the degree of skill possessed by the designer. Moreover, the addition of the rectangular extension to the trapezoid shape of the 1993 end cap produces a markedly different visual impression. In my judgment there was sufficient skill and labour to render the 1995 redesign of the end cap original in the copyright sense. Apart from the requirement that the end cap must not be smaller than the section of the top cap that it is designed to cap, there is no design constraint on its shape. Mr Savage's evidence was that there was "a multitude of different shapes and sizes of end cap that were used generally in the field." I think that, in essence, Mr Rick agreed with this in cross-examination. Indeed, it is not necessary to have end caps at all, because top caps can be sealed at the end by a fascia panel which extends the whole width of the roof. In my judgment the 1995 redesign was not commonplace.
142. Mr Rick said in his witness statement that the end caps "really just follow the outer profiles of the rafter." I do not consider that this is so. The rectangular extension on both the Ultralite 500 end cap and the Pinnacle 500 end cap hang down below the lower portion of the top cap, and protrude below the lower edge of the roof panel. However, as I have said the detailed dimensions of the Pinnacle 500 end cap are not the same as those of the Ultralite 500 end cap. The Pinnacle 500 end cap is broader and shorter than the Ultralite 500 end cap; the angle of the trapezoidal part of the cap is shallower and the top corners are rounded rather than angular. When one is overlaid on the other, they produce a different visual impression. Both designs are simple geometric designs. For one to be substantially the same as the other, the differences must, in my judgment, be very small. In my judgment, although by a narrow margin, the Pinnacle 500 end cap is not produced "substantially" to the design of the Ultralite 500 end cap. I reach the conclusion that Eurocell has not infringed Ultraframe's design right in the end cap, if considered in isolation.
143. The starter bar end caps are, if anything, less similar to each other than the intermediate end caps. In my judgment, the Pinnacle 500 starter bar end cap is not produced "substantially" to the design of the Ultralite 500 starter bar end cap. Again, I reach the conclusion that Eurocell has not infringed Ultraframe's design right in the starter bar end cap, if considered in isolation.
144. The 1995 redesign of the panel. The two changes made to the panel in 1995 were a reduction in the thickness of the panel and a change in the shape of the light-diffusing ribs. Mr Lancaster explained that he showed the new extruding company a Rodeca panel and told them that he wanted them to produce an identical panel but that he wanted the ribs changed "so that they would be like the "pips" on Ultralite 250" and he wanted "some of the "thickness" taken out of the top and bottom skins of the panel" The evidence relating to the ribs suggests either that the ribs were copied from Ultralite 250, in which case the redesign was not original; or, alternatively, that feature was commonplace. The mere instruction to the extruding company to reduce the panel thickness does not, in my judgment, amount to an original design either. I hold that no new design right in features of the panel came into existence in 1995.
145. Are my conclusions on infringement of design right in the designs of the top caps and end caps (no infringement) inconsistent with my conclusion on infringement in the assembly as a whole (infringement)? I do not think that they are. It depends on the level of generality in which the design is described. What may amount to reproducing an assembly in substantially the same form as that of a previous assembly may not be the reproduction of an individual part. You can create a copy of an Old Master even though every individual brushstroke is different. Viewed at the level of the picture as a whole, it is a copy; but viewed at the level of an individual brushstroke, no two are the same. In other words this is a legal form of "gestalt" in which the whole is more than the sum of its individual parts. In my judgment that is the position here.
146. The "U-shaped" channels. For the reasons I have given these are not protected by design right at all, because they are within the "must fit" exclusion.

Conclusions

147. I summarise my conclusions as follows:

i) The Ultralite 500 patent is valid;

ii) Pinnacle 500 does not infringe it;

iii) Pinnacle 500 infringes Ultraframe's design right in:

a) The assembly as a whole;

b) The panels, taken as a whole.

148. Pinnacle 500 does not infringe Ultraframe's design right in the top caps or end caps, viewed individually.
149. The "U-shaped" channels are not protected by design right at all.
150. I will discuss with counsel what order I should make to give effect to this judgment.