Autodesk Inc & Anor v Dyason

[1990] HCATrans 280

:

-!) 1 AUSTRALIA,&

-.).);,~ )> ~.««<.i.. '-

--

IN THE HIGH COURT OF AUSTRALIA

Office of the Registry

Melbourne No M29 of 1990

B e t w e e n -

AUTODESK INC. and AUTODESK

AUSTRALIA PTY LTD

Applicants

and

MARTIN PATRICK DYASON,
CHRISTINE DYASON and PETER

VINCENT KELLY

Respondents

Application for special leave

to appeal

MASON CJ
GAUDRON J

MCHUGH J

Autodesk 1 16/11/90

TRANSCRIPT OF PROCEEDINGS

AT MELBOURNE ON FRIDAY, 16 NOVEMBER 1990, AT 10.13 AM

Copyright in the High Court of Australia

MR J.M. EMMERSON, OC: If the Court pleases, I appear with

my learned friend, MR A.K. PANNA, for the applicant

in this matter. (instructed by Stephens

Solicitors)

MR J.W.K. BURNSIDE, OC: If the Court pleases, I appear with

my learned friend, MR I. WALLER, for the

respondent. (instructed by Nevett Ford)

MASON CJ: Yes, Mr Emmerson.

MR EMMERSON:  If the Court pleases, this application raises

as its principal issue the nature of the protection

given by the Copyright Act to computer programs.

The Court is familiar with the broad approach

adopted by the Copyright Act to various types of

works. It asks whether copyright subsists in the

alleged subject-matter and, if so, whether it is

infringed.

If one gets to the question of infringement then the tests which are laid down in the Act are

to ask whether the copyright work has been reproduced in a material form or whether a

substantial part of it has been reproduced.

Alternatively, the question arises whether what the

defendant has done is to make an adaptation of the

work or to reproduce an adaptation of the work or a

substantial part of it.

Now, on all those issues it is familiar law

that one considers two things; one is whether the

allegedly infringing work sufficiently resembles

the copyright work and the second is whether it was

produced by the use of the copyright work. Now,

the specific issues raised in the present
application are what is the type of resemblance, in
the context of computer programs, which is required
to constitute a reproduction and what is the type

of resemblance that is required to be an

adaptation. Now, in our submission, the resolution

of these two questions are the critical issues in

determining the nature of the protection given by

the Copyright Act to computer programs.

If the Court pleases, this is the first case

that has come before appellate courts dealing with the effect of the amendments which were introduced to the Copyright Act in 1984 for the express

purpose of protecting computer programs and so it

is the first case in which the type of resemblance
required to constitute a reproduction or an
adaptation has come up for consideration.

The Court will be familiar with the earlier

litigation in the Apple v Computer Edge case which
was determined under the old Act. There the

Autodesk 2 16/11/90

plaintiff party failed, in effect, because it

failed to satisfy the Court that there was a

sufficient resemblance between certain chips or

ROMs and a computer program considered as a

literary work. Now, between the time when the Full

Court of the Federal Court gave its decision and

this Court gave its decision, the 1984 amendments

were introduced for the purpose of giving such

copyright protection.

When this Court came to consider the Computer

Edge case, the Court recognized that future cases

would fall to be considered in a different

statutory context and, of course, that is what has

happened in the present case.

Now, the way in which Parliament chose to give

protection to computer programs is that it

introduced a number of definitions. It, in

particular, made computer programs fall within the

scope of the expression "literary work" and it

introduced a particular definition of "computer
program". It did not, in terms, lay down what is

the character of the resemblance which is required

in order to make a finding of infringement or a

finding of adaptation. But it is our submission

that the correct approach to this legislation is to

construe it as remedial legislation which was

intended to give effective protection to computer

pro~rams and was intended to make an infringement

the making and selling of devices which were

programed with the program in suit.

Now, it is the broad question of approach

which has divided the Federal Court. The learned trial judge, in our submission, correctly treated

the legislation as being broadly remedial and

looked to seeing the type of protection which was

given by it. The Full Court, in our submission,

incorrectly, having accepted that the legislation

gives protection to computer programs, then when it

came to consider what is the type of protection

given, that is to say, what sort of things

constitute reproductions or adaptations, the Court

fell back on the law as it had existed before the

1984 amendments and as it had been stated by the

majority of this Court in the Computer Edge case.

Now, in our submission, the Court erred in so doing because, in considering matters of copyright,

when you come to consider what is a sufficient

resemblance to lead to a finding of reproduction or

of adaptation it is of crucial importance to take

into account the nature of the subject-matter that

is being protected by that particular part of the

Copyright Act. Where, as in the Computer Edge

case, what is sought to be protected is a literary

Autodesk 3 16/11/90

work, and the cases had shown that a literary work

in this context meant something in print or

writing, then it is easy to find that a mechanical

device like a silicon chip is not a reproduction of

a literary work being a set of things written down

on paper and constituting a computer program.

Now, in our submission, the learned trial

judge correctly recognized that once one makes

express provision for new subject-matter, the new

subject-matter being computer programs, then one

has to recognize that this necessarily changes the

analysis of what is the type of resemblance that
one must look for in order to find that there has

been reproduction or adaptation.

MASON CJ:	Mr Emmerson, assuming for the moment that on the issue of reproduction or adaptation you have
identified a question of general principle
sufficiently important to warrant the grant of
special leave, are you none the less faced with
this difficulty, that there has been an adverse
finding against you on the question of copyright, ie, that the AutoCAD lock does not satisfy the
statutory definition of "computer program"? Now,
you would have to succeed on that issue in this
Court before we come to consider the question of
general principle that you have been elaborating so far.

MR EMMERSON: If Your Honour pleases, we would, with

respect, disagree with that. We say we should

succeed and we are entitled to succeed on that

point but we say it is not necessary for us to

succeed. Your Honour has, with respect, correctly

identified the second of the criticisms which we

make of the judgment of the Full Court.

MASON CJ: And you come back to the third issue then, do

you, as a response to the comment I have made?

MR EMMERSON:  As a response to the comment Your Honour has

made, the way that we say that this matter arises	is that the Court wrongly allowed itself to focus
solely on the nature of my client's lock and it ignored the fact that both the trial judge and,
indeed, the members of the Full Court accepted that the AutoCAD program plus lock was, in fact, a
computer program. So that when one comes to look
to see between what must there be resemblance, one
looks not only to resemblance to the AutoCAD lock considered in isolation but one looks to
resemblance between the whole complex because as
one member of the Full Court correctly pointed out, the so-called lock is really in the nature of a key
and the key and the lock are complementary and
inseparable. The court, in our submission, in

Autodesk 4 16/11/90

considering whether copyright subsisted in the
so-called lock of my client's product wrongly
focused on the key only and not the combination of

the key and the lock and not, in particular, the

combination of the lock in the computer itself

which, on any view, was a computer program. So,

the program called Widget C was accepted by all

members of both courts as being a computer program.

GAUDRON J: What, Mr Emmerson, may I ask, on that analysis

do you take to be the expression of the set of

instructions?

MR EMMERSON:  The expression of the set of instructions is

the Widget C program.

GAUDRON J: But that was not really reproduced, was it, or

adapted?

MR EMMERSON:  We say it was but there are perhaps two

elements to the answer to Your Honour's question.

As far as reproduction was concerned, we say there

are findings which amount to a finding of

reproduction of the expression in Widget C, that is

to say, there was a look-up table in Widget C,

there was a look-up table in the allegedly

infringing key, so, the expression is the same.

But, secondly, we say that it is sufficient for the

purpose of finding an infringement if one is to
give a practical effect to the 1984 amendments that

you have a device which has the right kind of

resemblance to the program. One does not have to

have something which itself has a set of

instructions.

GAUDRON J: Then it would be very simple, in terms of the

statutory definition, if you took the set of

instructions to be the emission of the signals from

the key.

MR EMMERSON:  The essential part of the set of instructions

on which we rely is the set of instructions

embodying the so-called 127 state machine and we	say that that set of instructions, together with
the wrap-around facility is something which was reproduced in the allegedly infringing key and, indeed, had to be reproduced because it was
absolutely essential to the activities of the
respondent that they did not go about completely
independently creating their own device. They had
to copy our's because they had to make it work. In
a sense, if you are making a key for a lock, you
have got to make a key which precisely fits the
lock, not any other lock.

As to what are instructions for the purposes

of this part of the law, we say that you have

Autodesk 16/11/90

sufficient instructions at least if, as in the

present case with the allegedly infringing key, you

have a wrong which is programed with a series of

numbers which operate so that when the device

functions one sends back the appropriate signal,

because that is what computer technology is all

about.

If the Court pleases, we say that the

essential question that one has to look at and the

one to which all these other matters really are
subsidiary is whether we are correct in saying that
the protection given by the Act extends to a device
which, in the words of the definition of "computer

program" performs the particular function which the

program is designed and intended to make it

perform. We say that is sufficient. We say that

is the critical point at which the analysis of the

Full Court went wrong.

The Full Court, however, also went wrong in

focusing too narrowly on our key while ignoring

that our lock was itself a computer program on any

view and we say, further, that the Full Court,

unlike His Honour the trial judge, in fact,

therefore, never got on to, really, considering

whether the respondent's key had in it a computer

program.

MASON CJ: Mr Emmerson, we will not trouble you further at

this stage and we will hear what Mr Burnside has to

say.

MR EMMERSON: If the Court pleases.

MR BURNSIDE:  If the Court pleases, on the first question

which Dr Emmerson identifies as arising on this

application, we would agree that it is a question

of great interest and ultimately one of real

importance but it is our primary contention that it

does not really arise appropriately in this case

because the decision below is not attended by
sufficient doubt.	In order to succeed on the point

which they seek to raise, what the applicants must

do is demonstrate not only that the test of

resemblance relevant in the field of computer

programs has been altered by the definitions but

that the test can be satisfied by a similarity of

function alone.

Now, it may be that the Parliament intended,

by the amendments in 1984, to introduce, albeit

only by implication, some different form of test

for resemblance and for adaptation where the

subject-matter of copyright is a computer program.

Assume that for the purpose of argument.

Nevertheless, the only resemblance which was found

Autodesk 6 16/11/90

by any judge, either at first instance or in the

Full Court, was that the two keys performed the

same function. Every member of the Full Court said

that each key adopted different algorithms, that is

to say, the theoretical outline of the logical

approach adopted, and each algorithm was

implemented differently and all they had in common

was that they performed the same function.

So, in order to succeed on the point they wish to raise, what the applicants must do is to satisfy

the Court that by reason of the definitions

introduced in the 1984 amendments the Parliament

has swept aside the dichotomy between idea and

been achieved by the use of the word "function" in

expression which had been fundamental to the law of

copyright for a century and can be traced back to

the definition of "computer program.	We say that
is plainly wrong.	Each member of the Full Court

dealt with it squarely and said it was plainly

wrong and we say, accordingly, there is no

sufficient doubt on that question. Now, without

repeating myself, it is difficult to enlarge on

that argument.

GAUDRON J: But that really does depend, does it not, on how you identify or what you identify as the expression

of a set of instructions?

MR BURNSIDE:  We would agree with that, with respect, but

making every assumption in favour of the

applicants, even if you assume that the AutoCAD

device is a program, even if you assume that the say that it is not possible to find any relevant

similarity between them except function.

GAUDRON J:  I do not follow that, I must say. The

difference between your hexadecimal expression and

your binary expression is really a difference of 1, is it not? And the function is achieved by a short

circuiting, is it not? I mean, is it really not as

if you took a large concerto, adapted the tempo,

and replayed the chorus?

MR BURNSIDE:  With respect, no. A program can go through

many phases. One first starts with the problem.

The problem may lend itself to a logical solution

in a variety of different ways. At that level of

generalization the solution adopted is conveniently

described as an algorithm, it is not a program.

The next step, after deciding on the logical

approach to the solution of the problem, is to

create a set of instructions which will, either

directly or after relevant transformation into a

different form, cause the computer to act in a

Autodesk 16/11/90

particular way and that is, effectively, what the

definition of "computer program" is and we say it

is an appropriate definition. But, in this case

what the court had in front of it was a description

of how each of the locks worked and each member of the court held that not only were they not similar

in their implementation, they did not even

implement similar algorithms, that is to say, they

took entirely different logical approaches to the

solution of the same problem.

GAUDRON J: But so, for example, is the different notation

of music.

MR BURNSIDE: But this is not a question of transformation

between one mode of representing a thing and

another mode of representing the same thing.

GAUDRON J: But that again comes back to what is the

expression of your set of instructions.

MR BURNSIDE: Yes.

GAUDRON J:  And if you look at it as the sequence of digits

sent back - - -?

MR BURNSIDE:  If that is said to be the copyright work, then

this is the first place in which that suggestion

has been raised.

GAUDRON J: Yes, I am aware of that.

MR BURNSIDE: It is not pleaded. It has not been argued

and, we would say, this is certainly not the time

at which it can be raised because to suggest that

would raise a whole range of quite different issues

including, for example, whether a sequence of 127

binary digits is capable of being a subject-matter

of copyright at all. We would say that it is as

trivial as the name Exxon, perhaps even more
trivial, and would fail for that sort of reason.

Furthermore, we would say that it cannot properly

be described as a set of instructions. But the

answer to that suggestion really operates at quite

a different level from the answer which we raise

against the case we have, in fact, met. The case

we have met is that there is in the lock or the key

a program and that there is in our key a program.

We say that is not so but we assume it in favour of the applicants.

If there is, in each of those devices, a set

of instructions, then one has to identify what the

set of instructions is. It is no good looking at

the devices. They are physical devices and they

give no information whatever. They can be

described in natural language because they are made

Autodesk 16/11/90

up of circuits and circuits are, by their nature,

capable of being described. When one describes the

circuits in the two devices, it becomes abundantly
clear that they comprise different components that

do different things and solve the underlying

problem in different ways, and that is what each

member of the court below held.

We would say that that is what was also held

by the trial judge although His Honour, at first

instance, thought that similarity of function was

sufficient resemblance between two devices which

were, in all other respects, different.

Now, the function of the two keys, in this

case, is to produce a certain output, that is to

say, a string of 127 bits, a particular string of

127 bits. That is the function, and we say that

that function is to be equated with the idea of the

device. That is its purpose; that is the notion

for which the device is embodied. The evidence at

first instance was that there is, in theory at

least, an infinite number of ways of producing that

same 127-bit string. The courts have held, at each

level, that the two locks each use different ways

to produce the 127 bits.

Now, the question, in our submission, which

really causes the applicants to fail in this case

and must cause them to fail is that the only

similarity between the two programs, if one calls

them that, is that they produce the same output and

if that is sufficient similarity to constitute one

a reproduction or adaptation of the other, then

quite fundamental principles of copyright law have

been swept aside in the amendments in 1984. It was
plain in the second reading speech that that was
not intended and we would say it is plain, in
looking at the definitions, that that was not

intended, but that is the result that the

applicants must achieve if their point is to have

any prospect of success on appeal.

If there is any suggestion that special leave

might be granted to argue a case for copyright of

the 127 bits, we would say that that is a procedural unfairness of very substantial

proportions. I do not think that we can take the

submissions any further unless there are any

matters that I have left unclear.

MASON CJ:  Yes, thank you. The Court need not trouble you,

Mr Emmerson. There will be a grant of special

leave in this case.

AT 10.42 AM THE MATTER WAS ADJOURNED SINE DIE

Autodesk 16/11/90