Data Access Corp v Powerflex Services Pty Ltd
[1999] HCATrans 3
IN THE HIGH COURT OF AUSTRALIA
Office of the Registry
Melbourne No M113 of 1997
B e t w e e n -
DATA ACCESS CORPORATION
Appellant
and
POWERFLEX SERVICES PTY LTD
First Respondent
POWERFLEX CORPORATION PTY LTD
Second Respondent
DAVID MEREDITH BENNETT
Third Respondent
MARGARET ANN BENNETT
Fourth Respondent
DEMILEIGH PTY LTD
Fifth Respondent
GLEESON CJ
GAUDRON J
McHUGH J
GUMMOW J
HAYNE J
TRANSCRIPT OF PROCEEDINGS
AT CANBERRA ON TUESDAY, 2 FEBRUARY 1999, AT 10.19 AM
Copyright in the High Court of Australia
______________________
MR J.M. EMMERSON, QC: If the Court pleases, I appear with my learned friend, MR A.K. PANNA, for the appellant in this matter. (instructed by Stephens Lawyers & Consultants)
MR R.C. MACAW, QC: May it please the Court, I appear with my learned friend, DR J.F. BLEECHMORE, for the respondents. (instructed by Trumble Szanto Lawyers)
GLEESON CJ: Yes, Mr Emmerson.
MR EMMERSON: If the Court pleases. The Court will have our written submissions. It is convenient to start with consideration of our first proposition which is set out in paragraph 6(a) of our written submissions. That is to say, we say that copyright subsists in each of the reserved words in the Dataflex language on the footing that each of them is a computer program within the definition in section 10 of the Copyright Act. We have set out that definition in our submissions.
GUMMOW J: And therefore, a literary work, you have to say.
MR EMMERSON: Exactly.
GAUDRON J: And you say that divorced from the underlying code, as it were?
MR EMMERSON: We say that divorced from the underlying code, yes; we say that each of the words is itself an expression and that expression has been taken.
HAYNE J: What is the set of instructions that is referred to in the definition of “computer program” in this respect?
MR EMMERSON: The set of instructions is the meaning and the syntax of the reserved word in each case.
McHUGH J: But this would lead to serious problems with computer programming generally, would it not, because, take a program like Access, with which I am familiar with, the syntax may be, “select expression” or “select from”, “where”, “order by”, for example, which will be a chain of instructions. Now, what about the word “select” or the word “where” or “from” or the phrase “order by”, are they expressions for this purpose? Are they programs?
MR EMMERSON: In this case each of the words, we say, is a program but we are not seeking to divorce these words from the meaning and syntax that they have in the context of data access. So, our complaint here is not merely a taking of the words simpliciter but a taking of the words, giving them the same meaning and giving them the same syntax and we say that when all of that is combined, then we have all the elements of the definition of “computer program” which appears in section 10 of the Act and that all of that has been taken.
HAYNE J: What is the set of instructions?
MR EMMERSON: The set of instructions, we say, is the meaning and syntax of each of the words so that each of the words is a way of expressing that set of instructions.
HAYNE J: Therefore the set of instructions is the function that is thus performed?
MR EMMERSON: The set of instructions is, as I say, meaning and syntax and in that sense it is the function that is performed.
HAYNE J: What meaning then, in the definition, do you give to the words “a set of instructions” as distinct from the last words in the definition, “to perform a particular function”?
MR EMMERSON: We say that the performing of the particular function is something which is here taken. The set of instructions is, as the majority held in Autodesk, capable of being expressed in a number of different ways but we say that what one has got with the definition of “computer program” that occurs in the Act is a practical way of looking at computer programming so that one is protected where what is taken is, in this case, both the expression and the meaning and it is our submission that that is sufficient to amount to an infringement. Now, the matter, of course, comes before this Court in a sense at an earlier stage of the argument, namely that the Full Court held that copyright did not subsist although it also held that if copyright had subsisted there was strong reason to say that the respondents had infringed it but it was on the point of subsistence that my clients succeeded before the trial judge but failed in the Full Court.
GAUDRON J: When you use the word “meaning”, can you give me a definition of that in the context in which you use it because what I really want to know is if you are using the word - the meaning of these words to mean the ultimate function performed?
MR EMMERSON: Well, I am using the word “meaning” comes to mean what the command does and for that - it is a matter of what it does. As the Court is, of course, only too well aware, the word “function” is a word from which claimants under copyright tend to shrink because it has a certain amount of intellectual baggage attached to it. If we can seek to put aside that intellectual baggage, one can say, “Well, of course, in the context of a computer program, the computer program must perform a particular function”. That is part of the definition of “computer program” and it is in this respect that the 1984 amendments which introduced protection for computer programs into copyright law are making somewhat of a shift in the concepts and usages that one has to follow.
GUMMOW J: There is this expression “intended to cause performance of a function”.
MR EMMERSON: Yes, exactly.
GUMMOW J: That is all revolutionary, really.
MR EMMERSON: It is, with respect, all revolutionary. And what we say, in a nutshell, is that if what is taken is the expression and the performance of the same particular function, then we say that in the way that computer copyright works, it is not to the point that at some stage or stages in between there may exist a code which does not look objectively similar.
HAYNE J: Does that mean, therefore, that copyright could exist in the word “print”, that is, you use the word, the expression, if you will, “print” which leads to the printer printing. The fact that there may be very different electronic impulses generated as a result of the entry of that word, does that matter?
MR EMMERSON: There are a number of things which could be said about that, yes, your Honour. Of course, in order for copyright to subsist the program itself must be original.
GUMMOW J: What is the force of this word “original” from section 32 transposed into computer program copyright?
MR EMMERSON: The force of the word “original” we would say, is the same as the traditional force, namely, that what is required is that the program should originate with the author. That is to say there is no novelty requirement in our respectful submission, but there is a requirement of origination.
GLEESON CJ: What is the difference between a set of instructions to cause the performance of a function and a description of the function?
MR EMMERSON: The difference is that you can describe each of these in a variety of different ways. The two expressions can be looking at different stages of the process or they can be looking at the same stage of the process, so one could get a form of words which satisfied both definitions, but they are not logically the same thing because the set of instructions is looking at the origin of the computer program whereas performing a particular functions is looking at the end use of that program.
McHUGH J: But you are claiming copyright in respect of each word. That is what I have difficulty – could I just put to you a very simple program? Supposing I have a database with names of people and their suburbs and I want to add area code. Now, if I am in Access I would have to use three words: update, set and where, and I would have to add a field for area code. My program would be then update names, where, set area code equals 2027, where – suburb equals Point Piper. Now, is each of those three words, update, set and where, a program?
MR EMMERSON: Yes, it may be, but to continue the answer that I was giving to his Honour Justice Hayne, there are built into the Act a number of safeguards against, as it were, protection being sought for the merely trivial and I was putting to his Honour one of the requirements which is the requirement of originality.
There is also of course the requirement that what is protected must be a work, and so that involves the expenditure of an element of skill and labour, which was found by his Honour the trial judge to have been present in this case. He said that there was substantial work in each of these words. So that requirement is satisfied. Then finally of course copyright itself does not confer a monopoly on the use of a particular expression. It merely requires that other people should not copy it. The notion of copying has in it not merely similarity in form but also the requirement that one be derived from the other. So, as with other branches of copyright, any possible fears about monopolisation of simple expressions simply does not arise because, unless there is copying, there is no infringement.
So it is our submission that in the present case all the usual safeguards that are built into copyright law and specifically into the Copyright Act will also operate. Where you get infringement, we say, is where, as it has happened in this case, that not only have the expressions been taken but they have been taken from our program with precisely the same meaning and syntax. It is at that stage that we say infringement arises.
GUMMOW J: Do you say you have a compilation of computer programs?
MR EMMERSON: That is an alternative way that we put the matter, your Honour.
GUMMOW J: Within that definition of “literary work”?
MR EMMERSON: Yes. So that we say that we have copyright in the reserved words and macro commands taken individually but, if we are wrong on that, we have nevertheless got a compilation here which is capable of being protected.
GAUDRON J: Does compilation have any requirement of a particular order or sequence or conjunction?
MR EMMERSON: We would say not, your Honour; that is not a word which has been considered by the courts in this context. It comes into the Act by way of the 1984 amendments, which were, of course, introduced after the Computer Edge Case, where a new definition of “literary work” was introduced into the Act to include “a computer program or compilation of computer programs”, but, at least so far as our reading is concerned, there has not been consideration by the Court of that latter point. It would seem likely that the word “compilation” here would be used in a rather wide sense, because one can get various sets of circumstances in which groups of computer programs might be brought together and we would say that all of those were compilations.
The learned trial judge dealt with this first point in particular at page 1337, starting at line 11 on that page and going through to line 4 on page 1338. His Honour states his conclusion at the start and then his Honour, in the second sentence says, as we would submit, that where you have a taking, both of the expression and the detailed meaning and syntax, it does not matter and, indeed, it is fairly common place that you have different forms of code at some stage in between. As his Honour says:
At the level of abstraction under consideration –
this is namely at the highest level of abstraction –
the objective similarity is complete.
And he then goes on to say that each is an expression of a –
set of instructions intended to cause the –
device to perform a –
particular function – - -
GAUDRON J: But does not that beg the question, in a sense, whether each is in fact a set of instructions? If you assume that each is a set of instructions, what his Honour says is perfectly correct.
MR EMMERSON: Well, each is an expression of a set of instructions. I have to draw that distinction because the Act itself draws that distinction.
GAUDRON J: Yes.
MR EMMERSON: In the definition of “computer program” what is technically protected is an expression of a set of instructions. On the language of that definition, it might appear that the set of instructions itself was not protected, but your Honour will recall that in Autodesk 1, the majority said that it was necessary to give a workable definition to this, and that there was therefore protection for every way in which that set of instructions could be expressed.
Here we have a different case because, unlike the case with Autodesk, here, the expression itself is taken. With Autodesk there was a problem as to whether the expression itself was taken.
HAYNE J: Could you, at some point in your argument, give me a simple example from this case of the expression and the set of instructions that is in issue?
MR EMMERSON: Yes.
HAYNE J: Identifying what are the instructions or what is the set of instructions?
MR EMMERSON: Yes, if your Honour pleases. We say that each of the reserved words and each of what are referred to in the papers as the macro commands is itself a set of instructions. The reserved words are set out in the First Schedule to the order made by the learned trial judge which is found in volume 8 of the appeal papers at 1382 and going on for the next couple of pages. The Second Schedule at 1385 sets out the macro commands.
Now, for the purpose of this part of the argument there is no relevant difference between the First Schedule words and the Second Schedule words. There is, however, a separate argument which I will come to in due course which applies to the Second Schedule words.
HAYNE J: Well, yes, what is the set of instructions? My question is, if you go to 1382, take any of the words in that First Schedule, can you give me an example of the set of instructions that is expressed by the word?
MR EMMERSON: Yes, the set of instructions is what the computer is told to do when these words in the context of the program are keyed into the computer.
HAYNE J: Yes, but at some point convenient to you, could you give me a simple, concrete example of it? That is, that if you take the word, say “clear”, can you identify for me precisely what is the set of instructions thus expressed, for may I say to you that my present impression, perhaps quite wrong, is that the expression “clear”, say, is of a set of instructions to create certain electronic impulses. In one program a particular set, in the other program a very different set and my difficulty at the moment, a difficulty of understanding no doubt, is to identify that the expression is of the same set of instructions. That is the best I can do to make plain my difficulty. You will deal with it, no doubt, at whatever time you find convenient.
MR EMMERSON: Yes. The set of instructions at that level is the meaning and syntax of these words. That is to say, each of these words has a definite meaning. It tells the computer to do something. It also has to do so under tightly defined conditions because this is the nature of computer programming. So, that meaning and syntax, we would say, are the set of instructions. That set of instructions is variously expressed at different stages. At the highest level, which is the level that we are concerned with, it is expressed by the reserved words and the macro commands that are in the schedules.
McHUGH J: Well, take a simple item in that First Schedule on 1382, “clearscreen”. I imagine what it does is it just clears the screen so it becomes blank.
MR EMMERSON: Yes.
McHUGH J: Now, the words “clearscreen” either is “cl”, “scr” or something similar is used in numerous types of programs. Do you claim that the word “clearscreen” in that First Schedule is a program for the purpose of the definition?
MR EMMERSON: Yes, I do, your Honour, and if the concern is that other people cannot use ordinary words then we say other people can use ordinary words but what they are not entitled to do is copy our expression together with the same meaning and syntax.
McHUGH J: But take a word like “loop” and “insert” and “total” and “save” I notice are among these words.
MR EMMERSON: Yes.
McHUGH J: They are found - “report” in the Second Schedule, or “enter”.
MR EMMERSON: Yes.
McHUGH J: You find those, in my limited knowledge, in a half a dozen, a dozen scores of programs, and they all perform the same functions.
MR EMMERSON: Well if people find them elsewhere they can use them. There is no difficulty. If they invent them themselves they can use them.
McHUGH J: But it is almost as if it has moved into the language of the computers, that it is just part of the body of knowledge that computer programmers generally would ‑ ‑ ‑
MR EMMERSON: With respect, your Honour, we are here dealing with a computer language which is devised by my clients for the very special purpose of the type of data management programs that we are here concerned with. It is a language. The totality of the words amounts to a language and part of the work that went into it was devising a language which was suitable for use for this particular purpose and which was a so‑called fourth generation language so that it was one which it was as easy to use as it was possible to make it.
So, yes there is a language here, and yes, we do say that we have copyright in that language, because we say that there is no reason why one should not. Other people are free to develop their own language. What they are not entitled to do, in our submission, is copy ours. And, of course, what has happened in this case is that the respondents simply have copied our language. So, in essence, in both the major parts of this appeal we are saying we have copyright in the language. The first proposition that I have been dealing with we say we have copyright in the words taken individually. In the second part we say, “Well, if that is wrong, we still have copyright in words taken collectively”. These are points which come further down.
But to say that some of these words may be familiar, or other people may want to use them for purposes independent of copying our system, is of course something that we would accept. But we say that if people derive them from somewhere else or invent them themselves, that is absolutely fine. It is the copying which we say is prevented by copyright law.
Could I also just add one thing: that although, of course, we accept that some of these words look fairly familiar, the fact is there is also a body of words which is not familiar. There is a body of some 57 words which are not found anywhere else.
HAYNE J: What is the significance of that? Does that mean that your argument divides, that there are additional arguments you deploy in respect of the uncommon expressions or words?
MR EMMERSON: It is rather, with respect, your Honour, that some matters which could be urged against us with some of the words simply cannot be urged against us in respect of the 57 words which are unique to the Dataflex language. The logic in each case is the same, but where it is said against us, “Well, some of these words are used elsewhere”, then we give a series of answers to that. We say, “Well that does not matter, independent creation is protected”, and so on.
HAYNE J: But if your argument is good, it is good in respect of all of the words, not simply those that are not found elsewhere.
MR EMMERSON: Yes.
HAYNE J: And if your argument is bad, it fails for all, does it?
MR EMMERSON: If our argument is bad on the ground that for some reason it fails to work where somebody else has used the words before, and we dispute that, but suppose it is bad on that ground, then it cannot, in our submission, be bad on that ground in respect of the 57 words which have never been used before. But the logic of the argument and our broad contentions are, of course, the same for both these types of argument.
McHUGH J: But your principal argument seems to depend upon the subjective state of mind of “the infringer”. Is that the way infringement works in copyright?
MR EMMERSON: There are elements of subjectivity but this is not a case in which those arise, with respect, your Honour. If I have suggested that the test for infringement was subjective in that sense, then I have been incorrect. The test for infringement is a test of unauthorised copying, and copying itself has two elements. One is the element of objective similarity and the other is the element of derivation. Could I, while on the subject, give what is a rather standard reference to that. That is in the Computer Edge Case (1986) 161 CLR 171 at 186 in the judgment of the then Chief Justice, Sir Harry Gibbs. If the Court goes to the passage a little bit above the halfway mark on that page, his Honour says this:
The notion of reproduction involves two elements – that the infringing work sufficiently resembles the copyright work and that it was produced by the use of the copyright work.
Those are the two elements, and of course they have to be present together and together they bring about the notion of copying.
GUMMOW J: But there is a debate in the music cases, is there not, about unconscious copying?
MR EMMERSON: There is some debate in music cases and indeed in some other cases about unconscious copying and the position seems to be that it is not necessary that copying should be conscious in the sense that someone might remember a passage of literature or a passage of music or, I suppose, a drawing and reproduce it, albeit that at the time that the reproduction occurs that person does not remember where the passage came from and hence the notion of unconscious plagiary and so on.
McHUGH J: Does the second element in what the Chief Justice says at 186 apply also to the publication of a work as opposed to the reproduction of a work, or the adaptation of a work? All of which are acts which are comprised in copyright.
MR EMMERSON: Yes. There must in every case be the necessary causal connection between the copyright work and the infringing copy and that is true if one works right the way through section 31 dealing with the various protections that are given in respect of copyright. Now, the result of all of that to round out what I was putting in answer to a question from your Honour Justice McHugh is that the test is not really subjective but there is a test of derivation and that no doubt involves the mind of the infringer, but there is no test that requires the Court to ask, “Did the infringers say, ‘Ha, I am going to infringe’”.
McHUGH J: Correct me if I am wrong, but if I understand you correctly, what you say is if somebody sits down and maybe from his or her head puts down a set of instructions which may replicate your program in detail and word for word, that there may be no infringement - or there will be no infringement in that particular case, but if they sit down with Dataflex in front of them and copy it, then there is.
MR EMMERSON: That is precisely right, yes, your Honour, and there are cases which make that point, perhaps fairly strikingly. For instance, one can have copyright in a set of logarithm tables albeit that if you sit down and properly calculate a set of logarithm tables you would always get the same numbers. What other people are not entitled to do, if you have copyright in a set of logarithm tables, is sit down with your set and copy them. People can do their own work independently and they can get exactly the same result but they will not be infringing.
McHUGH J: If I sit down on Central station in Sydney and clock the time each train leaves and I produce a program showing the times the trains leave, the Department of Railways has no cause of action against me even though my times may be identical with their timetable.
MR EMMERSON: Precisely, and again in the ‑ ‑ ‑
HAYNE J: Obviously a New South Wales comment where the trains run on time.
MR EMMERSON: There is, in fact, a 19th century case in which there was copyright infringement in Bradshaw’s Railway Guide which is a case directly in point and what was held there in accordance with principle, is that, yes, of course you can make your own railway timetable but what you cannot do is copy someone elses. In each case the aim of the law, in our submission, is to protect the labour and skill which the copyright owner has put into the developing of the copyright work and the moral rational is to say other people are not entitled to take that or to take the consequences of that exercise of labour and skill.
Could I now refer to our outline of argument and one or two additional references which we say ought to go in. In paragraph 17 at the top of page 7, we say that it was common ground that each of the reserved words and macro commands in dispute has the same meaning and the same syntax in the Dataflex language and the respondents’ language. The reference that we there give at page 1406 of the appeal book is to the Full Court. The matter is dealt with by the learned trial judge at pages 1336 to 1337 in his reasons for judgment. Also in respect of that paragraph, we would invite the Court perhaps to begin a little bit earlier than the passage that we have quoted. The mode of operation is dealt with by the Full Court at page 1409, in the paragraph running from line 7 to line 15. The court says:
the Dataflex program contains both a compiler and a run program.
The difference being that you use the compiler when you are setting up the system and you use the run program when you are using the system. The court goes on to say:
In other words, the compiler program recognises the particular command used in the simple program and substitutes for that a set of instructions written in source code. When the program is then run with the runtime program, typing in the particular command will send instructions to the computer, ultimately in assembly language, for the computer to perform the series of operations corresponding with the command.
So that is the way the system operates.
GUMMOW J: Now the Full Court is reported in 75 FCR 108 and the passage you have just taken to us is at 112 letters E to F.
MR EMMERSON: I am indebted to your Honour. I will perhaps consider over the adjournment whether we can somehow get a glossary which enables one to go from the appeal book to the report in the Federal Court Reports.
Another short passage from the Full Court, which is relevant to paragraph 17, is to be found at page 1412, lines 11 to 15, the court having said that:
A reserved word can only be used in connection with the program in accordance with the rules of the language of that program. In essence what this means is that the Dataflex program is written so that when one of these reserved words is used, an instruction is sent to the computer to perform a defined set of operations. The word can not be used in any way other than in accordance with that set of instructions.
Now, in paragraph 18 we have set out why we say that each of the reserved words and macro commands satisfies the statutory test, the definition of “computer program”.
At paragraph 21 we start dealing with the reasons why we say the Full Court went wrong in considering this issue. It is perhaps convenient to start a little bit earlier in the reasons for judgment of the Full Court and I would invite the Court to turn to page 1413 in the appeal book. The court having dealt with the reserved words and having said at lines 7 to 14 on that page that there were:
fifty five words used in the Dataflex language which were in fact unique to the Dataflex program.
The court then says, at line 15:
It goes without saying that, but for ease of usage, the precise words used in a particular computer language are irrelevant.
Now, in our submission, one can discern here the seeds of one of the errors into which we say the court fell. The precise words used in a particular computer language are what we say is protected by copyright. We say that a lot of work and skill went into choosing the words that we have used. The words, it may be said, are irrelevant in the sense that they do not have a necessary connection with the function that the program ultimately performs so you could choose other words to be your expression for a particular instruction or set of instructions and the court says, “You could use “XZB” and the answer is yes, you could.
However, what we complain of is that that was not what the respondents did. They did not say, “Right, we will invent our own language. We will do the work and we will determine a set of words which can make our computer language without copying the ones that are Dataflex’s computer language.”
Now, if they had done that then the particular complaint of taking the expression in each of the words in the Dataflex language would not arise but they have chosen not to do this. They have taken this thing and we say that that expression is protected and that to start with the proposition that the precise words used are irrelevant is to go in the wrong direction. The precise words used are highly relevant to the issues in this case. They are highly relevant to the work that we did in designing the Dataflex language and the taking of the words is a central aspect of our complaint.
HAYNE J: What is the relevance, then, that you attribute, say, to the choice of the word “total” as opposed, say, to the choice of the word “sum”? You have said that the choice of the word was highly relevant.
MR EMMERSON: Yes.
HAYNE J: What do you mean by that, by reference to that or any other example you care to choose?
MR EMMERSON: I am sorry, was your Honour taking the word “total”-
HAYNE J: From 1384.
MR EMMERSON: From 1384, right. Yes.
HAYNE J: What is the relevance of the choice of that word? You said it was highly relevant. I do not understand.
MR EMMERSON: It is highly relevant to the work that we did in setting up an expression of “a set of instructions”.
HAYNE J: The work you did in the set of instructions was to devise the program that would result upon entering of the word “total” in, I take it, the summing of a list of fields or values in lists of fields or whatever it may be. Perhaps my example is askew but the point perhaps is clear. Is not that the work that was done rather than choice of “total” rather than “sum” or “total” rather than “add” or some other expression?
McHUGH J: Could I just add to that that in “Dbase” they use “sum”, Access uses “total”. Both get the same result.
MR EMMERSON: Yes. In answer to the question from your Honour Justice Hayne we would say that both require a large amount of work, that it is, with all respect, asking the wrong question to say which was the work as if they could not both involve work.
HAYNE J: But my question was provoked by your proposition that the choice of the word was highly relevant. I do not understand the proposition.
MR EMMERSON: The choice of the word is highly relevant because it is our choice of the expression which is the computer program. So, so far as the issue before the Court, where the Court is considering protection of an expression of a set of instructions, then the choice of that expression is highly relevant, it is the central issue. Can I take an example which might help to clarify what we say here? You can imagine circumstances in which you were compiling a work of reference and a vast amount of work went into compiling the work of reference in finding out facts from various sources, but having found the facts there could and probably would also be a substantial amount of work in expressing the facts that you had found. Now, the amount of work in expressing the facts that you have found may be very much smaller than the amount of work that you would do by way of background work.
But that is not to the point. You would nevertheless get protection for your expression of those facts even if you could not prevent other people from using those facts. This, in our submission, is a distinction which appears throughout copyright law. Could I also give another reference to what we say in paragraph 24. This is dealing with our proposition that in order to be a computer program within the meaning of the Act, it is not necessary that the expression should be a computer program in the ordinary or conventional sense. We at the bottom of page 9 say it is not limited to computers at all. The expression “a device having digital information processing capabilities” is apt to describe a great many devices which are not computers.
GUMMOW J: That agitates the respondents, that statement.
MR EMMERSON: It does agitate the respondents, your Honour. Could we perhaps have a look at the explanatory memorandum.
GAUDRON J: I mean, is it so controversial? My washing machine has some digital information processing capabilities.
MR EMMERSON: Yes.
GAUDRON J: But that brings me to this point. The set of instructions associated with my washing machine is activated by my pushing a button. Are not the words with which this case is concerned the equivalent to my pushing the button rather than the set of instructions?
MR EMMERSON: We would say that one can give a set of instructions by pushing a button or buttons. We would say that the parallel ‑ ‑ ‑
GAUDRON J: Do you give the instructions by pushing a button or does the pushing of the button activate the set of instructions?
MR EMMERSON: Well, you could imagine circumstances in which either or both of those propositions would be correct, with respect. There always has to be a way of a human being communicating with a computer. You communicate a set of instructions in one of a relatively limited number of ways: either pushing keys on a typewriter, which would be pushing buttons, or using a mouse and perhaps clicking that, which would be pushing a button, and so on. In our respectful submission, one does not solve the issues in this particular case by saying, “Is it or is it not a pressing of a button?”. We say that what you have to do is first ask, “Has there been an expression?”.
GAUDRON J: Of a set of instructions.
MR EMMERSON: That is right, yes.
GAUDRON J: But that is a composite phrase, is it not?
MR EMMERSON: Yes, yes it is.
GAUDRON J: Yes, the pushing of the button can itself be an expression, if you like.
MR EMMERSON: Yes.
GAUDRON J: But there is the further question whether it is the expression of a set of instructions.
MR EMMERSON: There is the further question, your Honour, though it is not a question which actually arises in this case at this stage.
GAUDRON J: Why not?
MR EMMERSON: Because we are not, in this appeal, agitating the issue as to whether the pressing of a button amounted to a giving of a set of instructions. It was an earlier issue in this case, but that has not survived to this Court, so we are not dealing with that.
McHUGH J: What do you say about the word “set”? Supposing I have a row of figures in a spreadsheet and I want to find out the standard deviation and what I have to do is type in, say, “STR” as the opening command.
MR EMMERSON: Yes.
McHUGH J: Now, is that a set of instructions without identifying what the file or the field is that it is to operate on?
MR EMMERSON: Yes, it is because it defines what is to be done. When you are actually running a computer program, of course, you have a program and you have various things that it operates on, and in order to get a useful result you must, of course, have both. But the set of instructions is nonetheless a set of instructions, notwithstanding the time when it is prepared the argument that it is going to operate on has not yet been defined.
GUMMOW J: What were you seeking to draw from the Explanatory Memorandum?
MR EMMERSON: I was seeking to draw from the Explanatory Memorandum part of what we say in response to something which agitated our learned friends, namely that the words used in the definition of “computer program” are “a device having digital information processing capabilities”. We said that is not limited to computers, and we say that primarily on the language that is used. If Parliament had wanted to say that it was talking about computers, it would be easy to say so. It did not. But we also say that we are supported in this in the Explanatory Memorandum ‑ ‑ ‑
HAYNE J: Do you rely on paragraph 22, page 7?
MR EMMERSON: I am indebted to your Honour. The Explanatory Memorandum in paragraph 22 says this:
The phrase “having digital information processing capabilities” is intended to make clear that the device is not a device which merely processes information by analogue methods (e.g. a radio) but does include devices which, though considered as a whole might not be information processors, nevertheless have some such capability. Examples would be computerised telephone switching equipment and computerised ignition systems.
Presumably in motor cars. So it is clear not only from the language, but also from the Explanatory Memorandum, that the net is to be cast widely and protection is to be given to subject matter which would not ordinarily be regarded as a computer program because it would have nothing to do even with a computer.
Nevertheless, what is intended here, and what is the definition of the definition in our Act, is to get protection from ‑ ‑ ‑
GUMMOW J: How does this fit in with the circuit layout legislation which somehow stands on its own. If it had some interrelation - - -
MR EMMERSON: It rather stands on its own. The difference is that the circuit layout has a different type of expression which is ‑ ‑ ‑
GUMMOW J: Justice Gaudron’s washing machine has objects in it which fall within the Circuit Layouts Act, does it not?
MR EMMERSON: Yes, and your Honour is putting to me that, therefore, her washing machine may not be the subject of copyright because the Circuit Layouts Act would deal with the matter comprehensively?
GUMMOW J: Yes.
MR EMMERSON: Your Honour, I have not had occasion to think of that question specifically but it would seem on the face of it that, yes, Justice Gaudron’s washing machine might perhaps lie outside this for reasons irrelevant to the case but relevant to the provisions of the Circuit Layouts Act.
Could I now go on to what we say in paragraph 26 of our outline of argument? One matter considered by both courts and also dealt with by our learned friends in their written submissions is the relevance of United States authorities and we have set out in paragraph 26 what we say, broadly speaking, about those.
GUMMOW J: The United States definition does have this expression, does it not, “A set of statements or instructions”? It uses the expression “A set”; “A set” of instructions?
MR EMMERSON: Yes.
GUMMOW J: So, there is that - - -
MR EMMERSON: The details of the wording are, of course, quite different.
GUMMOW J: Well, to “be used in order to bring about a certain result”; it is not all that different, really, is it?
MR EMMERSON: Well, the first thing is that as is important in our law, the United States definition is not a protection for an expression of a set of instructions, which is our language. That is important in this case because, of course, it is the expression taken here in the context of the other matters that I have mentioned which we complain about. So, while they have the expression “set of statements or instructions”, they do not have the notion of expression. They also use different language in saying it is intended “to bring about a certain result” as distinct from causing a device having information processing capabilities to “perform a particular function”.
Now, no doubt there is room for argument about the precise ambit of what the United States definition means but we would say that the language is different and, on the face of it, it is not going to be helpful to construe the expression “computer program” in our law by reference to what United States courts have held about the meaning of the expression “computer program” under their definition.
GUMMOW J: So, is there some requirement of reciprocal protection under our law and their law?
MR EMMERSON: There is, of course, the requirement under the Convention that protection can be given or must be given.
GUMMOW J: Yes.
MR EMMERSON: But that requirement does not mean that we have to follow the precise workings of their law. The law remains a domestic one in each case, not governed by international convention but there are certain requirements for equivalent treatment of nationals of different countries and matters of that sort which must be there.
GUMMOW J: Yes, now what is the origin of your client’s program? Is it protected under this Act only because the Act picks up international arrangements?
MR EMMERSON: Yes, necessarily so; that is to say the author is a qualified person under our Act, because he is a resident and citizen of the United States of America and because the United States of America, by our Copyright International Arrangements Regulations, is one of the countries for which we give this sort of protection.
GUMMOW J: Yes, under section 184, under regulation 184.
MR EMMERSON: Yes.
GUMMOW J: So are you saying it does not matter whether or not this would fall within the US definition?
MR EMMERSON: No, it would not matter at all. The way the Copyright International Arrangements legislation runs is as follows. If you look at ‑ ‑ ‑
GUMMOW J: I mean, this is really the starting point.
MR EMMERSON: Absolutely, yes.
GUMMOW J: In this case.
MR EMMERSON: Yes, yes. Well, it is the starting point except that it is not, with respect, an issue in this appeal.
GUMMOW J: Maybe not, but if one is writing a judgment one likes to know one is starting at the right spot.
MR EMMERSON: As a matter of logic, of course, yes. Could I, perhaps, just take the Court to the way the logic works. If one starts with section 32 of our Act, that deals with original works in which copyright subsists and 32(1) is not relevant because it deals with unpublished works. Section 32(2) deals with the position of published works and it says:
Subject to this Act, where an original literary, dramatic, musical or artistic work has been published:
(a) copyright subsists in the work; or
(b) –
copyright would continue:
if, but only if:
and then there are three tests relating ‑ ‑ ‑
GUMMOW J: You say this is (d), do you, “qualified person”?
MR EMMERSON: If your Honour will bear with me, we have to go through two steps.
GUMMOW J: Yes.
MR EMMERSON: If:
(c) the first publication took place in Australia;
(d) the author of the work was a qualified person at the time when the work was first published –
which takes us to subsection (4) and a:
qualified person means an Australian citizen, an Australian protected person or a person resident in Australia.
GUMMOW J: Yes.
MR EMMERSON: Subsection (e) does not come into it. Now then, superimposed on that are the Copyright International Arrangement Regulations which, I am ashamed to say, in the text I have in front of me are not present, but the effect of them is ‑ ‑ ‑
GUMMOW J: Well they are made under section 184.
MR EMMERSON: Yes, and what they do is extend the category of authors for whom protection is given to persons who are citizens of named countries or residents of named countries or to works which were first published in named countries, so all the machinery of section 32(2)(c)(d) and (e) is taken in, but with the effect that if you come to the word “Australia”, it would also include the words “United States of America” and if you come to “Australian citizen” or “Australian protected person” you would also include a citizen or protected person of the United States of America. And so, under that, our domestic law has the effect of automatically extending to persons who are resident in, or citizens of, certain named countries, the same protection as they would get if they were residents in or citizens of Australia. It extends the protection given to works first published in Australia, to works first published in these other countries.
Now the logic of that, though, is that this operates - for the purpose of the Act, American citizens are treated as if they were Australian citizens. These provisions do not however have the effect of bringing American domestic law into our domestic law; it is an extension to their citizens and in respect of place of publication, rather than saying that if something is protected under United States law it will be protected here; that is not the step. The step is via looking at the author or the place of publication and, if it is in the United States of America, that is sufficient.
GUMMOW J: But there are two sides to the coin, I suppose. Section 185 would enable this country to say the foreign state question does not reciprocate sufficiently so we deny here. Presumably there could be mirror legislation in the United States, which would deny reciprocity in respect of us, if we did not provide sufficient protection for them.
MR EMMERSON: One would certainly expect to find limitations on the extent of protection that they would give. The main machinery for all of this is that a foreign country does not get to first base unless it is included in the Copyright International Protection Regulations and, broadly speaking, the countries that are there included are countries where our government has regarded it as being fair and sufficient to give equivalent protection to citizens of those countries.
Section 185, to which your Honour has referred, of course, would allow regulations cutting across that. I am not able to say whether that has ever happened. What, though, in our submission, does arise out of this is that there is no requirement that copyright protection should be coextensive in the United States of America before copyright protection is afforded to United States citizens.
I wanted to say a little bit about one United States case which is discussed by both the judge at first instance and by the Full Court and that is Lotus Development Corporation v Borland International Inc. What I have to say can conveniently be said with reference to the passages from that case which are quoted by his Honour the trial judge at 1338 in the appeal book and following.
GUMMOW J: Now, the correct citation of that is 49 F 2nd 807 and the Intellectual Property Reports are subsidiary series after all, and this Lotus Case went to the United States Supreme Court, did it not, and there the court equally divided and gave no decision. That appears from 133 Lawyers Edition 2nd 610.
MR EMMERSON: Well, I am indebted to your Honour, I had not realised that.
GUMMOW J: It is quite notorious, I think. And, in Nimmer’s Book on Copyright, paragraph 13.03[F](3)(e) there is a lengthy discussion of this. The ninth judge ‘recused himself” after argument and before judgment and that left eight, and they divided four-four and they decided not to give any reasons and they have been much criticised for not doing so.
MR EMMERSON: I see.
GUMMOW J: So, one is left with this decision of the first circuit but that will not be the end of the story, I would have thought.
MR EMMERSON: Well, it seems unlikely to be the end of the story and, indeed, it just gives another reason which we would say would support our submission that the United States law is important in relevant ways different from our own law here. I would just wish to focus though on one or two points. The effect of section 102(b) in the United States code is in our submission significant in looking at the Lotus v Borland Case. That is conveniently set out by his Honour the trial judge at 1338 beginning at line 17. The way that is expressed is it operates as a denial of copyright for an original work of authorship in certain categories and the important category here is the method of operation and that denial applies regardless of the form in which it is described, explained, illustrated or embodied in such a work.
In our submission, not only is there no corresponding provision to section 102(b) in our law but also the way in which our law operates is quite different. In section 32 to which I took the Court a moment ago, our starting question is in the present case: is this an original literary work? If it is and if certain other statutory tests are satisfied, then copyright subsists in it. The way that the United States code works is in certain circumstances to give protection for what they describe as an original work of authorship but then to provide for an exclusion of protection in certain circumstances regardless of the form in which the work is expressed. That particular machinery is, in our submission, no part of our Act and no part of our law.
GUMMOW J: Is not 102(b) in the US really declaratory – putting aside computers for a minute – namely, to make clear what is marked off from patent law, which is all about function, from copyright law, which is about other things? What has happened in Australia with computers is because of this definition, function is injected into the copyright law here.
MR EMMERSON: Yes, your Honour.
GUMMOW J: Here we are.
MR EMMERSON: Absolutely, your Honour. Function is injected in and under our law, at least so far as computer programs are concerned, the line drawn by the United States section 102(b) ‑ ‑ ‑
GUMMOW J: The question in the United States is, I suppose: does 102(b), that general provision, have some limiting effect on the other provisions they put in, I think at a later date, to do with computers?
MR EMMERSON: Yes, the sequence of events was fairly complex in the United States. In Lotus v Borland the court held that section 102(b) had a limiting effect.
GUMMOW J: On the computer provisions.
MR EMMERSON: On the computer provisions.
GUMMOW J: That is your point, really.
MR EMMERSON: Yes. Your Honour did put to me that section 102(b) is declaratory. In our respectful submission, one needs to take a little bit of care about that because one can reach a particular position by more than one route, as I think the development of the two systems of the law ‑ ‑ ‑
GUMMOW J: Well, a pure concept, principle or discovery would not even be patentable.
MR EMMERSON: Yes, and a pure concept of course would not be copyrightable under our law because it would not be an original literary work.
GUMMOW J: Yes.
MR EMMERSON: On the other hand, once you have reduced that pure concept to a particular form in a passage of prose, then one could get protection. The classic example which is quoted in Computer Edge and a number of earlier cases is the case of a cookery book. Mrs Beaton produces a recipe for rabbit pie. She cannot of course get copyright in that method of making rabbit pie. What she can, however, get is copyright in the form of words that she uses to describe the recipe for rabbit pie. So protection is given to the form, albeit that protection is not given to the underlying substance or message. That was a distinction which was considered and endorsed in this Court in Computer Edge.
GUMMOW J: But to stop other people making these pies, per se, she needs a patent.
MR EMMERSON: Absolutely, yes. But that is how we get there. We get to the position that a method of making rabbit pie is not copyrightable simply because it does not fall within the definition of “literary work” until it is reduced to print or writing. When it is reduced to print or writing, you do get protection for that form of expression.
Now, section 102(b) does not accord with our law, on one view, because the protection is denied regardless of the form in which the method of operation is described or explained. Now, if that means no more than that you cannot copyright a method of operation per se, no particular difficulty arises. But difficulty would arise if United States law went one further step and said, “Because Mrs Beaton is describing a method for making rabbit pie, that cannot be copyrightable because of the prohibition under section 102(b) and, accordingly, you cannot get copyright even in the form of expression”. Now, if that were the effect of section 102(b), in our submission, our law and the United States law in this respect, as in many others, would diverge.
This is important in looking at Lotus v Borland because Lotus v Borland was concerned with a difference of opinion between the court of first instance and the Court of Appeals precisely on the issue of how did section 102(b) work. The district court is dealt with at a passage which quotes it in 1339, beginning at line 25:
The district court held that the Lotus menu command hierarchy, with its specific choice and arrangement of command terms, constituted an ‘expression’ of the ‘idea’ of operating a computer program –
and then the Court of Appeals goes on to say:
Under the district court’s reasoning, Lotus’s decision to employ hierarchically arranged command terms to operate its program could not foreclose its competitors from also employing hierarchically arranged command terms to operate their programs, but it did foreclose them from employing the specific command terms and arrangement that Lotus had used.
So the court at first instance in the United States took the view that it was not possible to copyright the method of operation, but there was protection for the specific command terms which were there used.
Now, the Court of Appeals reversed that decision based on section 102(b) and based on a reading of section 102(b) which, in our submission, could not be taken as being declaratory of our law. It starts with the prohibition of a method of operation, which is part of the prohibited subject matter set out in 102(b), and the way that the court approached it was by saying, “Well, the first thing that you do is look to see whether the alleged copyright subject matter is within the prohibition of 102(b).” If that is the case, you need look no further. The start of the reasoning, or perhaps the compendious conclusion is at 1339, line 7:
We hold that the Lotus menu command hierarchy is an uncopyrightable ‘method of operation’.
Then, down at the bottom of the page at line 42 the court deals with the issue as to whether there were other possibilities for expressing the particular command hierarchy, and the court then at 1340 at line 8 on that page says:
The fact that Lotus developers could have designed the Lotus menu command hierarchy differently is immaterial to the question of whether it is a ‘method of operation.’
That is to say, regardless of whether the US doctrine of merger happens to apply, protection is denied at the outset because of the words “method of operation”, and the court goes on to say:
In other words, our initial inquiry is not whether the Lotus menu command hierarchy incorporates any expression. We think that the Altai test would contemplate this being the initial inquiry.
That said, the initial inquiry is, does this prima facie fall within copyright subject matter? The Court of Appeals said, “No, the initial inquiry is does it fall within the prohibition?” Now, in our submission, that reasoning cannot apply under our law. We do not have any section corresponding to section 102(b) in the United States code and the critical point about the way in which that section is interpreted by the Court of Appeals in Lotus v Borland simply does not apply under our law. Our law does not require that you first look to see whether allegedly copyright subject matter is within a prohibited class. Our law, as set out in section 32, looks first to see whether, in this particular case, what we have is an original literary work.
Accordingly, that reasoning cannot apply. The court then goes on to deal with a reference to Baker v Seldon ‑ ‑ ‑
GUMMOW J: Do they discuss the interrelation between the definition of
“computer program” itself?
MR EMMERSON: I do not think so, your Honour; they do not quite come to grips with the problem that your Honour was putting to me, namely, that once you introduce our definition of “computer program”, which involves function, then in effect you cannot maintain the hard and fast old-style line between form and function. Indeed, it was of course precisely to deal with this that the 1984 amendments were introduced into our law. They arose out of the Computer Edge Case. The plaintiff failed in Computer Edge precisely because of the reasoning that there was here no reproduction on the footing that words more or less in the English language of source program had ultimately ended up as electrical signals in a silicon chip, and the Court held that under the copyright law as it then stood, the distinction between form and function had to be applied and that was where the line was drawn.
McHUGH J: My recollection is the legislation was introduced into the Parliament before the decision of this Court.
MR EMMERSON: That is absolutely correct, your Honour. There is unfortunately a complicated history because the learned trial judge decided the matter in favour of the defendant on really the first issue, namely, whether computer programs could be literary works at all, given that they were not intended to amuse or entertain or instruct human beings. Then the Full Court reversed that and by majority held all points in favour of the plaintiff party and it was at that stage that the 1984 amendments were introduced. It was known at that time that the matter was going to this Court. It was not known what the outcome was; it was thought that the matter should be dealt with urgently, and so your Honour is quite correct. These matters had been agitated before both courts but the views of this Court were not known.
McHUGH J: Well, what do you say in relation to the Australian legislation? Is the position in relation to a subject matter mentioned in the American decision - the VCR machine with the word “record” or “play” on the button?
MR EMMERSON: Yes.
McHUGH J: Now, is there copyright in respect of that?
MR EMMERSON: I think one would have to define what copyright subject matter one was talking about.
McHUGH J: Is there a program for the purpose of the definition, because the word “record” or “play” issues a set of instructions which causes ‑ ‑ ‑
MR EMMERSON: Well, your Honour, the question is: does it issue a set of instructions or is it in the simple analogue device mentioned in the explanatory memorandum to the 1984 amendments.
McHUGH J: What is the difference then between “record” on the VCR and my typing in a question mark or “type” or “STD” or “calculate” or “sum” or “total” or whatever it is?
MR EMMERSON: The question is ultimately whether what you are typing in is a representation of a set of instructions to a device having information‑processing capabilities. That would appear to entail that there must be some relationship between the set of instructions and the information‑processing capabilities.
Now, one could therefore imagine two separate type of cases. One where what you are doing is no more than the equivalent to operating an analogue device, pressing a button and the bell rings, or something of that sort. That clearly would not be protected. On the other hand, if what you do is instruct a device having digital information processing capabilities and it uses those processing capabilities, then there may be programming involved. It must all, in our respectful submission, depend on the precise circumstance.
HAYNE J: Program is the expression of that set, is it not? First, catch your set, if we are talking Mrs Beaton language, and find the expression of that set and it is the expression in any language, code or notation which is the subject of protection, is it not?
MR EMMERSON: Yes, but it is ‑ ‑ ‑
HAYNE J: Which brings me back to my example of a set at some point, Dr Emmerson.
MR EMMERSON: But, your Honour, with respect, what Mrs Beaton or, some would say, Mrs Glass, wanted to catch was the hare and we would say it is first catch the expression. It is the expression which is protected.
GAUDRON J: But is it not. It is a composite expression, surely, an expression of a set of instructions.
MR EMMERSON: Yes.
GAUDRON J: You cannot isolate expression out, can you, from that definition. You cannot.
MR EMMERSON: You cannot ignore any part of the definition, we would, of course, fully accept that, but if you look to see what is actually protected then what is protected is the expression.
GAUDRON J: Of a set of instructions.
MR EMMERSON: Of a set of instructions, yes, yes. Intended ‑ ‑ ‑
GAUDRON J: There can be no doubt that the various words which are in issue in this case are expressions.
MR EMMERSON: Yes.
GAUDRON J: The question is whether they are of the expression of a set of instructions, not just the expression of an instruction ‑ ‑ ‑
MR EMMERSON: No.
GAUDRON J: ‑ ‑ ‑ because they might even be the expression of an instruction. I mean, it might be beyond argument that they are the expression of an instruction.
MR EMMERSON: Yes.
GAUDRON J: So you have got to go further.
MR EMMERSON: There are perhaps two things that we would have to deal with arising out of what your Honour has just put to me. Yes, there must be an expression of a set of instructions and, yes, it may be possible to express that set of instructions in a number of different ways and you might therefore get protection for all different ways of expressing that set of instructions.
Your Honour also, however, drew a distinction between an instruction and a set of instructions. It would be our submission that in the context of this particular technology that distinction is really not a useful one because it is always possible to expand any instruction in a higher level language into more than one instruction in a lower level language.
GAUDRON J: At the end of the day you have to get the lower level language, have you not?
MR EMMERSON: Yes, indeed we do.
GAUDRON J: Is that not what the protection is concerned with, that end line bit?
MR EMMERSON: The end line is though that it cannot be protected. You cannot protect only the lower level language.
GAUDRON J: No, not only because you have to have a set of instructions.
MR EMMERSON: The particular point that is perhaps relevant here is set out in Autodesk 1 (1992) 173 CLR 330 in a passage beginning at 335. There the majority of the Court, the then Chief Justice and Justices Brennan and Deane, while concurring with the leading judgment of Justice Dawson, deal with some aspects of the problem which are perhaps relevant to what your Honour Justice Gaudron was putting to me. In the paragraph which begins about two‑thirds of the way down their Honours say this:
When the definition of “computer program” is read in its context in the Act, however, it appears to us to be clear that the reference to “an expression, in any language, code or notation” should not be understood as referring only to an actual written expression or representation of a set of instructions with the result that copyright does not exist at the time of an alleged infringement unless such a written expression actually exists or has existed at that time.
So their Honours are concerned with the distinction between the set of instructions and the expression and they say:
The definition should be understood as conferring protection upon the set of instructions itself but as doing so in a way which is adapted to the nature of copyright. That is to say, the stored set of instructions in a non‑sensate form such as electrical impulses is itself protected on the basis that copyright actually subsists in any expression or description of it which can theoretically be made in language, code or notation.
So the majority of the Court there are saying that the protection lies in the expression but also adding that in order to make this work, protection is going to extend to a variety of different expressions – that is to say any expression – of the underlying set of instructions in the computer program.
The majority then go on to deal with another point. They say if you produce an original program, then you get copyright in any expression of it, but they also deal with a case in which the set of instructions does not itself satisfy the requirement of originality, that is to say, a set of instructions itself is not protectable, and then they say:
that requirement can, of course, be satisfied by the originality of an expression or description of it in language, code or notation with the result that copyright is confined to that particular expression or description alone.
So one can get a position in which the underlying set of instructions are not protected but an original expression or description of it in any language, code or notation can be protected.
Now, this, in our submission, supports the view that you have a broad practical protection here. Where as in the present case you have an original expression, then that original expression can be protected in copyright. You can also get protection for the underlying set of instructions, however expressed, or in some circumstances you might not get protection for the underlying set of instructions. But in any event, you get protection for the original expression. Our submission is that that is what happens in the present case.
I think just to round off what I was saying about Lotus v Borland, the quotation from Baker v Selden, an old and very well known United States case, is, of course, dealing with a much earlier law and it is dealing with a case in which Parliament had not intervened to blur the distinction between “expression” and “function”.
GUMMOW J: What were the facts of Baker v Selden?
McHUGH J: It was an accounting program case, was it not?
MR EMMERSON: That is right, yes, and the question was whether there could be protection for the underlying system of accounting, from recollection.
On the issue of United States law, could I refer the Court to Ibcos Computers Ltd v Barclays Mercantile Highland Finance Ltd (1994) 29 IPR 25. That is a decision of Mr Justice Jacob in England. It is relevant in that his Honour deals, amongst other things, with the applicability or, as his Lordship held, non-applicability of United States law in England. His Honour was dealing with a decision of a Judge Baker and it is dealt with in fairly unflattering terms. However, he does deal with a number of points – perhaps it is convenient to turn to page 38. At the top of page 38 he deals with the statement by Judge Baker:
The mere linking of several programs is not itself, in my judgment, an original literary or artistic work –
that is something we come to a little bit later but his Lordship dismisses that. He says:
It cannot be correct as a general statement. Indeed no counsel before me accepted it as accurate. There is no reason why a compilation of computer programs should not itself acquire a separate copyright.
He then deals with the proposition that if there is only one way of expressing an idea, that way is not subject to copyright.
His Lordship dismisses that at the bottom of page 38 and he says it is a dangerous statement because, as he notes at the top of page 39:
But of course as the late Professor Joad used to observe, it all depends on what you mean by “ideas”. What the respondents in fact copied from the appellants was no mere general idea.
He then deals with the way the English cases approach the question of protecting a principle. He says:
It is of course true that copyright cannot protect any sort of general principle, such as the principle of drawing a hand to show how to vote, but it can protect a detailed literary or artistic expression. Thus in the case of an exhaust pipe it was said that the copyright was protecting the engineering principles which went into its design. Nevertheless, there was copyright in the drawing and the copying of the drawing…..amounted to an infringement. This was so even though there was a copying of the engineering principles –
And he deals with that. Then in dealing with “idea” and “form of expression of an idea” at line 30 on page 39 he says:
The true position is that where an “idea” is sufficiently general, then even if an original work embodies it, the mere taking of that idea will not infringe, but if the “idea” is detailed, then there may be infringement. It is a question of degree.
And at line 39 he says:
It should be noted that the aphorism “there is no copyright in an idea” is likely to lead to confusion of thought. Sometimes it is applied to the question of subsistence of copyright: Is there a “work” and if there is, is it “original”? Sometimes it is applied to the different question of infringement: has a substantial part been taken?
Then at line 47 he says:
In this context there was some appeal to United States cases on copyright. The fact is that United States copyright law is not the same as ours, particularly in the area of copyright works concerned with functionality and the area of compilations. The Americans (many would say sensibly) never developed copyright, so that functional things like exhaust pipes could not be copied. This is partly due to their statute, which is different from our Act.
And then he quotes section 102, including section 102(b), and at page 40, line 21 he says:
Moreover, United States case law, ever since Baker v Selden (1879) 101 US 99, has been extremely careful to keep copyright out of the functional field, either by saying there is no copyright in, or that copyright cannot be infringed by taking, the functional.
He mentions Baker v Selden and says that is different but then he says that the United States case law and copyright cannot be said “never to be of assistance” but it did not assist him in that case.
Now, we would rely on the Ibcos decision as being an example of how an English court has dealt with the question of whether it should follow United States law in this particular area and the way in which it has emphasised that there are really major differences between the development of United States law and the development of common law.
GAUDRON J: This case is not about whether we follow United States law or not; we have got to apply our statute.
MR EMMERSON: Absolutely yes, your Honour.
GAUDRON J: The question in this case is simply whether what is in issue here falls within the definition of the “computer program”.
MR EMMERSON: Yes, your Honour.
GAUDRON J: It begins and ends with that, does it not?
MR EMMERSON: Well, it has traversed other territory on the way along and the courts below have dealt with other territory and our learned friends deal with other territory and, for that reason, I find myself on other territory, but we would accept, of course, that the underlying question is, what does our definition of “computer program” mean; does it extend to the present case? So, if I may say so, I agree with what your Honour has just put to me, but it would nevertheless appear to be necessary to deal with some of these other matters.
Our learned friends dealt with this first issue in their written submissions, beginning at paragraph 5 and while, for the most part, it may be necessary to wait until my learned friend, Mr Macaw, develops these submissions, there are some things that we would want to say about them.
The passage dealing with the reserved words and macros starts at paragraph 5. I should perhaps mention, though, that our learned friends, in their paragraph 2, have invited the Court to have regard to what they say is an industry convention of compatibility or interoperability. Now, in our respectful submission, there is no finding by the trial judge of any such industry convention as is here alleged and it is our submission that one should approach the question of construction of the definition of “computer program” in our Act without, as it were, a predisposition to say it cannot somehow deal with cases such as the present.
There is no reason for reading down our Act by reference to a so‑called industry convention, particularly in the absence of any factual basis to support that. Further than that, in our respectful submission, the expression “highly compatible” is something of a euphemism in the present case. What the respondents are seeking to do is compete with my clients by offering a rival system which uses the same language and we say they cannot do that.
Granted that computer programs interoperate, that does not involve the proposition that there can be no copyright in computer programs which might operate with some other computer program. One only has to think of obvious examples like operating systems, Windows and so forth, which are computer programs, are subject to copyright notwithstanding that they must operate with other things.
Going now to paragraph 5 of our learned friends’ submissions, that is mainly saying, as we read it, that the Full Court was right. Well, on that, of course, the parties are at issue of the Full Court. But our learned friends then raise what appears to be a different argument in their paragraph 6. They say that the reserved words and macro commands “are not, in the proper sense, ‘instructions’”. They say the word “‘SHOW’….. is not, in the relevant technological sense, an instruction”, they are merely “related information”. Now, the reference to “related information” comes from the definition of “computer program” in our Act because it refers to “a set of instructions (whether with or without related information)”.
Now, our position on our learned friends’ paragraph 6 is we simply challenge that. We say that the reserved words and the macro commands are sets of instructions, they are not merely related information. We would rely on passages which this Court has already had its attention directed and, in particular, we would rely on page 1336 where the learned trial judge, beginning at line 15, says:
And the use of each of those 192 words –
those are the words of the Dataflex language which had been taken by the respondents –
in the source code of either language contributes to cause a device having digital information processing capabilities to perform the same function. The same may be said of…..“the reserved words” –
and so on. Now, this is clearly not a basis, we would say, for the submission that it is only somehow related information and not constructions.
Similarly, at 1409, the Full Court dealt with that and deals with the translation of a source code program into a lower level source code, and at line 10 they say:
In other words, the compiler program recognises the particular command used in the simple program and substitutes for that a set of instructions written in source code.
So, we would say that there is no proper basis for saying that this is merely related information.
Our learned friends, in their paragraph 7, deal with the word “expression” and they say, in:
copyright law, where the law is concerned with distinguishing protectable expression from unprotectable ideas, the expression must obviously relate to the more particularised rather than the less particularised manifestation.
Now there we say that the definition of “computer program” protects any expression, it is not limited to a particular type of expression, more particularised or less particularised, provided it is otherwise within the Act. And so it is not, in our submission, a correct view to read down the provisions of the Act in the way that our learned friends suggest.
In paragraph 8 they say that the consequences of our proposition are startling. We would say no more startling than having copyright protection for computer programs in the first place. You have copyright protection for computer programs. Ours are as much worthy of protection as any others.
They also, however, go on to quote a statement by the Copyright Law Review Committee. Now, at the outset we would say the Copyright law Review Committee cannot assist in construing our statute as it stands, it is concerned with possible changes which might be, but have not been, enacted and is, therefore, irrelevant. But the particular passage that they quote is simply this. The Copyright Law Review Committee itself:
quoted a submission pointing out that the “protecting software structure, sequencing, commands and language could rapidly exhaust those alternatives –
Well, in our submission, there is no evidence of this, the judges’ findings are to the contrary, the judges’ findings were that substantial work and effort went into developing our particular language and there is no basis at all for saying that that is the only possible language that you can get.
Paragraph 9, they say:
The appellant’s argument that the expression “computer program” in the Act should not be limited to computer programs in the ordinary sense, is a dangerous one.
Now, in our submission, it is absolutely plain, both from the language of the definition of “computer program” and, if one gets to that point from the explanatory memorandum, that the expression is not limited to computer programs in the ordinary sense and we would, of course, deny that this consequence is a danger. They deal with United States law at paragraph 10 and I have already dealt with that. In paragraph 11, however, they say this:
the Reserved Words are not protectable because they represent a merger of the expression into the idea or function. Dr D Bennett had to appropriate those particular words, or reference to those words, if Powerflex was to be compatible with Dataflex, if Powerflex was to be able to execute programs written in the Dataflex language.
Now, in our respectful submission, a lot of the nub of the fallacy in the respondents’ case turns up in that passage. Dr Bennett did not have to appropriate particular words or commands in the Dataflex language; he could have developed his own words and commands. What he sought to do was to appropriate to himself some of the benefit of the Dataflex language. But, if we are right in saying that the Dataflex language is itself protected by copyright, there is no reason for denying copyright simply because somebody might want to use the same system. They can perfectly easily develop their own. They were not limited so that it was impossible to express ideas in any other way. They simply wanted to take our expressions.
Now, having said that, we do not of course quarrel with what was said by Justice Dawson in the first of the Autodesk Cases about merger of form and expression. That was dealt with by his Honour in Autodesk v Dyason (1992) 173 CLR 330 at 344 to 345. His Honour deals with this in the paragraph that spans those two pages. He says there is a:
traditional dichotomy in the law of copyright between an idea and the expression of an idea.
We accept that that is so subject to the difficulty that it is hard to draw the line very precisely in the case of computer programs because it is inherent in computer programs that they are expressions of instructions for performing a particular function so functionality comes in. His Honour then goes on to deal with the notion that you might get a merger of the idea so that its expression “is inseparable from its function” and if that is so then we accept that this Court has here held that that merger can take place.
What we say, though, is that there is simply no evidence in the present case to treat this as being such a case – on the contrary. It would have been open to the respondents, had they chosen to do so, to develop their own language and they simply did not. For that reason, we would say that the references to ideas and expressions, while no doubt raising interesting issues as a matter of law, are not issues which require to be resolved by this Court in the present proceedings.
There is a separate set of arguments on which we rely in the case of what is referred to in the papers as the macro command. These are commands of a complex nature. They are set out at page 1385 in the Second Schedule to the order made by the learned trial judge.
GAUDRON J: Am I correct in thinking that the macro commands result in the performance of more than one function by the computer? Am I correct in that?
MR EMMERSON: The macro commands are at the more complex end of the spectrum.
GAUDRON J: That does not help me. I find it all at the complex end of the spectrum.
MR EMMERSON: I think that, as the late Professor Joad might have said, again it all depends on what you mean by more than one function.
GAUDRON J: I want to know what you mean by “macro commands”, and I do not need to know it in terms of legal concept at this stage. I just want to know what the word means.
MR EMMERSON: I am simply using the expression “macro commands” to mean a particular subset of reserved words about which his Honour made certain findings.
GAUDRON J: That does not tell me anything, either.
McHUGH J: Well, a macro command is the sort of illustration I gave earlier where you say, update name, set area code to equal 2027, where suburb equals Point Piper. That is just a form of macro command, whereas you were relying on words like “replace” as having a particular syntax, were you not?
MR EMMERSON: Well, the macro commands themselves have a particular syntax actually, with respect.
McHUGH J: Well, of course they do, yes.
MR EMMERSON: In answer to your Honour Justice Gaudron, they are at the more complex end of the spectrum, but all these commands, as I understand it, can be analysed into groups of commands of a slightly simpler sort. This is the point of having a fourth generation language; you have a language which is capable of expressing very shortly something which is pretty complex and a lot of the reserved words are complex in this way, and that is why they have been reserved to have a particular meaning. Now, in the way that the judgments are expressed, one might get the impression that there is a very hard and fast line between macro commands and the rest of the reserved words. On my instructions, that is not so, so I, for that reason, baulk at putting that proposition to the Court.
McHUGH J: Well, correct me if I am wrong, but I just simply understood that a macro command can just be a single command, but more usually there are a series of commands within little programs.
MR EMMERSON: Yes. There is a little program which deals with all of these things and it takes in the reserved words of the macro commands and they can in turn be expressed in terms of simpler commands but at the same level of sophistication, that is to say, also in the Dataflex language. So you find if the thing is worked out that some of these commands are taken in by the program, expanded out, if you like, into something else in the Dataflex language. And you might also find that the thing is expanded out. It can also itself be expanded, and so the process can continue. Then, at a later stage, you really do go down one level in programming.
But at the very high level of which we are concerned, there is an element of sort of Chinese boxes about all of these, and the macro commands tend to be more complex than the rest, but unfortunately, from the point of view of expressing this in a simple way, there is not on my instructions anyway a really hard and fast line between the macro commands and the other reserved words. It is just that in the context of this particular case, we have a finding of objective similarity one level down in the case of the three macro commands. That is to say, when you take those macro commands and you analyse them into sets of instructions at a lower level, you find that in our system and in the respondent system you get a strong objective similarity between those two sets of instructions. So there is this additional argument which is available to me, but it is really only in that special context – which I am afraid is a legal context – that one can fairly say that the distinction occurs.
GLEESON CJ: Is that a convenient time, Mr Emmerson, at which to adjourn?
MR EMMERSON: Yes, indeed it is.
GLEESON CJ: We will adjourn until 2.15 pm.
AT 12.48 PM LUNCHEON ADJOURNMENT
UPON RESUMING AT 2.20 PM:
GLEESON CJ: Yes, Mr Emmerson?
MR EMMERSON: If the Court pleases, two matters by way of rounding off what I was saying this morning. Your Honour Justice Gummow asked me a question about the equivalent of protection for Australian and United States citizens. I should, perhaps, draw the attention of the Court to the fact that the learned trial judge did mention this point. He did it at page 1355 of the appeal book in paragraph 21 of his reasons for judgment, and the conclusion that he reached is that:
“the protection afforded by the Australian Act in relation to the subject matter and persons referred to in ….. reg. 4 ….. is no greater and no less than that provided for works first published in Australia by a qualified person”, and lack of conformity with legislative conditions of copyright protection in the foreign country where a work is first published is of no effect –
That was a slightly different point that was argued before his Honour but for completeness I draw attention to the fact that his Honour does refer there to the international protection regulations. We would also, in response to what is said in our learned friends’ submissions about the United States law, want to pass up without reading a further extract from Laddie, Prescott and Vitoria “The Modern Law of Copyright and Designs”, second edition 1995. That is a passage which begins at page 836.
The Court will note that the learned authors under paragraph 20.58 have a section on “Irrelevant or exotic tests”, and support the view that under the English law with which they are concerned one follows the traditional analysis of first looking to see what the work is, asking whether it is original, asking whether there is copying and if there is copying, whether a substantial part has been copied. We would say that our Act, which of course is based on the United Kingdom legislation, would lead one to adopt the same approach.
McHUGH J: Dr Emmerson, can I take you back to the definition of “computer program” and your contention that you can get the protection for a single word or whether the single word constitutes computer program, but let me put this to you. If I type in the computer the word “sort” or “sum” or “total” as the case may be and nothing else, the computer will say to me, “syntax error”. I have to then go and identify the file or the row that I want processed by that command. Now, does not that indicate that more is needed than the single word “replace” or “sort” or “sum” to be a set of instructions to the computer?
MR EMMERSON: No, with respect, your Honour. We would say that it is always true with computer programs, or almost always true, that you do need something else for that program to work on and to perform the particular function which is the subject matter of that program but that does not mean that if other things are right that each of the words that your Honour put to me cannot itself be a computer program.
In exactly the same way, if you have, say, an operating system in a computer, that does not do anything useful until it has something to operate on, and it may compile another program, it may do any one of a range of different things, it has to have the material to work on, as have these programs, of course, but, in our submission, to limit the notion of computer program to something which is freestanding in the sense of something which does not need something else to work on, would be to curtail the protection for computer copyright to the point at which it would be of relatively little value.
McHUGH J: Can I ask you another question arising to some extent from that, but also from what Justice Hayne put to you earlier: your arguments assume that the infringement here relevantly is with the word, as the case may be, “L trim” or “R trim”, or whatever the word may be, but what causes the device to perform a particular function is not merely the source code, but its compilation into a form of object code. Now, the object code may be different under both sets of programs. If that is so, how can there be an infringement on that basis?
MR EMMERSON: Your Honour, could I say this: what your Honour is putting to me must be true in general terms and, in fact, states a problem which the 1984 amendments were intended to overcome, because, of course, if one is talking about infringement of copyright in a source code, you might find that that source code could be run on a number of different computers with different systems, different configurations and the object code would necessarily be very different and it is for that reason, in our submission, that Parliament has drawn the net widely when it gives as its definition of “computer program” that:
a set of instructions…..intended, either directly or after either or both of the following:
(a) conversion to another language, code or notation;
(b) reproduction in a different material form;
to cause a device having digital information processing capabilities to perform a particular function.
In our submission, the reason for that is that it allows the courts to say that it does not matter that in actual use a program may be compiled into something in object code and the plaintiff’s object code may not be the same as the defendants’ object code. The aim is to cover both and it is for that reason that the Act expresses the expression “computer program” in the way it does. I should add that historically this definition of course arose out of the Computer Edge Case. In Computer Edge the problem was just what your Honour has put to me, namely, that copyright was claimed in source code as a literary work. The defendant had compiled it out of the country into the formal object code and it was only the object code and indeed a particular electronic storage of object code which came into the country. The intention, as we would say is clear from the preamble and the surrounding circumstances, is that Parliament intended that that sort of activity would be caught by these provisions.
GAUDRON J: Parliament said what it intended with its definition. There is very little scope for a court to read into or read out of the definition things that are or are not there.
MR EMMERSON: We would say that, your Honour, and for reasons that we have set out in our written submissions, we would say that if you work through the definition, as we have in paragraph 18, you find that all the elements are there.
GAUDRON J: Well, is that right? Does the word cause the operation? I mean, there may be scope for argument as to ‑ ‑ ‑
HAYNE J: The key, if I may interpose, is at 18(1), is it not, where you say:
the third respondent chose to use the same expressions for corresponding sets of instructions –
What do you mean “corresponding”? Perform the same function, or are you saying they are identical sets of instruction?
MR EMMERSON: I am saying that they are identical words.
HAYNE J: Yes.
MR EMMERSON: And identical things that they make the computer do.
HAYNE J: Function, but are the instructions, that is, are the electronic impulses thus generated the same or nearly so, leave aside the macros where other things emerge?
MR EMMERSON: Leaving aside the macros, we cannot show that they are the same and it is our submission we do not have to show they are the same.
HAYNE J: I understand that. But does it follow, to follow on from what Justice McHugh was saying, that not just object code is different, but source code is different?
MR EMMERSON: No.
HAYNE J: No?
MR EMMERSON: We way that the source code is the code at the level of the Dataflex language, and the source code is the same.
HAYNE J: So “save” or “enter” or “total”.
MR EMMERSON: Yes, the source code is the same. There are then various intermediate levels which, for the generality of reserved words, are different, though in the case of the specific macro commands we say are relevantly the same.
HAYNE J: But if in your client’s program the person enters “total” that generates a different set of electronic impulses from those that are generated if, in Mr Macaw’s client’s program, those same letters are entered?
MR EMMERSON: Yes it does.
GUMMOW J: And you say that falls within the definition, do you?
MR EMMERSON: Yes.
GUMMOW J: Because the words are:
intended, either directly or after either or both of the…..conversion to another language –
and you say it need not be the same language.
MR EMMERSON: It need not be the same language, no.
GUMMOW J: At the level of object code?
MR EMMERSON: No, indeed it need not, no.
HAYNE J: Well, it is a fact it refers to code as well as language, “conversion to another language, code ‑ ‑ ‑
MR EMMERSON: - - - “code or notation”. Yes, it uses an expression which, in our submission, is intended to draw the net very widely so that the code may be a human readable code, or it may not be.
McHUGH J: I was in Computer Edge in the Full Court and that was the difficulty. The difficulty for Apple was the problem of conversion of the source code into an object code because it had been copied in Taiwan or somewhere, and it was just the disc that came in.
MR EMMERSON: Yes, it was just actually a chip that came in, yes.
McHUGH J: Yes, a chip that came in, yes.
MR EMMERSON: And so no act of relevant copying in the country could be shown and in that particular case Apple had to rely on the importation provisions and the importation provisions only hit at the chip, not at what may or may not have been done in Taiwan.
McHUGH J: Yes.
GAUDRON J: But does not your submission now raise the meaning of “cause” in the definition? There is a definition of “cause” as being directly or after translation, if I can use a shorthand. The question is whether it has got to be the sole cause or whether it can be one of several causes, it seems to me in this case because, as I see it at the moment, the word is only part of what causes and what is intended to cause, to go back to the definition, the machine, we will call it, the device, to perform a function.
MR EMMERSON: That is correct, your Honour. We have necessarily here multiple causes. To take it at its most elementary, the computer program must have a set of instructions but there must also be something for those instructions to operate upon before the computer performs the function and so you do need other things but we say that the definition does not deny that, nor could it if it is to be given a construction which will be workable for this technology.
McHUGH J: But it is a problem, is it not, because if you just type in “total” - no set of instructions – it is not a set of instructions which will cause the computer to perform a particular function. It will not operate at all. It will tell you there is a syntax error. It is only when I add to the word “total” “field 1” or “field 8” or whatever the name of the field is, that it then performs the function.
MR EMMERSON: Your Honour, with respect, it must have something to operate on, yes, of course, and until it has that thing to operate on then it will not perform the function, but we would say that is always the case.
McHUGH J: Well, that may be, but the question is whether the word “total” in this example is itself a set of instructions and what at the moment seems to be against you is that the set of instructions is not the simple word “total” but to total something else which instructs –the set of instructions is to “total field 8” or whatever you would like to call your particular field.
MR EMMERSON: If you could have no infringement in a computer program until that computer program was actually in a position to run on the machine having actual data, then, of course, it would never be an infringement for somebody else to sell a program which was capable of operating on data to produce a particular result and that would, with respect, construe this definition in a way which would leave so serious a gap in the protection of computer copyright owners as to make the protection virtually worthless.
McHUGH J: I am not sure about that. I mean, there are many programs. It might be a program in respect of an accountancy function or agricultural function or bookkeeping. There are all sorts of things one can think of and there will be a very detailed set of instructions but it will contain more than the word “total” or the word “enter” or “L trim” or “R trim”. It might be thousands of lines of code or it might be three lines.
MR EMMERSON: There are perhaps a number of things involved in that, if I may say so, with respect, your Honour. The first is the issue as to whether a computer program must be freestanding before it can be a computer program or whether it is still a computer program albeit that in actual use it has to have something to operate on. We say it does not have to be freestanding in that sense and the usual form would be that it would need to have something to operate on.
McHUGH J: Can I just interrupt you so you can deal with it as part of your argument? I mean, since Windows and the use of mouses and buttons and so on, “click here” will just get a whole program going, but underneath the click are often hundreds of thousands of lines of code of instructions to the computer.
MR EMMERSON: Yes.
McHUGH J: In the old DOS days one had to write it all out.
MR EMMERSON: Yes, absolutely.
McHUGH J: But perhaps that is what the definition is aimed at, in that sense, rather than a single word which has got to have some material added to it.
MR EMMERSON: Well, in each case you start out with the expression in the highest language and it then goes into the machine and is compiled into one or perhaps a succession of codes at lower level language.
HAYNE J: And is that the set of instructions, that set of codes in successively lower languages?
MR EMMERSON: No, your Honour, that is not. That is a set of expressions of the set of instructions. It is not the set of instructions itself.
HAYNE J: Well then, in the “total” example, is the set of instructions “add the values of fields to be nominated”?
MR EMMERSON: Yes.
HAYNE J: That is the only set of instructions.
MR EMMERSON: Yes.
HAYNE J: Then how is that different from “perform a particular function”? What is the translation that is contemplated by conversion to another language, et cetera? Every database program that exists will have a function that enables you to sum up a series of fields.
MR EMMERSON: Yes, but it is giving you instructions to carry out certain operations ‑ ‑ ‑
HAYNE J: You or the computer?
MR EMMERSON: The computer.
HAYNE J: The computer is not told by the RSM on the parade ground, sum the values in fields to be nominated, the instructions to the computer are electronic and the metaphor and the analogy is at great risk of misleading. What is the set of instructions?
MR EMMERSON: The set of instructions in your Honour’s particular case is sum the various fields. The expression of the set of instructions is total; the function is the actual summation of the various fields but, of course, although there is a relationship as there must necessarily be between the set of instructions and the function, it does not mean that they mean the same thing. Conceptually they are different and we say that that is a distinction which is maintained in this definition.
McHUGH J: Do you throw the weight of your argument on the word “expression” rather than “set of instructions”?
MR EMMERSON: Yes. We throw it on all parts but we are told the computer program means an expression, so one has to look in each case for an expression before you get to the protected subject matter.
McHUGH J: In this sense, “expression” is used as a label in effect, is it, of a set of instructions? It replaces a label for a set of instructions on your argument, is it?
MR EMMERSON: Well, we prefer the word “expression” rather than “label”, your Honour.
HAYNE J: It is closer to manifestation rather than label, is it not? It is the manifestation of a set of instructions.
MR EMMERSON: Yes, because it is something that does not merely label something which is there but it is something which can be typed into the computer and which causes it to perform the function.
McHUGH J: So, a question mark could be, in an appropriate context, an expression of a set of instructions - in a number of programs it is.
MR EMMERSON: It could be. Your Honour’s knowledge of computer programming is, I am afraid, more recent and detailed than mine.
McHUGH J: I mean, in some programs in which I have some familiarity a question mark will calculate their calculations.
MR EMMERSON: Yes, it could be.
McHUGH J: You have to add the figures in but the question mark itself ‑ ‑ ‑
MR EMMERSON: If the Court pleases, I wanted next to go on to a different topic, namely, the topic of what we have called special considerations relating to the macro commands. We deal with that in paragraph 28 and following in our written submissions. The place of this in the overall scheme of our argument is simply this, that in the case of the three macro commands we have not only the same expression and the same meaning of that expression and the same syntax of that expression but there is a further similarity which is important. That is that we say these macro commands when they are compiled under the system of my clients and under the system of my learned friend’s clients lead to different expressions which are nevertheless so related that one is an adaptation of the other because there is a high degree of objective similarity between the two.
There is just one wrinkle in the way that that is shown, and that we have attempted to summarise in our paragraph 28. What happens is that in my client’s system the macro commands on the machine get encoded into an intermediate code which is referred to as I‑code. In the respondents’ program it is encoded into something which is called, perhaps not entirely helpfully, source code.
GAUDRON J: And “source code” there is being used differently from your reference to “source codes” in answer to Justice McHugh?
MR EMMERSON: Yes, your Honour. That is why I said “not entirely helpfully”. It is unfortunate that in this particular case that expression has been used specifically as well as generically. I have taken the words from those of the learned trial judge but it certainly, unfortunately, makes it a little bit more difficult to follow. However, the point perfectly simply is this, that when you translate the I‑code into source code you will find a high degree of objective similarity between the two. The result of that was that the learned trial judge held that in this particular case the respondents’ code was an adaptation of my client’s code. That raises the question of what is meant by “adaptation”.
His Honour dealt with that at appeal book 1345 beginning at line 9 and over the page to 1346, line 8. The gist of the reasoning is to be found beginning at the bottom of page 1345:
Each of the three commands cause the performance of a more complex function than many words. The objective similarity of he two versions of source code are in each of the three macros strong, and of course in each case compile to cause the computer to perform the same complex function. And that was the objective Dr Bennett sought to achieve by close study of FMAC.
FMAC is the program in my client’s system which interprets the various reserved words that are the basis of the Dataflex language. His Honour concludes:
In my opinion Dr Bennett made an adaptation of the expression in I‑Code of each of the three sets of instructions invoked by the three commands by his expression of that set of instructions in source code.
So the coding in ‑ ‑ ‑
GAUDRON J: If you are right in your argument, why does one go to the notion of adaptation? I am looking simply at the definition, “whether directly or”, et cetera.
MR EMMERSON: Yes, I understand the point, your Honour, and with the greatest respect, it is one that has slightly puzzled me as to whether -perhaps it might be said that we had conceded too much. The structure of the Act goes in this way, that if you look at section 31, it is an exclusive right which is given to a copyright owner by section 31(1)(a)(i) is:
to reproduce the work in a material form.
(vi) is:
to make an adaptation of the work.
Now that distinction had already been present in the Act, which had a definition of “adaptation”, which is set out in section 10(1) of the Act and the definition is an exclusive one, and it means: “in relation to a literary work in a non-dramatic form”, a certain thing; “in relation to a literary work in a dramatic form”, something else; in each case basically turning drama to other literature or vice versa. Before the present amendments - the next was (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 the story or action is conveyed solely or principally by means of pictures.
So we have built up here various types of adaptation going from a novel to a play, from a play to a novel, going to a novel to a translation of the novel, going from a novel to a cartoon strip, and then there was something also for musical works.
Now, what Parliament has done is to introduce into the definition of “adaptation” paragraph (ba) which says that adaptation means:
in relation to a literary work being a computer program – a version of the work (whether or not in the language, code or notation in which the work was originally expressed) not being a reproduction of the work.
Now, that leads to some difficulties in the formal analysis of a computer copyright infringement case because adaptation and reproduction are made to be exclusive, albeit that either of them would be an infringement.
GAUDRON J: Yes, but I am just wondering, if your argument is correct in relation to these more complex commands, if it is not a reproduction rather than an adaptation, and you say it does not matter ‑ ‑ ‑
MR EMMERSON: Your Honour, I can only say that is a very fair question.
GAUDRON J: But it does not matter.
MR EMMERSON: But it does not actually matter, no. Certainly one can imagine questions of degree arising and the two points having to be put in the alternative, as indeed they were, but the way things have fallen out his Honour held that here we had an adaptation. I ought to say that the matter was made no easier by the fact that there had not been consideration by the courts of the meaning of the word “adaptation” in the context of computer programs at that stage.
GUMMOW J: If you were right in your submissions about the macros, that the Full Court was in error there, but you were not successful in your earlier submissions this morning, what would the result be as to the practical outcome of this litigation, in terms of orders?
MR EMMERSON: Well, the practical outcome of this litigation is that my clients would succeed in respect of the macros but they would fail in respect of the reserved words which were not shown to have the same objective similarity at the second level of organisation that the macros have. The arguments are really, in a sense, independent because so far as the macros are concerned we rely on the source code, the macros themselves, for the first part of the argument but what we are relying on for the second part of the argument is that the third respondent actually constructed source code at a lower level ‑ ‑ ‑
GUMMOW J: Yes. Is there a particular passage in the Full Court judgment dealing with macros which you say pinpoints their error in analysis?
MR EMMERSON: Yes.
GLEESON CJ: Is it the second‑last sentence on page 1433.
MR EMMERSON: I was about to place it a little bit later, if your Honour pleases.
GLEESON CJ: Is that the only difference between the position of the macros and the reserved words?
MR EMMERSON: Not so far as the legal position is concerned, no. It seems likely that what the Full Court was seeking to do at this stage, however, was introduce the macros rather than make a finding adverse to my clients, and it was attempting the exercise which I attempted in answer to questions from her Honour Justice Gaudron before the adjournment to explain the significance of macros in this case. The place at which we would say that the court went wrong was where it dealt with the definition of “adaptation” beginning at page 1435, line 9, and then, in fact, the relevant part continues on through 1437. The way that, in our respectful submission, the Full Court got itself into difficulty was simply this: that the word “adaptation” in the computer program context requires that there should be a version of the work so that the adaptation itself is a version of the original. And the question arose as to what was meant by a “version” and the court attempted to deal with that towards the end of 1435. It says that:
Reference to the Explanatory Memorandum for the Bill which was enacted as the Copyright Amendment Act 1984, makes it clear that what was contemplated by the word “version” is meaning “a translation”.
Now, just to step forward to where this leads, the Full Court then goes on to say, “Well, if it simply means a translation, then this is not what happens here. What you get is objective similarity but it falls short of ‘a translation’.” At least so it would appear where “translation” is referring to the sort of mechanical operation which is carried out by computers themselves in which they either have code at a high level and they compile it into code at a lower level, or occasionally where you have decompilation.
Now, where the court went wrong, in our submission, was construing the explanatory memorandum as limiting the word “version” to two particular types of translation. One was the type of translation which is a computer translation going from code at a high level to code at a lower level or vice versa and the alternative was going from a high to a low and back again, that is to say a sequence of compilation followed by decompilation. And the Full Court said neither of those things happened in this case and that was the end of the story, notwithstanding that there was strong objective similarity in the expression of the source code that the third respondent made compared with my client’s I‑code.
GUMMOW J: This has all been decided as no infringement. What is put on one side is any characterisation of the macro itself.
MR EMMERSON: Yes.
GUMMOW J: As the subject of copyright.
MR EMMERSON: Yes, that is a fair comment, your Honour. It is treating the macro at least one level down at the I‑code level as being itself capable of copyright protection but it is saying copyright protection does not go as wide as this, so the reasoning runs, because this is not ‑ ‑ ‑
GAUDRON J: Can I take up there that they are treating what is capable of copyright at this level of your argument, are they not, as not only the expression of the word “expression” but the word plus the set of instructions that are called into existence by the command.
MR EMMERSON: The I‑code that is called into existence by ‑ ‑ ‑
GAUDRON J: It is a set of instructions in the I‑code, is it not?
MR EMMERSON: Yes, it is. If I can agree with your Honour for the purpose of this stage of the argument without saying that it is the only set of instructions involved ‑ ‑ ‑
GAUDRON J: No.
MR EMMERSON: Yes, I do.
GAUDRON J: Yes, on your view it is rather a compilation of programs than a single one, is it?
MR EMMERSON: No, actually, we do get to the compilation of programs slightly later in the way the argument goes because the way the argument goes here is that in my client’s case you have the macro commands at the top level, lower level you have the I‑code. The respondents chose to use the same command at the top level. They wrote a program in source code one level down but that source code itself had strong objective resemblance to the I‑code and so, in essence, what we are here talking about is infringement of copyright in the I‑code.
GAUDRON J: As distinct from the macro itself?
MR EMMERSON: As distinct from the macro itself.
GLEESON CJ: And is the I‑code what is referred to on page 1437 at line 13 as the “underlying set of instructions”?
MR EMMERSON: It is the next level down set of instructions. I hesitate to use the expression “the underlying set of instructions”, because you may have one or more further layers to go through before you get down to the machine code instructions.
So, the issue that arises in this part of the case is whether the Full Court was right in limiting the meaning of the word “version” to mean one of the following two things: either translation by compilation or decompilation or, alternatively, a process of both compilation and decompilation. The relevant part of the reasoning is to be found at the bottom of page 1436 beginning at line 24:
The evidence is clear that while Dr D Bennett carefully studied the Dataflex program so as to ensure that the PFXplus commands in question performed the same functions as the Dataflex commands –
PFXplus is, of course, the respondent system –
the expression of the source program as written by him was an original expression, albeit having much which was objectively similar to the expression of the source code in the Dataflex program. But it is clear that the process involved no translation from one form or language to another, nor did it involve the kind of process referred to in para 14 of the Explanatory Memorandum involving compilation followed by decompilation or vice versa.
Now, the flaw that we say arises in this reasoning is that the court construed the meaning of the word “version” in the definition of adaptation in connection with computer copyright, by reference to the explanatory memorandum but, in fact, over‑limited the ambit of the expression in the Act. What we say it did was take paragraph 13 and 14 of the explanatory memorandum set out towards the top of page 1436, and treat them as if they were an exhaustive statement of all possible meanings of the word “version” whereas, in fact, in our submission, they are simply examples of sets of circumstances which fall within those meanings.
GLEESON CJ: What do you say about the sentence that begins on the bottom of 1436 and goes over to the top of page 1437?
MR EMMERSON:
In our view, a process of devising a source code to perform the same function as is performed in some other source code expressed in original language does not involve creating a version of the original source code.
We say it either might or it might not. It depends whether, at the end of the day, you have sufficient objective similarity between the two codes. Now, here we had both the access to, and use of, my client’s code by the respondents and we also had the objective similarity between the two sets of codes, and where you have both of those then, we say that, prima facie, there one is an adaptation of the other even if it does not fall within the specific examples set out in paragraphs 13 and 14 of the explanatory memorandum.
GAUDRON J: You say in effect, do you - I may not have followed your argument precisely – that it is the same or similar expression of a similar set of instructions?
MR EMMERSON: Yes. It is sufficiently similar, we say, to ‑ ‑ ‑
GAUDRON J: Both in expression and in instructions?
MR EMMERSON: Yes. We say that is sufficient to make this an infringement. Just to deal with the actual form of words used in the explanatory memorandum which, in our submission, has been misconstrued, if one looks at paragraph 13 of the explanatory memorandum which starts on page 1436 at line 6:
A new definition is intended to cover translation either way between the various so-called ‘high level programming languages’…..Thus ‘adaptation’ is intended for example to cover the compilation of FORTRAN program to produce machine code…..Languages, etc of intermediate level would also be covered.
14. It is also possible for a program to be converted from object code into source code or between different languages of similar levels. In some circumstances this process will result largely in a substantial reproduction of the original program. In other cases, however, such as compilation followed by decompilation, the differences may be so substantial that one cannot speak of a reproduction although the final product is clearly derived from the original. The new definition of adaptation is intended to cover such situations.”
In our respectful submission, one could accept all of that. All these things are going to be covered by the expression “adaptation” in connection with computer programs but it does not follow from this that paragraphs 13 and 14 of the explanatory memorandum set out an exhaustive treatment of all possible cases which can be an adaptation. For that we say one goes to much broader considerations. In order to have an adaptation, of course you must have the use of the copyright work by the person who is making the adaptation, but that is clearly shown on the facts here. The third respondent made a close study of my client’s program.
But then one has to look to see what is sufficient objective similarity to make this an adaptation and in order to understand that, it is convenient, we say, to look at the words used in the definition, that was “a version of the work” and to compare it with the other parts of the definition which also used the word “version” and version can include, in other contexts, going from a novel to a play or vice versa or going from a novel to a - - -
GUMMOW J: But involved in all of that is some meaning of reproduction, is it not?
MR EMMERSON: I am sorry, your Honour?
GUMMOW J: Involved in all of that is some idea of what reproduction is not ‑ ‑ ‑
MR EMMERSON: It is, as it were – it lies outside reproduction.
GUMMOW J: Yes.
MR EMMERSON: But it has the same sort of intellectual underpinnings. It is trying to cover, we would say, what can fairly be covered in order to protect the copyright subject matter in the particular circumstances and so it is thought to be proper to give the owner of the copyright in a novel the right to prevent or licence the production of a version of the work in a dramatic form whether or not the actual words used are the same at all.
GUMMOW J: Does section 21 throw any light on this? It tells us something about reproduction.
MR EMMERSON: Section 21 gives a little bit of light in subsection (3) because it also uses the word “version”.
GUMMOW J: Yes.
MR EMMERSON: That is in the context, of course, of artistic works and it talks about two-dimensional artistic work in a version being produced in three-dimensions or vice versa. But, again, it is our submission that the notion of the word “version” is not used in our Copyright Act in as narrow a way as the Full Court seemed to have assumed in this particular case and it is not limited to translation and, indeed, it would be our submission that it would be very odd if it were limited to translation because it is certainly clear from the explanatory memorandum that there was a school of thought abroad that you did not need a new definition of “adaptation” simply to deal with translations. They would be reproductions following the language of the definition of “computer program”.
So, there are a number of indications present that the word “version” is to be given a wider connotation than the Full Court gave it and, in our submission, without attempting to be exhaustive, it is sufficient to have the findings in the present case where there was clearly access to and use of my client’s program so that there was the relevant causal connection and there is also a finding of substantial objective similarity between the two codes. We would say that is sufficient and you do not have to narrow the matter down to the mechanical process of compilation and decompilation or rendering something from one language code or notation to another.
GUMMOW J: Is there any separate pleading in respect of these macros?
MR EMMERSON: Yes, there is.
GUMMOW J: They were isolated in the further statement of claim, were they; page 28, I think it starts?
MR EMMERSON: Yes, the further statement of claim starts at page 28; the isolation – yes, it is at paragraph 6.7 in the further amended statement of claim, beginning at the bottom of page 31 in the appeal book:
The Respondents have reproduced and/or adapted the computer programs known as:
Report Macro command
Enter Group Macro command
Enter Macro command -
and those are the three macro commands which are the subject of the Second Schedule to the order made by the learned trial judge.
GUMMOW J: And was there any issue raised by 6.7 that these were not computer programs?
MR EMMERSON: I would have to, I am afraid, look into that to answer that question, your Honour.
GUMMOW J: It seems not ‑ ‑ ‑
MR EMMERSON: As far as I know, not; it does not seem to come through any of the judgments ‑ ‑ ‑
GUMMOW J: No, it does not.
MR EMMERSON: ‑ ‑ ‑but whether there was an issue somewhere along the line, I am not able to say immediately here.
GUMMOW J: Thank you.
HAYNE J: Could you remind me, Dr Emmerson, the defence at page 40 was that the source code for the macros was independently conceived and written by one of the respondents and was original and completely different.
MR EMMERSON: Yes.
HAYNE J: The finding below was substantial objective similarity. Was there any finding on the allegation of independent conception and writing?
MR EMMERSON: Yes, there was.
HAYNE J: Do not delay; perhaps you could mention it some time tomorrow.
MR EMMERSON: Yes. Perhaps if I could take that question on notice, your Honour. The matter is dealt with, so far as the macros are concerned, by the trial judge at pages 1345 to 1346. I am not able to say, though, immediately whether some of the earlier parts of his Honour’s judgment also bear on this issue.
The Full Court deals with that at 1436 to 1437, where they say that:
the evidence is clear that while Dr D Bennett carefully studied the Dataflex program so as to ensure that the PFXplus commands in question performed the same functions as the Dataflex commands, the expression of the source program as written by him –
and they say it was original. So there was a close study of the program or the way it worked. Whether I can take from that a further - I will have to check, your Honour.
GAUDRON J: Is there some incongruity in that finding? The evidence is clear there has been access to it.
MR EMMERSON: Yes.
GAUDRON J: “objectively similar to the expression”, but it was an original expression.
MR EMMERSON: It certainly cannot in these circumstances be entirely original. What we would understand the court to be saying, “Well, he put his own form of code there, but the form of code was there, nevertheless, objectively similar”. And we say that that is sufficient to amount to a reproduction.
GAUDRON J: Yes.
GLEESON CJ: Did it need to be objectively similar to the achieve his purpose?
MR EMMERSON: I do not think there is an expressed finding on that, your Honour. There are expressed findings of the need for similarity so far as the language is concerned, and that deals with the reserved words and the top tier of the macro commands.
GLEESON CJ: That is because of the convenience of the users, as I understand it.
MR EMMERSON: That is correct, yes. I do not think there is any finding of any need for objective similarity other than perhaps that was easier, which one might infer, at the second level down so far as the macros are concerned. There seems to be no argument either way, either it being said they had to do that, or it being said that it was somehow necessary to achieve interoperability.
GAUDRON J: Can I take you then to page 1437. I am having difficulty, Mr Emmerson, with this paragraph in the Full Court. It is the words “expressed in original language” at lines 1 and 2. I just cannot understand what is there being said. Is it because it is expressed in original language that it is not a version? Is that what is being said?
MR EMMERSON: The answer is yes, but it all depends on what you mean by “not expressed in original language”. As we read the Full Court, the Full Court is really confining itself to a connection of the sort which could be produced by using a computer; that is to say, compilation from source code to object code, decompilation or one after the other in succession.
It is saying once somebody sits down to try to do the same thing but writes code, then that is sufficient to break the nexus and we would say, “No, that is a bit like saying that in order for a play to be an adaptation of a novel it would be necessary for it to use the same words”. We say it is not necessary to use the same words but whereas here you have an identity of function and a strong similarity of form at the objective level, then that is sufficient to amount to an adaptation.
HAYNE J: You say strong similarity of objective form, that is revealed if, but only if, what is it, I‑language, I‑mac?
MR EMMERSON: I‑code.
HAYNE J: I‑code is first translated to source code, is it not?
MR EMMERSON: Yes, by a mechanical process.
HAYNE J: Yes, and is it suggested that Dr Bennett had done that as part of the process of construction of his program?
MR EMMERSON: I do not think it was, your Honour, but again I would have to look at precisely what is known about him.
HAYNE J: Was it put to him in evidence that that is how it was done.
McHUGH J: There was no finding against him of decompilation, was there?
MR EMMERSON: No, there was not. We would say, of course, that that, on ordinary copyright principles, cannot give him an answer. Where there is use of the copyright work and there is sufficient objective similarity then it is not to the point to say the actual infringer had never seen the original work. This occurs commonly in the case of copyright in engineering drawings, for instance. Engineering drawings are protectable as an artistic work. Someone makes something carrying out those. Somebody else copies what is made. They have no access to the original artistic work but they, nevertheless, can infringe where ‑ ‑ ‑
GUMMOW J: They can copy a copy.
MR EMMERSON: Exactly, yes, and this indirect infringement at least must have been the case with Dr Bennett in the present case.
GAUDRON J: In the context of this case, would it be a fair analogy to talk about transposition of music from one key to another, that seeming to be what the Court of Appeal had in mind, as distinct from someone who, say, sometime before has heard the tune but instead of transposing it writes a version of it in a different key, without going through the mechanics of transposition?
MR EMMERSON: With respect, your Honour, that probably gives a good analogy to the division between the two courts in this case, because the transposition to another key is very similar to the process of compilation or decompilation, whereas in order to make an arrangement of the music or do something which was slightly different, then you are going wider. That is in fact something which has some echoes in the definition of “adaptation” itself because in relation to a music work it is an adaptation if you make an arrangement or transcription of the work. That would seem to cover at least some of the ground that your Honour was putting to me as a second possibility.
McHUGH J: Dr Emmerson, I do not know whether this has much to do with it but it is something I may give some thought to at some later stage in this case when I am writing a judgment: many programs are so structured that they are able to read other programs and transfer them across. Is there any suggestion of infringement? You of course would say no, I suppose, would you? Supposing, for example, Access wants to read a program in D base or Foxpro or something like that or Excel and carry it across into its own program or allow the user to do it. Does it give a set of instructions in, say, Access to copy this or copy that?
MR EMMERSON: There is, if I may say so, an unresolved issue of law here. At the risk that what I am going to say might turn out to be out of date, the position is as follows: yes, if you input a program into a computer even for the purpose of compilation or doing various other things, at least in a transitory sense you store it in the computer. It is therefore relevantly in material form for the purposes of the Copyright Act because “material form” is defined as including:
any form (whether visible or not) of storage from which the work or adaptation, or a substantial part of the work or adaptation, can be reproduced.
The question arose – I think it was in Autodesk at first instance – as to whether an act simply of inputting a copyright program itself involved the act of reproduction of it because it was rendered into a material form within the definition of the Act.
The court at first instance said, no, it was not, this was too transitory to be in material form and there the matter, so far as I am aware, has rested. But it is a question which has not been considered by any appellate court as to whether that would or would not amount to a reproduction.
McHUGH J: So, if Word, for example, reads a Wordperfect program then arguably translates it into a Word program, it is arguable that there has been an infringement.
MR EMMERSON: Yes. One final matter just to round out the adaptation. We draw attention to the paragraph (ba) of “adaptation” in that it says:
in relation to a literary work being a computer program – a version of the work (whether or not in the language, code or notation in which the work was originally expressed) –
Now, that suggests that you could have an adaptation of a computer program by taking something into source code and adapting it, still in the same source code, but in such a way that it was not sufficiently close to be a reproduction but it was an adaptation and it suggests that Parliament had in mind something as wide as that in the context of the definition. I should say that no doubt it could also be contended that this was simply a longhand way of saying, “Well, you could have compilation followed by decompilation in which case you would get back to the original language code or notation”, but we would submit that this is a terribly longhand way of capturing a relatively small point if it is intended simply to be limited to the case where you have compilation followed by decompilation.
Can I next turn to another way in which we put our case, and that is the way set out in paragraph 6(b)? We say:
copyright subsists in the collocation of Reserved Words comprising the DataFlex language on the footing that this collocation is a computer program within the said definition;
Now, the distinction between this and what I have put to the Court so far in argument is that here we are concerned with the totality of the reserved words and we are saying even if individually they are not a computer program, collectively they are a computer program.
Now, this was dealt with by the Full Court at page 1432, lines 4 to 11, where the Full Court said:
Senior counsel for Data Access sought to argue in the alternative that if the individual words were not the subject of copyright, then the group of words together was. However, it is just as impossible to say of the entire set of commands that it is itself an expression of a set of instructions, as it is to say that in respect of any individual word. It is true that each of the words appears, as we have already indicated, in one form or another, in the source code of the Dataflex program. But it is not true to say that the words collectively are a computer program. They are merely the totality of commands available for use in the particular program.
Now, we take issue with that reasoning. We say that again one needs to go back to the definition of “computer program”:
an expression, in any language, code or notation, of a set of instructions (whether with or without related information) intended, either directly or after either or both of…..to cause a device having digital information processing capabilities to perform a particular function.
Now, here we would say that even if we are wrong in saying that each individual word, taken on its own, is an expression of a set of instructions, the group of words which together comprise the totality of the Dataflex language are, in our submission, just such, a set of words and we say that the Full Court, accepting as it did that the individual words were instructions, should also have said, “Well, if that is so then the entire group must be a set of instructions”. The Full Court had itself said at 1428:
It may not be inaccurate to describe each of the commands as itself an instruction. It is likewise not necessarily…..each of those words as representing the set of instructions –
and then we get on to the point of whether one is so general as to be merely what the court calls “a trigger”.
But once you have got to the totality of the Dataflex language, in our submission there is no room for argument that you do not here have a set of instructions and indeed, the set of instructions are quite central to the Dataflex system, because this is the set of instructions which together comprise the language of Dataflex.
Now, once again one can, as a matter of statutory construction, get to this point in two different ways: first, following out of our earlier submissions that each of the reserved words is a computer program, for reasons which obviously I do not restate; we say therefore the collocation of reserved words is a compilation of computer programs and that if you go to the definition of “literary work” in the Act, you find, under paragraph (b), that it can include:
a computer program or compilation of computer programs.
It follows, we say, that even if for some reason individually the reserved words do not qualify for copyright as computer programs, nevertheless, when you take the lot of them together, being the totality of the Dataflex language, here you have a compilation of computer programs and that immediately brings you within the definition of “literary work”, which immediately takes you to the subsistence of copyright via section 32 in the manner which I have already put to the Court.
HAYNE J: The hypothesis for this limb of the argument is that we have found that they are not computer programs, individually?
MR EMMERSON: No, the hypothesis is that they are computer programs, but for some reason we are denied copyright in them and if that is so, if it is said for some reason that, albeit computer programs they are not copyright, we say well that objection does not apply to the collocation of the whole lot.
GUMMOW J: But this word “collocation”, where does that come from in the statute?
MR EMMERSON: Collocation, I am sorry to say, your Honour, I think comes from our original notes of appeal. The word in the statute ‑ ‑ ‑
GUMMOW J: Is compilation, is it not?
MR EMMERSON: - - -so far as “literary work” is concerned, is compilation.
GUMMOW J: Yes. And what is the meaning of compilation? It involves some systematic connection or ordering?
MR EMMERSON: It does not involve necessarily a systematic ordering; it involves some systematic collection, as of course occurred in this case, because the reserved words with which we are concerned are the Dataflex language.
GUMMOW J: But, compiled to produce something integral, in some sense, surely?
MR EMMERSON: Yes, compiled to ‑ ‑ ‑
GUMMOW J: It is put against you that that is absent here.
MR EMMERSON: Well, in our submission, the totality of the language is something integral. It is integral in the same sense that a dictionary is integral. You set out the various words and you set out the meaning.
GUMMOW J: Yes, but the dictionary will be a dictionary of engineering terms, of legal terms, whatever.
MR EMMERSON: Yes.
GUMMOW J: It would purport to define its integration, as it were, or it may be the whole English language, but it is still defining its universe.
MR EMMERSON: Of course, we say that one of the special things about our system is that within this language you can do a whole lot of very useful things. That is to say, we say that the language itself is useful and it is integrated. It is not just a scattered concatenation of things. It is a set of words which have been constructed and compiled and which together can form the Dataflex language which allows people, of course, to carry out the data handling operations with which the system is concerned.
GUMMOW J: But this term “reserved words” is no more than the collection found in a particular schedule in an order, is it not?
MR EMMERSON: The order is unimportant. The totality of members of that ‑ ‑ ‑
GUMMOW J: Well, the reason why you got the order was that they had been filched, so you say.
MR EMMERSON: Yes.
GUMMOW J: But anterior to the filching, what integral character do they have?
MR EMMERSON: Anterior to the filching, these are the reserved words in my client’s language. The numbers are set out, I think, at the start of the trial judge’s reasons for judgment at 1336. At line 10, his Honour says this:
In the PFXplus language –
that is to say the respondents’ language -
use is not made of the 29 words of the DataFlex language which express commands for developing graphics.
So, you put graphics to one side. He then said:
Of the other 225 words of the DataFlex language which are listed in the DataFlex Encyclopaedia…..192 are in the PFXplus language…..And the use of each of those 192 words in the source code of either language contributes to cause a device having digital information processing capabilities to perform the same function.
So, we have here not just some sort of random selection; we have the totality of the Dataflex language which is 225 plus 29 words. Putting to one side those words which are concerned with graphics, we have 225 of which 192 are taken.
It is our submission that the totality of the words in the Dataflex language, each of which, in the sense in which I was arguing this morning, is a computer program, amounts to a compilation of computer programs. There they are, they are the words in the language, they are listed in the Dataflex encyclopaedia and that is the lot of them.
GAUDRON J: Can I take you back to what you earlier said, Mr Emmerson. You said this is on the hypothesis that the individual words are computer programs, but for some reason you were refused copyright.
MR EMMERSON: Yes.
GAUDRON J: Is there any issue in this case which would lead to your being refused copyright if the individual words satisfied the definition of a computer program in the Act?
MR EMMERSON: We, of course, would say no, but we have to face the fact that, as questions from the Bench this morning indicated, there is an argument against us that all this is somehow simple, and if that argument against our submission happened to succeed, then it is theoretically possible that the conclusion could be that the reserved words are computer programs but copyright does not subsist in them.
GAUDRON J: Because?
MR EMMERSON: Because ‑ ‑ ‑
GAUDRON J: Because they are simple.
MR EMMERSON: Either because they were simple, or because - I find myself embarrassed obviously putting an argument which we would not endorse but saying that this is an argument which we try to cover.
HAYNE J: Why is that? There are pleadings in the case. The case has gone to trial, to intermediate appeal. Why is it now not plain what are the issues in the case? Why can a simple answer not be given to Justice Gaudron’s question?
MR EMMERSON: The simple answer which we would, with respect, give to Justice Gaudron’s question is that if there is no issue here, we would be entirely happy. If there is, this is what we say about it.
GAUDRON J: If one were to find that the individual words are a computer program within the definition of the Act, and that copyright subsists, there is then no reason to look at this argument.
MR EMMERSON: Then one does not need to go further to the collocation.
GAUDRON J: If one finds that they are not computer programs, the argument does not arise?
MR EMMERSON: Yes, but if one finds that they are not computer programs, then one gets to the argument that we set out in our paragraph 34, which is the next paragraph that I am coming to, because we say that, in any event, whether or not the individual words are computer programs, the collocation of reserved words is itself a computer program within the definition in section 10. That is to say, it is an expression; the language code or notation is the Dataflex language; it expresses a set of instructions, namely the meaning and syntax of each of the words or commands – they are all instructions - and the set, of course, includes the totality of the Dataflex language ‑ ‑ ‑
GUMMOW J: But you only get into this if the integers, as it were, are each themselves a computer program. That just flows in the definition of “literary work”, does it not?
MR EMMERSON: Not necessarily.
GUMMOW J: In other words, the word “compilation” is used rather differently in paragraphs (a) and (b) of the definition of “literary work”. It has never been suggested, I do not think, that in order to have a compilation of answers (a), you have first to decide whether each element in the compilation itself is an independent work.
MR EMMERSON: No, that is correct. Your Honour, what we set out in paragraph 33 which deals with compilation of computer programs, does not depend on what we set out in paragraph 34. That is to say, they are two independent arguments and we say that either or both of them are true.
GUMMOW J: Well, that is giving this word “collocation” a different meaning to “compilation”.
MR EMMERSON: Yes. You do not get to 34 by way of compilation at all. Paragraph 33 is saying, well, look, one way we can get there is to say each of them is a computer program. The collocation of them is a compilation of computer programs within the meaning of “literary work”.
GUMMOW J: Yes, the devil is in the word “collocation” I suspect.
MR EMMERSON: Yes.
GLEESON CJ: Could I take you back to page 1432. The paragraph that summarises the reasoning of the Full Court on this issue is on that page, is it not?
MR EMMERSON: That is correct, yes, your Honour.
GLEESON CJ: Now, may I ask you this: if the reasoning of the Full Court up to that point is correct, is there anything wrong with the reasoning in that paragraph?
MR EMMERSON: If it is correct in its totality, that is to say if there are various reasons for rejecting my client’s submissions and those reasons are all correct, then probably we do not succeed under this paragraph. The alternative way in which we are putting the matter deals with the case where it is asked how can a single command be a computer program? We say first that it can be and it is for reasons we have set out. But if we are wrong on that, we would say at least collectively the collocation of commands is sufficient. The reasoning in page 1432 we say simply goes wrong because the Court is apparently able to accept that each of the commands individually is an instruction. It does that at the top of 1428.
If it says, “That is not enough. Where’s the set of instructions?”, our answer is that on that view the set of instructions is found in the instructions taken together. We say that the reasoning at 1432 is wrong because it is not correct to say that it is just as impossible to say that the entire set of commands is an expression of a set of instructions. It all depends for what reason one concludes, if one does, that the individual commands are not a set of instructions. So it is perfectly possible as a matter of logic to say one finds that the individual commands are individually instructions but they are not sets of instructions within the definition. If that, contrary to our submission, is the correct view, then we say that argument does not avail against our contention that collectively the commands are a set of instructions.
We have dealt with that in paragraphs 33 to 35 of our written outline. An alternative way that we put the case is that set out in paragraph 6(c) of our written outline that, even if the collocation of the reserved words is not itself a literary work, it nevertheless forms a substantial part of a literary work by way of adaptation, so that copying of it is an infringement of the Dataflex system.
GUMMOW J: Now, there is complaint about this not having been taken at the trial.
MR EMMERSON: Well, yes, your Honour, but we say that in fact this point is dealt with by the trial judge.
GUMMOW J: In a way the Act itself deals with it by section 14. Dealing is a dealing in relation to “a substantial part”.
MR EMMERSON: Yes.
GUMMOW J: It is a very odd copyright case maybe where it was totality rather than substantiality.
MR EMMERSON: Absolutely, your Honour, yes. It always must be an infringement if what has been taken is a substantial part of the copyright work, and in this particular case that has happened. Dealing with this in the court below, his Honour dealt with it at page 1354 beginning at line 27 on that page, where his Honour said:
If I have been in error in treating a single word of the DataFlex language as “a set of instructions”, within the meaning of that expression in the definition of “computer program” –
and there is a bit in parentheses saying why that might be –
I would be of the opinion that Dr Bennett’s copying of the words constituted an adaptation, within the defined meaning of that word, of a substantial part of each of the computer programs comprising the DataFlex application development system: an adaptation because PFXplus is a version of part of each of those programs and a substantial part because what has been copied informs virtually all of the expression of the instructions which cause the device having digital information processing capabilities to perform the functions of the development application system.
So his Honour there deals with the infringement here on the footing that the reserved words form a substantial part of the computer programs on the footing that virtually all the expression of instructions that caused the device to have digital information processing capabilities, to perform functions of the development application system.
So, practically all is taken of the expression of instructions, substantially the entire vocabulary is taken, and his Honour regarded that as a substantial part and, accordingly, an infringement. So, in our respectful submission, it is clear that this was a matter considered by the court and it is one which one can fairly say is still a live issue. Now, that being so, in our submission, the reasoning of the sort that was upheld by this Court in Autodesk applies. Here you have a system for which, if one may, with respect, adopt the expression “the linchpin” is the set of instructions, the vocabulary of the Dataflex language. If you find that substantially all of that Dataflex language is taken, we say that that forms a substantial part of what is on any view a protected computer program, namely, the Dataflex system.
There is acknowledgment by the respondents that copyright subsists in the Dataflex system. It is clear that this part has been taken. We say that this part is a substantial part and by reason of section 14 of the Act what has therefore been taken is a substantial part of a copyright system and it follows that copyright has been infringed.
GUMMOW J: The mere fact that an element is essential or in fact is substantial, is it?
MR EMMERSON: Well, the words are “substantial powers”.
GUMMOW J: Yes. In a sense, every element in all of this is essential, because without it will not work. That does not mean merely to extract one is to extract a substantial power.
MR EMMERSON: We would not necessarily accept as the fact that every element was substantial in the sense that without it, it would not work. I am conscious that this is an argument which has had some airing.
GUMMOW J: Yes.
MR EMMERSON: But it is not an argument which, in our submission, is founded on the facts as found in the present case, namely that you cannot take out any bit of it without it stopping it working. We would look at the question much more broadly and say, after all, this is a system which is designed to make it possible for people to develop their own database applications. There is the skill and ingenuity of setting up a set of reserved words which form the vocabulary which allow you to do that. This is truly the heart of the system and we do not really need to go to the second step of saying, “Well, anything which is essential to the operating of the system in the sense that it will not operate without it forms a substantial part”, because one could imagine there could be room for argument about that.
GUMMOW J: Yes, so in using the phrase “the heart of the system” you are drawing in some evaluative element.
MR EMMERSON: Yes, or, with respect, I would adopt what Justice Gaudron said in Autodesk. It is the linchpin of the system.
GAUDRON J: Yes, but that was in a different context, because the issue there was whether data could be a substantial part, could even be a part of a computer program.
MR EMMERSON: Yes, I fully accept that, your Honour, and as I think we said in submissions on my learned friend’s proposed cross‑claim, that was the context in which your Honour wrote those words. Nevertheless, if one is looking at the question of substantiality it is, in our submission, well settled that what one is looking at is something which is qualitative rather than quantitative and that here, in the context of the facts of this particular case, the trial judge was clearly right in finding that the totality of the words in the Dataflex language were a substantial part of the Dataflex system.
GUMMOW J: Granting all of that, what then do you say about Mr Macaw’s submissions at paragraph 16 on page 11? The thrust of that is that the trial judge is really flying on a wing and a prayer in saying what he did, when he did not have an evidentiary underpinning.
MR EMMERSON: What our learned friend says is you cannot have an evidentiary underpinning without a quantitative account of how much has been taken. Now, we start to diverge at that very fundamental level because, in our submission, it is clear on the authorities that the test for what is a substantial part is qualitative and is not quantitative and further that ‑ ‑ ‑
GUMMOW J: Well, it may have characteristics of both.
MR EMMERSON: Well, you may show that something is substantial by showing that it is quantitatively substantial, for instance, suppose it is the greater part of the program. But you may also show that something is substantial by showing that it is qualitatively substantial as indeed, of course, was the case in Autodesk when there were only, I think, 127 bits of information out of many tens of thousands in the program. But, we would say that it is just simply not true to say that a quantitative analysis is an essential element of an assessment of substantiality. That is point number one. But point number two: we dispute the proposition that there was no quantitative analysis below. There was evidence before his Honour of the proportions that the various components of the system bore to each other.
So, we would say that the factual underpinning from what our learned friends say in paragraph 16 has simply not been made out by them. The fact of the matter, in our submission, is it was before the trial judge. He decided the matter correctly on qualitative considerations but this neither
shows that he should have taken into account quantitative considerations; more that there were no quantitative considerations before the court.
GLEESON CJ: Is that a convenient time, Mr Emmerson?
MR EMMERSON: If the Court pleases.
GLEESON CJ: How long do you expect to require to complete your submissions?
MR EMMERSON: I would not expect to be more than about half an hour, your Honour.
GLEESON CJ: How long do you expect to be, Mr Macaw?
MR MACAW: I would expect to be finished by lunch time, your Honour, then I would ask my learned junior to present the case we wish to make on the cross appeal.
GLEESON CJ: Very well. Then, we will adjourn until 10.15 am tomorrow morning.
AT 4.16 PM THE MATTER WAS ADJOURNED
UNTIL WEDNESDAY, 3 FEBRUARY 1999
6
3
0