FEDERAL COURT OF AUSTRALIA ITW AFC Pty Ltd v Loi and Tran Pty Ltd [2008] FCA 552 PATENTS – validity – construction of claims – novelty – combination patent – inventive step – hindsight bias – commercial success – secondary considerations as proof of non-obviousnessness – infringement – functional advantage – purposive construction – Improver questions WORDS AND PHRASES – "drilling tip", "taper" AFCO Automotive Foam Components Pty Ltd v Woodbridge Foam Corporation (2003) 58 IPR 56 Aktiebolaget Hassle v Alphapharm Pty Ltd (2002) 212 CLR 411 Bristol-Myers Squibb Co v FH Faulding & Co Ltd (2000) 97 FCR 524 Catnic Components Ltd v Hill & Smith Ltd [1982] RPC 183 Firebelt Pty Ltd v Brambles Australia Ltd (2002) 76 ALJR 816 Gillette Safety Razor Co v Anglo-American Trading Co Ltd (1913) 30 RPC 465 Graham v John Deere Co 383 US 1 (1966) Haberman v Jackel International Ltd [1999] FSR 683 Improver Corporation v Remington Consumer Products Ltd [1990] FSR 181 Kirin-Amgen Inc v Hoechst Marion Roussel Ltd [2005] 1 All ER 667 Lockwood Security Products Pty Ltd v Doric Products Pty Ltd (No 2) (2007) 81 ALJR 1070 Olin Corporation v Super Cartridge Co Pty Ltd (1977) 180 CLR 236 Pfizer Overseas Pharmaceuticals v Eli Lilly & Co (2005) 68 IPR 1 Populin v HB Nominees Pty Ltd (1982) 59 FLR 37 Root Quality Pty Ltd v Root Control Technologies Pty Ltd (2000) 49 IPR 225 Sachtler GMBH & Co KG v RE Miller Pty Ltd (2005) 65 IPR 605 Sakraida v Ag Pro Inc 425 US 273, 282-283 (1976) Wellcome Foundation Ltd v VR Laboratories (Aust) Pty Ltd (1981) 148 CLR 262 Moy's Walker on Patents (2nd ed, 2007) ITW AFC PTY LTD and ITW AUSTRALIA PTY LTD vLOI AND TRAN PTY LTD, SPECIALISED SERVICES (AUST) PTY LTD, TO HA LOI and DAVID JOHN BATE VID 453 of 2003 FINKELSTEIN J 23 APRIL 2008 MELBOURNE IN THE FEDERAL COURT OF AUSTRALIA VICTORIA DISTRICT REGISTRY VID 453 of 2003 BETWEEN: ITW AFC PTY LTD and ITW AUSTRALIA PTY LTD Applicants AND: LOI AND TRAN PTY LTD SPECIALISED SERVICES (AUST) PTY LTD TO HA LOI and DAVID JOHN BATE Respondents AND BETWEEN: LOI AND TRAN PTY LTD SPECIALISED SERVICES (AUST) PTY LTD TO HA LOI and DAVID JOHN BATE Cross-Claimants AND: ITW AFC PTY LTD Cross-Respondent JUDGE: FINKELSTEIN J DATE OF ORDER: 23 APRIL 2008 WHERE MADE: MELBOURNE THE COURT ORDERS THAT: 1. The application and the cross-claim be dismissed. 2. On or before 4.15pm on Wednesday 30 April 2008, the parties may file written submissions of no more than 3 pages on the question of costs. Note: Settlement and entry of orders is dealt with in Order 36 of the Federal Court Rules. IN THE FEDERAL COURT OF AUSTRALIA VICTORIA DISTRICT REGISTRY VID 453 of 2003 BETWEEN: ITW AFC PTY LTD and ITW AUSTRALIA PTY LTD Applicants AND: LOI AND TRAN PTY LTD SPECIALISED SERVICES (AUST) PTY LTD TO HA LOI and DAVID JOHN BATE Respondents AND BETWEEN: LOI AND TRAN PTY LTD SPECIALISED SERVICES (AUST) PTY LTD TO HA LOI and DAVID JOHN BATE Cross-Claimants AND: ITW AFC PTY LTD Cross-Respondent JUDGE: FINKELSTEIN J DATE: 23 APRIL 2008 PLACE: MELBOURNE REASONS FOR JUDGMENT 1 This is a patent infringement action. The first applicant, ITW AFC Pty Ltd, is the proprietor of Australian Patent No 719725 with a priority date of 20 November 1997. The patent is for a type of fastening screw, known in the building industry as a Zips screw. The screw is manufactured by the first applicant's licensee, the second applicant, ITW Australia Pty Ltd. It is convenient to refer to the two applicants jointly as ITW. ITW alleges that the four respondents (directly or indirectly) infringed its patent by importing and selling within Australia screw fasteners known as Clearfix and Bat'N'Fix screws. With respect to the issues in this proceeding, there is no relevant difference between the Clearfix and Bat'N'Fix screws. The respondents deny that their screws infringe the patent and also cross-claim for the revocation of several of the patent's claims, alleging they are invalid. The grounds of invalidity are the common ones; viz, no inventive step, not a manner of manufacture, not novel. 2 The trial was listed to begin on 13 November 2006. It was put off because the respondents were not ready with their evidence. The hearing then commenced on 18 June 2007 and the trial occupied seven days. Without orders to that effect, many of the witnesses filed multiple affidavits. Dr Collinson, the applicants' chief expert, filed ten affidavits, and their director filed five. The respondents' witnesses also filed multiple affidavits. Mr Loi, a director of the first and second respondents, filed six, and Mr Hunter, the respondents' expert, filed seven. Most of this evidence turned out to be irrelevant. As events transpired, by the time of the trial there was little in dispute as regards the facts. There were issues of construction that had to be sorted out and only one or two significant factual disputes. If the parties and their solicitors had early on concentrated on those issues the case would have been over and done with in less than half the time it in fact took, and there would have been a massive saving in costs. 3 The specification of the patent sets out the background to the invention. Fastening screws take a different form if adapted for penetration into metal or timber. Fastening screws intended to drill through sheet metal have a drill tip similar to that of a conventional drill tip, where the end is not threaded. A fastening screw for timber is different. It has a tapered end portion and the thread extends substantially to the end or tip. The thread through to the end is important because it enables the thread to bite into the timber and pull the screw in, thereby requiring minimum endwise pressure. A fastening screw designed to drill through metal is difficult to use to fasten metal to a timber support. The thread does not "bite" into the timber. Substantial pressure is required to achieve self-tapping. 4 Before the priority date, two types of screws were in wide use. The first is known as a Type 17 screw, which was used for drilling and tapping into timber. These screws have shanks that are tapered and threaded. The second, known as an ASD or Teks screw, was used for drilling into metal. These have parallel (untapered) shanks and a wider drilling point with cutting edges designed to penetrate metal. 5 As at the priority date it was widely known that Type 17 screws had difficulty penetrating metal and often broke or caused burrs when used on metal. Similarly, it was also widely known that ASD screws, while able to penetrate metal easily due to the wider drilling point and cutting edges, failed to fasten optimally to wood due to the lack of taper and thread to draw them in. 6 For about ten years from the mid-1980s, ITW and its predecessors invested hundreds of hours in attempting to design a screw that would more effectively affix metal to wood. In 1997 the inventor, Mr Roberts, managed to solve the problem by combining the wider drilling point of the ASD screw with the tapered shaft and threaded shank of the Type 17 screw. The Zips screw, as it became known, was launched commercially in December 2002 to great success. 7 The patent for the Zips screw contains several claims. The first claim is for: A fastening screw for fastening metal sheet to timber, having an elongate shank, and a drilling tip and a head provided at respective opposite ends of said shank, said shank tapering to said drilling tip which has a pair of cutting edges disposed on opposite sides of the longitudinal axis of said shank for cutting through metal sheet and a flute extending longitudinally of said shank from each of said cutting edges, a thread being formed on said shank and extending from adjacent said drilling tip rearwardly to beyond the section of the shank in which said flutes are formed. Claims 2 to 6 are dependent. Claim 7 is a method of forming the fastening screw described in claim 1. Claims 8 to 13 are dependent on claim 7. The rolling die for forming the fastening screw is described in claim 14 and variations are the subject of claims 15 and 16. Claims 17, 18 and 19 are omnibus claims. 8 The interpretation of claim 1 was the subject of both debate and evidence, much of which was not admissible. There are three controversial aspects. These relate to the requirement that the shank of the fastening screw be "tapering to said drilling tip". The first controversial aspect is whether the "drilling tip" is located at the terminal end point (or chisel point) of the screw or whether the tip takes in the terminal end or chisel point of the screw, the cutting edges and either part or all of the flutes. The second aspect depends on the answer to the first. Here the question is: what does "tapering to the drilling tip" require? Does it require a taper all the way to the terminal end or chisel point of the screw, so that the screw can, with more or less accuracy, be described as conical in shape? Or does it merely require a reduction in diameter from the cylindrical or parallel portion of the screw to the point where the drilling tip can be said to begin? The third and most important controversy concerns the degree of taper that is necessary to satisfy the claim. It was common ground that any relevant taper is to be assessed by measuring the root surface of the shank at points at which the valleys lie between the crests of each thread. 9 I propose for the time being to put to one side any discussion about the degree of taper required. In the first instance I will deal with the other two constructions points. Read independently of the specification and the remaining claims, there is a degree of ambiguity in claim 1 as to what constitutes the "drilling tip". The requirement that the fastening screw have "a drilling tip and a head provided at respective opposite ends" of the shank does not clarify the problem, although the head is that portion of the screw used to propel the screw into a surface at the opposite end of the screw, rather than merely the surface of that portion. By analogy the drilling tip could also be a portion rather than the terminal point at the other end of the shank. The phrase immediately following the requirement that "shank [taper] to said drilling tip" is "which has a pair of cutting edges". The word "which" may relate to the shank or the drilling tip. If the word relates to the drilling tip itself, the drilling tip must consist at least of the cutting edges and part of the flutes. 10 The contention that the word "which" refers to the shank, rather than to the drilling tip is at least inconsistent with claim 7, which in large part picks up the wording of claim 1 (as its purpose is to teach a method of forming the screw described in claim 1), but specifically provides for "said drilling tip having a pair of cutting edges". This indicates that the word "which" in claim 1 relates to the drilling tip. 11 In the specification there are figures depicting one embodiment of the invention (figure 5) and the final stage of formation of one embodiment of the invention (figure 10), both of which label the chisel of the screw as the "drilling tip". On the other hand, there is much in the specification that indicates that the "drilling tip" includes the cutting edges and flutes. The drilling tip of the screw is described as being "designed to efficiently cut into metal and for collection and removal of swarf created during penetration of drilling tip". The flutes perform the function of removing the swarf, so this description makes it clear that the drilling tip encompasses not only the cutting edge but at least part of the flutes. Later there is a reference to the "drilling tip cutting edge" and a statement that the "drilling tip preferably includes two cutting edges and two flutes". In the description of the method of forming the screw, a "material displacement operation" is said to "create a drilling tip at an end portion" of the shank. This wording is used in claim 7 and indicates, again, that the drilling tip encompasses a portion of the shank. It is also worth noting that the specification uses the words "terminal end" when the reference is to the apex or chisel of the screw, suggesting that the "drilling tip" is something other than the apex or chisel. 12 When describing the prior art from which the invention derives, similar terminology is employed. Thus, in describing the conventional screws for drilling through sheet metal, the specification refers to the drilling tip as having a "shape". It is difficult to see how an apex or chisel could be said to have a "shape". In describing the conventional screw for penetration into timber, the specification refers to a "tapered end portion which terminates in a pointed terminal end" and a single flute "provided in the drilling tip". Figures 1 and 3 are described as being a "side view of the drilling tip" of the prior art fastening screws, and those figures clearly depict a portion of the prior art fastening screws incorporating both cutting edges and flutes as well as the chisel. 13 Finally, the specification states that the purpose of the invention is to create a screw that can both cut through sheet metal and penetrate into timber in a form "which is essentially a hybrid of the drilling tips of the two prior art fasteners". The invention is clearly not a hybrid of the chisels only of those prior art fastening screws. 14 When claim 1 is considered by reference to the specification as a whole it is clear, in my view, that the "drilling tip" incorporates the chisel, the cutting edges and at least part of the flutes. It is equally clear that the required taper is to be measured to the point where the drilling tip begins. 15 It is convenient next to deal with the attack on the patent. The invention, the Zips screw, is essentially a combination of features found in the generally known prior art. It is for this reason that the respondents argue that the patent is invalid for a lack of novelty and lack of inventive step. On the other hand ITW's case is that, while fasteners in the prior art may each have some of the features of the Zips screw, no one fastener has all of its features. Moreover, ITW says that the idea of combining the best features of the ASD and Type 17 screws into one screw was an inventive step because, although perhaps seemingly obvious in hindsight, no one had managed it (or apparently even thought to try it) before, despite the fact that the metal-to-wood problem had been an issue in the building industry for some years. 16 First of all, with respect to novelty, for a prior publication to destroy novelty: "[It] must give a direction or make a recommendation or suggestion which will result, if the skilled reader follows it, in the claimed invention. A direction, recommendation or suggestion may often, of course, be implicit in what is described and commonly the only question may be whether the publication describes with sufficient clarity the claimed invention or, in the case of a combination, each integer of it.": Bristol-Myers Squibb Co v FH Faulding & Co Ltd (2000) 97 FCR 524, 548; see also Pfizer Overseas Pharmaceuticals v Eli Lilly & Co (2005) 68 IPR 1, 67. 17 The respondents did not identify any prior art which referred to a screw that contained all the features of the Zips screw. The best example of a piece of prior art the respondents could come up with was UK Patent GB957626 (the 626 patent). The invention the subject of that patent was a combination hole-drilling and thread forming screw fastener. I do not propose to analyse the invention in any great detail. It is sufficient to say that the 626 patent does not describe with any clarity four integers of claim 1 of the Zips screw patent. First, the 626 patent neither recommends nor suggests a fastener suitable for affixing metal to timber. Instead, the 626 patent describes a fastener "advantageous in the securing together of light gauge metal work pieces". While there was evidence to the effect that a fastener described by the 626 patent was capable of affixing metal to wood, there was other evidence which showed that 626 fasteners are, as the 626 patent language would tend to suggest, neither designed nor suited for such purpose. Second, the 626 patent describes a fastener with either four or six cutting edges, while the Zips screw has only two. Third, the 626 patent specifies fasteners possessing flutes collinear to the cutting edges, in contrast to Zips screws, which have flutes extending from the cutting edges in a longitudinal direction. Fourth and finally, the 626 patent does not describe a fastener having a thread extending from adjacent to the drilling tip rearwardly to beyond the section of the shank in which the flutes are formed, such as described in claim 1 of the Zips screw patent. 18 As regards lack of inventive step, reference should be made to what the High Court said with respect to when the combination of features of known prior art can be said to constitute an inventive step: "The test is whether the hypothetical addressee faced with the same problem would have taken as a matter of routine whatever steps might have led from the prior art to the invention": Wellcome Foundation Ltd v VR Laboratories (Aust) Pty Ltd (1981) 148 CLR 262, 286. Steps taken as a matter or routine will be deemed an obvious or workshop modification of prior art and do not constitute an inventive step: Gillette Safety Razor Co v Anglo-American Trading Co Ltd (1913) 30 RPC 465, 480; Woodbridge Foam Corporation v AFCO Automotive Foam Components Pty Ltd (2002) 58 IPR 56, 58-59. 19 In Aktiebolaget Hassle v Alphapharm Pty Ltd (2002) 212 CLR 411, 433 the High Court restated the inventiveness question this way: Would the notional research group at the relevant date, in all the circumstances, which include a knowledge of the relevant prior art … directly be led as a matter of course to try [the step in question] … in the expectation that it might well produce a useful alternative to or better [product] … or a body useful for any other purpose? See also Lockwood Security Products Pty Ltd v Doric Products Pty Ltd (No 2) (2007) 81 ALJR 1070, 1095. 20 A point which is obvious but bears repeating is that the question of inventiveness must be approached from the standpoint of an inventor or screw expert working in the 1980s and early 1990s, not with the benefit of hindsight: Alphapharm 212 CLR at 423; see also Root Quality Pty Ltd v Root Control Technologies Pty Ltd (2000) 49 IPR 225, 247. 21 The evidence on inventive step is as follows. The starting point is the introduction of Type 17 screws in the early 1980s. In fixing metal to wood those screws suffered from the problems already mentioned. Accordingly, ITW and its predecessors began considering the development of a replacement product in the mid-1980s. A research project was started in 1992 to develop an improved Type 17 screw, on which one ITW engineer alone spent several hundred hours studying the fastening mechanism of Type 17 screws with high speed photography and experimenting with new designs. Over the years, various approaches to the problem were considered, including experiments with: (1) modifying Type 17 screws to place the start of the thread closer to the point to encourage cutting to occur earlier but then interrupting the thread in places to prevent burring; (2) using different flute or slot configurations on a Type 17 screw; (3) trying variations of the "Type AW" point, which is an ASD drill screw point but with threads rolled on the cutting edges; and (4) testing a smaller drill point like a Number 1 point on Type 17 screws. These efforts were unsuccessful. As Mr Roberts put it, ITW had "just totally c[o]me to a brick wall". 22 Despite having spent "years and years and years trying to work out a way to do this," Mr Roberts reached the conclusion "that the efforts that we'd been trying to solve this problem were … a waste of time and the only way it was going to be [solved was] to get a metal cutting point onto a Type 17 style of screw." In September 1997 Mr Roberts conceived the idea of combining the ASD drill screw point with the with the threaded Type 17 shaft. 23 Initially, Mr Roberts had "great reservations" whether the idea would work. In particular he "didn't think it was possible" to overcome the technical difficulties of designing a die for the manufacture of the hybrid screw. This design process included modifying a pair of Type 17 forging dies by reducing the length of the point cavity, creating pointed blanks, and rolling the thread onto the pointed blanks. Four variations of die were produced, each with a different thread location, but two were discarded as having the thread either too close or too far away from the point. The remaining two variants were tested with a Type 17 die used as a control. Manual tests of screw samples created from the dies showed improved drill performance. 24 Even though performance was improved, Mr Roberts felt that the threads at the point of the new screws were not well enough defined due to the configuration of the Type 17 die. It was necessary, therefore, to create a new roll point roll die specifically to address this situation. According to Mr Roberts: "The real difficulty was getting a die that would produce a screw that combined the qualities of a drill screw [ASD] point and a Type 17 screw." The development of such a die was "technically very difficult and took some considerable time." It was necessary for the die to gather material from higher up on the shank and reposition it closer to the point, by cold forming or cold rolling, in order to provide a sufficient volume of metal to allow a thread to be formed at the point. Mr Roberts estimated that ITW spent approximately one year and 250 man hours in developing a die that worked. The design was eventually completed in October 1998. 25 It is worth emphasizing that Mr Roberts' description of the design process was not challenged by the respondents. Their case is that his evidence is irrelevant because the ITW designers were working on the wrong problem. They seize on the following proposition with which Mr Roberts agreed during his cross-examination: [T]he difficulties [in creating a metal-to-wood screw] weren't in the realisation that what you needed to achieve was a hybrid screw, it was rather whether you could make one. This is the basis of the submission that the evidence of research undertaken prior to the development of the Zips screw patent is irrelevant. It is said to be irrelevant because the researchers were focused on the "wrong problem" and "once the correct problem was addressed the solution (the hybrid screw) was apparent and obvious." 26 It is clear the development of the screw the subject in the patent in suit took considerable time, skill, and effort to design. The combination of the best features of the ASD and Type 17 screws was a step that ITW embarked upon only after years of failed attempts; it was not something the inventor was directly led to try, nor did he even have the expectation it would succeed. It was, in short, an innovative breakthrough. 27 The respondents' argument in substantial part suffers from the hindsight bias warned against by the High Court. The fact that previous researchers were focusing on the wrong problem is of course only apparent in retrospect. Evidence of contemporaneous views, on the other hand, indicates that researchers focused on other approaches to the problem precisely because the so-called "obvious" solution of the hybrid screw was considered at the time to be not only not obvious, but perhaps even impossible. In Lockwood 81 ALJR at 1095, the High Court said that the question whether the combination of known prior art "is inventive will turn on what a person skilled in the relevant art, possessed with that person's knowledge, would have regarded, at the time, as technically possible in terms of mechanics, and also as practical." (Emphasis added.) 28 Finally on the question of obviousness there is the commercial success of the Zips screw. According to the cases the commercial success of an invention may contribute to the conclusion that it involved an inventive step. Initially, however, not much could be made of secondary indicia. In Firebelt Pty Ltd v Brambles Australia Ltd (2002) 76 ALJR 816 the High Court, by reference to the judgment of Laddie J in Haberman v Jackel International Ltd [1999] FSR 683, 699 accepted that secondary indicia such as commercial success should be approached cautiously. As Laddie J said: "[I]t is sometimes possible to make large profits by selling an obvious product well." But the law has moved on. In Lockwood 81 ALJR at 1096, the High Court referred with approval to the post Graham v John Deere Co 383 US 1 (1966) line of cases in the United States for the proposition that "the enquiry into secondary considerations of non-obviousness has been treated as being an important enquiry which must be taken into account because prior art cannot be evaluated in isolation." (Emphasis added.) 29 With all due respect, in my view the US precedents supporting that proposition go too far. In Graham, the Supreme Court certainly approved the use of secondary criteria as part of the enquiry into non-obviousness. The Supreme Court said (at 17-18): "[S]econdary considerations … might be utilised to give light to the circumstances surrounding the origin of the [invention] … As indicia of obviousness or non-obviousness, these enquiries may have relevancy." (Emphasis added.) But, as was made plain in a later decision of the Supreme Court, commercial success cannot establish patentability by itself: Sakraida v Ag Pro Inc 425 US 273, 282-283 (1976). 30 The High Court has side-stepped the many critics of the post-Graham approach (see eg Moy's Walker on Patents (2nd ed, 2007) § 9.62) and "upped the ante" by conferring on secondary criteria generally, and commercial success in particular, almost the status of proof of non-obviousness. Surely this is going too far. Unless there is actual evidence of invention, patentability cannot be made out. In this regard, secondary considerations such as commercial success are, at best, but factors which may be taken into account, and even then their relevance will vary. In no event should they be determinative factors, as some US cases suggest. 31 Giving criteria such as commercial success such a large role to play will fundamentally alter patent litigation. To prove commercial success the patentee will call evidence of market share, growth of market share, or displacement of existing prior art devices. The patentee will also be required to show that this success was due to the merits of his invention and not to other factors, such as advertising or significant sales efforts. Put another way, the probative value of secondary considerations, such as commercial success, must depend on the nexus between the criteria and the invention. The problem is that a good deal of trial time will now be taken up with attempting to demonstrate that nexus. Not only will we have evidence from the usual passing parade of scientists, chemists and engineers, we will now have jobbers, advertising executives and behavioural scientists. Those that bear the cost of litigation may not think this to be a useful development. 32 It is now convenient to turn to the question of infringement. 33 Development of the respondents' screws began in about July 2000 in Taiwan, with the first order for production quantities of Clearfix fasteners being delivered in February 2002, well before the commercial launch of the Zips screw in December of the same year. For this reason, no doubt, ITW did not contend that the developers of the respondents' fasteners had copied the patent in suit in creating their screws. In any event, for the respondents' screws to infringe the Zips screw patent, ITW must show no more and no less than that the respondents' screws possess each and every one of the essential features of the relevant claims of the Zips screw patent. Populin v HB Nominees Pty Ltd (1982) 59 FLR 37, 42; Olin Corporation v Super Cartridge Co Pty Ltd (1977) 180 CLR 236, 246. 34 ITW alleges that the respondents' fasteners infringe several claims of the Zips screw patent, but the heart of the case boils down to a dispute over whether the respondents' screws in fact possess one of the essential features of the claims of the patent in suit. Claim 1 describes a screw "for fastening metal sheet to timber" with the relevant essential features of (1) a shank tapering to the drilling tip (ie, narrowing in diameter as one moves toward the tip); and (2) a thread which is formed on the shank and extends from adjacent the drilling tip rearwardly to beyond the section of the shank in which the flutes are formed. ITW contends, and the respondents deny, that the respondents' screws possess both of these features. 35 It is convenient to focus first on the term 'taper' found in the first disputed integer. Initially, the existence of a taper in the respondents' screws was hotly contested. It is common ground that such a taper, if it exists at all, is not visible to the naked eye. A substantial part of the evidence was directed to the existence and extent of the taper. The dispute between the experts was resolved when the respondents' screws were submitted for measuring to Thales Australia's metrology laboratory, which maintains some of the most sophisticated measuring equipment to be found in Australia. Even the use of this equipment produced differences in the results of the tests. One thing, however, turned out to be absolutely clear. That is that the "taper" of the respondents' screws, that is the reduction of the thread root diameter to the point at which, on my contention, the drill tip begins, is in the order of 0.04 mm to 0.07 mm although at some point the reduction is about 0.12 mm. That is to say, the reduction in the thread root diameter is in the hundredths of a millimetre. The question in issue is whether a reduction in the diameter of the respondents' screws of that magnitude is, relevantly, a taper. 36 The respondents submit that the reduction in diameter (a taper of sorts) is not legally cognisable both because it is so miniscule and because it conveys no functional advantage. For example, they say that: [T]here cannot be any relevant taper if the presence of a requisite taper (or indeed any narrowing of diameter at all) cannot be visually observed in the accused products … as physical exhibits [or] as photographs including at considerable enlargements …. [T]he Respondents' fasteners are not surgical instruments or aircraft parts…. … [I]t defies common sense to judge the existence, or not, of an assessed taper in fastener products produced and used by the millions in building sites around Australia at a measurement level of tenths or hundredths of a millimetre. It is to be hoped that common sense does not stop at the Court door. … The Respondents' fasteners have a different configuration and work entirely differently to the fasteners described [in the ITW patent]. 37 According to the Oxford English Dictionary the literal definition of the verb 'taper' is, in its intransitive form, "to narrow or diminish gradually in breadth or thickness towards one end; to grow smaller by degrees in one direction". When used as an adjective, "taper" is defined as "diminishing gradually in breadth or thickness towards one extremity …; becoming continuously narrower or more slender in one direction; tapering". When used as a noun, "taper" is defined as "a spire or slender pyramid; a figure which tapers up to a point" or "anything that gradually diminishes in size towards one extremity, as a tapered tube". If the literal meaning is applied to the claims this suggests that all the invention requires is any reduction however slight in diameter along the shank of the screw to the drilling tip. 38 In my view, however, the true question is not whether there is a taper according to the dictionary meaning of the word. The taper in the screw has a purpose. The evidence shows that once the Zips screw penetrates metal and drills into wood, the threads engage to pull the screw into the wood as the pointed drill tip and taper work to force the timber apart. When the screw is removed, the displaced wood tends to reoccupy its original space. In other words, the roughly 0.3 millimetre taper of the Zips screw (which is visible to the naked eye) is an essential feature because it aids the screw in being pulled into the wood. 39 In contrast, the respondents' evidence (which I accept) shows that their screws work by a different mechanism: they simply drill a hole into the wood of the same diameter as the shoulder of the drill tip end of the screw. That is to say, the screws work by boring out the timber such that, if they are removed, there is no displaced timber to fill or reoccupy the space; they do not rely on a taper to draw the screw in but instead bull their way into the wood. 40 This finding is not consistent with the evidence of ITW's expert, Dr Collinson. He said that the taper in the respondents' screw does play a similar role to that of the taper in the Zips screw in the fastening process. But for reasons I will now explain I did not find his evidence to be persuasive on this important point. 41 First, his evidence that the taper in the respondents' screw had the same functional significance as that found in the Zips screw was not unequivocal. Dr Collinson began by conceding that where the taper of a screw is less than 0.3 mm, the fasteners tend to perform like a metal drilling point and bore a hole in the timber, unlike a tapered Zips screw, and that performance or function in the Zips sense "trails off severely with less" taper. 42 Although Dr Collinson attempted to maintain the position that the taper in the respondents' fasteners was not so minimal as to go past the "breakpoint" at which the taper has no functional significance, I am not satisfied that this view held up under cross-examination. First, Dr Collinson was asked to accept that, assuming that the respondents' fasteners were advanced into wood at an infinitesimally slow rate, they would bore a hole into the wood of the same diameter as the shoulder of the drill tip end of the screw. After being pressed, he did accept this. However, obviously a screw is not affixed at an infinitesimally slow rate in real life, so the question was whether Dr Collinson would accept that the same process occurred when the rate of advancement was increased. There then followed a crucial exchange during the cross-examination conducted by Mr Hess SC: [Q]: And indeed, if [a fastener such as the respondents'] advances at a greater than infinitesimal rate, the same thing will occur? [A]: [I] have taken fasteners into hard timber and for sake of comparison tried to compare their performance by putting them in infinitesimally slowly. As they go in – we've been talking a lot about the root diameter of the screw. When it's spun at a point, because of the threads there are[,] well, they're also touching the timber. The further the screw goes in, the more timber is pushed aside, moved around. With the threads, it can't escape the hole and so it just builds up and builds up, and it eventually grabs and drives itself home into the hole; the threads engage. [Q]: Yes. But that is the threads doing that work; do you accept that Dr Collinson? [A]: It's the function of the threads to pull it into the hole, yes. [Q]: It's not any part of the diametric shaft of the fastener after the shoulder has entered the piece that is doing the work that you have just described? [A]: I believe the taper is affecting the way the threads function. [Q]: Well, the next point I will put to you, Dr Collinson, is that for the taper on the respondents' fasteners, as you describe it, to have any effect analogous to the taper on the [ITW] fasteners …, then the taper must be at the leading end of the screw, otherwise the analogy doesn't work? [A]: I've tested the fasteners into timber, and I believe they work very well. And I've been through my explanation of why I believe that happens. [Q]: But leave aside whether they work, Dr Collinson. For the analogy,… the taper … has to be identified as a taper at the leading end of the screw, namely at the furthermost point forward of the screw? [A]: No, I don't believe so. [Q]: Otherwise there is just no analogous comparison? [A]: I believe that the taper performs the function. As this exchange shows, Dr Collinson's answers were unsatisfactory. In both this and other exchanges, each time Mr Hess led him to the point of a contradiction or inconsistency (in this case, that it was the threads rather than the taper doing the work), Dr Collinson would retreat into bald assertions such as "I believe the taper is affecting the way the threads function" or "I believe the taper performs that function.". 43 To some degree, these bare assertions were clarified on re-examination. What was not remedied, however, was a more general problem with Dr Collinson's evidence. Dr Collinson's reasoning was infected or underpinned by a faulty syllogism in that his assertions that the "the taper [in the respondents' fasteners] performs the [same] function [as a fastener described by the Zips patent]" were based on the fact that the respondents' fasteners "work very well." In other words, Dr Collinson appeared to be proceeding on the basis that: (1) ITW's fasteners perform well; (2) the respondents' fasteners also perform well; (3) therefore, the respondents' fasteners must perform in the same way as ITW's fasteners. 44 This type of reasoning was most clearly on display in the following exchange: [Q]: I accept [that the taper in the respondents' fasteners] is measurable, but I am putting to you that it is insignificant? [The Court]: In a functional sense? [Q]: In any functional sense at all? [A]: No, I would disagree. I think the performance of the [respondents'] product and the commercial success shows its significance. Again, an attempt was made to improve this evidence during re-examination, but resulted only in the clarification that in "performance" Dr Collinson was speaking of the "physical performance of the product." 45 I am of the view that Dr Collinson's opinion was in no small part formed by reasoning backwards to form a just-so story—"well, their screw works, so it must be due to the taper." However, as Mr Hess suggested in his questioning, that the respondents' fasteners perform well does not of itself prove they have a functionally significant taper. To put it bluntly, Dr Collinson's reasoning is unacceptable as the basis of expert opinion in the absence of a foundation that no fastener could adequately affix metal to timber without a taper (for only in such a case could an inference be drawn, without more, from the performance of a screw to the existence of a taper). No such foundation was laid here. It is thus unsurprising that when his opinion was subjected to scrutiny, Dr Collinson retreated into assertion and equivocation. 46 A second and related evidentiary point is the lack of testing and measurement to support Dr Collinson's opinion regarding the functional significance of a taper in the respondents' fasteners. Assuming for the sake of argument, and contrary to what I have in fact found, that his evidence on this point were otherwise satisfactory, I would nevertheless give it little weight. 47 Mr Caine SC accepted that Dr Collinson did not take any measurements. However, it was submitted that he did perform tests on the respondents' fasteners. The primary basis for this submission, apart from the passage earlier quoted, is the evidence of Dr Collinson given during cross-examination as follows: "What I have done is tested I guess the respondents' screws into timber and also other screws which have a large untouched drill point and tapered threads behind it and found that they did perform into timber very well." (Emphasis added.) As the emphasized portion suggests, this evidence was not particularly compelling even in its own right, having been delivered by Dr Collinson in a manner lacking in either conviction or confidence, and suggesting that the focus of testing was in fact screws other than those of the respondents. 48 This became clear when the evidence is considered in context. Immediately following the quoted passage, Mr Hess continued to press the point by essentially repeating the question whether Dr Collinson had done any measurements to support his observations. Dr Collinson responded: "What I have done is manufactured fasteners of that kind and noticed that they performed very well into timber." (Emphasis added.) Again the question was put as to the specific scientific basis for his observations, and Dr Collinson replied: [T]he best data I have on that [the functional significance of the taper], we have another competitor in an area of northern Queensland where the timber is particularly hard, like the end of the Red Gum hardness and above that. It's reaching the limit of where our current Zips fasteners will actually perform and so in a way to test the performance we've manufactured fasteners as a test to compete with theirs which have a wider area of untouched drill point and the taper behind them and they're performing extremely well. I have also tested other fasteners and they don't perform. (Emphasis added.) 49 These answers, and the emphasized portions in particular, make two things clear. First, the tests referred to by Dr Collinson were concerned with the performance of ITW's own screws, not the respondents' screws. Second, "the best data" backing Dr Collinson's observations as to the functional significance of the taper in the respondents' screws did not involve testing of the allegedly infringing fasteners; rather, it involved testing fasteners of Dr Collinson's own manufacture, which he considered to be similar to the respondents' fasteners. While these fasteners may well have been similar to the respondents', the failure to directly test the allegedly infringing screws means that an additional evidentiary foundation was required to be laid in order for this so-called "best data" to be relevant—namely, that the substitute fasteners were in fact similar to the respondents' fasteners in all relevant respects. No such evidence was forthcoming. 50 Moreover, given the hesitant, cursory, and qualified manner in which the reference to testing of the respondents' fasteners was made, I consider it likely that the only data backing Dr Collinson's observations involved tests of screws other than the allegedly infringing screws made for other purposes than the purpose of testing such screws' functions. Even accepting that he did carry out some tests on the respondents' fasteners, I am not satisfied in any event that Dr Collinson's opinion as to the functional significance of a taper in the respondents' fasteners is supported by adequate testing or measurement, and thus I give it little weight. 51 In short, I find that ITW failed to establish to the requisite standard that there is a taper in the respondents' screw—however defined, described, or measured—that performs any function at all, much less the same function as performed by the taper described in the patent in suit and embodied in the Zips screw, namely that of pulling the screw into the wood. 52 Having found as a matter of fact that whatever microscopic taper may exist in the respondents' screws conveys no functional advantage, it remains to consider the legal significance of that finding. In Improver Corporation v Remington Consumer Products Ltd [1990] FSR 181, 189 Lord Hoffman, elaborating on the purposive approach to patent construction outlined by Lord Diplock in Catnic Components Ltd v Hill & Smith Ltd [1982] RPC 183, 242 identified a set of three questions to serve as guidelines in determining whether a skilled addressee of a patent in suit would understand a subsequent product as infringing the claims of a patent. These questions are equally applicable under the Australian legislation: Sachtler GMBH & Co KG v RE Miller Pty Ltd (2005) 65 IPR 605, 613-614. 53 Most relevant for present purposes is the rule of thumb that a variant that has a material effect on the way an invention works will be deemed not to infringe: Improver Corporation [1990] FSR at 189. Because the microscopic taper of the respondents' screws serves no useful purpose and because the respondents' screws fasten metal to timber by a different mechanism, the respondents argue their screws are a materially different variant from the Zips screw described in claim 1 as possessing the essential feature of a tapered shank. 54 I should emphasize that the rule of thumb described in Improver and relied on by the respondents is just that—a guideline to aid in implementing the purposive approach to patent construction rather than a legal test in itself. This was a point made by Lord Hoffman himself in Kirin-Amgen Inc v Hoechst Marion Roussel Ltd [2005] 1 All ER 667 at 686 in the course of commenting on his reasons for decision in the Improver case: The [Improver] questions, which the Court of Appeal in Wheatly v Drillsafe Ltd [2001] RPC 133, 142 dubbed 'the Protocol questions' have been used by English courts for the past 15 years as a framework for deciding whether equivalents fall within the scope of the claims. On the whole, the judges appear to have been comfortable with the results, although some of the cases have exposed the limitations of the method. When speaking of the 'Catnic principle' it is important to distinguish between, on the one hand, the principle of purposive construction which I have said gives effect to the requirements of the Protocol, and on the other hand, the guidelines for applying that principle to equivalents, which are encapsulated in the Protocol questions. The former is the bedrock of patent construction, universally applicable. The latter are only guidelines, more useful in some cases than in others. I am bound to say that the cases show a tendency for counsel to treat the Protocol questions as legal rules rather than guides which will in appropriate cases help to decide what the skilled man would have understood the patentee to mean. 55 To put the point another way (a way which illustrates the connection between the Catnic and Improver cases alluded to by Lord Hoffman in Kirin-Amgen—namely, that the Protocol questions are merely proxies or a short-hand version of a full Catnic analysis that a court may employ to save time when considering alleged equivalents), I return to the basic question which requires resolution—how the term "taper" found in claim 1 of the patent in suit should be construed—and approach it from first principles. For example, should taper be construed literally to mean a narrowing in the diameter of the shaft? Or should it be construed purposively to mean any narrowing, no matter how great or small, provided the narrowing has a function or purpose similar to that of the taper described in the claim and found in the Zips screw, that is, to draw the fastener into the wood? 56 When framed in this way, the answer to the question becomes obvious, given that literalism as an approach to patent construction was long ago rejected in favour of the purposive approach: Kirin-Amgen [2005] 1 All ER at 679-684 (describing the history of the US and English rejection of a literalist approach to patent construction in favour of a purposive approach in England and the doctrine of equivalents in the US). As explained by Lord Hoffman in Kirin-Amgen, the goal of the purposive approach outlined by Lord Diplock in Catnic and now followed in this country is to give "fair protection to the patentee" (by avoiding the result of a literal approach under which a potential infringer might copy the mechanism of an invention and yet escape liability by making an "immaterial variation"—for example, by making a small change in the taper of a screw that does not affect the screw's function yet causes the variant to fall outside the literal meaning of the patent's terms) on the one hand while still recognising some limits on the scope of the monopoly on the other: Kirin-Amgen at 682-683 (noting, in the course of describing the history of the judiciary's struggle to police the boundaries of patent law, that "once the monopoly had been allowed to escape from the [literal] terms of the claims, it is not easy to know where its limits should be drawn"). 57 Returning to the instant case and applying the principles enunciated by Lords Diplock and Hoffman, it is clear that the term "tapered" in the patent must be understood by focusing on the purpose or function of the taper. In other words, if the purpose of the taper is to draw the screw into the wood, the term "tapered" in claim 1 must be construed to encompass any narrowing in the diameter of a fastener's shaft, regardless of how great or how small, that serves to draw the fastener in. Because the microscopic narrowing in the diameter of the shaft of respondents' fasteners does not serve that purpose, the respondents' screws cannot be said to be "tapered" in the manner described in claim 1. Not possessing this "essential feature" of the claim, it follows that the respondents' screws do not infringe. 58 For the foregoing reasons, I would dismiss both the application and the cross-claim. Given that each side has succeeded and failed in more or less equal measure, my present view, subject to what the parties may have to say, is that there should be no order as to costs. I certify that the preceding fifty-eight (58) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Finkelstein. Associate: Dated: 23 April 2008 Counsel for the Applicants and Cross-Respondent B Caine SC and B Fitzpatrick Solicitor for the Applicants and Cross-Respondent: Davies Collison Cave Counsel for the Respondents and Cross-Claimants: B J Hess SC and F Marks Solicitor for the Respondents and Cross-Claimants: Alfred Tatlock Solicitor Date of Hearing: 18, 19, 20, 21, 22, 25 and 26 June 2007 Date of Judgment: 23 April 2008