Apple Computer Inc v Computer Edge Pty Ltd

Re: APPLE COMPUTER INC. and APPLE COMPUTER AUSTRALIA PTY. LTD.
And: COMPUTER EDGE PTY. LIMITED and MICHAEL SUSS
No. G 405 of 1983
Copyright
(1984) AIPC para 90 - 132 / (1984) ATPR para 40 - 453 / 53 ALR 225
2 IPR 1 / 1 FCR 549

COURT

IN THE FEDERAL COURT OF AUSTRALIA

NEW SOUTH WALES DISTRICT REGISTRY
GENERAL DIVISION
Fox(1), Lockhart(2) and Sheppard(3)

CATCHWORDS

Copyright - Computer programmes - Original literary works - Adaptations - Infringement by sale and importation of reproductions of adaptations.

Trade Practices - Misleading or deceptive conduct - Computers of similar appearance - Labels - Respondents' goods supplied with appellants' handbook - False representation of sponsorship, approval or affiliation.

Trade Practices Act 1974 - ss. 52, 53(c), 53(d), 55.

Copyright Act 1968 - ss. 10(1), 36, 37, 38

Copyright - Nature and subject matter - Computer programmes - Programmes in source code and object code - Original literary works - Meaning of "literary" work - Adaptation of original literary work - Definition of "adaptation" - Adaptation not visible to human eye - Importation and sale of microcomputers containing reproductions of adaptations - Infringement of copyright - Copyright Act 1968 (Cth), ss 10(1), 31, 37 and 38.

Copyright - Infringement - Computer programmes - Adaptation of original literary work - Reproduction of adaptations in microcomputers imported for sale - Comparison of codes in adaptation and reproduction - Proof of knowledge - Right to injunction - Copyright Act 1969 (Cth), ss 37 and 38.

Trade Practices - Misleading or deceptive conduct - Physical similarity - Labelling - Promotion of products as "compatible" - Provision of competitor's operating manual - Trade Practices Act 1974 (Cth), ss 52, 53(c), 53(d).

HEADNOTE

The first appellant makes and sells "Apple II" microcomputers which the second appellant distributes in Australia. An employee of the first appellant wrote the source code for its computer programmes including "Applesoft" and "Autostart", the two programmes in issue in the proceedings. The source code was then assembled by a computer into machine readable code and permanently stored in ROMs of the Apple II.

The respondents imported from Taiwan a microcomputer known as "Wombat" which contained ROMs and an EPROM which had been copied from the Apple II ROMs. Both microcomputers were similar in physical appearance. Although distinctly labelled, the Wombat was promoted as compatible with the Apple II and Wombat computers were sold with the Apple manuals. The appellants alleged breaches of the Copyright Act 1968 (Cth), ss 37 and 38 and the Trade Practices Act 1974 (Cth), ss 52 and 53.

Held: (1) A literary work is not limited to something which is intended to afford either information or instruction or pleasure in the form of literary enjoyment. There is no necessity for a literary work to be of any literary quality but it needs to be a "work", to have had some skill, even if very small, applied to its preparation.

Hollinrake v. Truswell (1894) 3 Ch. 420 at 428; Exxon Corporation v. Exxon Insurance Consultants International Ltd (1981) 3 W.L.R. 541 at 551-552, explained.

(2) The computer programmes in source code are within the concept of a literary work for the purposes of the Copyright Act 1968 (Cth).

(3) (Sheppard J. dissenting). The machine readable codes in the Apple II ROMs are adaptations, (as translations) within the meaning of s. 31(1)(a)(vi) and s. 10(1) of the Act, of the original literary works constituted by the "Applesoft" and "Autostart" programmes in source code.

(Per Fox and Lockhart JJ.)

(4) The resemblance to be considered for the purposes of infringement of the provisions of the Copyright Act 1968 (Cth) is that between the code embodied in the adaptation and that in the alleged reproduction and it is not as between the materials in which they are embodied.

(5) Upon that comparison, the Wombat ROMs and EPROM contained reproductions of the Apple machine readable codes within the meaning of ss 31(1)(a)(i) and 31(1)(a)(vii).

(6) There is no infringement of ss 37 and 38 of the Act unless knowledge is shown. The knowledge required is actual though it may be inferred from the circumstances.

Albert v. S. Hoffnung and Co. Ltd (1921) 22 S.R. (N.S.W.) 75 at 81; R. C. A. Corporation v. Custom Cleared Sales Pty Ltd (1978) 19 A.L.R. 123; Sillitoe v. McGraw-Hill Book Co. (U.K.) Ltd (The Times, 18 January 1982), followed.

(7) At least from the commencement of the proceedings the respondent had knowledge and since there is a continued threat to the rights of the appellants an injunction quia timet ought to issue.

Trade Practices Act

(Per Lockhart and Sheppard JJ.)

(8) The decisive labelling of the two computers ensured that despite a similar physical appearance and shape there was no possibility of a purchaser being misled or deceived within the provisions of the Trade Practices Act 1974 (Cth) s. 52.

Parkdale Custom Built Furniture Pty Ltd v. Puxu Pty Ltd (1982) 56 A.L.J.R. 715, followed.

(9) The word "compatible" suggests a difference in trade source rather than a common source therefore its use by the respondent was not likely to mislead or deceive within s. 52.

(10) The supply of Apple manuals with the sale of Wombat computers was misleading and deceptive conduct within s. 52 and constituted an untrue representation that the Wombat computer is approved by the manufacturer or supplier of the Apple II computer within s. 53(c) and an untrue representation that the first respondent is approved by, or has affiliation with, the manufacturer or supplier of the Apple II computer within s. 53(d).

HEARING

Sydney, 1984, February 20-24, 27, 28; May 29. #DATE 29:5:1984

APPEAL.

Appeal from the decision of Beaumont J. dismissing the appellant's claim for injunctions and damages for infringement of its copyright in computer programmes pursuant to the Copyright Act 1968 (Cth), ss 37 and 38 and for breaches of the Trade Practices Act 1974 (Cth), ss 52, 53(c), 53(d) and 55.

M. McHugh Q.C. and D. Catterns, for the appellants.

J. McL. Emmerson and N. Burns, for the respondents.

Solicitors for the appellants: Abbott Tout Creer & Wilkinson.

Solicitors for the respondents: Herbert Geer & Rundle.

B.A.G.

ORDER

The appeal be allowed and the judgment of Beaumont J. set aside.
The first respondent by itself its servants and agents and the second respondent be restrained from infringing the first appellant's copyright in Applesoft Source and Autostart Source programmes.

3.(A) The first respondent by itself its servants and agents and the second respondent be restrained from, without the licence of the first appellant, importing into Australia ROMs or EPROMs or other firmware or software containing programmes which are reproductions of the first appellant's programmes, Applesoft Object and Autostart Object, for the purpose of:-

(a) selling, letting for hire, or by way of trade offering or exposing for sale or hire, the said articles; or

(b) distributing the said articles for the purpose of trade or for any other purpose to an extent that will affect prejudicially the first appellant or by way of trade exhibiting the said articles in public.

(B) The first respondent by itself its servants and agents and the second respondent be restrained from, without the licence of the first appellant,:-

(a) selling, letting for hire, or by way of trade offering or exposing for sale or hire, the said articles; or

(b) by way of trade exhibiting the said articles in public.

If the appellants wish to pursue claims for damages in relation to infringements of s. 37 or s. 38 of the Copyright Act 1968 the subject of this proceeding the appellants shall file and serve upon the respondents or their solicitors within 21 days of today written notice to that effect WHEREUPON without further order the matter shall be referred to Beaumont J. for findings on the evidence as it presently stands in respect of the knowledge, if any, of the first respondent, which includes so far as relevant the knowledge, if any, of the second respondent, of the matters mentioned in s. 37 or s. 38 of the Copyright Act 1968 and in issue in this proceeding AND, if his Honour's findings are favourable to the appellants, for the assessment of damages in relation to the infringements aforesaid.
The appellants bring in short minutes of orders to give effect to the Court's findings on the issues relating to the Trade Practices Act 1974 and the matter is adjourned to a date to be fixed for this pupose.
The respondents shall pay the costs of the appellants of this appeal and of the proceedings before Beaumont J. The costs of any further proceedings before Beaumont J. pursuant to any of the orders aforesaid shall be as ordered by his Honour.
Liberty to apply is reserved generally to any party on seven days' notice and, in particular, if any variation of or addition to orders 2, 3 or 4 above is sought.

Orders accordingly.

JUDGE1

This appeal from a judge of the Court relates to computers and computer programmes in respect of which the appellants claim rights. They allege breaches of ss.52. 53(c). 53(d) and 55 of the Trade Practices Act 1974, and infringement of copyright. The first appellant, a Californian corporation, makes and sells micro-computers under the name "Apple", the relevant computer in this case being known as Apple II. The second appellant is the Australian distributor of these computers. The first respondent, a Victorian company, has imported into and sold in Australia under the name "Wombat", a micro-computer manufactured, or assembled, in Taiwan. The second respondent is the managing director of the first respondent. For convenience, I will refer to both those respondents collectively, without distinction between them, as "the respondents". There were two other respondents to the proceedings as commenced, but the cases as against these have been disposed of, and they are no longer parties. It is claimed that computer programmes installed in the Wombat are unauthorized copies of those of the first appellant, and complaints are made respecting the equipment and the way it is represented to the public. A cross-claim was filed by the respondents claiming relief in respect of claims in circular letters sent by the appellants, but the learned judge, in the belief that the matter was not a live issue, did not make a finding in respect of it, and stood it over generally. Argument concerning the cross-claim has not been presented to us.

In relation to what he called the primary case under s.52, the learned trial judge found that there was no likelihood that a purchaser would be misled or deceived into believing that the respondents or their goods were in any way associated with the appellants. He said:

" . . . on the contrary, the respective products are clearly distinguished from one another by the use of very different brand names and the reference, in advertising, to the 'compatibility' of Wombat software with programmes for the Apple II computer only serves to reinforce the distinction."

He said that for similar reasons the cases under ss.53(c), 53(d) and 55 should fail.

The secondary claim under s.52 related to three silicon chips sold by the first respondent in its Wombat computer which contain computer programmes. The appellants claimed these to be "infringing copies" of programmes of the first appellant within the meaning of the definition in s.10(1) of the Copyright Act 1968 ("the Act"). In that event it was submitted that the respondents misled or deceived purchasers of Wombat computers into thinking that they would take title and enjoy possession of the machines free of any claim such as that in conversion which the first appellant here asserts.
The Claim in Copyright

The appellants further sought injunctions and damages for infringement of copyright. These aspects of the case depended upon copyright being proved in the relevant programmes of the first appellant.

His Honour held that there was no copyright in the first appellant's programmes, because they were not literary works. The essence of his reasons is contained in the following passage:

"In my opinion, none of the programmes are literary works within the meaing of the statute. In my view, a literary work for this purpose is something which was intended to afford 'either information or instruction or pleasure in the form of literary enjoyment' (see Hollingrake v. Truswell (1894) 3 ch.420; Exxon Corporation v. Exxon Insurance Consultants International Ltd. (1982) R.P.C. 69 at p.88; cf. D.P. Anderson & Co. Ltd. v. Lieber Code Company (1917) 2 K.B. 469; Mirror Newspapers Ltd. v. Queensland Newspapers Ltd. (1982) 59 F.L.R. 71; Northern Office Micro Computers (Pty.) Ltd. v. Rosenstein (1981) (4) S.A. 123(c); Sega Enterprises Ltd. v. Richards (1983) F.S.R. 73 at pp.74-5). The function of a computer programme is to control the sequence of operations carried out by a computer. In this sense, as Dr. Emmerson submitted on behalf of the respondents, a contrast may properly be drawn between something which is merely intended to assist the functioning of a mechanical device and literary work so called. The position is even stronger in the case of the object programme, as Rosenstein, supra, recognises: this type of programme as Dr. Emmerson submitted, is at a more advanced stage of the process of controlling the sequence of operations carried out by a computer."

It is convenient to examine first the claim to copyright. The parties very helpfully presented to his Honour a statement of agreed facts, dealing with both the trade practices and copyright cases. I set them out as a schedule to these reasons. Paragraphs 17-32 and 40-41 are those more directly concerned with the copyright claim. There was also a large body of oral and other evidence. While this evidence could not be used to contradict the agreed facts, it was available both to help in an understanding of them, and to supplement them.

The appellants claim that the computer programmes in question are original literary works in which copyright subsists by virtue of s.32(1) and s.32(2) of the Act. The works relied upon were written by United States citizens and were first made and published there. Regulation 4 of the Copyright (International Protection) Regulations which were made under s.184 (and s.248) of the Act is relied upon to secure the same protection here as if the works were made and first published here. No question has been raised concerning publication. The programmes are pleaded as being the original "source" codes and the original "object" codes, but, in the alternative, the latter are pleaded as adaptations of the source codes.

It is alleged that the respondents' conduct amounted to infringement of the first appellant's copyright. Sections 37 and 38 of the Act were relied upon by the appellants at the trial, but on this appeal it was sought also to place reliance on s.36. I shall return to discuss this last-mentioned application. The sections referred to are as follows:

"36.(1) Subject to this Act, the copyright in a literary, dramatic, musical or artistic work is infringed by a person who, not being the owner of the copyright, and without the licence of the owner of the copyright, does in Australia, or authorizes the doing in Australia of, any act comprised in the copyright.

(2) The next three succeeding sections do not affect the generality of the last preceding sub-section.

. . .

37. The copyright in a literary, dramatic, musical or artistic work is infringed by a person who, without the licence of the owner of the copyright, imports an article into Australia for the purpose of -

(a) selling, letting for hire, or by way of trade offering or exposing for sale or hire, the article;

(b) distributing the article -

(i) for the purpose of trade; or

(ii) for any other purpose to an extent that will affect prejudicially the owner of the copyright; or

(c) by way of trade exhibiting the article in public.

where, to his knowledge, the making of the article would, if the article had been made in Australia by the importer, have constituted an infringement of the copyright."

. . .

"38.(1) The copyright in a literary, dramatic, musical or artistic work is infringed by a person who, in Australia, and without the licence of the owner of the copyright -

(a) sells, lets for hire, or by way of trade offers or exposes for sale or hire, an article, or

(b) by way of trade exhibits an article in public.

where, to his knowledge, the making of the article constituted an infringement of the copyright or, in the case of an imported article, would, if the article had been made in Australia by the importer, have constituted such an infringement.

(2) For the purposes of the last preceding sub-section, the distribution of any articles -

(a) for the purpose of trade; or

(b) for any other purpose to an extent that affects prejudicially the owner of the copyright concerned.

shall be taken to be the sale of those articles."

The first of the sections (s.36) is commonly referred to as involving "direct" infringement, the other two as involving "indirect" infringement.

The terms "literary work" and "adaptation" are defined in s.10(1) as follows:

"10.(1) . . .

'literary work' includes a written table or compilation;"

"'adaptation' means -

(a) in relation to a literary work in a non-dramatic form - a version of the work (whether in its original language or in a different language) in a dramatic form;

(b) in relation to a literary work in a dramatic form - a version of the work (whether in its original language or in a different language) in a non-dramatic form;

(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) a version of the work in which a story or action is conveyed solely or principally by means of pictures; and

(d) in relation to a musical work - an arrangement or transcription of the work;".

It is to be noted that the definition of "literary work" is not exhaustive. It may well take account of modern means of communication and of recording information, which have moved so much (and so rapidly) into the electronic field.

A typical Apple II computer, when set up for use, is seen to comprise the following main parts:

(a) the micro-processor, roughly the shape, dimensions and appearance of a portable typewriter;

(b) a keyboard, similar to that of a typewriter, but with some adaptions, which is at the front of, and integral with, the micro-processor;

(c) a disc drive unit, which is smaller and separate, has a slit to accommodate magnetic ("floppy") discs, and can be connected to the micro-processor by cable;

(d) a visual display unit or "monitor"; and

(e) sometimes, a "printer".

There are various models of the Apple II computer, and progressive improvements have been made thereto, but the references in this case have been to Apple II computers in general, without attention to any variations.

The monitor and the printer enable the computer's operations to be displayed and, with the disk drive, are referred to as "peripheral" equipment. The micro-processor contains:

(i) a central processing unit or "CPU" for the performance of calculations and other tasks; and

(ii) various "memory" devices in which the information and the instructions for the carrying out of those tasks can be stored.

A programme is commonly referred to as an ordered set of instructions, but this term can be misleading. For present purposes it is unnecessary to do more than say that what is primarily sought to be protected is the skill and care involved in the preparation of a code for a computer, using code in its ordinary sense of an abbreviated, but understood, form of expression. Programmes exist in great variety, so far as concerns subject matter, and different sets of programmes have different functions in relation to the setting up and running of the computer.

Some programmes can be stored permanently in a memory device called a "ROM" (Read Only Memory), the essence of which is a silicon chip. These are attached inside the computer at the time of manufacture, although they are readily removable and can be replaced. Their "memory" is not erased when the computer is switched off. They cannot be altered, and are therfore sometimes called "firmware". There are six ROM's in an Apple II micro-processor. As I think there is some lack of understanding of what a ROM comprises, I set out two passages from the Apple II Reference Manual, which is in evidence.

"The Apple, in its natural state, can hold from 2K (2,048 bytes) to 12K (12,288 bytes) of Read-Only memory on its main board. This ROM memory can include the System Monitor, a couple of dialects of the BASIC language, various system and utility programs, or pre-packaged subroutines such as are included in Apple's Programmer's Aid 1ROM.

The Apple's ROM memory resides in the top 12K (48 pages) of the memory map, beginning at location $D000. For proper operation of the Apple, there must be some kind of ROM in the uppermost locations of memory. When you turn on the Apple's power supply, the microprocessor must have some program to execute. It goes to the top locations in the memory map for the address of this program. In the Apple, this address is stored in ROM, and is the address of a program within the same ROM. This program initializes the Apple and lets you start to use it."(p 72)

The explanation of the term given in the glossary to that publication is:

"Read-Only Memory (ROM): This type of memory is usually used to hold important programs or data which must be available to the computer when the power is first turned on. Information in ROMs is placed there in the process of manufacturing the ROMs and is unalterable. Information stored in ROMs does not disappear when the power is turned off." (p.183)

The computer and all the physical devices attached to it are known as "hardware".

Other programmes are fed into the micro-processor when a magnetic disc comprising a programme is placed in the disc drive unit or when one is typed on to the keyboard. The keyboard and disc drive unit are referred to as "input" devices. While the computer is being operated, the further programmes are stored temporarily in a memory device known as a "RAM" (Random Access Memory). The programmes will be erased from the RAM when the computer ceases operation, but usually remain stored on the magnetic disc for later use. Programmes of this kind are called "software" as they form no part of the micro-processor itself.

A distinction is sometimes drawn between "operating" programmes and "application" programmes. The Apple II ROMs fall more into the first category than the second, but the distinction, for copyright purposes, is rather too arbitrary and not very useful. The distinction was not a focus of argument in the present case.

In this case we are concerned with two "firmware" programmes that have been installed in the six ROMs of the Apple II computer. "Applesoft" is stored in five of them (identified as D0, D8, E0, E8 and F0) and "Autostart" is stored in the remaining one (identified as F8).

Both these programmes were originally written by persons in the employ of the first appellant. On one view of the law the fact that there was original writing may be regarded as a critical element in the case. Between August and December 1977. Kenneth Randall Wigginton wrote "Applesoft" by modifying a programme called "Microsoft Basic". He did so using considerable skill, time and effort. Wigginton expressed each step or operation of the programme in a shorthand form called "6502 Assembly Code". That language consists of mnemonics each comprising three letters of the alphabet, and can be readily understood by people skilled in its use. A programme in this form is said to be expressed in "source code". In like fashion Stephen Gary Wozniak in 1977 wrote, and John Arkley in 1978 modified, "Autostart".

These programmes included labels (identifying particular parts of the programme) and comments (explaining particular parts thereof), written next to many of the mnemonic statements.

A micro-processor cannot store or operate in accordance with a programme expressed in source code. As an electrical device it can only register the presence or absence of electrical impulses and performs all its functions according to a process which can be described in terms of binary arithmetic. According to the evidence an "0" indicates the presence of an electric current, or impulse, at a particular place, and a "1" indicates the lack of a current, or an electric impulse, there. It makes no differnce if this indication is reversed.

The fundamental aspect of computerization is not, however, the use of electricity or of micro-chips, but the conversion of figures, words and symbols into a code and, in the case of digital computers, such as those here in question, a code expressed in terms of binary arithmetic, that is to say, arithmetic to the base 2. Results could and can be achieved by manual and mechanical means, usually with much labour, but the importance of the process in recent years is due to means being established for doing it electronically. The silicon chip has become the successor of more elementary electric devices.

Programmes in source code must therefore be converted into sequences of electrical impulses. This is achieved by the use of a computer (a PROM programmer) which is equipped with an "assembler" or "compilation programme". A source programme's mnemonics are typed into the computer. Each keystroke generates a series of electrical impulses which the assembler will recognize and convert into impulses which detail each step of the programme. The sequences so produced are said to express the programme in "object" or "machine readable code". Assembler programmes operate in a relatively standard manner, so that programmes in a particular source code (e.g. Applesoft Basic) comprising the same terms will produce object codes made up of the same sequences. Both "Applesoft" and "Autostart" were converted into object code in this way.

Initially, representations of these sequences of on a magnetic disc or tape. Later, they were regenerated, transferred to silicon chips contained in the ROMs, and stored or retained there.

The silicon chips contain numerous fusible connectors and insulated capacitors. Using a device referred to as a "ROM Burner", the impulses are passed through the chips, burning out the connectors and leaving the capacitors electrically charged. The pattern of connectors and capacitors thus affected corresponds to the sequence of electrical impulses so that the same sequence of electrical impulses can be regenerated. The impulses are regenerated in this way each time the programme is used or "run" in the micro-processor. The ROMs in each Apple II computer contain silicon chips bearing patterns which represent "Applesoft" and "Autostart" in object code. These can be seen, but only with an electronic microscope.

An Apple II computer has the capacity, when the appropriate keys are pressed, to display a description of the object code or sequence of impulses, this being reproduced from the patterns in its ROMs. The sequences can most accurately be described with binary notation, but a display made up this way would be extravagantly voluminous. The Apple II uses what is called hexadecimal notation when a description of the sequences is displayed on its monitor or printer. Either notation is a description of the object code programme, and is intelligible to persons trained in the use of that form of notation. There is evidence that Wozniak originally wrote part of what became "Autostart" in hexadecimal notation.

The computer has the further capacity to "disassemble" the sequences of impulses and display the results in source code, in this case 6502 Assembly language. A print-out of the disassembled source code does not entirely reconstruct the programmes as originally written. The mnemonics reappear, but the labels and comments are not reconstructed.

In the present case the paper on which was set out the source codes, and other details, was destroyed. This is a common practice. As indicated, the essential matter that is fed into the computer as source code can be recovered.

The respondents import Wombat computers, the micro-processors of which contain two ROMs and one memory device called an "EP ROM" (Erasable Programmable Read Only Memory). A programme stored in an EP ROM is embodied therein as a pattern of electronic charges stored on insulated "floating gates" (referred to as "MOS" technology). The absence of an electronic impulse is indicated by an uncharged gate. If an EP ROM is illuminated through a "window" in its outer surface by ultraviolet light the pattern will be erased, otherwise it remains. An EP ROM, unlike a ROM, is a variable memory device.

The computer programmes which are embodied in the Wombat ROMs and EP ROM are said to be "reproductions" of those stored in an Apple II's six ROMs. It was not submitted, nor was it pleaded, that the Wombat components comprised "adaptations" of the programmes comprised in the Apple ROMs (see definition, s.10(1) above).

The Wombat computer, like the Apple II computer, has the capacity to display, either on a monitor or a print-out, a description of the sequence of impulses contained in its ROMs and EP ROM. It can disassemble the sequences and display the results in source code. Print-outs describing the sequences of impulses generated within the Apple II and Wombat computers were made available to the trial judge (Ex. "AB", "AC" and "AF"). The print-outs are substantially the same and indicate that the sequences of impulses are the same. One minor difference between the print-outs indicates that one of the Wombat programmes enables the word "Wombat" to be displayed on the monitor or printed out when that programme is used, although notably in the same place as "Apple" appeared, and appears, in the Apple programme.

Whilst the Wombat EP ROM reproduces similar sequences to part of the relevant Apple ROMs, its pattern of charged gates is probably not the same as the pattern of burnt-out fuses and capacitors in the Apple ROMs. It may well be that the Wombat ROMs are also programmed using "floating gate" or MOS technology, but the submissions left this matter at large. There is apparently no evidence as to the technology used to store their respective programmes. It therefore cannot be decided positively that there are patterns in the Wombat ROMs which are closely similar in appearance to the burnt-out fuses and capacitors in the Apple ROMs.

The evidence clearly establishes that the degree of similarity between the Apple and Wombat programmes in the components mentioned could only be achieved by direct copying. Whatever the exact process used in bringing it about, the Wombat ROMs, and the EP ROM, contain representations of the same sequences of impulses as are represented in the Apple ROMs. The conclusion of copying is reinforced by the fact that the initials of the author of Applesoft are contained in both the Apple ROMs and the Wombat ROMs. Electrical devices are available to enable the patterns on a ROM to be copied readily on to another ROM or EP ROM, and it seems to be accepted that this is the easier or better way to do the copying. Another way would be to recover the object code of the ROMs, translate back to source code, or hexadecimal notation, and to programme forward again. The Wombat ROMs and EP ROM were, then, copied from the Apple ROMs. The fact that six Apple ROMs have been copied on to two Wombat ROMs and one Wombat EP ROM demonstrates an advance in micro-chip technology unrelated to the programmes or codes contained therein.

Being copies of the programmes contained in the Apple ROMs, the Wombat programmes do not include the labels and comments which were part of the original Apple programmes as written in source code.
The Subsistence of Copyright

The question whether copyright subsists was dealt with by the judge briefly on the basis that neither the programmes in source code, nor in object code, were literary works. I have understood this conclusion to flow from a general view concerning copyright in computer programmes. It has been suggested, on the other hand, that he was deliberately confining his observations to the particular programmes, but he did not indicate what relevantly distinguished the present from other computer programmes, nor did he draw a distinction between the "source" and "object" codes.

We have not been invited in this case by the respondents to find separately in relation to either Applesoft or Autostart on some basis such as function, and in what follows I will not always discuss them separately. Although we were addressed as to the function of the programmes, and particularly their use in the Apple II computer, the essential concern of copyright is expression and not function, or concepts. Similarly, while "purpose" is doubtless relevant in many cases, it is not by itself critical.

I do not myself doubt that the programmes in source code were literary works. There is no necessity for a literary work to be of any literary quality. It is accepted that the term includes mathematical tables, codes, and, in general, alphanumerical works. One limit doubtless is that it needs to be a "work" and to have had some skill, even if very small, applied to its preparation. Meaningless rubbish would plainly be excluded. Copinger and Skone James on Copyright, 12th ed., para. 156 has a succinct passage which I believe to be supported by the authorities and to be applicable:

"But so long as something in 'writing' exists, it is sufficient, and it is not necessary that what is written should express a meaning in ordinary language. Thus, there may be copyright in a list of words used as a telegraph code, or in a catalogue of type, or in a system of shorthand."

Halsbury's Laws of England, 4th ed., Vol. 9, para. 835 says, in a passage applicable to our Act:

"The expression 'literary work' covers work which is expressed in print or writing, irrespective of whether the quality or style is high, and is used in the sense of written or printed matter."

In the present case, there is no doubt that a great deal of skill went into the preparation of what was written, and that what was written could be read and understood by suitably trained people. Even applying the test his Honour thought appropriate, and which I have set out as part of his reasons, it is certainly difficult to exclude the possibility that a reader may have got instruction and information from reading it.

It is clear that the first appellant owned, and owns, the copyright in these literary works, and that they were "original" in the relevant sense.

The statement of what is a literary work which his Honour gave, in reliance principally on Hollingrake v. Truswell (1894) 3 Ch. 420 and Exxon Corporation v. Exxon Insurance Consultants International Ltd. (1982) R.P.C. 69, is not, nor was it intended to be, exhaustive. There has I think long been a tendency (not invariably observed) to apply the language of the statutory law governing copyright in a practical manner, consistently with the needs of the time, and the then current concepts.

In my view the source codes before us express meaning as to the arrangement and ordering of instructions for the storage and reproduction of knowledge. It is incorrect to describe them simply as components of a machine. With respect, I do not agree with the statement apparently accepted by the learned judge which suggests an analogy with "something merely intended to assist the functioning of a mechanical device". There is a distinct and recognised difference between a computer programme and the electro-magnetic functioning of the machine. The codes are in my view original literary works.
Adaptation of an Original Literary Work

I am satisfied that the object codes in the Apple II ROMs are adaptations, within the meaning of s.31(1)(a)(vi) and s.10(1), of the original literary works constituted by the programmes in source codes. This is, I think, because they can fairly be described as translations. Transliteration may more precisely explain what happens, but this is plainly comprehended within "translation". This term doubtless normally suggests translation from one language to another, but its ordinary meaning is wider and it is necessary to apply it with due regard for modern technology. The object codes contained in the Apple ROMs are a straightforward electronic translation into a material form of the source codes, and it would be entirely within ordinary understanding to say that they are translations of the source code. An indication of the natural application of the word to the production of an object code is found in the 1979 report of the United States National Commission on New Technological Uses of Copyright Works (p.21, n.109):

"A source code is a computer program written in any of several programming languages employed by computer programmers. An object code is the version of a program in which the source code language is converted or translated into the machine language of the computer with which it is to be used."

(The emphasis is mine)

It is apparent from the definition of "adaptation" in s.10(1) that neither it, nor, I suggest, its respective ingredients, are to be given a narrow or confined meaning. See also Laddie Prescott and Victoria, The Modern Law of Copyright (1980) p.96 and Sega Enterprises Limited v. Richards (1983) F.S.R. 73 (per Goulding J.).

It is plain from the language of the Act itself that an adaptation of a literary work does not itself have to be a literary work, and it does not seem to me that it must by itself be capable of being the subject of copyright. Whether the Apple II object codes could be regarded as literary works is not a matter which has to be considered. The making of an adaptation is one of the exclusive rights comprised in copyright.

I should say that I do not find myself in agreement with the expressed view of his Honour to the effect that computer programmes were deliberately omitted from the provisions of the Act. It is certainly true that no special provision was made for them, but this is a different matter. Apart from the history of the matter, it can be borne in mind that at the time the 1968 Act was in the course of preparation (going back to the time of the Spicer Report (1959)), computers were not widely used, and micro-computers were almost unknown.
Infringement

I have already stated that the Apple object programmes were copied into the Wombat components. So far as concerns the Applesoft object programme, at least, it was early in the field and was of considerable value. The copies, when inserted in the Wombat computers, meant not only that Wombat software was "compatible" with Apple programmes (which they were advertised to be), but that the programming facility of the two machines was the same or virtually the same. This was undoubtedly done deliberately, for the financial benefit of the respondents. The question which arises is whether this offends against the copyright laws.

It was submitted for the appellants that the Wombat ROMs contained reproductions of either the source or object code expression of the Apple programmes. It was submitted on behalf of the respondents that the evidence does not establish that they copied the appellants' programmes from source code, or any written derivation thereof, into source code, nor that they copied a representation of the object code in hexadecimal notation. What they did, however, was to copy that which the source code programmes had been converted into, that is to say, the programmes expressed in object code and embodied in the Apple ROMs. As I have said, this manner of copying seems to be the most likely. If the object codes in the Apple ROMs were "reproduced" in the Wombat ROMs and EP ROM there would in my view be reproductions of the adaptations and hence the commission in each case of an act comprised in the copyright (see s.31(1)(vii)).

The subject matter is not the chips but the code. The code is embedded in the chips, in the sense that it is in their components that the electric charges are to be found, arranged according to the code. Infringement in the present case involves the reproduction in "a material form" of an adaptation of a work (s.31(1)(i) and (vii)). It is not necessary to reproduce the material form of the work or adaptation. The resemblance to be considered is that between the code embodied in the adaptation and that in the alleged reproduction, and not one as between the materials in which they are embodied (see on the distinctions involved, Re Dickens (1935) 1 Ch. 267, per Romer L.J., at pp.292-295, and Pacific Film Laboratories Pty. Ltd. v. Federal Commissioner of Taxation (1970) 121 C.L.R. 154, per Windeyer J., atpp.167-169).

From the evidence of the print-outs it is plain that the arrangements of the object codes stored in the Wombat ROMs are the same, or virtually the same as the arrangements of the object codes stored in the Apple ROMs. Moreover, the Apple codes could be reproduced, with substantially their same content, from a print-out recovered from the Wombat ROMs. It is of no consequence that the Wombat ROMs and chips probably bear different patterns from those appearing on the Apple ROMs or chips. The fact is that the former embody the same arrangements of impulses and produce the same result, and are directly copied from the Apple ROMs. In my view it does not matter, if it be the fact, that the code cannot be seen, even with special apparatus. The adaptation of the work is rendered perceptible with a machine. It is sufficient that the code has its existence in, and is ascertainable from, the chips.

It seems to me that the correct view is that the Wombat ROMs do contain reproductions of the Apple object codes, within the meaning of s.31(1)(a)(i) and (vii). I should add that the question has not been raised of direct adaptation of the source code by production of the Wombat object code.

The Wombat computers contained the ROM chips in question at the time of importation and of sale, and ss.37 and 38 are applicable if knowledge is proved. Print-outs were not, as such, imported, nor sold, by the respondents and need not be considered. There will not be infringement under ss.37 or 38 unless knowledge of infringement at the relevant time or times is established. This is not a matter upon which the learned judge made a finding. Counsel for the appellants relied upon the documents in evidence and some of the cross-examination of the second respondent, a decision on the matter may turn on matters of credibility.

Section 36, which deals with primary infringements, does not require knowledge. A case might be made out thereunder, but counsel for the applicants (the present appellants) disclaimed reliance upon it at the trial. Counsel for the respondents objected to the section being relied upon before us, and said that if it had been a consideration at the trial the respondents' case might have been conducted differently. In the circumstances I do not think that the appellants should now be allowed to raise it.

If knowledge going back to some date prior to the commencement of the proceedings has to be proved, the matter will have to be referred back to the trial judge. The primary difficulty, as I see it, is that we do not know what weight, if any, or, indeed, what effect generally, to give to Mr. Suss' evidence. It may be difficult to deny that the relevant knowledge existed having in mind the action commenced on 11 March 1983 in the Supreme Court of New South Wales against people, admittedly agents of the respondents, claiming, amongst other things, breach of copyright. The summons in that case refers to a Golden II computer but in his evidence Mr. Suss says that this computer is the same as a Wombat. He also states that he had knowledge of these proceedings in March 1983. However, no finding of the learned judge as to Mr. Suss' explanations as to the extent of his knowledge is available.

It is plain, nevertheless, that if there is copyright and the making of the Wombat computers containing the Applesoft and Autostart programmes otherwise constitute an infringement, the necessary knowledge has existed at least since the commencement of the proceedings. Reference is often made to the test of knowledge propounded by Harvey J. in Albert v. S. Hoffnung and Co. Limited (1921) 22 S.R.(N.S.W.) 75, at p.81:

"'Knowledge' in the section cannot mean in my opinion any more than notice of facts such as would suggest to a reasonable man that a breach of the copyright law was being committed."

(See also Copinger and Skone James 12th ed. para. 530). The statement is to be understood as interpreted in R.C.A. Corporation v. Custom Cleared Sales Pty. Ltd. (1978) 19 A.L.R. 123 (see also Sillitoe v. McGraw-Hill Book Co. (U.K.) Ltd reported in "The Times" newspaper of 18 January 1982).

The appellants on more than one occasion put their argument in support of injunctions on the basis of knowledge existing from at least the commencement of proceedings. The case then becomes substantially a quia timet one. In the circumstances it seems to me that the appellants should be able to put their case this way. The fact is that the respondents continue to claim the right to import and sell as they have been doing. Agreed fact 12 is doubtless expressed elliptically, but it is as follows: "The first respondent has imported and sold in Australia and, unless restrained, will sell in Australia micro-computers under the name Wombat". It seems to be the fact that they have never been sold other than with the copies of the appellants' ROMs. The case thus put relies on ss.37 and 38, and not s.36.

I am therefore of the view that there was copyright, and that, after the time of commencement of proceedings, if not before, a threat of infringement existed. Some infringement may have occured during the course of proceedings, but this is not a matter which has been explored, or argued. I am therefore of the view that injunctions should go. If damages are sought, the matter should be referred to the trial judge for findings concerning knowledge, as well as other relevant matter.
Claim under the Trade Practices Act

In relation to the case under the Trade Practices Act, passages have already been cited from his Honour's reasons. He was of the view that the distinct label "WOMBAT", combined with small but observable differences in appearance of the computers, and some other factors, meant that there was no likelihood of deception.

His conclusion in relation to s.52 (i.e. the "primary case" under that section) was based on absence of product similarity. He immediately went on to deal with other provisions of the Trade Practices Act as follows:

"The case sought to be put by the applicants on the other provisions of the Trade Practices Act, namely, ss.53(c), 53(d) and 55 must also fail, for similar reasons. Given the conclusion reached that the respondents' products have been sufficiently distinguished from the Apple range of computers, it must follow that no false representation of the kind contemplated by s.53(c) or (d) has been made and that no misleading conduct of the kind proscribed by s.55 has been established. Again, this is essentially a question of fact that must be resolved adversely to the applicants. In the circumstances, this part of the claim is also rejected."

It does not seem to me to follow from his Honour's expressed reasons with regard to s.52 that the case sought to be made under the other provisions should fail, for similar reasons, or at all. His Honour did refer to the supply of Apple II manuals, but, only, as it were, as a minor ingredient on the aspect of product similarity when considering the application of s.52. For the purposes of s.53(c) it is necessary to examine more closely the effect of the supply of Apple II and Applesoft manuals, and copies thereof, in association with the Wombat computers. Section 53(c) is as follows:

"53. A corporation shall not, in trade or commerce, in connexion with the supply or possible supply of goods or services or in connexion with the promotion by any means of the supply or use of goods or services -

. . .

(c) represent that goods or services have sponsorship, approval, performance characteristics, accessories, uses or benefits they do not have."

Agreed facts 16 and 34 relate to this matter. The following facts also appear, and so far as I can see, are not disputed:

(a) In a series of telexes (undated, except for one dated 4 March 1983) orders for one hundred Apple II User's Guides were placed with orders for one hundred Wombat computers, and their shipment was notified to the second respondent from the same Taiwanese supplier. The orders may have been renewed once or on a number of occasions, or there may only have been one order, before it was acted on.

(b) By another telex, notification was given that fifty Apple II User's Guides were being shipped with fifty Wombat computers.

(c) There was an urgent plea on 3 and 8 November 1982 to the Taiwanese supplier to provide manuals on the Apple computer.

(d) On 1 July 1983 an order was placed for eighty Apple User's Guides with ten Wombat computers.

The manuals in question were of two types. One bore the name in typescript of a Taiwan company, under that, in print, of Osborne/McGraw-Hill, with whose agreement the book was expressed to be published. The book also contained, on a fly leaf, and on the back cover page, some Chinese typescript, or print. It was prominently described as the Apple II User's Guide and the only direct evidence of the supply of such a book with a Wombat was of its inclusion in a box with a Wombat purchased from the respondents by an employee of the second appellant.

The other manual was issued under the name Guan Haur Industrial Co. Ltd., of Taiwan, and was supplied from Taiwan and admittedly distributed by the respondents in association with their supply of Wombat computers. The manual purported to be for the Apple II and "Golden II Plus" personal computers, but in fact was a complete copy, in the one volume, of two manuals of the appellants, described respectively as the Applesoft Basic Programming Reference Manual, and the Apple II Reference Manual. The Guan Haur manual bears an endorsement on its first page, in large capitals:

"A REFERENCE MANUAL FOR THE APPLE II AND THE GOLDEN II PLUS PERSONAL COMPUTERS".

There are many references throughout to Apple II computers and, all the illustrations seem to be of them. In many places the words "Golden II" are inserted (rather crudely) where, in the original manuals, the words Applesoft or Apple II appeared. The manual appears to me to be a pirated copy of the Apple computer manuals to which I have referred.

There is no evidence of the existence of separate Wombat manuals. So far as appears, the abovementioned orders were fulfilled. The presence of the Apple II books, and their supply with the Wombat computers, was explained on the basis that the Wombat software was advertised as "Apple compatible". There is ambiguity about these words, and it is not always clear how they are used. I have generally taken them to mean that software designed for use in an Apple II could be used in the Wombat. This is the explanation given by Mr. Suss in his affidavit and at the commencement of his oral evidence. I quote a passage from his affidavit:

"I therefore obtained a machine, which I call the Wombat, which was, on my understanding designed to run Apple compatible programs. That is to say, a computer that is capable of running programs which were designed for Apple computers without substantial modification to those programs."

On the other hand, some advertisements contain a prominent note: "The Wombat software is compatible with programs for the Apple II". It is understandably an important matter for the producer of a programme (on a floppy disk or cassette) to be able to say that it could be used in a popular computer, such as Apple. It is also important, no doubt, to the producer of a computer to be able to say that popular programmes could be used in its machine. What would be understood, in the note I have set out by "Wombat software" is also not clear. The term "software" may refer to the programmes in the EP ROMs and the Wombat, although that would not, I think, be a normal or general use. It does not seem that the respondents were engaged in producing, or selling, something under the name or description "Wombat software". The Apple II manuals went far beyond dealing with the use in other machines of programmes designed for use in the Apple II computer. Indeed, they do not seem to me to deal with this aspect at all. As I see it, the respondents were more or less compelled to provide the Apple manuals. They had none of their own, and they needed a manual which would adequately deal with that which they were selling. As they were selling a unit programmed with the Apple II programmes, the Apple manuals would obviously go a considerable distance in providing information and guidance to their purchasers. They were manuals for the understanding and operation of Apple computers, and dealt with Apple II ROMs.

The immediately relevant question is whether there was a representation that the Wombat had a sponsorship or approval (namely one by Apple Corporation Inc. or its associates) which it did not have.

It seems to me that the conclusion is open that a purchaser would have believed that there was a closeness of association between the makers or suppliers of the Apple computer and the Wombat computer. The meanings of "sponsorship" and "approval" were discussed by Franki J. in McDonald's System of Australia Pty. Ltd. v. McWilliams Wines Pty. Ltd. (1979) 41 F.L.R. 429, at p.434. I find it difficult to conclude that here there was a representation of "sponsorship", which suggests a degree of support or assistance by a sponsor. No oral evidence was given to this effect, nor, for that matter, to the effect that in someone's understanding there was an "approval". The trouble with this latter word, in the present case, is that it is not the most apt word to use, but in the context I do not think that it should be regarded as connoting anything formal. The best way I can put it with relative accuracy is to say that there would appear to be a measure of agreement or concurrence between those producing or supplying the respective computers.

The respondents sold computers other than the Wombat, and provided manuals other than the Apple II, and it is possible that they had the Apple II manuals for sale, or supply, to persons other than Wombat purchasers. There is, however, evidence that some were supplied with the Wombat computers, and, in the absence of evidence to the contrary, the inference seems to me to be irresistable that this was a common course.

It is necessary, however, to take account of his Honour's finding, as follows:

"In the present case, in my view, no purchaser of the respondents' goods is likely to be misled or deceived into believing that the respondents or their goods are in any way associated with the applicants: on the contrary, the respective products are clearly distinguished from one another by the use of very different brand names and the reference, in advertising, to the 'compatibility' of Wombat software with programmes for the Apple II computer only serves to reinforce the distinction. In my opinion, the primary claim under s.52 must fail."

I have earlier made reference to the use of the term "compatible", and, in effect, suggested that, at the least, it amounts to a half-truth. In view of the fact that the learned judge was dealing with s.52, and focussing attention principally upon the appearance and possible deceptiveness of the Wombat computers, I do not think I should regard this statement as a finding of application to s.53(c). On the other hand, I have found particular difficulty in arriving at a conclusion on the matter, in the absence of discussion thereon by his Honour and his assessment of relevant aspects of the evidence. The situation is one in which an appellate court finds benefit and assistance from the findings and analysis of a trial judge. In my view the least undesirable course is to set aside his Honour's decision as far as it relates to the application of s.53(c), and to remit the matter to him for his further consideration and decision.

This brings me to what his Honour described as the secondary claim under s.52.

The appellants submit that the Wombat ROMs and EP ROMs are "infringing copies" as described in s.10(1) of the Act and that the rights and remedies established by s.116 in respect of them are not only available against the respondents, but also against persons purchasing Wombats from the respondents.

They argue that by selling the Wombats the respondents have, in breach of s.52 of the Trade Practices Act, misled or deceived those purchasers into believing that they will get good title to and enjoy quiet possession of the ROMs and EP ROM, free of any claim such as one under s.116.

The respondents argue that s.116 is only available against actual infringers and not against persons in possession of infringing copies. As to whether it can be implied in the supply of goods by the respondents that they made the relevant misrepresentation, attention should be had to s.69(1) of the Trade Practices Act. This section implies a condition as to title and a warranty as to quiet possession in contracts between a corporation and a consumer. See also s.53(g). I doubt therefore whether an absence of title upon the sale of goods is a case to be dealt with under s.52. I need say no more.

In my view the appeal should be allowed and the judgment of his Honour set aside as far as concerns the copyright claim and the application of s.53(c) of the Trade Practices Act. In lieu thereof there should be injunctions and declarations in accordance with the foregoing. The claim under s.53(c) should be remitted to be considered further. If the appellants wish to pursue claims for damages in relation to infringements under ss.37 and 38 of the Copyright Act notice should be given in twenty-one days and in that event the matter should be referred to the trial judge for findings on the evidence as it stands in respect of "knowledge", and if favourable to the appellants in relation to a particular time or times, there should thereafter be an inquiry as to damages under those sections.

JUDGE2

This appeal raises important questions relating to the law of copyright. It also concerns issues under the consumer protection provisions of the Trade Practices Act 1974 (the "Trade Practices Act").

The first appellant, a Californian corporation, manufactures and sells computers under the name "Apple". Large numbers of Apple computers have been sold throughout the world. The second appellant, a New South Wales company, is the Australian distributor of the first appellant's computers. Apple computers are made and sold under names which include "Apple II", "Apple II Plus", "Apple Euro Plus" and "Apple IIE". The Apple computer with which this case is concerned is the APPLE II. The appellants have acquired a substantial reputation in Australia in respect of APPLE II computers, and more than 20,000 of them have been sold here.

The first respondent, a Victorian company, imports and sells in Australia computers under the name "WOMBAT", a computer manufactured or assembled in Taiwan. The second respondent is the managing director of the first respondent.

The first respondent was formerly an APPLE dealer until shortly before it commenced to sell WOMBAT computers. The first respondent sells WOMBAT computers from the same premises as it previously used for the sale of APPLE II computers.

The APPLE II computer and the WOMBAT computer are generally referred to as microcomputers or personal computers. With the development of the microprocessor (a Central Processing Unit on a silicon chip) there has been a substantial reduction in the size and cost of computers and at the same time a dramatic increase in their speed and capacity. These developments have led to the use of microcomputers in small businesses, homes and schools. A microcomputer is one of three classes of computers, each defined by its size and capacity. It is the description given to that class of computer which is smallest in capacity and slowest in speed of computation. The other classes, in ascending order, are minicomputers and mainframe computers.

Within the microcomputer market there is a degree of standardisation of both equipment and programmes, despite a plethora of competing manufacturers and suppliers. Products manufactured by companies other than the first respondent are compatible with the APPLE II computer. Similarly, some programmes produced by the first appellant and designed for use in the Apple II computer are compatible with the WOMBAT computer.

The appellants sued the respondents in this Court alleging breaches of s. 52, paras. 53(c) and 53(d) and s.55 of the Trade Practices Act and infringement of copyright under ss. 37, 38 and 115 of the Copyright Act 1968, (the "Copyright Act"). The respondents cross claimed for an injunction to restrain the appellants from sending circular letters making various claims which the respondents contend are misleading and deceptive and therefore contravene s.52.

The learned primary Judge dismissed the appellants' claim under s.52 of the Trade Practices Act on the ground that no purchaser of the first respondent's goods was likely to be misled or deceived into believing that the first respondent or the WOMBAT computer was in any way associated with the appellants. His Honour said that for similar reasons the claims under ss. 53 and 55 must also fail. His Honour also dismissed a secondary claim of the appellants, a blend of s.52 of the Trade Practices Act and the Copyright Act, that the first respondent had sold WOMBAT computers containing three silicon chips and that the chips were "infringing copies" within the meaning of that expression in s. 10 of the Copyright Act. The appellants argued that in those circumstances the first respondent misled or deceived purchasers of WOMBAT computers into believing that they would take title to, and enjoy quiet possession of, the computers free from any claim in conversion or detinue, so that the respondents were engaging in misleading or deceptive conduct contrary to s.52. His Honour dismissed this claim on the ground that no infringement of copyright occurred.

The learned primary judge also dismissed the appellants' application for injunctions and damages for infringement of the first appellant's copyright in certain of its computer programmes. His Honour found that the first appellant did not have copyright in those programmes because they were not "literary works" within the meaning of the Copyright Act. His Honour said that in his view a "literary work" for relevant purposes is something which was intended to afford either information or instruction, or pleasure, in the form of literary enjoyment. He looked to the function of a computer programme and held that its function is to control the sequence of operations carried out by a computer, especially in the case of programmes written in object code (an expression to which I shall refer later), being some of the programmes the subject of this case. His Honour therefore decided that none of the programmes in suit were "literary works" within the meaning of the Copyright Act.

As to the respondents cross claim, his Honour did not regard it as being any longer a live issue between the parties and stood it over generally without making any findings about it.

I propose to consider first the claim for infringement of copyright and then the claim under the Trade Practices Act. However, before I consider the legal issues involved in the claims, it is necessary to say something about computers in general and the products in this case in particular.
Computers

A computer is an interconnected and sophisticated system comprising four parts- input, processing, output and storage. The machines in a computer system are called its "hardware". "Software" is the term given to the computer's programmes and it is distinct from hardware. A programme is a concise series of instructions that directs the computer to do the tasks required of it step by step and to produce the desired result.

The copyright aspect of this case concerns software or computer programmes. The Trade Practices Act claims relate more to the hardware aspect of the computers in question.

Although in one sense all programmes instruct the computer, they are frequently categorised by the function they perform. In computer language, they may be "operating" programmes or "application" programmes. Operating programmes are programmes that, as a system, instruct the computer to manage its own resources, execute application programmes and manage data. Application programmes are programmes written and used to solve specific problems, such as, in business, to calculate the payroll or the accounts receivable.

Operating programmes are usually supplied by computer manufacturers, whereas application programmes, especially for microcomputers, can be bought "off-the-rack", or can be "tailor-made" to suit the users needs.

A microcomputer with an operating programme but without an application programme is, in most cases, of little more use than a literate calculator. A microcomputer with only an application programme and without an operating programme will not function. An operating programme is fundamental to the working of the computer. It is the programme that tells the computer how to function and how to recognise and execute the application programme. programme.

The distinction between an operating and an application programme is not a rigid one because a single programme may answer both descriptions.

This case concerns two programmes that answer the description of operating programmes. "Applesoft", in whatever form it is written, is one of the main operating programmes for the Apple II microcomputer. It is an interpreter programme, with the function of processing application programme statements entered into the computer in the high level language of BASIC (about which I will say more later), one statement at a time, and then causing the computer to execute the instruction contained in it. The Apple II computer is designed to operate on application programmes "keyed in" with the BASIC language. Without "Applesoft" to interpret and process that language, the computer would not be able to perform the tasks required of it. "Autostart", in whatever form it is written, is the programme that sets the Apple II computer in motion and prepares it for the demands made of it by the various application programmes.

A computer is thus controlled and operated by programmes. Before looking at how the computer interprets and obeys programmes fed to it, it is necessary to look at how the programmes progress through the computer system.

A programme has to be placed into the computer by some means. This may be done by inserting a tape or disk with the programme recorded on it, or by keying the programme in, line by line, through a keyboard terminal. It may also be done "automatically" by computer, when the programme has been permanently encoded on a silicon chip within the computer's main memory. This last method is usually adopted for operating programmes, and applies to the "Applesoft" and "Autostart" programmes in the APPLE II and WOMBAT computers.

Unless it has been stored there permanently, all programme instructions must be sent to the computer's "main storage", a part of the microcprocessor in the computer. From there each item of instruction is "fetched", processed or executed, and returned in sequence by the Central Processing Unit, (the "CPU"), which is another part of the computer's microprocessor. All of this is done according to the operating programm's instructions.

Generally, the main storage units hold the programme only temporarily while the CPU is executing the programme. Main storage is not to be confused with "secondary storage" which holds data in a permanent or semi permanent state on, say, disk or tape.

The CPU is the operating centre of every computer. All the manipulation of information entering a computer system takes place in its CPU. For example, when the computer system is used for the retrieval of legal information - a form of data processing - the CPU is where all the sorting and rearranging takes place.

It is necessary to understand how information gets into the primary storage or CPU so that the computer can process it. The CPU is a complex and extensive set of electrical circuitry. As such, it knows only the presence or absence of electrical impulses. The semiconductors in the microprocessor can be in one of two possible states, similar to that of an electric light globe, namely either "on" or "off".

For convenience, this two-state system (known as a binary system) is represented by the digits 0 or 1. The zero ("0") indicates the presence of electric current at a particular place in the CPU and the one ("1") indicates the absence of current there. The digits 0 and 1 are called binary digits or "bits", and represent the only units of information that can be recognised by a computer namely, the presence or absence of electric current.

Since a single bit itself cannot store all the letters, numbers and special characters which must be processed by a computer, a coding scheme has been devised, by which each character is represented by a combination of binary digits. For example, in the Apple II computer the letter "A" may be represented by a bit combination of 11000001 (that is, eight semiconductors each in the following electrical states; on on off off off off off on). Thus, when the user of the Apple II computer depresses the key "A" on its keyboard, (or reads the letter in by other means) the computer systems converts this character into a binary code of electrical ons and offs and is then able to understand it as the letter "A". Similarly, when it has finished processing that information, and is instructed to issue a print out, that process is repeated in reverse, so that the internal binary code of ons and offs is converted back to the letter "A".

It is important to note that different internal binary coding systems are used in different computers. Many conventions have developed over the expression of this internal binary code. The code is rarely written out by humans in binary notation, that is as a series of groups (or bytes) of bits, or 0's and 1's. More usually, a base 16 number system called "hexadecimal notation" is used. This is a number system that has base 16 and is thus able to have a different character (by using the numbers 0-9 or the letters A-F) to represent each of the absolute values 0 to 15. Considerably fewer characters are needed to represent values using this code than binary notation. For example, whereas in binary notation the letter "A" may be represented as "11000001", in hexadecimal notation, it may be represented as "C4", which makes a programme considerably more manageable. The hexadecimal notation convention is well defined and intelligible to persons trained in its use. The relevance of this is that the computer understands programmes only when they are stored in the CPU as electrical ons and offs, whereas the human user or programmer understands programmes only when written out in some language known to him. Since the electrical states cannot be written down, they have to be represented in some other notation before a human is capable of reading it. The closest representation of those states is the binary notation in which each character is assigned a special configuration of 0's and 1's directly corresponding to the on or off state of the electrical circuits. More conveniently, hexadecimal notation may be used to make the same representation. When in either of those notations the programme is said to be in "machine language" or in "machine readable form". Machine language is the lowest of all computer languages.

Programming languages are said to be "lower" or "higher" depending on whether they are closer to the language of the computer itself (0's and 1's - low) or to the language of the user (akin to English - high).

In the early stages of computer development, it was necessary for every programme to be written and read into the computer in machine language which proved tedious, cumbersome and liable to error. These inconveniences led to the early development of the next level of languages, the "assembly languages", which by today's standards are fairly "low-level", that is, they are oriented more to the computer than to the human user. By using them, however, the trained programmer is released from the burden of 0's and 1's or from hexadecimal notation, and is able to write his instructions by way of abbreviations and mnemonic codes and other symbols more meaningful to him.

Although assembly language is more readable to the programmer, and therefore more convenient for him to work with than machine language, the computer cannot understand it. Thus a programmer who uses an assembly language requires a translator to convert the assembly language programme into machine readable form. The translator is called an assembler programme, and is usually referred to as an "assembler". An assembler takes the programme written in assembly language and converts it into machine language. The assembler is usually automatically included in any operating system.

It is important to note that, as with machine languages, assembly languages are not universal. In situations where one manufacturer's assembly language can be applied to another's computer, the first's language is said to be "compatible" with the second's computer.

Assembly languages presented difficulties to the burgeoning computer industry which led to the development of "high level languages". These are highly refined languages which are akin to English in both vocabulary and style, making them very convenient to use. There are many high level languages; some are very specific and limited in their design and use; whilst others, such as BASIC, are general and universal in operation. FORTRAN, COBOL and PASCAL are three other well known examples of high level languages. Of course, whichever type of high level language is used by the programmer, a translator is needed to translate the symbolic statements in that language into machine language. In performing this function the translator is called a "compiler".

This case involves two programmes, each written in a form of assembly language known as 6502 Assembly Language and each also reduced to a form of machine language. As neither programme in this case is written in any form of high level language, no issue concerning programmes in that language is involved.

A further description is often applied to programming languages. When a programme is written in machine language it is often said to be in "object code". When it is written in any language higher than that, i.e. in a language which needs to pass through either an assembler or a compiler, it is said to be in "source code".

These descriptions have been employed in the present case, so that when the "Applesoft" programme is written in 6502 Assembly Language it is known as "Applesoft Source", and when in machine language it is known as "Applesoft Object". Similarly when written in 6502 Assembly Language the "Autostart" programme is known as "Autostart Source", and when in machine language it is known as "Autostart Object".

There are slight differences between the programmes involved in this case in their source code and in their object code. These differences concern only the labels and comments placed on the source code programme to assist the programmer which were not translated or assembled into object code as they are not needed by the computer.

Once a programme in machine readable form has been entered into the CPU the computer is able to process it. However, no programme is able to be executed until stored in the CPU's primary storage whence each instruction is processed sequentially. Most modern computers use semiconductors for storing data. The semiconductor typically used is silicon in the form of a silicon chip. A silicon chip is made up of thousands of integrated, complete electronic circuits etched onto a small chip of silicon. Each of these circuits can be in one of two states, either carrying electrical current (on) or not (off). In computer language, these storage or memory chips, in which data and programmes can be written in and read from any main storage location readily and speedily, are called RAMs, meaning "Random Access Memory". All RAMs share the feature of being volatile, that is, they require a continuous supply of electric current to maintain the stored data. If the current is stopped or even interrupted, all the data represented in the memory is lost. In some cases, this volatility is desirable, for instance where the CPU is using the RAM as its "scratchpad", to do the task at hand. But in cases where the same programme is being used all the time, such as an operating programme, that volatility is a disadvantage.

To overcome this potentially disadvantageous situation, a permanent form of main storage was developed. This is also a form of silicon chip storage, but is designed so that any programme or data encoded on it at the time of manufacture is permanently recorded. This type of storage is called a "ROM", meaning "Read Only Memory". ROMs typically store operating programmes for the microcomputer, as those programmes are always required whenever the computer is being used. The permanence of the programme encoded on a ROM has two effects. First, the programme cannot be changed by the user of the computer. Its contents can only be "read" each time, not erased, amended or rewritten. Second, the storage is not liable to power losses; if there is a break in the supply of electricity the data is not lost. These features are the result of the method of manufacture of the ROMs. The ROM silicon chip, with its thousands of electronic integrated circuits, is manufactured by a "ROM Burner" which passes electrical impulses through the chip. Those impulses burn out the connectors between the many capacitors, leaving certain of them electrically charged. The resulting pattern corresponds to the sequence of electrical impulses produced by the programme as represented by its object code or machine readable form. Each time the ROM is read, the same sequence of electrical impulses is generated, regenerating the programme precisely.

Because the ROM may be seen as software in hard (permanent) form, it is often referred to as "firmware" or "hardwired software". The programme contained on a ROM is still software, it is still a programme; but it now has the hardware characteristic of permanence, making it much more convenient and inexpensive to use.

An "EPROM", meaning "Erasable Programmable Read Only Memory", is a special type of firmware storage. It is like a ROM in that its contents can be read (and only read) during the execution of a programme. However, it is possible for the contents of an EPROM memory to be altered, rewritten or erased completely by the user or programmer before is assembled into the computer system.

The Apple II computer, like most personal computers, has its operating system in ROM form. The "Autostart" programme is encoded on one chip or one ROM, while the "Applesoft" programme is encoded on five chips or five ROMs. In the case of both programmes, the pattern of the electrically charged integrated circuits on the ROMs corresponds to "Autostart Object" and "Applesoft Object" respectively. The inscribed patterns on the chips can be seen by a human being with the aid of an electronic microscope, but not otherwise.

"(a) labels identifying particular parts of the programme;

(b) mnemonics each consisting of 3 letters of the alphabet and each corresponding to a particular operation expressed in 6502 Assembly Code;

(c) mnemonics identifying the register in the microprocessor and/or the number of the instruction in the programme to which the operation referred to in (b) relates; and

(d) comments intended to explain the function of the particular part of the programme for the benefit of a human reader of the programme."

Partly because of the presence of the labels referred to in paragraph (a) and the comments referred to in paragraph (d) the programme in the form in which it was written was readily understandable by persons expert in programming. Thus if another programmer had picked it up, it would have had meaning to him and he would have been able to make use of it.

The programmes did not come into existence without a great deal of time, effort and skill on the part of those who wrote them. The programme referred to in para. 21 of the agreed statement of facts is known as APPLESOFT. The other programme with which the case is concerned is referred to in para. 28. It is known as AUTOSTART.

Both the APPLESOFT and AUTOSTART programmes as originally written have been destroyed. But they were used, before their destruction, to programme the appellants' Apple computer. The physical process by which this was done involved the use of a device known as an assembler which converted the mnemonics of the Assembly Code (otherwise source code) into object code. Object code is machine language, that is, language which can be understood by the microprocessor in the computer. Object code can be written. If it is, it will appear in binary notation (that is, mathematical notation to a base of 2) as a series of noughts and ones. But the noughts and ones are only symbolic. They are symbols designating the presence or absence of an electrical charge. Each such indication is a bit and a group of 8 (which is equivalent to a character, that is, a letter or digit) is a Byte. A kilobyte is 1,024 (not 1,000) Bytes. A kilobyte is designated by the letter K and a computer or a component thereof is said to have a capacity in terms of the number of kilobytes which it is able to store, for example, 6K or 12K and so on.

It should be mentioned that, although the process by which the assembler converted the source code into object code was remarkably quick, it involved two steps. The first was the conversion of the source code into hexadecimal notation. Hexadecimal notation is mathematical notation to a base of 16. The notation used comprised the digits 0 to 9 inclusive and the letters A to F inclusive, the latter representing respectively the numbers 10 to 15. The conversion to hexadecimal notation enabled the mnemonics of the source code to be shortened or abbreviated and the storage capacity of the ROM chips consequently increased.

By the time the conversion to object code had taken place the programme in that code was stored in ROM chips. By a process which is described in the evidence the programme was fixed or "burnt" into the chip. A programme so fixed is not erasable. It will not disappear when the power is turned off.

It should be emphasised that the process whereby the source code is converted into object code is not a manual or written process such as was involved in the compilation of the programme when it was originally written in the assembly or source code. The person carrying out the process does not see the transition into object code. None of the process is visible to the human eye.

Nevertheless it is possible once the programme has been fixed in the ROM in the way that I have described for it to be partially retrieved. The operation which will enable this to be done is not without its complications but central to it is the calling up, by the use of the keyboard, of the number -151. This will result in the programme appearing both in hexadecimal notation and in the mnemonics of the source code on the VDU or, if desired in permanent form, on a print-out. The programme will not be retrieved with the labels referred to in sub-para. (a) of para. 21 of the agreed statement of facts above quoted nor with the comments referred to in sub-para. (d) thereof. But it will show, instead of the labels referred to in sub-para. (a), certain address lines or memory locations. With difficulty an experienced programmer, after many hours work, is able to make some reconstruction of the programme with the aid of the mnemonics, but there is not possible any recovery or reconstruction of the comments referred to in sub-para. 21(d).

The Apple computers each contain six ROMs, five containing the APPLESOFT programme and one the AUTOSTART programme. Implanting the programmes on to further ROM chips is apparently a comparatively simple process. The procedure earlier described is not repeated. There is available a copying device by which the programmes are fixed into as many ROM chips as are required. Presumably this is how the manufacturing process takes place.

The evidence plainly establishes that the Wombat ROMs are almost exact replicas of the Apple ROMs. I say that not unmindful of the fact that the Wombat computers contain three, rather than six, ROMs and that one of the Wombat ROMs is not a ROM but an EP ROM. These letters stand for, "Erasable Programmable Read Only Memory". As the name implies it is possible to erase the memory fixed into an EP ROM. But this involves the application of ultraviolet light to its outer surface. The programme will not be erased merely because the power is turned off. For practical purposes the fixing of the programme into the EP ROM makes it as much part of the permanent memory of the computer as would have been the case if a ROM had been used.

A further matter of difference relied upon by counsel for the respondents was that the programmes were apparently fixed in the Wombat ROMs by the use of a technology different from that used to fix the programmes in the Apple ROMs. Again, I do not regard this as significant.

The plain fact of the matter is that the programmes in both sets of ROMs (that is, the hexadecimal notation and the mnemonics referred to in sub-paras. (b) and (c) of para. 21 of the agreed statement of facts) are, when printed out, virtually identical. Even the initials of one of the persons who wrote one of the programmes have been carried across. One difference provides positive evidence of piracy. It is the substitution of the name Wombat for the name Apple.

I have stated the facts sufficiently to come to the legal questions which are involved. One may have views about the morality of what those who manufactured the Wombat computer have done, but morality is not the question. The only question is whether there has been a breach of the law.

For the reasons given in the judgments of Fox and Lockhart JJ. I am of opinion that the programmes as originally written in source code were literary works and that they were owned, at all relevant times, by the first appellant. The learned primary Judge, relying principally upon Hollinrake v. Trusswell (1894) 3 Ch. 420 and Exxon Corporation v. Exxon Insurance Consultants International Limited (1982) R.P.C. 69, expressed the view that for relevant purposes a literary work was something which was intended to afford either information or instruction or pleasure in the form of literary enjoyment. I agree with the other members of the Court in thinking that this cannot be an exhaustive statement or definition of what a literary work is. The purported definition seems to contemplate that there must be an intention to make available the work for the information, instruction or literary pleasure of others. The words come from the judgment of Davey L.J. in the Hollinrake case (see p. 428). The case was decided before the United Kingdom Act of 1911 came into force. That Act applied in Australia - see Copyright Act 1912. Until the United Kingdom Act of 1911 there was no statutory copyright in unpublished works; see Copinger and Skone James on Copyright, 12th Ed. ("Copinger") para. 51. Under the present Australian Act, copyright subsists in unpublished works by reason of the operation of sub-sec. 32(1); see also s. 29.

It seems to me that many instances could be given of literary works in which copyright subsists but which are not embraced within Davey L.J.'s definition. I instance a diary kept for one's personal use or pleasure recording, perhaps. some period of one's life or a series of related events. As it is written, there may be no intention of publishing it so that it is not, at that time, for the information, instruction or pleasure of anyone but the writer. Years later the author may decide to publish it or to reproduce it. He has the copyright in it from the outset. If before his decision to publish, anyone took it and reproduced or published it without his consent, there would be an infringement.

So here the programmes in question were not, as I understand the evidence, ever published nor was it intended that they should be. They were not, so far as the evidence discloses, reproduced, but they were nevertheless literary works. Copyright subsisted in them as unpublished works. If they had been taken from one of the programmer's desks by a cleaner and reproduced or published by a competitor without the first appellant's consent, there would have been an infringement. During the argument, much was made of the fact that the programmes as originally written could not be fully retrieved or reconstructed. That was because neither the labels nor the comments could be "recalled" by the computer and thus resurrected or retrieved. But I am satisfied on the evidence that the two series of mnemonics referred to in sub-paras. (b) and (c) of para. 21 of the agreed statement of facts were, in the case of each programme, sufficient themselves to constitute literary works. Without the labels and comments they were not as intelligible as they were with them. Nevertheless, experienced people could understand and make use of them, albeit with a great deal of time and effort and with the need to assign comments probably different from those forming part of the original programmes to them. In any event, it seems beyond question, that if one were to have copied only the mnemonics, there would have been a reproduction of a substantial part of each programme for the purposes of s. 14 of the Act; see Copinger para. 465.

The programmes in source code being literary works, the next question is whether they are also literary works when converted in the manner earlier described into object code. If not, there arises the alternative question, as to whether the programmes when so converted are adaptations of the programmes in source code within the meaning of sub-para. 31(1)(a)(vi) and sub-sec. 10(1) of the Act. It is to be observed that it will not assist the appellants to show that the programmes when converted into object code became reproductions of the programmes in source code. That is because there is no infringement if what is done by the alleged infringer is to make a reproduction of a reproduction. That is no doubt because the draftsman of the Act considered that any reproduction would be a reproduction of the work itself, notwithstanding that it was derived from another reproduction. It follows that infringement will occur, for relevant purposes, only if there is a reproduction of the literary work itself (sub-para. 31(1)(a)(i)) or a reproduction of an adaptation of the literary work (sub-paras. 31(1)(a)(i) and (vii)).

In my opinion the programmes in object code are not literary works. Fixed as they are in the ROMs, they are unable to be seen in that code. True it is, someone could write them out so as to show them symbolically in binary notation or hexadecimal notation. The computer itself can show them symbolically in hexadecimal notation. But all of that is irrelevant. The important point is that it is only the machine itself, that is, the microprocessor, which can "understand" or "see", and thus deal with, the object code.

It is to be emphasised that the appellants need to persuade the Court that the ROMs with the programmes fixed in them are literary works because it was the ROMs which the manufacturer of the Wombat computer copied. Unless that copying constituted a reproduction of a literary work (or of an adaptation thereof to which I have later to come), there will be no infringement. I recognise that it is trite law that copyright subsists in the order of words and not in ideas. It is the form, not the idea, which is of paramount importance. But if what is alleged to constitute a literary work (here the programmes in object code) cannot be seen even with the aid of the screening or printing devices attached to the computer, they cannot in my opinion, amount to a literary work. A consideration of a number of the principal sections of the Act, particularly ss. 31, 32, 33 and 36 confirm me in this conclusion. In this connection it should be mentioned that the music recorded on the perforated rolls which were the subject of the controversial decisions in Boosey v. Whight (1900) 1 Ch. 122 and White-Smith Music Publishing Company v. Apollo Company (1908) 209 U.S. 1, could be heard when played on a pianola.

That leaves the difficult question of whether the programmes, as fixed into the ROMs in object code or machine language, are adaptations of the literary works constituted by the programmes in source code. The definition of "adaptation" in the Act is an exhaustive one. Although it is set out in the other judgments, I feel the need to set it out again. It is as follows:-

"In this Act, unless the contrary intention appears - "adaptation" means -

(a) in relation to a literary work in a non-dramatic form - a version of the work (whether in its original language or in a different language) in a dramatic form;

(b) in relation to a literary work in a dramatic form - a version of the work (whether in its original language or in a different language) in a non-dramatic form;

(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) a version of the work in which a story or action is conveyed solely or principally by means of pictures; and

(d) in relation to a musical work - an arrangement or transcription of the work;"

The immediately relevant part of the definition is found in sub-para. (c) (i) which constitutes a translation of a literary work an adaptation thereof. I see no reason to give the word "translation" any narrow or restricted meaning. Nevertheless, the context in which it appears in the definition of "adaptation" and the wider context in which the word "adaptation" itself is used in the Act must be taken into account in coming to a conclusion as to what it means. Generally speaking a translation denotes a change from one form to another. Often, this will be in relation to language, speech or writing. But a meaning of the verb "translate" in the Shorter Oxford Dictionary is, "To interpret, explain; also, to express (one thing) in terms of another. . . " The corresponding meaning of the noun "translation" is, "The action or process of turning from one language into another; also, the product of this; a version in a different language . . . The expression or rendering of something in another medium or form." It also means "transformation, alteration, (or) change." Literally then, it is by no means an abuse of language to describe the conversion of the source code into object code or machine language as a translation.

But as I have said, one needs to take into account the context in which the word is used both in the definition and in the Act itself. It is to be observed that paras. (a), (b), (c)(ii) and (d) are all plainly concerned with adaptations of literary works which will themselves be capable of being seen or heard. They encompass versions of literary works in non-dramatic form in dramatic form, versions of literary works in dramatic form in non-dramatic form, versions of literary works in which the story or action is conveyed solely or principally by means of pictures, and, in relation to a musical work, an arrangement or transcription of a work. It would seem odd to me if the draftsman had intended that the translation of the work to which he referred in sub-para. (c)(i) was to encompass something which could neither be seen nor heard. I do not say that it is impossible that that was his intention, but it would seem to me to be unlikely that it was.

My view in this regard is confirmed by a consideration of a number of the principal sections of the Act, particularly ss. 31, 32, 33 and 36 earlier referred to. There seems to be running through these various provisions the idea or notion that what is the subject of copyright (whether a work or an adaptation thereof) will, although not immediately published and perhaps never published, be capable of being published and thus being seen or heard. The very idea of publishing is that something should be seen or heard. The same may be said in relation to performing a work, broadcasting a work, causing a work to be transmitted to subscribers to a diffusion service and, I think it follows, reproducing a work; see sub-sec. 31(1). In short, adaptations of literary works, like literary works themselves, must, in my opinion, be capable of being seen or heard.

In those circumstances I have reached the conclusion, not without some hesitation, that the programmes in object code are not adaptations, that is, translations of the programmes in source code. Notwithstanding that my sympathies are with the appellants, I regret to say that I think the fact must be faced that the legislation in question does not extend to protect the appellants in the present circumstances. If what had been copied had been the programmes in source code, I would have had another view, but that is not what occurred. To some, this may seem to be a distinction without a difference, and I can understand that point of view. But in my opinion, the draftsman of the Act did not contemplate this kind of piracy when he drew the relevant provisions.

In reaching my conclusion I have not left out of account the fact that it is possible, as was indeed done before his Honour in the present case, to retrieve or resurrect the mnemonics of the source code by manipulating the keyboard. That I regard as irrelevant. It has nothing to do with any act done by those responsible for the manufacture of the Wombat computers. If it had been established that they had retrieved the programme in that way and then used the programme in source code to do what they did the result may have been different. I express no concluded view. But that is not what they did and the matter is not therefore appropriate to be taken into account.

I have also taken into account the reliance placed by senior counsel for the appellants upon certain paragraphs in one of the Apple manuals which inform an Apple user how to obtain access to the mnemonics of the programmes in question. The purpose of this was explained by the first appellant's Australian General Manager, Mr. Strong, in the following evidence:-

"Is there any usefulness in being able to examine the contents of ROM? ---- Yes, most definitely. The various sub-routines being small programmes that are stored in the ROM memory are worthwhile to examine, firstly, to determine exactly what they do and how they do it which is an invaluable way of determining good programming technique. Further, should a programmer wish to produce a very similar routine but not identical then the best way to do that is to examine the routine you wish to copy and display it on the screen and then subsequently make the small changes necessary to it."

Counsel said that the user was thus intended to be able to resurrect or retrieve at least parts of the two programmes in question. I agree that that is so, but have difficulty in understanding how this can be relevant to the case which the appellants make. That case is that the copying of the Apple ROMs by the Wombat manufacturer was an infringement of the first appellant's copyright. No question of the retrieval of all or part of either programme was involved. The vice of what the manufacturer was alleged to have done was to have copied the Apple ROMs for the purpose of facilitating the operation of the Wombat computers.

For the reasons I have given the ROMs are not, in my opinion, adaptations of the programmes in source code. It follows that there can be no infringement of the first appellant's copyright. The consequence is that the appellants' claim in copyright should fail. It is thus unnecessary for me to express a view on the submissions which were made concerning knowledge for the purposes of the operation of ss. 37 and 38 of the Act.

Before leaving the question of copyright, I wish to make this clear. The views I have expressed relate only to the programmes in question in this case. Those programmes are operating systems programmes and form the permanent memory of the computer. The position may be entirely different in relation to application programmes such as the accounting programme earlier mentioned. I do not express any view on this. Before one could do so, one would need detailed technical evidence about what was involved. Moreover, the position may not be the same in relation to all application programmes. And it may vary depending upon the circumstances of each case.

It seems to me that the learned primary Judge did not intend his decision either to apply to programmes other than those about which he had heard evidence. Although the matter is not spelt out in terms in the passage from his judgment cited by Fox J. (p. 4), I think it follows from the way he has cited the various authorities to which he has referred and from some of his discussion about them. I refer particularly to his mention of Northern Office Micro Computers (Pty.) Ltd. v. Rosenstein (1982) F.S.R. 124 and Sega Enterprises Limited. v. Richards (1983) F.S.R. 73, both of which concerned application programmes. In each case, although in Sega only on an interlocutory basis, it was held that an infringement of copyright had occurred. Nothing that the learned primary Judge has said leads me to think that he thought that those cases would necessarily have been decided differently under the Australian Act. That is certainly my own view. Whether they have application here or not is not a matter which now arises for decision.

As to the claims made pursuant to the various provisions of the Trade Practices Act, I have reached the conclusion that the appellants are entitled to relief pursuant to the provisions of s. 52 and paras. 53(c) and (d). I agree in substance with the reasons of Lockhart J. I would only add that in my consideration of the problem, I have had regard to the fact that the claims made pursuant to paras. 53(c) and (d) involve the appellants establishing, albeit in civil proceedings, the commission of a series of misdemeanours. The oft referred to caution in Briginshaw v. Briginshaw (1938) 60 C.L.R. 336, must be observed. I am nevertheless satisfied that the appellants have made out their case.

The injunction to which in my view they are entitled may require some consideration. That is particularly because the restraint to be imposed will be one which prevents the respondents from selling Wombat computers and, at the same time, giving or making available to purchasers copies of the Apple manual. The form of the injunction was not the subject of argument before us. My judgment in relation to the copyright claim will not prevail. The other members of the Court consider that an injunction to restrain continued infringements of copyright should go. Of necessity, that injunction will be in terms wider than any injunction under the Trade Practices Act. I do not therefore consider it necessary to attempt to formulate an appropriate injunction in these reasons. I would prefer to allow counsel to consider the reasons of the Court and to speak at a later time as to the form of orders to be made in relation to the claims under the Trade Practices Act upon which the appellants have succeeded.

In the result I would allow the appeal in so far as the Trade Practices claims are concerned but only to the extent indicated by what I have said. The appeal in relation to the copyright claims should be dismissed.

[1984] FCA 137

29 MAY 1984