Coogi Australia Pty Ltd v Hysport International Pty Ltd & Ors (includes corrigendum dated 7 October 1998) [1998] FCA 1059 (21 August 1998)

Last Updated: 19 January 1999

FEDERAL COURT OF AUSTRALIA

COPYRIGHT - copyright subsists in an entire work, not in a part or element of a work - copyright in fabric as a "work of artistic craftsmanship" found to exist in the appearance created by the combination of the stitch structure and selection of yarn colours - allegation of infringement of copyright by the taking of a substantial part of the stitch structure - allegation not proved because of copyright owner's failure to identify the subject matter of its copyright and consequent failure to show that what was taken was a substantial part of that subject matter.

COPYRIGHT - "work of artistic craftsmanship" - what constitutes such a work - copyright held to exist in the first item in a line of mass-produced items - Copyright Act 1968 (Cth), s 10.

COPYRIGHT - computer program - whether program used in a computerised knitting machine to knit fabric to a particular pattern a "computer program" within the Act - infringement of copyright in a computer program by unauthorised adaptation - infringement not established unless infringer has access to and makes use of the copyright program in developing its own program - no infringement of copyright in computer program where alleged infringer reverse-engineers a sample of the copyright fabric and from the information as to the construction of the fabric so obtained writes its own program for use in its computerised knitting machines to make a similar fabric - Copyright Act 1968 (Cth), s 10. INFRINGEMENT OF COPYRIGHT IN COMPUTER PROGRAM - where objective similarity relevant to question of infringement - not shown by similarities identified by a comparison of the two programs at the functional level of expression of each.

Copyright Act 1968 (Cth) - ss 10, 14, 22, 29, 31, 32, 75, 77

Trade Practices Act 1974 (Cth) - s 52

Fair Trading Act 1985 (Vic) - s 11

Copyright Act 1911 (UK) - s 22

Designs Act 1906 (Cth)

George Hensher Ltd v Restawhile Upholstery (Lancs) Ltd [1976] AC 64, discussed

Cuisenaire v Reed [1963] VR 719, discussed

Merlet v Mothercare PLC (1984) 2 IPR 456, referred to

Bonz Group (Pty) Ltd v Cooke (1994) 3 NZLR 216, discussed

Attorney-General for NSW v Trustees of the National Art Gallery of New South Wales (1944) 62 WN(NSW) 212, referred to

Ladbroke (Football) Ltd v William Hill (Football) Ltd [1964] 1 WLR 273, followed

Powerflex Services Pty Ltd v Data Access Corporation (1997) 37 IPR 436, followed

Leslie v J Young & Sons [1894] AC 335, referred to

Hollinrake v Truswell [1894] 3 Ch 420, referred to

Pantone Inc v AI Friedmann Inc (1968) 294 F Supp 545, referred to

Autodesk Inc v Dyason [No 2] [1993] HCA 6; (1993) 176 CLR 300, followed

Catnic Components Ltd v Hill & Smith Ltd [1982] RPC 183, followed

Interlego AG v Tyco Industries Inc [1988] UKPC 3; [1989] AC 217, referred to

Rose Plastics GmbH v William Beckett & Co (Plastics) Ltd [1989] 1 FSR 113, followed

Computer Edge Pty Ltd v Apple Computer Inc [1986] HCA 19; (1986) 161 CLR 171, followed

Lend Lease Homes Pty Ltd v Werrigal Homes Pty Ltd [1970] 3 NSWR 265, referred to

Inglis v Mayson (1983) 3 IPR 588, referred to

Hanfstaengl v HR Baines & Co Ltd [1895] AC 20, referred to

Bauman v Fussell [1978] RPC 485, referred to

Autodesk Inc v Dyason [1992] HCA 2; (1992) 173 CLR 330, followed

Green v Broadcasting Corporation of New Zealand [1989] 2 All ER 1056, referred to

Stowe v Thomas (1853) 23 F Cas 201, referred to

Computer Associates International Inc v Altai Inc (1992) 23 IPR 385, referred to

Horne v Comino [1966] Qd R 202, referred to

Lewis Falk Ltd v Jacobwitz (1944) 61 RPC 116, referred to

SW Hart & Co Pty Ltd v Edwards Hot Water Systems (1980) 30 ALR 657, referred to

Avel Pty Ltd v Multicoin Amusements Pty Ltd [1990] HCA 58; (1990) 18 IPR 443, discussed

Mechanical Services (Trailer Engineers) Ltd v Avon Rubber Co Ltd [1977] RPC 66, referred to

Pearson v Holden (1948) 65 RPC 424, referred to

COOGI AUSTRALIA PTY LT v HYSPORT INTERNATIONAL PTY LTD & ORS

VG 284 of 1993

DRUMMOND J

21 AUGUST 1998

BRISBANE

IN THE FEDERAL COURT OF AUSTRALIA
VICTORIA DISTRICT REGISTRY	VG 284 of 1993

BETWEEN:	COOGI AUSTRALIA PTY LTD Applicant
AND:	HYSPORT INTERNATIONAL PTY LTD First Respondent GEOFFREY GRAY JONES Second Respondent JOAN MARGARET JONES Third Respondent TERENCE GRAY JONES Fourth Respondent HARRY GEOFFREY BROWN Fifth Respondent JENNIFER ANNE BROWN Sixth Respondent BROMSGROVE PTY LTD Seventh Respondent

JUDGE:	DRUMMOND J
DATE:	21 AUGUST 1998
PLACE:	BRISBANE

REASONS FOR JUDGMENT

The issues

The applicant has long been a designer and manufacturer of machine knitted fabrics and garments made up from its fabrics. In the early 1990s it made and sold garments from a fabric it manufactured which embodied a design described as its "XYZ design". The first respondent (of which the second to sixth respondents are directors) is a competitor. The applicant claims that the first respondent has "since at least May 1992" made and sold three styles of knitted garments (together with variations of those styles) which are identified as HY95, HY82 and HY96. The applicant claims that, by doing this, the first respondent has infringed its copyright:

(1) in two "artistic works", viz, its fabrics and its garments that bear the XYZ design. It is said that each piece of fabric and each garment is an "artistic work" for the purposes of ss 31 and 32 the Copyright Act 1911 because it is a "work of artistic craftsmanship" within the definition of "artistic work" in s 10 of that Act;

(2) in a "literary work", viz, "the computer program and/or data comprised in the computer program, made in or about 1989, used for the manufacture by computerised knitting machines of garments bearing the XYZ design, including the print out of the programming codes, known as the XYZ control program, and the XYZ graph in symbol form contained therein". It is said that this program is a "literary work" for the purposes of ss 31 and 32 of the Act because it is a "computer program" within the meaning of that term in the definition in s 10 of "literary work", ie, it is within the definition of "computer program" in that section of the Act.

Allegations of infringement of copyright in other works associated with the XYZ design, initially made, were abandoned by the applicant during the hearing.

While Coogi, in its pleading, identified the subject matter of the first of its copyright claims as works of artistic craftsmanship comprising "fabric and/or garments manufactured by the applicant since 1989 bearing the XYZ design", at trial it put this part of its case more narrowly. In the course of the hearing, Coogi's counsel limited the subject matter of its copyright claim in the XYZ fabric to the form of the stitch structure of the fabric: Coogi produced the fabric in a total of eight colour ranges, but it made no attempt to identify them and disavowed that its colour selections comprised any part of this particular copyright claim. At trial, Coogi put its claim that Hysport had infringed its copyright in the XYZ fabric on the basis that there were so many similarities between the stitch structures of the two fabrics that Hysport had reproduced not the whole, but a substantial part, of the structure of the XYZ fabric.

Coogi did not sell any of the XYZ fabric, only finished garments made from that fabric. Mr Chapman, its general manager who also has technical qualifications in textile manufacture and who was involved in production of the XYZ design with Mr Walker and two others of the applicant's staff, Mr Taranto and Mr Sheltawi, says that Coogi's manufacturing process ensures that all its garments are distinctive, although they may be made from fabric bearing a particular design, because each is cut from a different part of the fabric and Coogi is careful to ensure that no two garments are identical: he says it regards each of its garments, including those made from the XYZ fabric, as a unique piece of wearable art. Coogi, however, made no attempt to identify the subject matter of the copyright that might, in view of this evidence, be claimed in particular garments. As I understand its case, Coogi only asserts that Hysport infringed Coogi's copyright in its garments by reproducing in its own fabrics and garments a large part of the stitch structure of the XYZ design fabric incorporated in each copyright Coogi garment. Pirating by Hysport of a large part of the stitch structure of the XYZ design fabric is the core of Coogi's complaint.

The applicant's case, as refined in counsel's closing submissions, of infringement of copyright in its computer program is one of infringement by unauthorised adaptation, rather than reproduction. Counsel for the applicant also said that no claim was made to copyright in the data associated with the program, save in so far as that data actually forms part of the Coogi computer program. There are also claims by Coogi that the first to sixth respondents contravened various provisions of the Trade Practices Act 1974 (Cth), including s 52, and passed off Hysport's business and garments as those of the applicant by making and selling garments in circumstances that involved representations that Hysport was the owner or authorised user of copyright in garments made to the XYZ design; appropriate relief is sought. The applicant acknowledges these claims stand or fall with its copyright claims.

The first to sixth respondents, as well as defending Coogi's case against them, cross-claim for a declaration under s 202 of the Act that Coogi's threats of copyright infringement were unjustified; they claim consequential damages. A cross-claim is also made for relief based on s 52 the Trade Practices Act and s 11 the Fair Trading Act 1985 (Vic), on Coogi's claim of copyright in the XYZ design as a work of artistic craftsmanship and on its assertions that Hysport infringed that copyright by making and selling its own garments.

The seventh respondent is a retailer. It has only ever stocked a few of Coogi's garments, selling the last of them in late 1991. It has, however, stocked and sold Hysport's garments, including garments of its HY82 and HY96 styles, since ceasing to carry the applicant's garments. Coogi's contention is that this respondent has thereby infringed its copyright in these "artistic works" and in its "computer program" and relies on s 38 of the Act. Coogi claims that the seventh respondent has also contravened s 52 and other provisions of the Trade Practices Act by selling Hysport's garments in circumstances that involve a representation that they are sold with Coogi's approval.

Coogi's case focuses entirely on the XYZ design, which is said to be a particularly important item of the company's intellectual property although it is but one of many designs it has applied to fabrics made by it. It only produced garments bearing this design for about twelve months in 1990-1991; they generated less than 2 per cent of the applicant's entire sales income in the period. However, there is no reason to doubt what Mr Taranto, Coogi's managing director, said: "it is a core design" that remains "the inspiration" for subsequent Coogi designs.

The XYZ fabric

The applicant produced its XYZ design fabric in large rolls, about two metres wide, on its Stoll brand computerised flat bed knitting machines. This fabric was made of a number of different yarns, including wool, cotton and silk, of various colours. Within a roll of the fabric, the design is repeated every one metre or so. The effect on the eye is of a piece of fabric with a highly textured surface, comprising a complex, abstract or non-representational multi-coloured design, with three dimensional elements. Each run of this design in a roll of fabric is made up of a number of bands or stripes of knitting. In appearance, some bands are of uniform width across the run eg the band of the stitch structure shown in panels 3 and 7 and the band of striped tube knit shown in panel 5 of Mr Chapman's comparison of Coogi's XYZ fabric with the first respondent's HY82 fabric (exhibit RLC23). Other bands in the run vary in width, eg, various of the bands in the photograph of panels 14 to 27 of the XYZ fabric in exhibit RLC23, an effect achieved, in part at least, by varying the tensions of the various threads as they are knitted into the fabric. The stitching making up some of the individual bands is of a uniform structure, although the use of different coloured yarns gives them a patterned effect, eg, the band shown in panels 3 and 7; other jacquards of differing designs in the two colours used in various bands are also a feature of the fabric, of which panels 2 and 8 of the exhibit are examples. Other bands, however, are made up of segments of differing stitches and stitch structures, with, eg, one such structure resulting in a flat section of knitted fabric, another resulting in a rolled section of fabric, and yet another in a complex three dimensional structure in the fabric of one of a number of designs, all three in the one band. An example of this is in panel 12 of exhibit RLC23. The final production stage involved the hand-cutting of the XYZ fabric into garment shapes, particularly sweaters and cardigans, and the sewing of those garments.

Mr Walker is a computer-programmer and knitting technician employed by the applicant. He says that the XYZ design was novel for the applicant in so far as it involved the use of different stitch structures within a single band in the fabric: previously, Coogi had only produced fabric in which the one stitch structure was used throughout the whole of a particular band. During the trial, Coogi, by its counsel, disavowed any claim that the stitches or stitch structures incorporated in the XYZ design were innovations made by Coogi: its case is that it has copyright in the particular selection of stitches and stitch structures and in the sequence in which those stitches and structures are arranged which make up the XYZ design fabric. It seems clear, eg, from the unchallenged evidence of Mr Batson, that stitch structures of the kind incorporated in fabric produced by Coogi to the XYZ design had been employed by others in the knitwear industry prior to the development by it of its XYZ design; these effects were made possible by stitching procedures and computerised knitting machine technology developed in the late 1970s and early 1980s. Coogi, in producing fabric to the XYZ design, did not make any innovations in knitting technology but rather used existing, albeit modern, technology to produce its XYZ design fabric.

It is necessary to say a little more about how Coogi came to develop its XYZ design fabric and something about how Hysport came to make the various garment styles that have provoked complaint by Coogi.

The development of the XYZ design

The development by Coogi of fabric made to the XYZ design was the result of discussions between Messrs Taranto, Chapman, Sheltawi and Mr Walker; Mr Walker made by far the most substantial contribution. At the early stage of the development of the design, he set about planning and then writing the program from which the program for the XYZ design was ultimately developed. He experimented with Coogi's computer-controlled Stoll knitting machine, using the graphs and control programs which he created, to produce trial runs of fabric. These were the subject of discussions between the four of them in which the design was further refined. Mr Walker had a significant input into the process which resulted in the final XYZ design, in so far as he himself chose, in the course of his experimentations, various stitch structures and sequences of various of the structures to obtain what he said was an "order of stitch structures which I thought looked right from a design point of view, based on the general concept of bringing together various stitch structures across the one knitting line ... No one presented to me in any visual sense an overall look of the design, but rather the design team discussed with me the design objectives and aspects of the design elements, which I brought together in my own way, exercising my own artistic judgment, and then refined in the course of discussions with" the other three. By late 1989, he had written a program comprising a control program and a graph for the XYZ design which had by then been developed to an advanced stage. Before any production runs of fabric bearing the XYZ design were made, however, Mr Walker ran the program in the knitting machine to produce further samples of the fabric; in the course of this final phase, he made adjustments to both the program and the knitting machine in order to arrive at the stage where he had the program in final form and the machine set up to produce fabric of acceptable design for manufacture. (He also said that later, during production runs of the fabric, he made further adjustments and changes from time to time both to the program and to the knitting machines to deal with problems arising from variations in yarn fibre due to temperature fluctuations and other causes. I do not regard Mr Walker's activities in making these adjustments to accommodate materials handling problems during the production phase as involving anything other than routine monitoring of a mass production process: they formed no part of the work of developing and finalising the XYZ design fabric.)

Mr Walker was never given a completely realised design for which he just wrote a program. He contributed to the development of the XYZ design fabric significantly more than his skills as a computer programmer: he had a number of years experience in the knitting industry and his knowledge of the materials properties of the yarns and his own artistic judgment all assisted him in producing, by a refining process, a program that realised the design objective of Coogi's design team. This objective was, I think, the development and setting up, by a process of trial and error experimentation, of a computer-controlled mechanical process for mass production of a multi-coloured fabric which, when made up into garments, would appeal to consumers of fashion clothes.

It does not appear that anything like a record of the development of the design was ever produced. Nor did Coogi produce any documentary evidence of the progress of this process. Mr Walker said that his own notes, programming design work done by Mr Sheltawi and the trial knitdowns (save for one) were not kept, although some rudimentary sketches made by Mr Sheltawi are in evidence. (They once were, but are no longer, the subject of a claim of copyright by Coogi.) However, it is plain that Coogi had the XYZ fabric in production in 1990 and the respondents did not challenge its claim to ownership of any copyright that may exist in either the XYZ design as incorporated in fabric and garments made by it or in the computer program.

The development of the Hysport fabrics

I turn now to the development by Hysport of its fabric designs.

Its key witness was its Managing Director, Mr Terence Jones, the fourth respondent. I approach his evidence with caution for a number of reasons. For example, he displayed a deliberate want of frankness in the account he gave in his affidavit about what he relied on to generate Hysport's HY82 design and he deliberately decided not to discover important correspondence passing between himself and Shima Seiki immediately after Coogi's complaint in March 1993 that the Hysport's own HY95 design was copied from Coogi's XYZ design. But I am still prepared to accept his account in oral evidence of how and when he developed the HY82, HY95 and HY96 fabric styles: he is corroborated in important respects by Mr Dawes, whose credibility was not attacked, and his sequence of production of the three styles fits the probabilities.

It is likely that, some time in 1990, Shima Seiki, a Japanese knitting machine manufacturer whose machines are used by Hysport, obtained one of Coogi's XYZ design garments and, working back from that garment by a technique not exposed in the evidence but which probably involved an examination of the stitching structure of the Coogi garment to identify the sequence of needle positions involved in knitting the XYZ fabric, produced a computer program which would enable its own computerised knitting machinery to make fabric of the same design. Shima Seiki included a sample of this fabric in its library as Pattern F823. It offers what is called a pattern service to users of its machines, which involves just such a process. In late 1990, Mr Jones attended a training course at Shima Seiki in Japan. He saw a sample of the F823 pattern fabric in Shima Seiki's library and obtained from it a disk copy of the program for that pattern. Mr Jones discovered on his return to Australia in early 1991 that this disk was defective. But he was able to obtain a replacement disk from Mr Dawes, who was then working for Ramsey McDonald, Australian agent for Shima Seiki. Although he only conceded as much in cross-examination, Mr Jones used this disk to produce the HY82 program and fabric. This was the first fabric made by Hysport which bore any similarity to Coogi's XYZ design. As Mr Jones admitted, he produced the first version of HY82 by electronically cutting and pasting from the disk of the Pattern F823 program he got from Mr Dawes. He also incorporated in this version of the HY82 program design data for a range of jacquards, including kangaroo and koala motifs and various word motifs such as "Cooee", which Hysport already held in its own computer files and which it had earlier used in producing garments of its own design about which Coogi makes no complaint. Mr Jones says he created the first version of HY82 fabric in March 1991. Garments made to this initial HY82 design were sold by Hysport in August 1991 and have been sold ever since, although every three or four months a modified HY82 design is produced which incorporates minor variations to the preceding version. The seventh respondent commenced to sell these HY82 garments in mid 1992 and has continued to sell them (and later variants of this design) ever since.

It is a feature of the knitwear industry in which both Coogi and Hysport are involved that manufacturers constantly produce garments to a large number of designs, some new, others variations of earlier designs. Hysport in late 1992 produced as one of its new designs, its HY95 style garments to the unmodified F823 pattern, which it had earlier modified as outlined above to generate its HY82 design. It used its copy of the F823 program to do this.

There is no basis in the evidence for accepting Coogi's submission, initially made, that Shima Seiki in some unidentified way produced its program for Pattern F823 from a copy of the applicant's own computer program for the XYZ design. Coogi did not attempt by evidence to suggest how a third party may have obtained access to the XYZ program without its permission and, in final submissions, contended instead "that what most likely happened is that someone had Shima copy some XYZ fabric" - not the XYZ program - and "[b]y copying XYZ fabric, Shima made an adaptation of the XYZ program by reverse engineering". Nor is there any basis for a finding that it was anyone associated with Hysport who caused Shima Seiki to produce the program for Pattern F823. This case therefore raises an entirely different question from that which arises where it is said that someone has infringed copyright in a computer program by devising his own program in source code by reverse engineering the original program expressed in object code.

In March 1993, soon after the first respondent put its HY95 design garments on sale, the dispute erupted: Mr Chapman early in that month bought one of Hysport's HY95 garments in a New Zealand shop. It was that garment about which he made complaint at his meeting with Mr Jones on 29 March 1993; he pointed to similarities between that garment and a Coogi garment which he had had specially knitted in colours similar to the Hysport HY95 garment. At this meeting he confirmed, at Mr Jones' request, that HY95 was the only one of Hysport's styles about which Coogi had complaint. Mr Jones no doubt made this enquiry because he must then have appreciated the similarities between his HY95 garment and the specially prepared example of the XYZ design garment Mr Chapman showed him and because Mr Jones well knew that Hysport's HY82 design, which was still on sale, originated in the same source as its HY95 design, viz, Shima Seiki's F823 pattern program. In the correspondence which he failed to discover, he tried to elicit confirmation from Shima Seiki that the disk he got from Mr Dawes and which he used to develop the HY82 program and to produce HY95 fabric contained the program for the Shima Seiki Pattern F823 fabric. I infer that Shima Seiki, in responding as it then did to Mr Jones, was reluctant to be drawn into possible litigation. It is now, however, financially backing Hysport's defence, probably out of concern that victory for Coogi may well restrict the use to which its customers can put its own computerised knitting machines. Mr Jones was unable to procure from Shima Seiki an acknowledgment that it bore responsibility for Hysport's innocent use of the F823 program to produce its HY95 fabric that was so like Coogi's XYZ fabric. As part of an attempt to settle the dispute that had arisen with Coogi, he caused the first respondent to circularise its customers in late April 1993 with the advice that it accepted that Coogi was the true owner of copyright in the design embodied in both Coogi's XYZ garments and Hysport's HY95 garments, admitted that the XYZ and HY95 design fabrics were "produced from the same knitting program" and that it was therefore withdrawing all its HY95 garments from sale. Mr Jones in oral evidence acknowledged that there was no difference between Shima Seiki's F823 pattern program and the program used to produce Coogi's XYZ design (save that they were written in different computer languages). But after April 1993, Hysport still continued to sell its HY82 garments, without complaint from Coogi, until May 1994.

Coogi was deeply concerned in March 1993 that Hysport was involved in pirating its XYZ design. After that, it was on the look-out for garments that might infringe its intellectual property rights in its products. However, it was not until 5 September 1993 that Mr Chapman bought two examples of the first respondent's HY82 style garment, in the seventh respondent's shop. Mr Chapman said this was the first occasion he recalled seeing that particular design. He claimed it caught his eye and he immediately identified it as "a derivative of the XYZ" design. He was closely involved in the development of the XYZ design and thoroughly familiar with it. Hysport's HY82 garments are of a fabric with some similarity to the fabric in Coogi's XYZ garments: both are made up of bands of various colours some of which are made up of different, quite striking stitch structures. But Mr Chapman had been to this same shop on five occasions between 20 May 1993 and 2 September 1993, on each of which he bought an HY95 design garment (or a variant of that design). He obviously went there on the look-out for pirated copies of the XYZ design. It is probable that HY82 design garments were on sale there throughout this period. But until September 1993 they were ignored by Mr Chapman. Even more significantly, Coogi did not complain about the HY82 design for a further eight months, despite Mr Chapman's claim that, immediately he noticed the HY82 garment in early September 1993, he recognised it as derived from Coogi's design. The inference is that neither Mr Chapman nor any one else at Coogi realised that there was sufficient similarity between the HY82 style and the XYZ design garments to support a charge of possible infringement of Coogi's rights, until long after Coogi's people were aware of the presence of HY82 garments in the market and long after they were concerned to identify any such infringements.

HY96 design garments first went on sale some time after March 1993 and were certainly on sale by October 1993: there is in evidence an invoice for a HY96 style garment sold by the seventh respondent so dated. I find that Hysport produced the HY96 design only after withdrawing HY95, ie, only after it realised, as a result of Mr Chapman's complaint, that Shima's F823 program was probably produced as a result of Shima Seiki copying Coogi's XYZ design and after Hysport accepted as valid Coogi's complaints as to the similarity Hysport's HY95 design bore to the XYZ design. In oral evidence Mr Jones admitted that he produced the HY96 design "to get around the problem [with the HY95 design] that had been revealed in March of 1993" and that he wanted to have a garment he could sell which would be close in general appearance to the successful HY95 garment. Even though there are differences between the stitching of the samples of the HY95 garment sold by Hysport to March 1993 (exhibit H17) and the first version of the HY96 (exhibit C38), both are in broadly similar colours and they have a similar banded appearance. Despite Mr Jones' evidence that he did not work from the F823 or the HY95 program, I infer that he developed the HY96 design from the F823 pattern he still had on the disk he got from Mr Dawes or from his disk version of his HY95 design, but that, in devising his HY96 program and fabric, he made fairly extensive modifications to the F823 program (or the HY95 program) to enable him to make another design of garment, which he believed would escape a charge of infringement of Coogi's rights in the XYZ design.

The XYZ fabric as a work of artistic craftsmanship

A substantial argument by Hysport against Coogi's claim that its XYZ fabric is a work of artistic craftsmanship turns on the proposition, which I accept, that Coogi's design effort was directed entirely to setting up a system for the mass production of fabric having particular features. The product of such a process, so Hysport says, is the antithesis of a work of artistic craftsmanship. In order to deal with this argument, it is first necessary to dispose of Coogi's contention that each of the many mass-produced runs of fabric embodying the XYZ design (and each of the many garments made from that fabric) is itself an "original work of artistic craftsmanship" within s 32.

Such a claim is not sustainable. Copyright only subsists, pursuant to s 32 of the Act, in an original work of which the author was a qualified person "when the work was made". It is clear from s 22 that, since a work is made for the purposes of s 32 when it "was first reduced to ... some ... material form", it is only the first expression or fixation of the author's skill and labour that can constitute the work in which copyright subsists. If the fact that Coogi produced fabric to the XYZ design in eight separate colour ranges is ignored for the moment, Coogi can, at most, have copyright only in that portion of the first roll of fabric produced by it in which the XYZ design was first completely embodied: it cannot sustain its claim to copyright in every length of fabric bearing the XYZ design.

But the question remains: can Coogi claim copyright in the first embodiment of the XYZ design as a "work of artistic craftsmanship", given that it too was the result of Coogi's efforts in setting up a process for the mass production of fabric to that design. (If Coogi can make out this claim, it would be an infringement for Hysport to reproduce that copyright subject matter by indirectly copying it from any one of the 2,500 or so Coogi garments that incorporated it.)

The law as to what is a work of artistic craftsmanship is not in a very satisfactory state: cf Lord Kilbrandon's comment in George Hensher Ltd v Restawhile Upholstery (Lancs) Ltd [1976] AC 64 at 98. But current authority supports, I think, a number of propositions:

(1) The phrase "works of artistic craftsmanship" is a composite phrase that must be construed as a whole. There is nothing to suggest that any of the words is used in other than one of its ordinary senses: Hensher, per Lord Simon, at 91.

(2) As is suggested by a comparison of paras (a) and (b) with para (c) of the definition of "artistic work" in s 10, a work will qualify as one of artistic craftsmanship only if it has an element of real artistic, ie, aesthetic, quality, whether or not it is a utilitarian article: Cuisenaire v Reed [1963] VR 719 at 730 ll 30 to 35; Hensher at 77F and 78F-G, 81H, 85H and 86E, 96G-H; Merlet v Mothercare PLC (1984) 2 IPR 456 at 465 and Bonz Group (Pty) Ltd v Cooke (1994) 3 NZLR 216 at 224.

(3) Implicit in the proposition that the existence of utilitarian features will not disqualify an article from being a work of artistic craftsmanship is the further proposition that the article need not have such a high level of aesthetic quality as to make it a work of fine art. In Hensher, Lord Reid at 78H and Lord Simon at 90F-G said as much. Lord Simon explained why (at 90): the object of giving copyright protection to "works of artistic craftsmanship", first conferred by the Copyright Act (UK), was to extend to works of applied art the protection formerly restricted to works of fine art, an extension made under the influence of the Arts and Crafts Movement.

(4) The level of aesthetic appeal required for a work of artistic craftsmanship is higher than mere visual appeal (one of the tests for design registrability); Hensher at 79C-F, 81E-F, 84F-G and 93E-F, 95F-H.

(5) In determining whether an article has sufficient aesthetic quality to be a work of artistic craftsmanship the Court is determining a question of fact on the basis of the evidence before it: Hensher at 82B, 87A-B, 94H and 97C-D.

It has been said that in deciding whether a work of craftsmanship has artistic quality the Court must not act on its own aesthetic judgment: Hensher at 78CD, 87BC, 94G and 96H-97A; Bonz at 224 ll 1-5. Accordingly, Laddie, Prescott and Vitoria, in The Modern Law of Copyright and Designs, 2nd ed, at 208, favour Lord Reid's test in Hensher and suggest that the necessary artistic element in a work of artistic craftsmanship will be shown by evidence that the visual appearance of the article is such as to cause at least some members of the public to wish to acquire and retain the article "on especial account" thereof. That, the authors say, is a fact capable of objective ascertainment by the Court and the Court is not required to base its decision on its own aesthetic opinion. Viscount Dilhorne at 87D rejected this sort of test as did Lord Simon, at 93D-H. It cannot be right. The statute by its definition of "artistic work" requires that a work of artistic craftsmanship be "of artistic quality"; in this context, that seems clearly to require that it have some aesthetic value. That a segment of the public can be found which may have been persuaded by advertising or by a transient fad to want to own the work because of its visual appeal does not necessarily mean that it must have the requisite minimum level of aesthetic value. I do not think the Court can, in determining whether a work is one of artistic craftsmanship, avoid the task of making a judgment, on the evidence before it, whether the work has a sufficient level of aesthetic appeal to be "of artistic quality". Forming a view, on conflicting evidence, as to what is essentially a matter of aesthetic taste is not a novel task for a court: see, eg, Attorney-General for NSW v Trustees of the National Art Gallery of New South Wales (1944) 62 WN(NSW) 212.

(6) Whether a work has the requisite aesthetic quality must be determined objectively: Hensher 81F-G (and cf 95B-C, where it was said that it is "the intent of the creator and its result" which will determine whether the work is one of artistic craftsmanship); Cuisenaire at 730 ll 50-55 and Bonz at 223-224.

(7) In determining whether an article has the requisite aesthetic quality, evidence that the creator intended to make an article possessing aesthetic quality is important, although not essential: Hensher 78F-G, 81G, 95B-C; Merlet at 465 ll 25-30 and Cuisenaire at 730 ll 50-55. Because the test is objective, whether the work has aesthetic quality cannot depend solely upon the subjective intent of the creator, since he may have failed to realise his intent, if it was an aesthetic one: Bonz at 223-224 and Merlet at 465.

In Hensher, Lord Kilbrandon, at 96F-H, appears to suggest that the presence of such an intention in the creator is determinative of whether the work will have the necessary artistic quality, although he also seems to say this is an objective test at bottom, in so far as he says that "[i]t must be possible to deduce the conscious purpose of artistic creation from the work itself or from the circumstances of its creation". But the way he then qualifies this statement makes it difficult to be certain just what he means.

(8) Expert evidence as to whether a work possesses the requisite aesthetic quality is admissible: Hensher at 78G, 82B-C, 87C, 94G-H (although the members of the House place differing emphases on the cogency of such evidence).

(9) Beyond this, it is difficult to find any guidance as to the considerations relevant to identifying the existence of the requisite aesthetic quality. But one thing is clear enough: the presence of non-functional features cannot be the test since, as Lord Simon observes in Hensher at 93B-C, the anthesis between function and beauty is a false one, especially in the context of the Arts and Crafts Movement.

(10) Before a work will qualify as a work of craftsmanship, it must be a manifestation of pride in sound workmanship and the result of the exercise of skill on the part of its creator in using the materials of which the article is made and the devices by which those materials are turned into the article. Lord Simon in Hensher said, at 91:

A work of craftsmanship, even though it cannot be confined to handicraft, at least presupposes special training, skill and knowledge for its production: see Cuisenaire v Reed [1963] VLR 719; Cuisenaire v South West Imports Ltd [1968] 1 ExCR 493, 514. `Craftsmanship', particularly when considered in its historical context, implies a manifestation of pride in sound workmanship - a rejection of the shoddy, the meretricious, the facile.

See also Merlet at 460, Bonz at 224 and Cuisenaire at 729.

(11) There is a difference of judicial opinion as to whether craftsmanship connotes only something hand-made. In Hensher Lord Reid at 77E referred to a work of craftsmanship suggesting "a durable useful handmade object" and Viscount Dilhorne at 84D said: "[a] work of craftsmanship is, in my opinion, something made by hand and not something mass produced". Pape J, in Cuisenaire at 729, spoke of craftsmanship as involving the display of manual skill.

In Hensher, Lord Simon, however, observed at 90F that Parliament deliberately used the words "artistic craftsmanship" not "artistic handicraft" and, in discussing the reason why copyright protection was given by the 1911 Act to works of artistic craftsmanship, viz, to give protection to works of applied art, in response to the influence of the Arts and Crafts Movement, said, at 91:

... although `works of artistic craftsmanship' cannot be adequately construed without bearing in mind the aims and achievements of the Arts and Crafts movement, `craftsmanship' in the statutory phrase cannot be limited to handicraft; nor is the word `artistic' incompatible with machine production: see Britain v Hanks Brothers and Co (1902) 86 LT 765.

Lord Simon, in this part of his speech, spoke of prominent members of the Arts and Crafts Movement who, by 1911, were applying its principles to machine-produced articles. As his references to printing books on machines and to Benson's metal work being produced by machinery on a commercial scale show, when he said at 91 that "craftsmanship" in the statute could not be limited to handicraft and that the word "artistic" was not incompatible with machine production, he was not referring to the use of machine tools, such as the lathe on which the cabinet-maker turns his chair legs, but to machines set up for the mass production of finely made books and metal articles. Lord Simon did not think that, in the context of the copyright legislation, there is any necessary incompatibility between artistic craftsmanship and mass production. There is much discussion in the literature on the impact of the Arts and Crafts Movement on industrial design and of the compatibility of the craftsman ethic with mass production. For example, Naylor in The Arts and Craft Movement, Trefoil Publications, 1991, refers to the later views of William Morris himself and of other leading figures in that Movement on the importance of applying, as Naylor puts it at p 177, "the Arts and Crafts ideals of fitness and truth to material to the machine-made [ie, mass-produced] product". See also the discussions of the opinions and approaches of Morris at p 161, Meynell at p 112, Crane at p 148, Benson at p 160, Frank Lloyd Wright at p 174, Ashbee at p 176 and of the success of the Scandinavians in integrating craftsmanship with industrial mass production at p 193.

In my opinion, Lord Simon provides good reasons for concluding that a mass-produced article can still be a work of artistic craftsmanship within the Copyright Act. The dicta to the contrary of Lord Reid and Viscount Dilhorne equate "craftsmanship" with "handicraft" when, as Lord Simon points out, there is good reason for thinking that Parliament did not intend the two concepts to be treated as equivalents. Pape J, in Cuisenaire, found support for his view that manual involvement was necessary to craftsmanship in the 1952 UK Copyright Committee's comment that copyright protection given to works of artistic craftsmanship was not intended to be given to mass-produced articles, which could obtain protection under the registered designs legislation, even if they might otherwise qualify as works of artistic craftsmanship: s 22 the Copyright Act 1911 (UK) then in force (the legislation with which Pape J was concerned - see [1963] VR at 725) denied copyright protection to designs capable of registration under the designs legislation. Pape J therefore read down the expression to cover only unique, hand-made objects. Lord Simon, in Hensher at 89, however, said that such a reading-down of the phrase "works of artistic craftsmanship" was not justified by provisions such as s 22 of the 1911 UK Act: in his opinion, they had their natural and ordinary meaning, which, as I have mentioned, he went on to explain could include certain mass-produced works.

Hysport put a similar argument to that which Pape J accepted. It contended that, since the object of Div 8 of Part III the Copyright Act was to deny copyright protection to artistic works applied in mass production and to permit only design protection for such works, achievement of that object would be impeded if the exemption of works of artistic craftsmanship from the general rule in Div 8 against dual protection for artistic works were widened by widening the meaning of the expression "a work of a artistic craftsmanship" to include mass-produced articles. But the Copyright Act treats works of artistic craftsmanship differently from other artistic works. Unlike the position reflected in s 22 of the 1911UK Act and that which obtained under Div 8 of Part III the Copyright Act 1968 prior to the 1989 amendments, protection under both the Copyright Act 1968 and the Designs Act 1906 (Cth) is now available to works of artistic craftsmanship. It matters not whether such a work is intended to be the subject of or used in mass production. Copyright protection is only lost for a work of artistic craftsmanship if the owner chooses to obtain design registration for that work. Otherwise, he can enforce his intellectual property rights in such a work by relying only on the Copyright Act, even though it was always his intention that the work would be industrially exploited. See ss 75 and 77 of the Act.

The Explanatory Memorandum to the Bill which became the 1989 amending Act suggests that the reason why works of artistic craftsmanship were exempted from the general rule in Div 8 that copyright should not be available to protect artistic works capable of industrial application was that, even though such works can embrace a wide range of works well capable of industrial application, they are a special case: "it is considered these articles are more appropriately protected under the [Copyright] Act whether industrially applied or not" (para 24). What may justify the special status conferred on works of artistic craftsmanship by ss 74 - 77 is recognition that the real artistic quality that is an essential feature of such works and the desirability of encouraging real artistic effort directed to industrial design is sufficient to warrant the greater protection and the accompanying stifling effect on manufacturing development that long copyright gives, in contrast to relatively short design-protection. Div 8 therefore provides no foundation for arguing that an expression "a work of artistic craftsmanship" should be narrowly construed.

(12) There is also a conflict of judicial opinion whether a work can be one of artistic craftsmanship only if it is the product of a person who combines both artistic and craft skills in making it or whether it is enough that the work meets the two criteria of being a work of craftsmanship and a work possessing the requisite minimum degree of aesthetic quality.

In Hensher at 94E-F, Lord Simon said that whether a work was one of artistic craftsmanship depended on whether it was produced by a person who is an artist-craftsman, not on whether it had two separate characteristics, viz, first, that it was produced by a craftsman and, secondly, that it had artistic quality: see also 95B-C. Statements in Cuisenaire at 729 and 730 suggest Pape J was of the same opinion. I have difficulty in accepting what is said in Cuisenaire at 730 ll 25 - 30 to the effect that the word "artistic" refers to some quality in the acts performed by the maker in the physical operation of making the article as distinct from "earlier cogitation and thought which produced the idea upon which the work was based". It must be possible for many works of artistic craftsmanship to come into existence as a result of an artistic design fully realised before any of the skilled work to implement that design is undertaken, even where it is all the work of a single person. Despite what he said in Bonz, at 224, in apparent acceptance of Lord Simon's view on the point, Tipping J in the passage immediately following considered that a work could be one of artistic craftsmanship even "[i]f two or more people combined to design and make the ultimate product". He held that the garments there in question were works of artistic craftsmanship because "the designer of the garments can fairly be called an artist and the handknitters can fairly be described as craftsmen".

Once it is accepted, as I think it should be, that "works of artistic craftsmanship" are not confined to hand-made objects, there is no reason why it should be essential for such a work to be the product of the efforts of a single person, ie, of an artist-craftsman. It is enough that it satisfies the two criteria of craftsmanship and aesthetic quality.

It cannot be contended that the XYZ fabric does not have aesthetic value sufficient to show that it is "of artistic quality". The way Coogi has used stitch structures and colour to produce an unusual textured and multi-coloured fabric for fashion garments suggests this. The experts, Mr Sonday, Mr Buckingham and Dr Dutton, agree that the fabric has aesthetic quality.

Mass production is not inconsistent with the articles so produced each having aesthetic value. Many mass-produced articles, such as nails, screws, etc made as purely functional fasteners, will not be capable of being works of artistic craftsmanship simply because they lack any aesthetic element in themselves (even though their display by Marcel Duchamp might be said by some to transmute them into works of art). But other mass-produced articles (such as the works of Charles and Ray Eames, "American designers best-known for the beauty, comfort, elegance and delicacy of their mass-producible furniture": New Encyclopaedia Britannica, 15th ed, Vol 4, p 317) have real aesthetic quality.

Nor do I accept that the only article capable of attracting copyright here, viz, the first production run of the XYZ fabric that was the first piece of fabric that the Coogi design team accepted as realising their design objective, was not also a work of craftsmanship.

There is no necessary difference between a skilled person who makes an article with hand-held tools and a skilled person who uses those skills to set up and operate a machine which produces an article. Such an article can still be a work of craftsmanship even though the creator has used a highly sophisticated computer-controlled machine to produce it, if nevertheless it is a manifestation of the creator's skill with computer-controlled machinery, knowledge of materials and pride in workmanship. The expert witnesses, Dr Dutton, Mr Sonday and Mr Buckingham did not suggest the contrary. To hold otherwise would be to import a Luddite philosophy into copyright legislation enacted against a background of modern industrial organisation and intended to regulate rights of value to persons involved in that area of activity.

Dr Dutton, a Senior Lecturer in the Philosophy of Art in the University of Canterbury, said that Coogi's garments could not be works of artistic craftsmanship, essentially because there was a sharp distinction between the design and production phases in which Coogi engaged to produce those garments. He accepted that artistry could be involved in the use of a computer in the design phase of the garments, but that neither artistry nor craftsmanship was involved in the mass production phase: that was simply an automated process in which the creator of the design had no relevant involvement in his capacity as an artist-craftsman. In view of Coogi's claim to copyright in each length of XYZ fabric and each of the 2,500 or so garments made from that fabric, as a separate work of artistic craftsmanship, it is understandable that Hysport focused on demolishing that claim and that much of Dr Dutton's evidence was directed to refuting this proposition. Mr Buckingham's opinion was that a work can be one of artistic craftsmanship, even though the designer does not participate in the operation of the machinery which makes the work, provided the designer consciously takes into account the production process and how it can be used in creating the effects embodied in the product. Mr Sonday held a not dissimilar view; but he did agree that the person who uses a computer to design knitwear is doing something quite different from the person who later uses a computer-operated knitting machine to manufacture the fabric.

It is unnecessary, however, to form any conclusion on whether any one of these differing views should be accepted as throwing light on the reach of the expression "work of artistic craftsmanship" in the Act.

The procedure by which the first run of fabric in the XYZ design which was accepted as meeting Coogi's objective was made cannot be segmented into two discrete processes, the one design and the other production, given the way Mr Walker went about producing that first run of fabric. He can, in my opinion, be fairly described as having used the computer-controlled knitting machine in that first exercise as a tool to enable him to realise, in the form of the first run of acceptable fabric, the creative ideas developed by the Coogi design team, including himself. He is skilled in textile machine operation, including program writing for such machines and in fabric design and he is knowledgable in the material properties of textiles; he used these skills and knowledge in setting up the machine and monitoring and adjusting it and the program he wrote to control it to produce the trial knitdowns, as he worked towards the stage where he had the machine set up to produce the desired effect. He is therefore entitled to be described as a craftsman and the product of his work as a work of artistic craftsmanship.

The first acceptable run of the XYZ fabric was not a prototype, ie, something developed separately from the mass production manufacturing process itself as a model for a line of mass-produced articles; it was as much a mass-produced item as each subsequent run of the fabric. But in a real sense, it was also "a work of artistic craftsmanship" within the Copyright Act because, firstly, it was the first fixation in tangible form of the idea of the XYZ design; in that one respect, it was different from every subsequent run of the fabric. Secondly, it was the end result of the efforts of Mr Walker in developing a program over a period during which he had regard to how he could program the knitting machine to produce particular effects, a process that involved trial and error use by him of the knitting machine until he produced a run of fabric that was accepted as meeting their objective by the Coogi design team, of which he was an important member.

Counsel for Hysport submitted that the notion of artistic craftsmanship required that the mind of a craftsman must be engaged during the act of production. Once it is recognised that the relevant act of production is that of making the first acceptable run of fabric to the XYZ design, the involvement of the Coogi design team and, in particular, Mr Walker's involvement in producing that particular article, answers this requirement. If Mr Walker had been given a fully developed design for the fabric and had simply written a program to ensure that the knitting machine would produce fabric to that design, he would not, in my opinion, have acted as a craftsman and no part of the fabric so produced could be said to be a work of craftsmanship, however much artistic quality it may have. He would have had no involvement with the materials from which the fabric was made and would not have displayed any skill in the use or understanding of those materials. But Mr Walker performed an entirely different role in the development and production of the first acceptable run of the XYZ fabric. That the initial one, like every subsequent run of fabric produced, is simply one of many identical products of a particular mass production process does not detract from the special quality of the first run of fabric in the circumstances of this case.

In my opinion, the first run of the XYZ fabric is a "work of artistic craftsmanship" for the purposes of the Copyright Act, notwithstanding that it was made as, and intended by those involved in its creation to be, the first of many identical runs of fabric bearing that design. None of the runs of the fabric subsequent to the first one to bear the XYZ design is, however, capable of being such a work (subject only to the consideration that needs to be given to the significance to its copyright claim of the fact that Coogi produced XYZ design fabric in a limited number of different colourways).

Coogi's claim of copyright in respect of the XYZ fabric

As I have indicated, Coogi's claim to copyright in the XYZ fabric was limited at trial to copyright in the assembly of stitches and stitch structures that make up the XYZ design: it makes no claim to copyright in the XYZ fabric or any of the garments embodying that design which extends to the colour combinations it used in producing that fabric. It is now necessary to consider whether Coogi can so limit the subject matter of its copyright claim.

By s 32 the Copyright Act, it is provided that: "[s]ubject to this Act, copyright subsists in an original literary, dramatic, musical or artistic work that is unpublished and of which the author: (a) was a qualified person at the time when the work was made ...". Section 10 defines the term "artistic work" to mean:

(a) a painting, sculpture, drawing, engraving or photograph, whether the work is of artistic quality or not;

(b) a building or a model of a building, whether the building or model is of artistic quality or not; or

(c) a work of artistic craftsmanship to which neither of the last two preceding paragraphs applies; ...

Many of these terms (but not "a work of artistic craftsmanship") are themselves defined in s 10. Section 22(1) provides:

A reference in this Act to the time when, ... a[n] ... artistic work was made shall be read as a reference to the time when ... the work was first reduced to ... some ... material form.

Section 29 provides:

(1) Subject to this section, for the purposes of this Act

(a) a literary, dramatic, musical or artistic work ... shall be deemed to have been published if, but only if, reproductions of the work ... have been supplied (whether by sale or otherwise) to the public;

...

(2) In determining, for the purposes of paragraph (1) (a), whether reproductions of a work or edition have been supplied to the public, section 14 does not apply.

..."

Section 31(b)(i) defines the nature of the copyright which s 32 declares subsists in an original work as including (in the case of an artistic work) "the exclusive right to reproduce the work in a material form". But copyright in an artistic work, unlike the position in respect of literary, dramatic and musical works, does not extend to the exclusive right to make an adaptation of the work. By s 14(1) of the Act, it is provided:

(a) a reference to the doing of an act in relation to a work or other subject matter shall be read as including a reference to the doing of that act in relation to a substantial part of the work or other subject matter; and

(b) a reference to a reproduction, adaptation or copy of a work shall be read as including a reference to a reproduction, adaptation or copy of a substantial part of the work, as the case may be.

Section 32 of the Act declares that copyright subsists in "an original ... work"; it does not confer copyright on elements or part only of an entire work. As Lord Hodson said in Ladbroke (Football) Ltd v William Hill (Football) Ltd [1964] 1 WLR 273, at 285, of a provision not materially different from s 32:

Copyright is a statutory right which by the terms of section 2 of the Act would appear to subsist, if at all, in the literary or other work as one entity.

Ladbroke was concerned with infringement of copyright claimed to exist in a compilation of existing material, which was said to be an original literary work. Since originality sufficient to attract copyright in such a case can be found in the skill and labour involved in making the selection from the mass of existing material that comprises the compilation, there is special reason, where the issue is infringement in a compilation of existing material, to focus on the entirety of the compilation in determining whether copyright can exist in it. But that consideration does not in my opinion qualify the fundamental principle enshrined in s 32 that copyright can only subsist in a work considered as a whole, not in elements or component parts of the work. Cf Powerflex Services Pty Ltd v Data Access Corporation (1997) 37 IPR 436 at 455.

What that entity the subject of copyright is in any given case cannot be governed by what the copyright claimant chooses to say it is. It must be a matter for objective determination by the Court. Otherwise a person suing for infringement of copyright could, by narrowing the subject matter of his claim, deprive a defendant of what might be a good defence that what the defendant took did not involve the taking of a substantial part of the true object of the claimant's expenditure of skill and effort.

The fact that copyright, once established, can be infringed by the copying of something less than the entire work, ie. by copying a substantial part of the whole, does not detract from the basic principle that before an infringement action can succeed, the claimant must, where its claim to copyright is put in issue, prove copyright in the particular entity that constitutes the "work": Lord Reid pointed out in Ladbroke (Football) Ltd v William Hill (Football) Ltd at 276 - 277 that it is wrong to conflate the exercise of identifying the subject matter of the copyright claimed with the exercise of determining whether there has been an infringement of the copyright by the reproduction of a substantial part of that subject matter.

It may be possible in some circumstances to claim copyright in what might be said to be only an element or part of a larger entity. But, in my opinion, that can be so only if the part can fairly be regarded as so separable from the material with which it is collocated as itself to constitute an "original ...work" for the purposes of s 32 of the Act. The publication in Leslie v J Young & Sons [1894] AC 335, which Lord Reid in Ladbroke at 277 said was an "apparent exception" to the rule that copyright can only exist in an entire work, consisted of a timetable and a tour guide. The first part of the publication lacked sufficient originality to attract copyright; the second part, and only that part, attracted copyright because it was both original and, as Lord Herschell said in [1894] AC at 342: "may be properly treated as an independent work and protected by the copyright law". In Hollinrake v Truswell [1894] 3 Ch 420, at 426, Lindley LJ denied copyright to instructions printed on a measuring device (which was not itself within the definition of "book" in the statute conferring copyright) on the ground that it was not "separately published ... not a publication complete in itself" but only a direction for the use of the instrument and so meaningless apart from the instrument. Parts of publications, such as titles and headings, have occasionally been accepted as having copyright protection, although that is not the general rule. Where the title or heading can be said to constitute a substantial part of the entire publication, the copying of it will of course be an infringement of the copyright in the whole work. But consistently with what I think is the law, Lahore, Copyright and Designs, Vol 1, para 6015 and Copinger and Skone James on Copyright 13th ed (1991) paras 21-31, suggest that an item such as a title or heading can itself be the subject matter of copyright only if sufficient original work has gone into its formulation to make it an original literary work on its own. See also Powerflex Services at 455.

It is therefore necessary to determine whether in the circumstances of this case Coogi can confine its claim in respect of its fabric to the assembly of stitches and stitch structures that make up the XYZ design and which is embodied in that fabric and say that the eight colour selections it made, one of which is incorporated in each production run of the XYZ fabric it made, are irrelevant to determination of the subject matter of its copyright.

As pleaded, Coogi's claim is to copyright in an artistic work is to copyright in "fabric and/or garments ... bearing the XYZ design". Whether it is made up of the assembly of stitching by itself or in combination with the colour selection adopted by Coogi, the XYZ design is not borne on or applied to the surface of the fabric in the way a two-dimensional screen print image is applied to the surface of fabric: that design is incorporated in and forms part of the fabric itself. It is clear that the XYZ fabric has the appearance it does in part because of the unusual stitch structures and the unusual assembly of different stitches and stitch structures. Dr Dutton, Hysport's expert witness, said that it is this which results in the garments made from the XYZ fabric being "robustly three-dimensional". Because all this stitching contributes to the appearance of the textile, it is, at the very least, an element of the work (the fabric) in which copyright subsists. Contrary to the submission of Hysport's counsel, this element is not able to be characterised as merely the principle upon which the fabric is constructed.

But there is no suggestion that Coogi ever produced that fabric in a single colour or that it ever produced or had the slightest interest in producing fabric to the XYZ design without regard to the colours of the fabric. The evidence is all to the contrary. Colour selection was an important part of Coogi's XYZ design process. Mr Taranto, a member of the design team, said it was in fact he who played a key role in its development. He said:

In developing any design, special emphasis is given by the Applicant to the structure of the knit of the fabric in order to create a bodied and highly textured look. I believe that the Applicant's distinctive image in the marketplace is based on its unusual and creative fabric designs, combined with its use of the colour palette and distinctive colours. Throughout our domestic and international marketing, it is continuously the colour and patterning in the Applicant's designs which are remarked upon by buyers as being unique features of the Applicant's garments. I believe that the originality of the colouring and of the fabric patterning in the Applicant's garments are very distinctive.

In describing the actual development of the XYZ design itself, he also said:

Throughout the design process, there was a constant review of the overall sense of balance of the fabric, involving the selection and balancing of colour, texture and patterns which combine to give the fabric design a unique feel of movement and flow.

Mr Walker said that: "[o]ne of Mr Taranto's prime stated objectives was to achieve the concept of movement and colour within the fabric" and it was Mr Taranto himself who selected the final colours adopted in the completed version of the design. Mr Walker spoke of the attention he gave to colour in describing his own work in writing the XYZ control program. He said:

the writing of this program for the XYZ design required careful attention to the image I had in my mind of the overall look of the XYZ design. The construction of each line in the control program was influenced by the overall look I wanted to achieve. This is done for example, by the selection of jacquards or other features in the graph, or the direction to knit with particular tensions and colours.

The applicant's expert, Mr Sonday, the Curator of Textiles in the Smithsonian Institution's National Museum of Design in New York, spoke of the samples of Coogi fabrics, which included an example of the XYZ design provided to him for the purpose of swearing his affidavit, as being of artistic quality "insofar as they combine a vivid pictorial quality, carefully constructed structures and a striking use of colour". In oral evidence he said that when he first saw Coogi fabrics he was struck by "the rhythm of the line, the rhythm of the colours, the fact that there are symbols as well as structures created by the knitting. It's the combination of that which gives you a very graphic visual quality." Although he did not suggest close familiarity with the fabrics bearing the XYZ design, he said that his comments as to the significance of Coogi's use of colour applied to the XYZ design, as he knew it, as well as to Coogi's other designs, which he had seen and some of which were held in his Museum. Dr Dutton pointed out that the standard tag attached to each of the Coogi garments which he had inspected in a New Zealand store contained the statement: "Coogi combines artistry with colour and technology to create extraordinary knit structures." That tag also appears on the only production sample of a garment made from fabric bearing the XYZ design in evidence, exhibit RLC4B.

The evidence is that the assembly of stitches and stitch structures that came to be incorporated in the XYZ design took its final form as part of a continuous design activity directed to the production of a fabric which, as a result of the chosen combination of the assembly of stitching with a carefully selected range of colours, would have consumer appeal. This evidence shows that colour selection is just as integral an element of the design of the fabric used to make the Coogi garments as is the assembly of stitching that gives the fabric its peculiar three-dimensional appearance. It is, I think, plain from the evidence of Mr Taranto, Mr Walker and Mr Sonday, that the eight colourways in which the applicant actually produced fabric embodying the XYZ stitching design are of as much importance as the one assembly of stitches and stitch structures that is embodied in every length of fabric produced by the applicant to that design, in fulfilling the aesthetic and commercial objectives of the Coogi people who were responsible for devising the XYZ fabric, in conferring objective artistic merit on the XYZ design fabric and in giving that fabric its distinctive appearance.

Copyright inheres in the form in which the author of an artistic work has chosen to express his idea. A deliberate selection of colours can therefore constitute part of the author's chosen form of expression. In Ladbroke at 282, Lord Evershed considered that selection of the colours in which the coupon was presented involved part of the skill and labour put into devising the coupon that attracted copyright. US copyright law is based on the same concepts as the Australian law: see Epstein's Modern Intellectual Property 2nd ed (1989) section 102 (although registration of copyright in works originating in the US is necessary if the owner is to be able to enforce all the rights comprising its copyright. See section 106, pp 170 - 172). It is well-established there that "intellectual labour of sufficient originality to confer copyright on the form of expression of that labour" can be found in the effort put into selecting the colours incorporated in a pictorial, graphic or sculptural work. See Lazzaro, Color It Copyrighted: Protection for Colorized Motion Pictures, Journal of Law & Commerce, Vol 8, at 405 - 406; Pantone Inc v AI Friedmann Inc (1968) 294 F Supp 545 at 548.

For the reasons given, the subject matter of Coogi's copyright is the first run of XYZ fabric in the particular colour selection first adopted by Coogi. Given the importance of colour to the visual appearance of the XYZ fabric, it may well also have a separate copyright in the first run of fabric embodying the XYZ design in each of the other seven colourways in which it subsequently produced the fabric. But Coogi can have copyright in nothing else, so far as such a claim is made in respect of its XYZ fabric and in respect of the garments made from that fabric.

Did Hysport infringe Coogi's copyright in the XYZ fabric?

Any copyright Coogi may have in the XYZ fabric and garments can be infringed only if Hysport in its own fabrics reproduced a substantial part of Coogi's copyright works. It remains to consider whether Coogi can make out its contention that that is what Hysport has done, when Coogi has not attempted to identify the whole subject matter of its copyright in respect of the XYZ fabric.

Coogi led a lot of evidence to show that, while Hysport did not reproduce the whole of the assembly of stitches and stitch structures of the XYZ design in its HY82 fabric, it reproduced a substantial part of that assembly of stitching. All this is consistent with Mr Jones' oral evidence as to the generation of both HY95 and the HY82 fabrics from the F823 electronic pattern and with my finding as to the likely source of that pattern. A causal link between the Hysport and the Coogi fabrics therefore being shown, whether there has been an unlawful taking depends on whether the portion taken is a substantial part of the XYZ fabric. This is turn depends upon the quality of the part taken, ie, on "the importance which the taken portion bears in relation to the work as a whole...": Autodesk Inc v Dyason [No 2] [1993] HCA 6; (1993) 176 CLR 300 at 305. So far as concerns an artistic work (which includes a work of artistic craftsmanship), whether the taken part is a substantial part of the original depends on "how important that part is to the recognition and appreciation of" the work as an entity: Catnic Components Ltd v Hill & Smith Ltd [1982] RPC 183 per Buckley LJ at 223. For the taken part to be a substantial part of an artistic work, it must be that which is visually significant: Rose Plastics GmbH v William Beckett & Co (Plastics) Ltd [1989] 1 FSR 113 at 123. These approaches were approved by the Privy Council in Interlego AG v Tyco Industries Inc [1988] UKPC 3; [1989] 1 AC 217 at 265-266. Although these comments were made in cases which were concerned with technical drawings, they are, in my opinion, of general relevance to the question: is the part of an original artistic work that has been taken a substantial part of the work? They emphasise that the answer depends on whether the part taken is a visually important part of the original work, ie, on the significance the part taken plays in creating the work's distinctive appearance.

There is also authority that, in order to show that the one is a reproduction of the other, Coogi must not only show that the HY82 fabric was derived from the XYZ fabric, but as well, that there is "a sufficient degree of objective similarity between the two works": Computer Edge Pty Ltd v Apple Computer Inc [1986] HCA 19; (1986) 161 CLR 171 at 186. This raises the question: "Sufficient for what purpose?" The answer, I think, is that there must be such a degree of objective similarity as is sufficient to show that what has been taken is part of the form in which the author of the copyright work has expressed his idea and not only a representation, in different form, of the idea itself. Where A copies the whole of B's original work, objective similarity will necessarily exist. Where, however, A copies only what is said to be a substantial part of the original work, since substantiality depends upon the quality rather than the quantity of what is taken, it does not logically follow that objective similarity need necessarily exist between the two works for infringement to be made out. But it is settled that proof of such similarity is essential. Proof of objective similarity provides an important protection against a finding of copyright infringement being made where substantial reproduction only is alleged, when all that a person has taken is the author's idea rather than any significant part of the author's form of expression of the idea, ie, that which alone attracts copyright. The rule has particular relevance where the question is whether there has been infringement of copyright in a work which possesses a degree of real novelty: there is always a risk that an element of the allegedly infringing article may so evoke the novel idea which is the subject of expression in the original work that that will be treated as the taking of a qualitatively important part of the original, even though the two forms of expression are quite different. This point is clearly made in Lend Lease Homes Pty Ltd v Werrigal Homes Pty Ltd [1970] 3 NSWR 265 at 273, one of the cases cited in Computer Edge at 186. See also Inglis v Mayson (1983) 3 IPR 588 at 605. This rule is, I think, a rule of general application in copyright. So far as it applies to whether an original artistic work has been substantially reproduced, it calls for much the same approach as that outlined in different words in cases such as Catnic.

Coogi has deliberately confined its case that the HY82 fabric reproduces a substantial part of the XYZ fabric, to showing that it reproduces a large number of sections of stitching and of the stitch structures of the XYZ design. In my opinion, this exercise misses the point: Coogi has not attempted to show that what was taken is visually significant, ie, sufficient to give to the HY82 fabric the "look" of the XYZ fabric. To show infringement by the taking of a substantial part of Coogi's XYZ fabric, exact replication of the appearance of the whole of the XYZ fabric is not necessary. It might, eg, be possible to make a fabric with the essential "look" of the XYZ fabric by making a simplified version of the XYZ fabric. But Coogi has to show that Hysport has taken for its HY82 fabric sufficient of the elements of the XYZ fabric that give the XYZ fabric its distinctive appearance to create in the HY82 fabric that same distinctive appearance.

I have referred to the evidence which I consider shows that it is the interaction between colour selection made by Coogi and the assembly of stitching devised by Coogi that gives the XYZ fabric its distinctive appearance. Whether that part of the XYZ fabric design which Hysport took for its HY82 fabric design is a visually significant part of the former is something that can only be determined by a comparison of the visual impacts of the inter-relationship of Coogi's colour selection with the XYZ stitch structure, on the one hand, and of the colours selected by Hysport in combination with its HY82 stitch structure, on the other.

However, Coogi never attempted to show by evidence the likely outcome of such a comparison. It is by no means self-evident that the entire assembly of stitches and stitch structures that undoubtedly forms an important element of the fabric embodying the XYZ design, let alone that portion of those stitches and stitch structures which Hysport indirectly copied from the XYZ design, is a substantial part of the XYZ fabric considered as a copyright work, so that a taking only of that part by the first respondent would be an infringement of the applicant's copyright. In Hanfstaengl v HR Baines & Co Ltd [1895] AC 20 at 28, Lord Watson spoke, in the context of determining whether a tableau vivant constituted an unlawful copying of "the design" of an artistic work (a painting), of the need to compare all "the particular forms and arrangements (whether of lines or colouring) which the copyright author has selected as the vehicle for conveying his idea to those who see his work" with all the details of the allegedly infringing work. The impact that colour can have on the structure to which it is applied when the question is one of copyright infringement in an artistic work is illustrated by Bauman v Fussell [1978] RPC 485, where Somervell LJ considered that the difference in colour treatment between the photograph and the painting it inspired was such as to prevent the latter constituting an infringement of copyright in the photograph, although both the subject-matter of the photograph and the positioning of that subject-matter in the photograph were copied by the painter. The fact that a change of colour scheme can powerfully affect the visual impact of a piece of textile used to make fashion garments was recognised by Coogi in a striking way: it went to the trouble of making lengths of XYZ design fabric in colourways it had never used when it had that design in production, but which were deliberately selected to be close to the colourways of the Hysport HY95 and HY82 garments about which Coogi complained. This was done to suppress the impact different colour schemes have when a comparison is made of the Coogi and Hysport fabrics, in order to highlight the similarities in stitching structure between them.

If, contrary to my view, the requirement of "objective similarity" involves a different test, so far as artistic works are concerned, from that outlined in cases such as Catnic, I do not think a finding that HY82 is objectively similar to the XYZ fabric can be made. Given that Coogi's copyright inheres in the combination of particular colour selections with a particular assembly of stitches and stitch structures, it is not open to the Court to reach any conclusion on whether the taking by Hysport of part only of the assembly of stitching of the XYZ fabric involves the taking of a substantial part of Coogi's copyright work. Coogi has not asked the Court to undertake that particular task. And by failing to identify the eight colourways in which the fabric was produced which are an integral part of the subject matter of the applicant's copyright, Coogi has, in my opinion, left the Court in the position where it is not able to judge whether that part of the assembly of stitches and stitch structures copied by Hysport in truth qualifies as a substantial part of the whole copyright work.

The difficulty of making such a judgment is emphasised when the only example of production run XYZ fabric in evidence (exhibit RLC4B) is compared with the first respondent's HY82 pullover. The latter is in quite a different colourway from the former, eg, it has very much more red in it and it is in generally brighter colours than the former; unlike Coogi's entirely abstract XYZ design, HY82 prominently incorporates animal symbols and word motifs in a variety of colours. In my opinion, the Court cannot sensibly judge whether the one involves reproduction of a substantial part of the other, in the absence of evidence on the basis of which the Court could assess the significance of the adoption by the first respondent of part only, albeit a large part, of the applicant's assembly of stitches and stitch structures, when it is not suggested that it took the whole or a part of any of Coogi's eight colour selections and when it has incorporated in its HY82 styles coloured representational motifs which sharply distinguish it visually from Coogi's fabric.

In view of my conclusion as to what can be the only subject matter of copyright in respect of Coogi's fabrics and garments, it is not necessary to analyse in any detail the evidence relied on by Coogi to show that Hysport, in its HY82 fabric, has incorporated a large number of the stitches, stitch structures and assembly of stitching that are a feature of the XYZ design fabric. The fairly cursory evidence comparing the stitching of that fabric with the stitching of Hysport's HY96 fabric can be passed over for the same reason.

Coogi fails to make out its case that Hysport (and thus the seventh respondent) infringed its copyright in respect of its XYZ fabric and garments by producing and selling its own HY82, HY95 and HY96 fabric and garments.

The Computer Program Infringement Claim

Is the XYZ program a "computer program"?

Hysport's first point is that the XYZ "program", ie, the XYZ control program and graph, is not a "computer program" within the meaning of that term in the Copyright Act. Hysport submits that the Act is intended to give protection, as "computer programs", only to data processing instructions and that to be a "computer program", the set of instructions must be intended to cause a device which has digital information processing capabilities to perform a digital information processing function. Otherwise, so it says, the absurd result would be that whenever anyone operates a device that has some form of digital computer incorporated in it, such as a word processor, an automatic petrol pump or a microwave oven, the user would be creating a computer program whenever the user instructed the device to perform its end function, eg, by typing text on the word processor keyboard or by operating the lever handle of the pump or the control buttons of the oven. Reliance was placed on Dr Forward's evidence. He said that the instructions comprising the XYZ "program" do not cause the digital processor controlling the knitting machine to perform a digital processing function, but rather the mechanical function of knitting, by causing the needles and other parts of the knitting machine to respond in particular ways to signals emitted from that processor. Dr Forward also distinguished between data and a computer program and said that the Coogi XYZ "program" was only data.

Both Coogi's expert, Professor Goldschlager, and Mr Walker said that the Stoll (and the Shima Seiki) knitting machines are highly flexible in so far as they can be set mechanically to produce a virtually limitless range of fabric patterns, including intricate and complex patterns. In each machine, the movement of individual needles in the needle beds and the movement of the carriages back and forth over the needle beds, carrying and placing the different yarns onto the needles, is controlled by a digital computer incorporated in the machine. This computer is operated by the machine-readable version of what can be called for the moment the XYZ program and the various Hysport programs.

Professor Goldschlager agreed that some kinds of material can be classified only as data and never as a computer program or part of a computer program. But he denied that there is a rigid dichotomy in computer science between data and computer programs, ie, he denied that the two can never overlap and said that, in particular settings, material can be classifiable as both data and a computer program (or part of a program). He said that, in order to determine the character of material:

It is necessary to examine the intended and actual effect of data input. The purpose of the Coogi XYZ computer program is to cause a computerised knitting machine to create a fabric knitted in the form of the Coogi XYZ design. The data input provided by Coogi's computer programmer, Mr Gary Walker, is fundamental to that purpose, because the data constitutes the very set of instructions necessary to cause the knitting machines to knit a fabric in the form of the Coogi XYZ design.

Professor Goldschlager said that each of the Coogi XYZ program and the various Hysport programs is written in what he described as a computer language, "Sintral" in the case of Coogi and "Shimatronic" in the case of Hysport and that "[t]here are astronomically large numbers of different programs which can be written in these languages". He said that both programs "specify, line by line, the movements which the needles and carriages should make in order to knit a piece of fabric, line by line". Each of these programs contains both a control program and graph. Professor Goldschlager said that the graphs "can be classified as tables of data which the control programs use to help effect the movement of the needles and the pattern of coloured yarns which are placed on the needles". He said they are "an integral part of the computer programs". Such a graph is created by the person writing the program; whether a particular component of the program is expressed as part of the graph to be used in conjunction with the control program or whether that component is written directly into the control program is a decision which the programmer can make in respect of each individual component of the fabric product. Professor Goldschlager said that in the Coogi XYZ graph the programmer has set out the instructions for the placement of needles that is necessary to make all the fabric patterns that together comprise the XYZ fabric, save for the instructions for some of the patterns of a less complex nature, which are contained in the control program. The structure of the Hysport programs differ from the Coogi XYZ program in so far as in the former, the control program element of the entire program performs a less sophisticated function than does the control in the Coogi program; most of the needle-placement instructions in the Hysport programs are contained in the program graphs. Each of these programs exists in electronic form stored on a disk and in hard copy print-outs from the disk.

Cross-examination extracted the following concessions from Dr Forward as to the proper characterisation of the XYZ program (which he called "the XYZ knitting pattern"):

No?---So just because this knitting pattern is knitted by a machine which is controlled by a micro-processor doesn't mean - doesn't turn it into a computer program. It still remains a knitting pattern.

A knitting pattern which contains a large number of instructions, do you agree?---Yes.

Yes, a large number of instructions which will be complied into a low level electronic form, do you agree?---Yes.

Those electronic instructions will be received in the machine and processed by a micro-processor, do you agree?---Yes.

And the micro-processor in turn will cause the machinery to operate, correct?---Yes.

And (speaking of the content of the control portion of the XYZ program, he said):

Right. You have accepted there are a set of instructions, right?---Yes.

Those instructions are designed for use by a machine, are not they?---Yes.

And the machine has digital information processing capabilities, does it not?---Well, in my understanding of what digital information processing capabilities are a knitting machine does not have that ability.

Does the micro-processor in the knitting machine have digital information processing capabilities?---Yes, it does.

Yes. And it is the micro-processor that receives these instructions, is it not?---It is the micro-processor that receives these instructions, yes.

And it receives those instructions and causes other bits of machinery to do things?---Yes.

Yes. And that is the purpose of the instructions?---Yes.

He also conceded that "the computer controls the knitting machine in accordance with the instructions that are contained in" the XYZ program and that the version of the XYZ program, which a computer in the knitting machine compiled into the machine-readable version of that program and which was fed into the other computer in the knitting machine that controlled needle placement and all the other operations the knitting machine had to perform to knit the XYZ fabric, was itself a computer program. Dr Forward acknowledged too, that the XYZ program contained individual instructions of the kind which, when aggregated, are capable of constituting a computer program; that the control portion of the program contains a large number of instructions and that it includes sub-routines (ie, particular sequences of instructions) and makes provision by a number of brief instructions for repeatedly calling up those sub-routines at various times in the running of the XYZ program, a feature he acknowledged is a characteristic of a computer program.

By s 32(1), copyright subsists in an original, unpublished literary work. Section 10 of the Act defines the term "literary work" to include a "computer program or compilation of computer programs". The same section defines "computer program" to mean:

an expression, in any language, code or notation, of a set of instructions (whether with or without related information) intended, either directly or after either or both of the following:

(a) conversion to another language, code or notation;

(b) reproduction in a different material form;

to cause a device having digital information processing capabilities to perform a particular function;

Section 31(1) provides:

... copyright, in relation to a work, is the exclusive right:

(a) in the case of a literary ... work, to do all or any of the following acts:

(i) to reproduce the work in a material form;

...

(vi) to make an adaptation of the work;

(vii) to do, in relation to a work that is an adaptation of the first-mentioned work, any of the acts specified in relation to the first-mentioned work in subparagraphs (i) to (v), inclusive;

Section 10 defines the term "adaptation" to mean:

(a) ...

(b) ...

(ba) in relation to a literary work being a computer program - a version of the work (whether or not in the language, code or notation in which the work was originally expressed) not being a reproduction of the work;

(c) in relation to a literary work (whether in a non-dramatic form or in a dramatic form):

(i) a translation of the work; or

(ii) ...

That an essential element of the Coogi XYZ program is a body of data in the form of the graph, as distinct from instructions, is no impediment to that graph and the control program together comprising a "computer program". This term is defined in s 10 to mean "an expression, in any language, code or notation, of a set of instructions (whether with or without related information) intended ... to cause" a digital information processing device to perform a particular function. The definition expressly envisages that a body of data - information, as distinct from instructions - may be an integral part of a "computer program". The Explanatory Memorandum accompanying the 1984 legislation that introduced the definitions relating to "computer programs" set out above states that the intention of the expression in the definition in emphasis is to make it clear that a "computer program" may include material other than instructions for the computer, such as "data to be used in connection with the execution of the program". Recognition of this intent was central to the decision in Autodesk Inc v Dyason [No 1] [1992] HCA 2; (1992) 173 CLR 330: see 346 - 347. In Autodesk [No 2] 176 CLR at 329 - 330, Gaudron J said:

Ordinary usage and the language in the context of the definition of `computer program' in s 10 of the Act compel the conclusion that the words `set of instructions (whether with or without related information)' extend to comprehend information as well as commands."

Both Gaudron J (at 329 to 330) and Brennan J (at 311) in Autodesk [No 2] said that, in the context of the Copyright Act, there is no necessary dichotomy between data or information, on the one hand, and instructions or commands, on the other. Both expressed the opinion that the phrase "computer program" in the Act extends to the whole body of material that causes the computer to perform the function in question and it does not matter that a part of that whole may be described as data or information, rather than instructions or commands. Applying what Brennan J said (at 311) in this regard to the present case, it is the entirety of the electronic signals generated by the computer embedded in the knitting machine when the XYZ program is running that makes the embedded computer perform the function of causing the knitting machine to execute the sequence of movements that are involved in knitting out the XYZ fabric: it is that group of signals that answers "the description of a `set of instructions' and an expression of those electronic signals answers the description of a `computer program' ..." On the evidence here, the graph, though in one sense only data, is as much a source, and also an expression of part of these signals, as is the control program. It is both together that constitute an expression of a set of instructions and thus a "computer program" for the purposes of the Act. The evidence of Professor Goldschlager and the concessions made by Dr Forward show that the Coogi XYZ "program" (graph and control program) falls within the words of the definition of "computer program" in the Act, so construed.

In my opinion, the words of the statutory definition of "computer program" do not readily convey the meaning contended for by Hysport, viz, that a set of instructions can only be a "computer program" if the function it causes a device having digital information processing capabilities to perform is a digital information processing function. That such a limitation is not to be implied in the definition is confirmed by the 1984 Explanatory Memorandum. This states (in para 21) that the phrase "to cause a device ... to perform a particular function" is intended to make clear that the device is one the performance of which is ultimately controlled by the "expression ... of a set of instructions", ie, by that which constitutes the computer program, whatever may be the function of the particular device; in para 22, it is said that the intention of the phrase "having digital information processing capabilities" is to make clear that devices which only process information by analogue methods are excluded and that this phase is intended to include devices which "though considered as a whole might not be information processors, nevertheless have some such capability". The Memorandum gives a computerised ignition system as an example of such a device. Such a system is controlled by a computer program and it incorporates a digital processing capacity. But that system's function is plainly not confined to processing digital information: a computer program designed to cause the computerised ignition system of a motor vehicle to function by injecting and then exploding different mixtures of air and petrol vapour in the combustion chambers of the engine as the accelerator is depressed or elevated does not cause that entire system to perform only a digital information processing function, even though that system is a device which must incorporate that processing function as part of its overall function. These statements in the Memorandum are inconsistent with reading the definition of "computer program" as a program limited to causing a device with a digital information processing capability to perform a digital information processing function.

There is a difference between what is involved in making a computer program and what is involved in using a device that incorporates a computer. All that is generated when the user operates the lever-handle of an automatic petrol pump or the buttons of a microwave oven is a transient sequence of electronic signals. This is sufficient to explain why, as Hysport correctly contends, such activities cannot amount to the making of a "computer program" within the meaning of that term in the Copyright Act. But accepting that this is so does not detract in any way from the entitlement of the Coogi XYZ program to recognition as a "computer program": it is important to note that the Coogi XYZ program exists in permanent form, both stored electronic and printed. It is clear from ss 32 and 22 that copyright can only subsist in a work that is "made" by being reduced to some material form. And it is clear from the definitions of "computer program" and "material form" in s 10 of the Act that copyright can only subsist in forms of expression that are fixed in some permanent embodiment. See Autodesk [No 1] at 335 and Lahore para 6,035 and cf Green v Broadcasting Corporation of New Zealand [1989] 2 All ER 1056 at 1058. On the assumption that the activity of manipulating the lever of a computerised petrol pump or pressing the buttons of a microwave oven does nothing more than generate transient electronic signals that do not continue to exist in any permanent form after generation, neither of those activities can constitute "computer programs" within the Act, or involve the making of such a "program" because they are never fixed in a material form. This is so even though the one causes the microprocessor in the pump to do a number of things including regulate the flow of petrol through the pump and the other causes the microprocessor in the oven to regulate the intensity and duration of radiation, among other things. (Whether Hysport's references to the word processor and the flight plan advance its case cannot be evaluated in the absence of evidence explaining how each functions. But if all that is generated when the user strikes the keys of a word processor, with the intention of causing the computer in which a word processing application program is running to produce the desired arrangement of text in stored electronic form, is a transient sequence of signals, the word processor example cannot assist Hysport here and, as to the flight plan, I note that para 20 of the Explanatory Memorandum in respect of the 1984 amending Act envisages that a "computer program" may consist of a handwritten set of instructions intended to be keyed onto magnetic disk and so converted into a machine-readable form of expression.)

Do the Hysport programs infringe Coogi's copyright in its XYZ program?

Hysport's major answer to the charge of infringement of copyright in the Coogi XYZ computer program, however, is that there is no objective similarity between the XYZ and the Hysport programs so Coogi cannot prove an essential element of its infringement claim. Alternatively, it was submitted that if there is such similarity between them, as Professor Goldschlager asserts, and which is revealed by his manipulation of the XYZ program in order to compare it with the Hysport programs, that must be the consequence of such form of expression as is common to both being entirely determined by function. That is, if there is any objective similarity between the forms of expression that constitute the two programs at a particular level of programming language, that is a consequence of it being impossible to write the instructions comprising a program to cause any computer-controlled knitting machine to knit fabric to the XYZ design in any other form because that particular form of expression is wholly dictated by that particular function. Such a program cannot be the subject of copyright for the reasons given in Autodesk [No 1] at 345.

Counsel for Coogi's response was that its case was that the question whether each of Hysport's programs infringes Coogi's copyright in its XYZ program depends upon whether each is an "adaptation" of Coogi's XYZ program, ie, whether in terms of para (ba) of the exclusive definition of "adaptation" in s 10(1) the Copyright Act any of those programs is "a version of the [Coogi program] (whether or not in the language, code or notation in which the [Coogi program] was originally expressed) not being a reproduction of the [Coogi program]". Counsel explained that Coogi's complaint was one of unauthorised "adaptation" and not unauthorised "reproduction" because the Coogi and Hysport programs are in different languages with the consequence that, while the programs contain the same instructions, they take "a different literal form", they "look different". However, counsel did not dispute the need to prove objective similarity between the copyright program and the programs said to be unauthorised adaptations of it. A central part of Coogi's case was the evidence of Professor Goldschlager, which was said to show exactly this objective similarity.

Counsel for Hysport conceded that it may still be possible for Coogi to have copyright in a part of the Coogi XYZ program, ie, in the part not wholly dictated by function which is constituted by the way Walker chose to organise part of the entire program in sub-routines that comprise the control program element of the XYZ program. Reference was made to the dictum in Autodesk [No 1] at 335 - 336. I doubt that the concession is well-founded since copyright can only subsist in that which constitutes an entire work. It seems to me that, so far as the XYZ program is concerned, the entire work in which alone copyright can subsist is the control program (which comprises the sub-routines) together with the graph. So far as the evidence indicates, unless used in conjunction with the graph, the control program is not capable of being put to any use by itself; it was not intended to, and cannot, serve any purpose other than to be used in conjunction with the graph to produce fabric to one particular design. The point, however, is academic since Coogi did not assert a copyright in the control program that was separate from the copyright it claimed in the XYZ program.

As to the first prong of the argument, turning on want of objective similarity, Hysport contended that its validity could be assessed by comparing the programs for XYZ and HY95 (which is a program for a fabric of a structure very similar to the structure of the XYZ fabric) to see if there is objective similarity between the two programs of the kind relevant to the question of copyright infringement. Hysport correctly submitted that "if you do not get infringement where you are producing identical fabrics [with these two programs], you [certainly] do not get infringement where you are producing very different fabrics [with the Coogi program and each of the HY82 and HY96 programs]".

Despite the criticisms that have been made of the accuracy of the proposition, it is settled law in Australia that copyright protects not the underlying idea, but only the form of expression of the idea. See Autodesk [No 1] at 344 - 345 and Powerflex Services at 450. It is also the law in Australia that merely because one program performs the same function as another, that is insufficient to make the one an infringement of copyright in the other: Autodesk [No 1] at 344; Autodesk [No 2] at 304 and Powerflex Services at 455. Consistently with this, a computer program can only be an "adaptation" of another when it is "a version of the work", ie, a version of the particular form of expression of that other in which alone copyright can subsist, and not merely an expression of the idea or function of that other program.

In Powerflex Services, the Full Court placed a precise and limiting gloss on the statutory definition, holding, at 454, that, while the word "version" in the statutory definition of "adaptation" is used in its ordinary English sense, the legislature intended that it should bear one particular meaning of those which it has in ordinary speech, viz, a translation; at 457 it repeated that "the word `adaptation', as used in s 10, means `translation'". In ordinary usage, the term "version" is applicable to a wider range of renderings of the original literary work than is the term "translation": a version can apply to as literal as possible a rendering of the original in another language, as well as to a work in the same or another language that is only recognisable as related to the original because that is the identifiable source of the theme of the version. The term "translation" has a much narrower reach. In ordinary speech, Piave's libretto in Italian for Verdi's opera "Macbeth" could fairly be called a version of Shakespeare's play; it could not, however, be properly described as a translation of the English original. The Full Court supported its conclusion that "adaptation" means "translation" with references to various paragraphs of the Explanatory Memorandum in respect of the amending Act of 1984 which introduced para (ba) into the definition of "adaptation" into the Copyright Act including:

11. Copyright in literary works includes exclusive rights to reproduce or adapt such works and computer programs will be treated as literary works. However, the present definition of adaptation in relation to literary works only includes translation, conversion between dramatic and non-dramatic forms, and conversion to a pictorial form.

12. Of these, only translation is likely to be relevant to adaptation of programs and there are legal doubts as to whether this refers only to translations between human languages.

13. The new definition is intended to cover translation either way between the various so-called `high level programming languages' in which the programs may be written by humans (often called `source code') and languages, codes or notations which actually control computer operations (often called `machine code' or `object code'). Thus `adaptation' is intended, for example to cover the compilation of FORTRAN program to produce machine code which will directly control the operation of a computer. Languages, etc of intermediate level would also be covered.

The Court also referred to paragraph 14 of the Explanatory Memorandum which states that the legislative intent was to include within the concept of a "version" of a computer program that which results from a process such as compilation followed by de-compilation, ie, by translation from a higher to a lower level of computer language and by translation the other way, even though "the differences may be so substantial that one cannot speak of a reproduction although the final product is clearly derived from the original".

The discussion in Powerflex Services about whether the three "macro" commands in the PFXplus program were an unauthorised "adaptation" of the corresponding commands in the DataFlex program is important to a proper understanding of what the Court considered that the concept of "adaptation" in s 10 of the Act comprehends. That litigation arose out of Powerflex's actions in creating an application development system designed to use the same commands, file structure and function keys as the Data Access system, so as to be an attractive alternative for users familiar with the widely popular Data Access system. The Full Court rejected the notion that the macros themselves could constitute computer programs (at 453). But it held that none of the sets of instructions underlying, and activated by, the PFXplus macro commands was an "adaptation" of any of the sets of instructions underlying the DataFlex macros. The trial judge said (in 33 IPR, at 201):

When PFXplus was first published each of the three DataFlex macros was encoded in a program component of DataFlex called FMAC, in an intermediate code called I-Code. The expression of the three commands in PFXplus was in source code. A translation of each command from the DataFlex I-Code instructions in FMAC to source code was compared by a number of expert witnesses with the PFXplus source code instructions. ... The objective similarity of the two versions of source code are in each of the three macros strong, and of course in each case compile to cause the computer to perform the same complex function. And that was the objective Dr Bennett [of Powerflex] sought to achieve by close study of FMAC. In my opinion Dr Bennett made an adaptation of the expression in I-Code of each of the three sets of instructions invoked by the three commands by his expression of that set of instructions in source code.

The Full Court overturned this ruling. It referred to the uncontradicted evidence that, while there was much objective similarity between the source code in which the offending set of instructions were expressed and the source code into which the relevant instructions in the DataFlex program could be translated, Dr Bennett, the author of the PFXplus instructions underlying the PFXplus macro commands, wrote them as an original expression of his own, although before doing so, he had carefully studied the DataFlex program so as to ensure that his commands would perform the same functions as the DataFlex commands. Even though there was ample proof of a causal connection between the offending instructions (which were published in source code) and the DataFlex instructions (which were published in I-Code form) and thus ample proof of a causal connection between the offending instructions in source code and the source code form into which the DataFlex instructions could be converted, the Full Court still concluded (at 454 - 455):

... it is clear that the process involved no translation from one form or language to another, nor did it involve the kind of process referred to in para 14 of the explanatory memorandum involving compilation followed by decompilation, or vice versa. In our view, a process of devising a source code to perform the same function as is performed in some other source code expressed in original language does not involve creating a version of the original source code. It is an original work, albeit that the function to be performed is the same. It is the concentration upon the performance of function rather than the expression of the set of instructions which is the fallacy in the submissions of Data Access. That this is so can be seen from the following passage from the respondents' written submissions, where the reference to `form' is a reference to the higher level code, the I-code form and the compiled code:

In each form, the same set of instructions exists but the expression of that set of instructions differs. (emphasis added)

Dr Bennett, the author of the PFXplus macro-activated instructions or programs, had access to the corresponding DataFlex programs; he closely studied them and wrote his own programs to ensure they would perform the same function as the DataFlex programs. Moreover, the particular form (in I-Code) of the DataFlex macro activated program which Dr Bennett made use of and in which Data Access claimed copyright could be translated into another form of expression (ie, source code) and there was a high degree of objective similarity between that other form of expression and the form in which the PFXplus macro-activated program had been written. Yet none of that was sufficient to require the conclusion that either of the PFXplus programs was an adaptation of the corresponding DataFlex program. Neither of these PFXplus programs was a translation, in the sense in which the Full Court used that term, and so neither was an "adaptation" of the corresponding DataFlex programs because, while Dr Bennett of Powerflex deliberately took the ideas expressed in the two DataFlex programs, he had not attempted to convert, in whole or part, the expressions of those ideas in the language in which the DataFlex programs were written into corresponding expressions in another language. Instead, he expressed the ideas in the DataFlex programs, which he needed to take to achieve his purpose, in a form of expression that he produced by the use of his own skill and effort.

Translation, as that term is ordinarily used, connotes more than conveying in a different language the same ideas expressed in another language. It describes a closer connection between the original and the translated text than that. I do not think an activity could be described in ordinary speech as a translation unless it involves the expenditure of effort on the original words or text to render them into words or text in a different language that conveys with precision the same meaning as that conveyed by the original. A person can probably be said to make a translation, in the ordinary meaning of that term, indirectly, eg, by translating into English from a French rendering of the original German text. But there can be no translation without the expenditure of effort, directly or indirectly, on the original text. Further, translation into a second language of instructions for achieving a result expressed in another language involves more than describing in the second language some method for achieving that same result: to be a translation, as that term is ordinarily used, of the original instructions, the new text would have to describe the particular sequence of steps for achieving the result that are expressed in the original text. Assume A writes in German a set of instructions for the making of a device, eg, a clock, and that someone uses those instructions to make a clock. B does not make a translation of A's original instructions by taking the clock, breaking it down into its parts and then writing his own set of instructions in English for reproducing the clock. The reason is that he has not produced his manual of instructions in English by working directly or indirectly (via a translation of the original in a third language) from the original text in German.

It is, I think, in this narrow sense which the word "translation" has in ordinary speech that the Full Court in Powerflex Services considered that the definition of "adaptation" applicable to computer programs should be understood. One computer program will therefore be an unauthorised "adaptation" of another firstly, only if the whole or a substantial part of the particular form of expression of the program in which copyright is claimed appears in the allegedly infringing program in a different computer language, either at the same language level or at a different language level and, secondly, only if the allegedly infringing program has been produced by direct or indirect use, in the sense described, of the copyright program. It is a meaning of "adaptation" that is supported by paras 13 and 14 of the Explanatory Memorandum and one which preserves, so far as concerns copyright in computer programs, the function/expression dichotomy that is currently entrenched in Australian copyright law.

The correctness of the proposition central to the respondents' case that Hysport's programs could only be infringements of Coogi's program if there was objective similarity between them can now be considered. That question will necessarily have to be determined when the allegation is that one program is an unauthorised reproduction of the other: Computer Edge Pty Ltd v Apple Computer Inc [1986] HCA 19; (1986) 161 CLR 171 at 186. But in view of the definition of "adaptation" in the Copyright Act, a reproduction of a computer program and an adaptation (ie, a translation) of such a program are mutually exclusive concepts. A translation of a literary work into a different language from the original will necessarily be in a form of expression different from the original. The making of a translation also necessarily involves the expenditure of original skill and effort by the translator: the making of a translation involves much more than the mere taking of the original author's own skill and effort. It was this consideration, coupled with the perceived need to respect the idea/expression dichotomy, that led courts, until the matter was dealt with by legislation, to hold that a translation of an original literary work into another language did not infringe copyright in the original: see, eg, the discussion in the old American case of Stowe v Thomas (1853) 23 F Cas 201. It is clear from paras 13 and 14 of the Explanatory Memorandum in respect of the amending Act of 1984 and from the Full Court decision in Powerflex Services, that para (ba) of the definition of the expression "adaptation" in s 10 of the Act was intended to cover computer programs derived from the original program, but expressed in quite different forms from the original, ie, even though there is no objective similarity between the forms in which they are expressed. It is therefore difficult to see how, as the respondents suggest and the applicants accept, that there can be any room for a requirement, where the allegation is that one computer program infringes copyright in another because it is an adaptation, ie, a translation, of that other, that there be objective similarity between the two forms of expression, if the first is to be an infringement of copyright in the second.

I consider that a computer program will be an infringing "adaptation", ie, translation, of an original program only if the Court can be convinced that, although the two programs are quite different so far as their forms of expression are concerned, use has been made of the original not just to identify the idea of the original or the function it performs, but as an aid to devising the descriptions of the activities (and any accompanying data) which together comprise the allegedly infringing program. Unless the expression "objective similarity" as used in copyright infringement law is to be given a new, highly artificial and technical meaning, instead of the ordinary usage meaning it has hitherto had, so that it will import into infringement law tests like that adopted in the American case, Computer Associates International Inc v Altai Inc (1992) 23 IPR 385 and favoured by the Copyright Law Review Committee Report (Canberra, 1995) para 9.27, it cannot sensibly be employed, in my opinion, in order to determine whether one program is an unauthorised adaptation of another.

In my opinion, the F823 program is not an adaptation, in the sense of a translation, of the particular form of expression of instructions that constitutes the Coogi XYZ program. I have explained why I am not prepared to find that when Shima Seiki produced its F823 program which Hysport used without modification to produce its HY95 fabric, Shima Seiki had any access to Coogi's XYZ program. It is, in my opinion, a misuse of language to say that, by examining a Coogi XYZ garment to identify how the fabric was produced and by then writing a program of its own to cause its computerised knitting machine to perform the sequence of steps necessary to produce a like fabric, Shima Seiki made an adaptation, ie, a translation of the particular form of expression constituting the Coogi XYZ program, which alone can be the subject matter of Coogi's program copyright. All it seems to me that Shima did, and all it could do without access to an XYZ program, was reproduce, in the program which it wrote in its own Shimatronic language, the same function that Coogi's XYZ program written in Sintral language performed, viz, the function of causing an appropriately computerised knitting machine to produce fabric of one particular design. Copyright in a computer program protects that skill and effort of the author that is expended in devising only the form in which the program is expressed: I am not prepared to find that Shima in producing its F823 program took any of that skill or effort. Hysport's programs derived from the Shima F823 program cannot therefore be infringements of the Coogi program.

I would therefore dismiss Coogi's claim against all respondents for this reason.

Autodesk [No 1] shows that you can infringe copyright in a computer program by making a reproduction of the program (or a substantial part of it) by copying it indirectly. But what occurred here is materially different from what occurred in Autodesk [No 1]. There, Dyason did not copy directly the form in which the relevant part of Autodesk's widgetC program was expressed; instead, he copied that particular form of expression indirectly by retrieving an exact replication of it from the AutoCAD lock. Dyason expended no skill or effort of his own in devising his infringing work, only skill and effort in identifying the embodiment of Autodesk's own skill and effort in the relevant part of its program, which he then took for his own work. Here, on the view I have taken of the evidence, Shima Seiki, and thus Hysport, never had access to any literary form in which the XYZ computer program was expressed, only to a garment manufactured by running that program in the computer control of a knitting machine. They could never therefore be in a position to make a translation, directly or indirectly, of the XYZ program. Instead, the F823 program and the Hysport programs based on it were the product of the extensive skill and effort of their own which they put into writing those programs.

In my opinion, it will never be possible to make out a case that one program constitutes an infringement of another computer program where the purpose of the original is to control the manufacture of an object and the alleged infringer has produced its own computer program to enable it to manufacture a like object by reverse engineering the original object, ie, by analysing it to see how it has been constructed and then by writing its own program to identify the steps that have to be gone through to make the object. I reject Coogi's submission to the contrary.

Given my conclusion that Hysport produced its HY95 fabric from the Shima F823 program which was itself produced by the reverse engineering of a sample of XYZ fabric, rather than by making use of the XYZ program itself, the letter of 22 April 1993 that Mr Jones wrote, in an attempt to settle Hysport's dispute with Coogi, and the undertakings that Hysport gave in the course of these proceedings cannot do the work that Coogi relies on them to do. They cannot provide proof by way of admissions that the F823 program is an "adaptation" of the XYZ program. Mr Jones, in the letter, explained why he was prepared to admit that the two fabrics were produced from the same program: that was the inference he drew from his examination of the two garments and their similarities. For the reasons given, he was wrong in drawing that inference. His admission, so based, is therefore not probative of the critical issue: was the F823 program produced by making use of the XYZ program. See Horne v Comino [1966] Qd R 202 at 207 - 208. There is no ground in the evidence for thinking that the undertakings were based on anything other than the same opinion that caused Mr Jones to write the letter.

However, if I am wrong in reaching my conclusion, I am of the opinion that, if it is a consideration relevant to infringement by adaptation, Coogi has failed to prove objective similarity between its program and any of the programs Hysport used to produce its own fabrics. I do not accept that Professor Goldschlager's evidence provides any foundation for a contrary conclusion. (Professor Goldschlager's evidence is really directed to what became, in the course of the trial, an irrelevant issue: he prepared his report to assist Coogi in proving program to program copying. The extensive similarities he identified would be relevant to that issue. But during the trial, it was revealed that the Shima F823 program (and thus the Hysport programs derived from it) was developed not by program to program copying, but from an XYZ garment, ie, in a way that would still ensure that there would exist many similarities between the XYZ program and the Hysport programs, at least at the levels expressing needle positions.

However, on the assumption that proof of program similarity remains relevant, Professor Goldschlager's evidence requires consideration. I have set out above his description of the content of the Coogi and Hysport programs as specifications of needle positions. Professor Goldschlager said that the various Hysport programs were mere mechanical unravellings of Coogi's XYZ program; his evidence was that the similarities between "the Coogi graph and the Hysport graph", ie, between "the patterns created directly by the Coogi control program compared with patterns created by the Hysport graph" (and its accompanying control program) are so remarkable as to show substantial copying. He reached this conclusion by first identifying similarities between portions of the XYZ program contained in the graph and portions of the Hysport graphs in, among others, the F823, HY96, HY95 and HY82 programs: he rewrote the printout version of the Coogi graph by replacing the symbols in the Coogi graph with the symbols used in the printout versions of the Hysport graphs. Many areas of close similarity were then obvious. He next identified those parts of the Coogi program contained in sub-routines in the control program element of Coogi's program that were similar to parts of the Hysport programs: he did this by expanding the particular Coogi sub-routine to state, in sequence, each individual needle placement operation which was condensed in the brief instructions comprising the sub-routine and by comparing the expanded expression of the sub-routine with the relevant Hysport graph to identify similarities. Finally, he completed his comparison of the Coogi and Hysport programs by comparing those parts of the Coogi and Hysport programs which controlled the sequence of stripes that make up what he called the overall structure of the fabrics.

His conclusion was that the similarities were so extensive as to preclude independent development.

Mr Walker gave Professor Goldschlager a number of programs on disk relating to various of the Hysport fabric styles, which he obtained on discovery in the action from Hysport. But Mr Walker confirmed that each was in a format that resulted from the processing of the original programs produced by Shima Seiki and Hysport into a format that could be run directly in a Shima knitting machine, ie, in the computer embedded in that machine which controlled the knitting out by the machine of the relevant fabric. Mr Walker also obtained on discovery disks containing higher level versions of the Hysport programs, including HY95, HY82 and HY96, ie, in the forms in which they were written by their author. But he did not give those disks to Professor Goldschlager.

It is apparent from Mr Walker's evidence that what Professor Goldschlager compared was his manipulation of the high level version of the Coogi XYZ program with the versions of the F823 program and the other Hysport programs produced by processing each from the form in which the author wrote it into either a machine readable form or a form in a lower level language that runs in the Shima knitting machine to produce the Hysport fabric.

Even if each of the forms of the F823 and Hysport programs given to Professor Goldschlager was in a low level but not actually machine readable language, that form was one which largely obliterated the forms in which the F823 and Hysport programs were originally written in the high level Shimatronic programming language, forms of expression that reflected the choices made by the Shima Seiki programmer and Mr Jones from the wide range of forms of expression available in Shimatronic when they produced these programs. Each of the forms in which these programs were given to Professor Goldschlager was a form which was therefore, in large part, an expression of the intended function of the particular program, viz, to cause a computer-controlled Shima knitting machine to knit fabric to a particular design. Professor Goldschlager's manipulation of the XYZ program into a form in which it could be compared with the Hysport material he was given by expanding the instructions in the XYZ control program and graph into an expression of the sequence of the needle positions necessary to produce the XYZ fabric also obliterated, in large part, the original form of that program which similarly reflected the choice made by Mr Walker of a particular form of expression from the wide range of possible forms in which he could have expressed his programming intentions in the high level Sintral programming language.

It follows from this that, since both the XYZ fabric produced by running the Coogi XYZ program in the Stoll machine essentially has the same structure as the HY95 fabric produced by running the F823 program in the Shima machine, Professor Goldschlager's exercise will necessarily throw up very extensive similarities between the manipulated form of the XYZ program and the low level version of the F823 program he compared since both programs perform much the same function. Where Professor Goldschlager compared the Hysport programs for HY82 and HY96, he also based his comparison with the manipulated Coogi XYZ program on processed, ie, low level, versions of those two Hysport programs: given the similarities between these fabrics and the XYZ fabric, his comparison of the program material will necessarily also identify similarities between the Coogi and Hysport material at the particular levels at which he made it.

The deficiency in Professor Goldschlager's evidence of similarity was most clearly exposed in cross-examination about the parts of the Coogi program and of the Hysport H82 program that govern the twenty-sixth stripe in the XYZ fabric and the corresponding stripe in the HY82 fabric. The effect of his evidence was that, while the forms of the XYZ program and the HY82 program here were expressed quite differently, the XYZ program could be expanded into a form expressing sequence of needle positions which was very similar to the form in which the relevant part of the HY82 program, also stating needle position sequences, was expressed. But the XYZ program was not written by Mr Walker, and did not exist in that expanded form of expression, until the program was being executed in the knitting machine computer, ie, until the program was performing its intended function. Although Professor Goldschlager denied that the comparison he made was of program output with program output, he went on to give this evidence when being questioned about a particular sub-routine in the Coogi program and the part of the Hysport graph that corresponded to it:

Well, what I suggest to you is that while it may be that they produce the same output the form is quite different. The form in the case of Coogi starts "F Beg Human" and then there are open brackets and s's and r's and all that sort of thing. And that the Hysport graph does not look at all like that?---Yes. I think the mistake you are making is you are looking at it at the wrong level of comparison. You are looking at it, you know, these two things are written in two different languages. So it's like taking a French translation, say, of an English book and saying: Well, look, the first letter here is different to the first letter there, therefore, the books are different. These are two different languages so naturally on the surface, if you look at it character by character, they're different because it's two different languages. But if you try to understand what the computer program is doing and what is inside the computer program you can see that, for example, in line 1525 you have got the dot eye there which is repeated. And, basically, what that part of the program does - basically look at, say, lines 1520 through to 1535.

Yes?---That part of the program takes the dot eye and replicates it across the row and also replicates the rows vertically. So you have got the fundamental pattern which is dot eye actually spelt out explicitly in the program.

And what that means is that the program will give the same output as the Hysport graph, is not that right?---That - that's not what I'm referring to, no. What I'm referring to is that - - -

Well, when you say dot eye is replicated, where, physically, is dot eye replicated in the Coogi system do you say?---Well, in the Coogi system the particular lines that I have referred to which are, say, 1520 to 1535, are instructions to a computer to say: Take this dot eye pattern and replicate it. That's what the instructions say.

...

But until the program is actually executed you never have in the computer dot eye replicated across the entire row, do you?---That's correct. (Emphasis added)

The Professor here acknowledges that he did in truth compare the two programs at the level of functioning: he identified what the form in which the XYZ program was expressed in Sintral would cause the computer controlled knitting machine in which it was run to do and what the form of the Hysport program that he was given would also cause a knitting machine to do; he then compared those two outcomes. Since the function of both was to produce fabric of similar design, it is inevitable that there will be, as Professor Goldschlager found, many similarities between the two programs at that level of expression.

I therefore accept Hysport's submission that Professor Goldschlager, in effect, made a comparison of the two lots of program at a level where the forms in which they were expressed were dictated by the common function each served, ie, the function of causing a knitting machine to produce fabric of similar structure. The high degree of similarity between the versions of the XYZ and Hysport programs at the levels he made his comparison is therefore inevitable and not probative of program copying or other impermissible use by Shima Seiki or Mr Jones of the XYZ program. Professor Goldschlager's comparison cannot provide any evidence to justify a finding that the F823 program (or any of the other Hysport programs) in a version in a higher level of language than the low level version he was given is an infringement of the copyright Coogi has in its XYZ program. Cf Autodesk [No 1] at 345.

On any view, Coogi's claim of computer program infringement must fail.

It follows that the first to sixth respondents are entitled to relief claimed under s 202 of the Copyright Act. The threats the subject of the cross-claim being proved, the persuasive onus is on Coogi to justify those threats: see Lewis Falk Ltd v Jacobwitz (1944) 61 RPC 116 at 118. By failing to prove infringement of its program copyright, Coogi has failed to discharge this onus of proof. It is no defence for Coogi to show, as it has, that it made the threats complained of in the bona fide belief that the respondents' conduct constituted infringements of its rights. See SW Hart & Co Pty Ltd v Edwards Hot Water Systems (1980) 30 ALR 657 at 661; Lahore, Patents, Trade Marks and Related Rights, para 18,070. What Dawson J said in Avel Pty Ltd v Multicoin Amusements Pty Ltd [1990] HCA 58; (1990) 18 IPR 443 at 454 does not constitute a departure from this long established rule, as reference to the English case cited by his Honour, Mechanical Services (Trailer Engineers) Ltd v Avon Rubber Co Ltd [1977] RPC 66 at 74 - 75 shows: it is there made clear that the only effective answer to an action in respect of unjustified threats of infringement is proof that the making of the threats was justified, ie, that the activities that attracted the threat constitute or, if done, would constitute infringing conduct. Nothing was there said to suggest that a bona fide but unjustified belief can be a defence to such an action or that the mere bringing of an infringement action, irrespective of its outcome, is a good answer to a groundless threats action. (The latter was once a good defence to a threats action under earlier forms of legislation corresponding broadly to s 202, but it is not a defence under provisions that are in the form of the current s 202: see Pearson v Holden (1948) 65 RPC 424 at 427.)

The first to sixth respondents are entitled to the declaration they claim in paragraph A of the prayer in the cross-claim. The evidence also indicates that they withdrew their HY95-style garments from sale in consequence of the applicant's threat to bring infringement proceedings. I accept that the respondents would have suffered some loss in connection with withdrawing those garments, which may include wasted costs of production and distribution, as well as the costs of withdrawal of those garments, together with profits lost on them and also gross profits that were foregone as a result of stopping production of that particular style. No other head of loss appears from the evidence. There should be an inquiry therefore as to the damages any of these respondents may have so suffered.

I do not think that, in order to vindicate the rights of these respondents, it is, on the evidence before me, necessary to grant any injunctive relief in respect of their s 202 claim. Coogi in all probability has copyright in its XYZ fabric and program. The threats of infringement that Coogi made were all made bona fide to protect those rights. Coogi has failed to show that Hysport has infringed its rights. But there is no reason to think that Coogi will make claims of infringement in the future which it now knows are without any foundation. Hysport's position is vindicated by this judgment. Moreover, there are, I think, real difficulties in devising the terms of the limited injunction to which Hysport would be entitled, if it were otherwise appropriate to grant it injunctive relief, given that Coogi does have copyright in respect of its XYZ fabric and in its XYZ program.

The respondents also cross-claim for damages and injunctive relief as a result of the representations made by Coogi in respect of its claimed ownership of copyright relating to the XYZ design. This cross-claim must be dismissed. Although Coogi has failed to establish any copyright infringement by the respondents, the core representation in this cause of action is neither misleading nor deceptive: Coogi has copyright in certain artistic works, viz, the first run of XYZ design fabric in each of the eight colourways it used and in its XYZ computer program.

In view of Coogi's failure to prove that the articles sold by the seventh respondent were infringing copies of artistic works in which Coogi has copyright, its action against the seventh respondent must also be dismissed.

The respondents must bring into Court minutes of the orders to which they are entitled, in accordance with these reasons, once the question of costs has been dealt with: I will receive written submissions on that.

I certify that this and the preceding fifty-two (52) pages are a true copy of the Reasons for Judgment herein of the Honourable Justice Drummond.

Associate:

Dated: 21 August 1998

Counsel for the Applicant:	JWK Burnside QC and CD Galvan
Solicitors for the Applicant:	Peter Speakman & Co
Counsel for the First to Sixth Respondents:	Dr JM Emmerson QC and AK Panna
Solicitors for the First to Sixth Respondents:	White Cleland Pty
Counsel for the Seventh Respondent	IA Phillips
Solicitors for the Seventh Respondent	Purves Clarke Richards
Date of Hearing:	17 February 1997 - 7 March 1997; further written submissions 26 April 1998, 9 and 19 June 1998
Date of Judgment:	21 August 1998

IN THE FEDERAL COURT OF AUSTRALIA
victoria districT REGISTRY	VG 284 of 1993

BETWEEN:	COOGI AUSTRALIA PTY LTD Applicant
AND:	HYSPORT INTERNATIONAL PTY LTD First Respondent GEOFFREY GRAY JONES Second Respondent JOAN MARGARET JONES Third Respondent TERENCE GRAY JONES Fourth Respondent HARRY GEOFFREY BROWN Fifth Respondent JENNIFER ANNE BROWN Sixth Respondent BROMSGROVE PTY LTD Seventh Respondent

CORRIGENDUM

Amendment to the Reasons for Judgment of the Honourable Justice Drummond delivered 21 August 1998:

Page 8, line 4, replace "Ramsey" with "Ramsay".

ASSOCIATE TO JUSTICE DRUMMOND

7 October 1998