Autodesk Inc & Anor v Dyason

[1991] HCATrans 95

_.

"".J, AUST1'~1A,,.r

-~»~~--~

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

Appellants

and

MARTIN PATRICK DYASON,
CHRISTINE DYASON and PETER

VINCENT KELLY

Respondents

MASON CJ
BRENNAN J
DEANE J
DAWSON J

GAUDRON J

Autodesk(2) 1 17/4/91

TRANSCRIPT OF PROCEEDINGS

AT CANBERRA ON WEDNESDAY, 17 APRIL 1991, AT 10.20 AM

Copyright in the High Court of Australia

MR J.McL. EMMERSON, QC: If the Court pleases, I appear with

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

apQellants in this matter. (instructed by Stephens

··so!ici tors)

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

my learned friend, MR I.G. WALLER, for the

respondents. (instructed by Nevett Ford)

MASON CJ:  Yes Mr Emmerson. Yes Mr Burnside. Mr Emmerson.

MR EMMERSON:  If the Court pleases, we hand up copies of our

outline of argument.

MASON CJ:	Thank you.
MR EMMERSON:	There is a minor misprint at page 5, paragraph 12. In the first line of that paragraph
the word should be "adaptation" not "adaption".
MASON CJ:	Yes Mr Emmerson.

MR EMMERSON: If the Court pleases, I propose in these

submissions to introduce our appeal by giving to

the Court a broad outline of the facts that we say

are important and the relationship between these

facts and the points on which, in our submission,

the Full Court of the Federal Court went wrong. I

would then propose to take the Court in some detail to the 1984 amendments to put our submissions as to what we say is the purpose and the effect of those

amendments. I would then propose to deal with the

issues which necessarily arise in this case in the
order, first dealing with subsistence of copyright;
then with infringement, infringement first in the
straightforward sense and secondly, infringement by

way of making and selling an adaptation, and then I

propose to deal with one or two general

considerations which arise out of this case.

If the Court please, it will be clear that in

this case a great deal of technical evidence was

given and, perhaps since this was the first case to explore the workings of the 1984 amendments, it was necessary for the Court to hear and consider that

technical evidence. However, it will be our case
that one does not need for the purpose of

disposition of this appeal, to go very deeply into

the technical evidence, and that arises because of

the first major point that we set out in our

outline. We say that in any copyright case,

whether expressly or implicitly, a court bears in

mind two distinctions and that one way of looking

at the problem that has arisen in this present case

and the way that problem should be resolved, is to

consider what is the application of those

Autodesk(2) 2 17/4/91

distinctions to the law relating to computer

copyright.

The first of the distinction is the one which

is often referred to between an idea which, it is

sometimes said, is not subject to copyright

protection, and the working out of that idea in a

copyright work. Now, that is a distinction which

is familiar in several other parts of copyright

law. It will be our submission that it is not

always an absolutely hard and fast distinction and
that one of the points on which the Full Court went

wrong was to classify too much of our program as an

idea, rather than, as we say, the working out of

that idea in the copyright work.

The second distinction is one which is less often adverted to expressly but which, in our

submission, is present in all copyright cases, that

is to say, the distinction between a work and the

actual mechanism used to implement or reproduce the

work.

Now, that is a distinction which is ordinarily

taken for granted, and one can give some examples

which come readily to mind. If one is dealing with

copyright in a musical work one is concerned with

the work itself and one draws a distinction between

that work and the mechanical workings of the

instrument or instruments on which it is played.

So, if somebody has a copyright work and it is

first instance played on the piano, no doubt the

mechanics are quite different if it is alleged that

there is an infringement by reason of someone

singing the work or playing it on a trombone or

whatever it might be.

But none of that is taken into account in

considering whether there is or is not infringement

because what one is looking at is whether the work

or a substantial part of it has been reproduced,

not how that takes place.

Now, similarly, it is familiar in the case of

literary works that in considering whether

copyright in a literary work has been infringed one

does not take into account details of the typeface

used, the arrangement of type on a page or the

printing machinery used to reproduce the work
because that is not what one is concerned with in

the ordinary sort of literary work.

In our submission, the Full Court went wrong

in law because it ignored or misapplied these distinctions so that it sometimes treated the

program in suit as if it were an idea only and

therefore not subject to copyright, and sometimes

Autodesk(2) 17/4/91

it concentrated, as we would say, wrongly on the
actual machinery used to implement or reproduce the
work and, of course, since it was the case that
three different machines are involved in this case,
each of which we say reproduce the work,

nevertheless the mechanics of reproduction vary

from machine to machine. It will be necessary for

me to address submissions supporting the

proposition that the Full Court went too far in its

emphasis on the actual machinery, and in so doing

ignored what was the work itself.

Now, in our submission, it is important to

recognize that the underlying state of facts which

led to the making of the copyright work was this,

that the appellants had a problem that they wished

to stop piracy of their AutoCAD which, of course, was a very large and important group of programs, but for the purpose of selling the AutoCAD it had

to be put into a form which could be run on the

user's own computer, so it was not restricted to a

particular computer. In order to do that, the
programs were put on disk but where programs are

put on disk they are, of course, notoriously easy

to copy.

So, the basic idea which underlies the program

in suit was that the appellants would make and sell

a device which is described as a lock, though as

the courts have pointed out, the word "key" might

have been a more apt description - they were to

make and sell this device and this device was to be

such that it could be attached to any computer
which was capable of running AutoCAD but without

which the AutoCAD would not run. So that all the user would notice when he tried to run AutoCAD is

that it would work when the lock was in place and

it would not work when the lock was not in place.

Now, that is the basic idea behind the

programs in suit, and it is to be noted that in

working out that idea a number of problems were to

be solved: the lock must be difficult to copy;

neither the lock nor AutoCAD itself must be

specific to any particular computer and the AutoCAD

lock must be capable of being in position in such a

way that AutoCAD itself can run and the presence of

the lock will not interfere with that running.

Now, the idea of having a lock in the sense in

which I have been using it is one which could have

been worked out in a substantially unlimited number

of different ways and each of these different ways

could involve a different computer program.	Now,
we would stress, at this stage - as I will be
stressing at one or two later points when

considering the judgments of the Full Court - that

Autodesk(2) 4 EMMERSON, 17/4/91

we are not claiming here that any way of working

out the idea, whatever it might be, is something

which is necessarily within our copyright

protection. But we do claim that the specific way

in which we worked out that idea is within
copyright protection and that in so submitting we
are in no sense departing from the established

principles which have been worked out for drawing a distinction between an idea and a working out of an

idea.

Now, the way we worked out the idea was as

follows. The appellants devised a program which

received digital information, chosen at random, and

processed it to give a particular answer. Now,

this program was placed, in our submission, both in

the lock and in AutoCAD itself and the digital

information was processed both by the lock and by

AutoCAD. The test that was run was that so long as

the two answers were the same the AutoCAD continued
to run, whereas if the answers were different then

the AutoCAD stopped running.

Now, the program in suit is the one that

processed the incoming digital information a_nd gave

the answer; that being an answer which was given both by the lock and by AutoCAD. This particular
program could, of course, in principle, have used

any combination of arithmetic and/or logical steps

to process the data. The appellant could have

chosen a program which simply performed

straightforward arithmetic operations. They could

have chosen a program which carried out logical

operations but not the specific logical operations

which were in fact chosen. They could, even within

the type of logic that they had chosen, make a

,great deal of variation in the way in which that

logic was worked out. There would, in fact, have

been an astronomical number of different ways in

which the program could have been devised.

In fact, the particular program chosen, and

the one which we claim is subject to copyright

protection, has two aspects. It receives incoming

digital information and detects whether each digit

is the same as or different from the immediately

preceding digit. If the two are different and the

change is in a particular direction, then the

program steps on to the next stage, otherwise it

does nothing. Now, that first aspect of the

program is what is referred to in one or two places

in the judgments as "acting on transitions", and I

may use that as a shorthand notion for the matters

set out in paragraph S(a) of our outline.

Now, where there is a transition, in this

sense, then the program puts out a digit which may

Autodesk(2) 17/4/91

be nought or one, according to a specific,

predetermined sequence. Now, the sequence is

chosen so that it is essentially impossible for an

outsider to duplicate it by guesswork or by

accident. The evidence was that even within the

framework of the logical structure of this program, it would have been possible to choose an absolutely

astronomical number of different sequences. My

clients chose, and claim protection for, one of

those sequences in combination with the other

aspects of the program. It is a sequence which

is 127 units long and is followed by what is

described in some of the judgments as "a wrap-

around feature", which means that when the program

has stepped through the first 127 steps in the

program, the immediately succeeding step takes you

back to step number one.

Now, that is the program for which we claim

that we have copyright protection, and the

difficulties that have arisen in the present case

have arisen by reason of something which we say is

likely to be common to all, or almost all, cases in

which it is necessary to consider computer

copyright, that is, that the program can be

actually implemented in different ways depending on

what is the device which is carrying out the

program.

BRENNAN J:  Dr Emmerson, I do not wish to interrupt your

presentation of the facts because I need to

understand them, but you have used the term "computer program" without reference to the

statutory definition. I understand why, but could

you tell us when you speak of a computer program by

reference to a particular digital sequence whether

you are speaking about an electrical impulse or a

number or what?

MR EMMERSON:  Yes, if the Court pleases. Could I say by way

of preface to my answer that the structure of these

submissions is intended to be as follows: I am

using the word "program" in a general sense at this stage simply to point to what we are talking about.

I will at a later stage of these submissions deal

in some detail with the definition of "computer

program" which appears in the 1984 Amendments to

the Copyright Act and I will put some submissions

directed to showing that in each of its various

manifestations, so to speak: the program that I am

now talking about is a computer program within the

meaning of the definition in the Act. For the

purpose of the part of the submissions I had

reached, I am drawing a distinction between the

program and the detailed representation or

implementation of that program. The program is the

combination that I have described which may be

Autodesk(2) 6 17/4/91

loosely referred to as a combination of acting on

transitions and a particular sequence with a

wrap-around feature.

Now, it is our submission that that program is

protected, regardless of any specific way in which

it might happen to be written and regardless of any

specific way in which it might happen to be

implemented, but in order to raise the distinctions

which we would rely on in this appeal, I must, of

course, deal with what we say about the fact that

where you try to run that program on various

hardware, the exact way in which you implement it

is different. So that you might have different

methods of storage, whether it be electrical

impulses or whatever, and you have different

methods of working it out so that the machinery

carries out the required instructions, but the

program in the sense in which I am using it at the
moment is the program as I have described it in

paragraph 5 of the outline of argument and it is in

that sense that I am using the word in this part of

the submissions.

DEANE J: Is there any significance in the different

spelling in clause 5?

MR EMMERSON:  The only significance is an error on my part,

I am afraid, Your Honour.

DEANE J:	I see.
MR EMMERSON:	When I am not thinking I spell it with a "mme"

but in the context of computers the normal usage is

to just end it with a single "m" and that is, of

course, the way it is spelt in the Act.

DEANE J: Except, the different spelling could have

reflected the sort of distinction you were drawing

to Justice Brennan.

MR EMMERSON:  Yes. have attempted to do it in these notes, and I hope In making the distinction, the way I

consistently, is to refer to "computer program" and

put it in inverted commas when I am referring to

something within the definition in the Act and

simply to use the word "program" when I am talking

about the matter in general without, at that stage,

considering the question of whether it is a

computer program within the definition.

DEANE J:	Thank you.
MR EMMERSON:	The various implementations of the program

that are referred to in various stages in the

judgments are, first, the program which forms part

of a larger program or group of programs described

Autodesk(2) 7 17/4/91

as WIDGET.C. Now, putting the matter loosely, that

implementation of the program is a software

implementation. You find the program physically,

so to speak, in the disks which form part of

AutoCAD and the way in which the software

implements the program is that the digits to be

output are stored in the memory of the computer in

what is described as a "look up table" and each

time a transition is detected then a new digit is

put out, though I should add, that when I say "a

new digit", that new digit may or may not be the

same as the immediately preceding one.

Now, in the appellant's lock the program is

implemented in hardware. The digits to be output

are generated by electrical circuitry and then

stored temporarily in a small memory unit called a

shift register and from that shift register, in due

course, they are put out. But, as with the
implementation of the program in the WIDGET.C

software, one gets both the essential features in

paragraph 5. It is, in our submission, an

implementation of the same program albeit that it

is one in hardware rather than in software.

Now, the respondent's lock, which is of course the one that one has to look at in testing the

question of infringement, is much more similar in

implementation to the implementation that takes

place in the appellant's software. Once again it

detects what I have called transitions and, once

again, the essential digits are simply stored in a

memory unit in a look-up table. However it is to

be noted that in each of these cases the program

acts on what I have .described as transitions and in

each case the sequence of digits that is put out is

identical.

Those are what we say are important

distinctions which should be borne in mind in

considering the application of the 1984 amendments	to the facts of this case and, indeed, considering	the construction of those amendments, and one can
illustrate the significance of the distinctions
that I have been seeking to make by considering
just what it was that the respondent, Kelly, actually did.

He was well aware that if you wanted to run

AutoCAD you needed a lock and that if you had the

lock in place it ran and if you did not, it did not

run. Now, if he had simply paused at that point

and devised his own lock or own device to stop

piracy, truly independently and without reference

to AutoCAD or to the AutoCAD lock, then there would

be no cause for complaint because what he would

have been doing was picking up no more than a

Autodesk(2) 17/4/91

general idea and working the matter out in full on

his own.

But that, of course, was not what Kelly did.

What he did was carry out a project extending over

several weeks which involved first making a minute

analysis of the digital information which passed

between the computer and the lock, and from the

lock to the computer, and this was necessary

because what Kelly ultimately did was discover and

copy the first aspect of the program, namely, that

it was concerned with transitions in the input data

as distinct from any of the other possibilities,

any other arithmetic or logical operations.

It was not concerned then with adding up the

first two numbers and dividing by the third, or any

more complex or other type of arithmetic operation.

It was concerned with detecting and acting on

transitions and Kelly, in order to discover that,

used both AutoCAD and the AutoCAD lock.

BRENNAN J:	What are transitions?
MR EMMERSON:	I am attempting to use "transitions" as a

shorthand for the matters that I have referred to in paragraph S(a) of my outline of argument, that

is to say, what happens is this: the lock is

attached to a part of the computer which sends out

signals to devices which are attached to the

computer, so-called peripherals. The lock can

either be in a mode described as transparent - that
means it just transmits the signals to anything
else that may be attached to the wire - or it can

act in a way that we are concerned with because it

can be running the program.

Data comes into the lock necessarily in a

series of electrical impulses which can be either

positive or zero, and the lock and indeed, the

program in AutoCAD act on transitions between
charge or no charge or no charge and charge.	That
is what I mean by transition.	So that if one were

simply as an electrical engineer to look at the

wire, you would find that you had a fluctuating

voltage level and that was how the digital

information is transmitted. It is when that

voltage level changes in a particular direction

that the lock detects that and acts.

DAWSON J:  So that the charge, or no charge, represents

nought or one, or vice versa?

MR EMMERSON: Exactly. Yes, Your Honour. So it is, as it

were, that aspect of the incoming data that the

lock, and indeed the program in the computer, look

for and act on, and simply because it is rather a

Autodesk(2) 17/4/91

mouthful to talk about it in any other way, and

because the judgments tend to use this shorthand,

could I say that when I use the work "transitions"

in this context, that is what I am talking about.

It is a transition from nought to one, or vice

versa, or from charge or no charge, depending on
whether you are considering the string of

information that comes from the electrical point of

view, or from the digits that it represents.

DAWSON J:  Why would it be simplistic just to say that the

program is the sequence of digits?

MR EMMERSON:  I am sorry, Your Honour, I do not think I

actually heard that question.

DAWSON J: Could you say that the program is merely the

sequence of digits?	I rather gather you cannot;
why not?

MR EMMERSON:  Yes. We would say, "No". That is one aspect

of the program, but the program is the combination

which involves the detection of transitions; the

response if, but only-if, you have a transition,

and the transmission back of the digits.

DAWSON J:  Why is that not just a way in which the sequence

is used?

MR EMMERSON:  It is just the way in which the sequence is

used.

DAWSON J:  Yes. I thought you said the program did not

include the way in which you work it out.

MR EMMERSON:  I am sorry, Your Honour. The question that I

was- attempting to answer was whether the program

was just a sequence of digits - - -

DAWSON J: Yes.

MR EMMERSON:  - - - and the answer perhaps stated

compendiously, is that the sequence of digits is

obviously an important part of the program, but it

is not just the sequence of digits simpliciter; it

is the sequence of digits in that particular

program context which is relevant.

DAWSON J:  Yes. The distinction I had in mind was the one

which you made first of all, between a work and the

actual mechanism used to implement or reproduce the

work.

MR EMMERSON:  Yes.

Autodesk(2) 10 17/4/91

DAWSON J: Well, all the time you are reproducing the

sequence in one form or another, why do the

mechanisms matter?

MR EMMERSON:  You are, but there is the important point that

you are not just simply reproducing that sequence

but what you are doing is receiving an incoming

signal and responding to that signal when you get a

transition. So that it is not only the order in

which the digits are delivered in the sequence, but

the timing of delivery of those digits which is

relevant for the operation of the program. Now,

that is important because Your Honour will recall

that the essence of the thing is that you have got

to have a program which can be both in the computer itself and also in the lock, and you have got to be able to compare precisely and it would be no good

if the two got out of kilter, for instance. So,

both aspects are, we say, very important, and

indeed the respondent, Kelly, studied both aspects.

The first thing he found out was that the program was concerned with transitions and not with something else, but then what he did was - - -

DEANE J:  Do you suggest that you could have copyright

simply of a sequence of numbers in this field?

Would you not end up with a result that nobody

could do maths any more?

MR EMMERSON:  A sequence of numbers simpliciter is not a

computer program. You can have copyright in a

sequence of numbers as a literary work. For

example, if you compute, as someone did, a set of

logarithm tables and you publish that set of

logarithm tables, then you can have copyright in

that set of tables and if somebody else wants to

produce a set of logarithm tables, then they have

got to work out their own and not copy yours.	I
think perhaps the question that Your Honour is

putting to me, as a worry, has two aspects really:

one is, can you stop people doing arithmetic? The

answer is no you cannot, because copyright never

stops people doing anything independently.

Copyright is not a monopoly right; it is a right

simply to prevent copying. So what is objected to,

in any case, is that the defendant party, instead

of working the thing out for himself, has made use

of the work done by the copyright owner, and that

must always be so.

So, it could never be the case that you could

prevent somebody from ever using a particular
sequence of digits or ever carrying out arithmetic.

That is not what copyright is talking about at all.

On the other hand, as in the logarithm table case, if somebody does work and publishes a set of

tables, then that can be a literary work; it is a

Autodesk(2) 11 17/4/91

table of compilation which is within the definition

of literary work in the Act and it is something

which can be protected.

BRENNAN J: It is right to say, is it, the amendments to the

Copyright Act, in a sense, introduced a completely novel conception?

MR EMMERSON:	We would say, yes.
BRENNAN J:	Up until then, copyright could only subsist in

something which was susceptible of sensory

perception?

MR EMMERSON:  Yes.

BRENNAN J: After that and indeed with respect to copyright

programs, it was with respect to something which

was not susceptible - sensory perception.

MR EMMERSON:  Yes. were introduced in the 1984 amendments and that, of
course, is the single critical point at which it
changes the law as applied by the majority of this That is one of the major changes that

Court in the Apple Computer case, because it

removes that need for one to be able to perceive

the copyright work through the senses, whether

visually, orally or however, and that is a major

change and it is one of the changes that one needs

to work out. It will be part of my argument, a

little bit later in these submissions, that the

Full Court of the Federal Court did not give

sufficient weight to that aspect of the change.

BRENNAN J:	I do not want to take you out of your - - -
MR EMMERSON:	If Your Honour pleases. Now, Mr Kelly

recorded precisely the sequence of digits, so this

was not a case in which he either devised the

logical structure of his program independently, nor

What he did, in each case, was copy those and - - - did he devise the sequence of digits independently.

DAWSON J:	He copied the self same sequence?
MR EMMERSON:	He copied the self same sequence, yes, and he

did not devise it independently, he devised it by

copying it from the appellants' lock.

Now, in our submission, that, when embodied as

it was in the respondents' lock, is sufficient to

establish the necessary elements for reproduction.

There is both the access to and use of the

copyright work and there is the taking of the

essence of the copyright work.	And that is what
the respondent Kelly did in this case.	He then

simply placed the sequence in the memory unit in

Autodesk(2) 12 17/4/91

his own lock and from that memory unit it was

reproduced as the lock detects transitions.

DAWSON J:  You have not explained - and maybe there is no

need to - how he detected the sequence. You said

that he found out the program was concerned with

transitions ..... sornething else.	The next step,
what was it?

MR EMMERSON:  The next step is that having worked out that

it was concerned with transitions, he took the

appellant's lock, connected it to electrical

machinery and subjected it to a sequence of

transitions.and then he recorded the digit that was

output from the appellants' lock at each

transition. So he, in effect, just simply copied

down the output from the appellants' lock on

transitions. He found, as was the fact, that there

was a sequence of 127 digits so he copied those

down and he found that there was a wrap-around

feature - as it is called - whereby when you come

to step No 128 you go back to step one and he

copied that too.

This he did simply by having electrical

equipment attached to the appellants' lock and we

would say that this is no different from someone

who copies something from a computer memory by

attaching wires to the computer memory, stimulating the memory and seeing what numbers come out. He in

fact mapped each of the numbers that came out of

the appellants' lock.

BRENNAN J: Is the possible number of transitions infinite?

MR EMMERSON: 

Yes, and indeed at almost every stage of the analysis the number of possibilities is literally

infinite. The number of different sequences that

you can have, even if you restrict yourself as we

did to 127 digits, in decimal terms is 10 raised to

about the power 40.  And when you remember that 10

raised to the power 6 is a million;  10 raised to

the power 12 is a million million; it is many
million million millions. Therefore this is not a

number that you are going to stumble on by

accident. It is something which has to be copied

in order to be arrived at and that was indeed

exactly what happened in this case.

Now, in our submission, these facts raise the

question of the operation of the 1984 amendments.
We say that on a true construction of those

amendments, the copying by the respondent, Kelly,
amounts to an infringement of the appellants'

copyright and I will be corning to those 1984

amendments in just a moment.

Autodesk(2) 13 17/4/91

We would also put, more generally, that if it

were open to the respondents to escape a finding of

infringement- by copying the essential features of a

program but simply by using different

implementations of the program, then this would
deprive the 1984 amendments of most of their
practical value because all one would need is to
make the modest change to the program which is
necessary if one goes from one implementing device

to another implementing device while keeping the

essence of the program and copying and reproducing

that.

It would be our submission that one should

bear in mind, when looking at the 1984 amendments,

that it is reasonable to suppose that that is not
the consequence which Parliament had in mind when

proposing those amendments.

BRENNAN J: Before you go to the law again, could I just

delay you for one further moment. If there is a

transition which is transmitted from the computer

to the lock, that then stimulates a transition

being sent from the lock to the computer?

MR EMMERSON:  No. It is, in fact, more complicated than

that, Your Honour, for this reason: that the

computer works not on transitions but on digits, so

that so far as the computer is concerned it may

have a digit, say 111, which will keep the

electrical signal constant for three units of time.

That is how it transmits 111. If the next digit is

zero, then you may have a transition, depending on

which direction transition you are looking for.

But the numbers the computer works with and

the ordinary operation of the computer is not

normally done by being triggered on transitions.

It is being triggered on numbers.

BRENNAN J: Then let me ask you this question.

MR EMMERSON: There is another aspect of Your Honour's

question which I think might get me to the point

that Your Honour was seeking to elucidate.	When a
transition arrives so that the computer is
triggered, that in - - -

BRENNAN J:  So that the lock is triggered?

MR EMMERSON: Well, the computer or the lock, yes, the

program. That does not itself necessarily involve

a transition either, but it involves a

predetermined number one or zero which may be the

same as the number before, or may be different.	So
that you do not find that every time a transition
goes to the lock the transition comes back, and

Autodesk(2) 14 17/4/91

there is a very good reason for that. That would

be very easy for an amateur to work out and perhaps

duplicate the lock. Whereas, in fact, what you

have is digital information going out to the lock,

digital information corning back, but in a way which

looks to the casual observer, even if he has the

electronic equipment to detect it, at first sight

as though it is completely random.

That really is an important aspect of what we say was the ingenuity that went into this program,

that instead of having just transition out,

automatic transition back which you could detect,

you have transition out, sometimes transition back.

Whether or not you get a transition depends on

whether the two succeeding numbers in the sequence

put out by the lock are the same or different.

BRENNAN J:  So the lock scrambles the information?

MR EMMERSON: In effect, yes.

BRENNAN J: But the essential notion of it is that digital

information in the form of a transition goes from

AutoCAB -the computer which is running AutoCAB to

the lock -

MR EMMERSON:  Yes.

BRENNAN J: - - - and digital information, but not in

transition form, comes in a predetermined though

apparently random form from the lock back to the

computer.

MR EMMERSON:  Yes, that is correct, Your Honour.

BRENNAN J: And the purpose - I will not say function, but

the purpose of the lock is to affect the

predetermined scrambling?

MR EMMERSON:  Yes. Scrambling is perhaps not quite the word

to use, but the purpose of the lock is to receive
the information, process it in the way I have

described, give back information and, of course,

the intention is that it would be very difficult to

work out any connection at all between the

information that comes back from the lock and the

information that goes out to the lock, and that is

what the program is supposed to do. It is supposed

to take digital information in one form, put it

into another form, which is very difficult to

relate back to the first form.

Just before corning to the 1984 amendments,

there is a further consideration which we say is

relevant and that is this: the program in the

present case was a relatively simple one and yet,

Autodesk(2) 15 17/4/91

as the Court will be aware from the judgments in

this case, the technical details that were inquired

into, become rather complex. In our submission, if

the true effect of the 1984 amendments was that it was necessary to inquire minutely into the sort of

technical details considered in the present case
then, for any program which was more than a

relatively simple one, this part of the law would

become essentially unworkable.

It is our submission that it is difficult to

envisage that Parliament in enacting those

amendments intended minute inquiries into

mechanisms. What Parliament was concerned with, we

would say, is no more than something which is

analogous to the sort of inquiry that you have when

you are asking about copyright in a piece of music,

you look at the piece of music or, in the present case, you look at what is done by the device, but
you do not inquire into the minute technical

details of the implementation because that would

become, for any program of a substantial size, an

impossible inquiry. So we say that not only when

you look at the actual language of the 1984

amendments, but also when you look at the

background and the structure and purpose of those

amendments, the analysis that was carried out by

the Full Court wrongly concentrated on details of

the mechanical implementation of the programs when

what it should be looking at was the program

itself.

That concludes what I want to say by way of

introduction. I would now wish to take the Court

to the 1984 amendments. It is hardly necessary to

remind the Court that these amendments were

introduced in order to deal with the facts, or

facts of a type that had arisen in the Apple

Computer case. The Court will recall that their

decision at first instance, in that case, held for

the defendant.	Then the Full Court of the
Federal Court reversed that decision by a majority	and it was at that stage, that is to say after the
Full Court of the Federal Court but before the High Court decision, that the 1984 amendments were
brought into effect.
In each of the courts that heard the case the court was divided, and the view which was
ultimately .successful in this Court, as a majority
view, turned on the issue of visibility. That was
the point that Your Honour Justice Brennan was
making to me a little bit earlier on, that the type
of programs in suit were not in the respondents' device something which could be perceived by the
human senses at all, and it was the view of the
majority in this Court and the minority in the Full

Autodesk(2) 16 17/4/91

Court of the Federal Court that this was sufficient

to ensure that literary copyright, as it then

existed, did not apply.

At that stage, the definition of literary

copyright was simply an inclusive definition, in

that it included a written:

table, or compilation -

but the Act did not deal at all with copyright in

computer programs.

Now, it is unnecessary for the purpose of this

argument to deal in detail with the Apple case

simply because the changes that were introduced in

1984, amongst many other things, clearly removed

the requirement for visibility. But we would

observe that the majority view in the Federal Court

in that case and minority view in the High Court in

that case was even then looking broadly to what was

being done by a computer program and in rejecting

the requirement of visibility was finding some of

the other elements of computer copyright there.

So that, for example, if one goes to the Full

Court of the Federal Court - this is reported in

(1984) 53 ALR 225 - Mr Justice Fox, at the top of

page 237, is dealing with an argument that arose in

that case, that the way in which the programs were

stored in the Wombat computer, which was the

defendant party's computer, and in the Apple

computer, which is the plaintiff party's computer,

those ways were different. The reference to ROMs

is a reference to the memory units, and His Honour

is-here dealing with the argument that there was a

different mechanical arrangement, and he says this:

From the evidence of the print-outs it is

plain that the arrangements of the object	codes stored in the Wombat ROMs are the same,	or virtually the same, as the arrangements of
the object codes stored in the Apple ROMs.
Moreover, the Apple codes could be reproduced,
with substantially their same content, from a
print-out recovered from the Wombat ROMs.

So you use another piece of equipment; you apply

electrical signals to the ROMs, and you find that

you get the same result for the Apple and the

Wombat. And His Honour continued:

It is of no consequence that the Wombat ROMs

and chips probably bear different patterns

from those appearing on the Apple ROMs or

chips. The fact is that the former embody the

same arrangements of impulses and produce the

Autodesk(2) 17 17/4/91

same result, and are directly copied from the

Apple ROMs.

Now we have, in the present case, the same

feature here. The lock of the appellants and the

lock of the respondents have the same impulses and

they produce the same result and, in the sense in

which His Honour was using the word, the

respondent's lock is directly copied from the

appellant's lock. His Honour continues:

In my view it does not matter, if it be the

fact, that the code cannot be seen, even with

special apparatus. The adaptation of the work

is rendered perceptible with a machine. It is

sufficient that the code has its existence in,

and is ascertainable from, the chips.

We would say that while that view was not

ultimately successful in that case, under the

amended Act that is a very useful test where one is

looking at a programmed device then one does not

ask whether the adaptation is perceptible to the

human eye; one looks to see whether the code has

existence in and is ascertainable from the chips.

His Honour Mr Justice Lockhart is to similar effect

at page 263, lines 6 to 10:

very same sequence of impulses as are represented in the Apple ROMs.

the inescapable conclusion is that the Wombat

His Honour then, not as it happens for the last

time, deals with the. possible need for reform in

this area of the law at page 269.

Now, in the High Court, reported in

(1986) 161 CLR 171, Your Honour the Chief Justice

and Mr Justice Wilson were, or course, in a

minority but -

MASON CJ:  I do not know that you need use the expression

"of course", Mr Emmerson.

MR EMMERSON:  It was unhappily chosen, if Your Honour

pleases. It is unnecessary to remind this Court

that on that particular occasion Your Honour and

Mr Justice Wilson were in the minority. But

Your Honours did say, at page 190, at the end of

the paragraph corning over from the preceding page:

We are broadly in agreement with the reasons

and conclusions of Fox J. and Lockhart J. -

and so there the same approach is taken. At

page 194, in the first full paragraph on that page

Autodesk(2) 18 17/4/91

beginning at line 5 on that page, Your Honours said

this:

It is not correct to describe an object

program as merely a sequence of electrical
impulses within the computer. Electrical

impulses there are, but those impulses serve

to identify a set of instructions in machine

readable language designed to guide the

machine in its basic operations.	They do not
form part of the computer itself,

chips in a machine.

notwithstanding that they may be embodied in a

So the question of whether the program is

permanently in the machine is not relevant.

They might equally well be contained in a

magnetic disk or tape which can be inserted

into or removed from the machine at will.	The
language may be recorded in visible form in
binary or hexadecimal notation.	The same

result as is achieved by the use of an object

program could be achieved by the manual

operation of a computer in accordance with

instructions written out in ordinary English

and contained in a manual. The only problem

would be that such a process would be

impossibly tedious and therefore wholly

impracticable.

Now, in our submission, what Your Honours were

there dealing with was an aspect of the distinction

I was attempting to draw at the start of these

submissions between the program itself and the

specific way in which it happens to be implemented

in a particular device and Your Honours were there

saying that the specific way in which it happens to

be implemented is not the point and Your Honours

were looking to the result that was achieved.

Similarly, at page 197 at the end of the

paragraph beginning after the quotation from the

Cuisenaire case at the top of the page,

Your Honours said:

Here the Wombat ROMs and EPROM contain in

substance the same sequence of electrical

impulses as are contained in the Apple ROMs.

And so, again, that is regarded as what is being

important, rather than the detailed manner in which

the program happens to be stored.

Now that case was decided before the

Autodesk(2) 19 17/4/91

1984 amendments and I have drawn attention to those

passages for two reasons: one being to stress that

the majority view depended on visibility, a

requirement which has been removed by the

1984 amendments, and the other is to show that the

alternative view was, in our submission, already

dealing in a way which, with respect, we say was

the correct way, with the distinction between the

program·itself and the specific way in which it

happens to be implemented.

Going now to the amendments that were

introducing a new set of definitions and the

introduced in 1984, the major way in which by

structure is as follows: there was a new

definition of literary work which was now made to

include a computer program or compilation of

computer programs; there was a definitions for the

first time of "computer program" and I will come to

that in detail in just a moment; there was an

addition to the definition of "adaptation" so that
it dealt with, specifically, the meaning of

"adaptation" in the context of computer programs;

there was a definition of "material form" and then

there were definitions of "infringing copy".

So, it was done by way of definition rather

than by altering the other aspects of the Act.	So,
it still leaves one in the position, as
traditionally in copyright law, that copyright is

given to specific forms of copyright subject-matter

and these are enumerated in the Act, and then one

looks to see whether there is infringement, and for
the purpose of the present case, the relevant form

of infringement is reproduction in a material form.

Now, it is true that as some members of the

Full Court pointed out, there was not before

amendment nor is there after amendment, a

definition of the word "reproduction". However, in

our submission, the effect of these new definitions	is that one now has a new statutory context in
which the word "reproduction" is used and that that
word takes its colour from the statutory context
and, in particular, it is necessary to consider
what one looks for in reproduction of a computer program having regard to the changed statutory
context. Now that, I hope, will become more clear
as I go to look at the various definitions.

Could I take the Court first to the definition

of "computer program". I will read it out and then

draw attention to a number of features of this

rather complex definition:

Autodesk(2) 20 17/4/91

"'computer program' means an expression, in

any language, code or notation, of a set of

instructions (whether with or without related

information) intended, either directly or

after either or both of the following:

(a) conversion to another language, code or

notation;

(b) reproduction in a different material form,

to cause a device having digital information

processing capabilities to perform a

particular function;".

By way of introduction to a construction of

that provision, I would draw attention to the fact
that Parliament has seen fit to introduce a

definition of "computer program" and that neither

the word "computer" nor the word "program" is

treated as a term of art. A consequence of that

is, we say, that it is wrong to seek to import into

this expressed definition limitations which are not

there and which are drawn from conventional usage

of the expression "computer" or the expression

"program".

Now, in our submission, the trial judge was

correct in the way that he approached this matter.

I will just refer very quickly to the page at which

he did. He deals with it at the bottom of page 970

in volume 5 of the appeal books, and he says this:

In considering the proper construction of

this definition, the Court accepts the proposition that the amendments to the Copyright Act had the purpose of extending the

rights of copyright to owners of computer

programs. In so doing, it is appropriate for

the Court not to take a narrow view of the

definition of a computer program but to

approach the question on the basis that as

wide a construction should be given within the limits of the meaning of the words used in the

definition.

That was one point on which the Full Court differed with His Honour but, in our submission, His Honour

is right in that approach.

Now, the fact that there is no direct

reference to "computer" as such in the definition

is relevant because the test is not whether the
object in question is a computer as conventionally

understood, but whether it is, in the words of the

definition -

Autodesk(2) 21 17/4/91

a device having digital information processing

capabilities.

That, in our submission, necessarily takes one to a

much wider field than the conventional notion of

computer. It is true, of course, that a computer

is itself "a device having digital information

processing capabilities", but it is not the only

device which has those capabilities and, in our

submission, it is important to bear in mind that

Parliament chose to use the expression that it did rather than to use the expression "computer".

If it is necessary to demonstrate this

further, the explanatory memorandum dealing with

the 1984 amendments makes it clear at paragraph 22

that this is intended. Paragraph 22 is to be

found at the bottom of page 7 on the explanatory

memorandum, and 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) -

well, that first point is not relevant here,

because we are concerned solely with digital

information. They go on to say -

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 computerized

ignition systems.

So, there is not, either on the language or on the

intention, a limitation there to computers. It is

our submission that both any computer on which

AutoCAD is run and both the locks in the present

case are devices having digital information

processing capabilities within the meaning of those

words in the definition of computer program, and

the learned trial judge was correct in holding that

both fall within that definition.

The next point that we would make about this

definition is that while it refers to performing a

particular function, the definition of program does

not say that the program itself must be limited to

a particular device. The definition is wide enough

to encompass programs which may, at various stages,

be run on different devices. Now that is, if one

may say so, obviously a wise precaution, because it

would be clearly unsatisfactory if the only

computer programs that were protected were ones

Autodeck(2) 22 17/4/91

which could be run and run only on a single

computer. As the Court is well aware, a number of

means are available to allow people to write

programs to be run on various computers. Now that

is again relevant in the present case because the
definition of computer program does not exclude

cases such as the present where we say that the

program in suit can be run both on the computer and

on each of the locks.

The corollary of all of this is that,

depending on the particular hardware of the device,

the actual method of implementation of a program

may vary from device to device but it would still

be a computer program within the meaning of the

definition here.

BRENNAN J:  What is the function in this instance?

MR EMMERSON: In this particular instance the particular

function is to receive incoming digital information

and process it in the manner that I have attempted

to describe and send back digital information as a

result of that. That is the particular task which

this program performs.

BRENNAN J: Is there anything in this program which is

relevant to the reception of digital information?

MR EMMERSON:  Yes, because the program acts and acts only

when digital information is sent to it and, as I

have sought to explain, it acts on what I have

described compendiously as transitions. But it is

the reception of digital information and the

detection of a transition within that digital

information which causes the program to act and it

causes it to move on to the next stage of its
operation and the next stage of its operation

involves outputting digital information. So there

is both the reception and the outputting and,

indeed, it is essential that this should be so

because Your Honour will recall that the underlying

function of the lock is that it should be possible to send out digital information to it and get back

digital information and check that you have got the

correct digital information back. Now, that means

that the function of receiving and also sending

digital information according, of course, to the

quite specific rules of the program, is the

function that is performed by this program.

BRENNAN J:  Can I put that to what might be a quite

inaccurate and simplistic test? If, on my

computer, I type "The quick brown fox jumps over

the lazy dog" and I wish to obtain a printout of it

or to have it on the screen, that may be affected

Autodesk(2) 23 17/4/91

by having Wordperfect or Microsoft or Primeword as

your software - as your program.

MR EMMERSON:	Yes.
BRENNAN J:	Do you say that the program there is the reception of the digital information created by the
keying in the translation of that then into a
digital impulse which produces the sign?
MR EMMERSON:	I think, with the greatest respect,

Your Honour, there is an underlying assumption in the question that Your Honour has just put to me

which we would challenge. Your Honour refers to

"the program" there. Now, we say, that one can

only talk usefully about "the program" in the

context of looking at a particular function, and it

is quite wrong to assume that there is only one

thing which would answer the description "the

program" there.

Now, if I were to say that there may be "a

program" there which performed certain operations
on the signals that came in, that we would accept,

but we would not, with respect, accept that it is

"the program" because, clearly, what one has got is

a whole complex structure of programs and, for some

purposes, you may be looking at a very large group

of instructions and say, "Well, for the purpose of

our present inquiry that is the program I am

interested in". Sometimes you may be looking at a

more limited one. But, clearly, if you have got

something like Wordperfect, that is a very complex

group of programs and you might, at some stage,

want to refer to "the program" as the totality of

that or sometimes you might be interested in part

only. But we would say that just as in, say, a
compilation of poems, you might have a number of

different people owning copyrights in various bits

of the compilation and you might have somebody else

owning copyright in the compilation itself so, with

a complex computer program or group of programs

like Wordperfect, you might have several different

copyrights present.

BRENNAN J: Really, what my question was directed to was

this: having regard to the definition of "computer

program", is a computer program anything other than

the means by which one stimulus is translated into
either another stimulus or the consequences of

another stimulus?

MR EMMERSON:  At one level it is probably not, that is to

say, digital information processing almost

invariably takes place by a series of electrical

impulses being stimuli coming in and stimuli coming

out.

Autodesk(2) 24 17/4/91

BRENNAN J: Well then, does one then define the relevant

function as the translation of the introduced

stimulus to the outputted stimulus?

MR EMMERSON:  One would define the relevant function by

defining the specific relationship between those

two; that is one way that you could analyse the

thing as a matter of logic, yes, but of course you

would have to know the specific relationship

between the two for the notion of relevant function

to have any useful meaning.

BRENNAN J: Well quite, but if I can come back to my word

"perfect notion".

MR EMMERSON:	Yes.
BRENNAN J:	The inward stimulus is generated by the touching

of the key - - -

MR EMMERSON:	Yes.
BRENNAN J:	- - - and the function which the programs, whether Wordperfect or whatever, that they are
intended to perform, is the production of the
visual screen or the output stimulus - - -
MR EMMERSON:	Yes, that is correct.

BRENNAN J: Stimulus that goes to the printer or the screen.

MR EMMERSON:  Yes.

BRENNAN J: Well now, what implication does that have for

your argument in relation to the inputted stimulus

that comes from AutoCAD and the outgoing stimulus

that goes back to the computer?

MR EMMERSON: Well, we would say that it is entirely

consistent with our argument. It is just that

Wordperfect is a different program and so it	processes the string of incoming information
differently to produce outgoing information, but
they are both programs.

BRENNAN J:  The point is that in neither case, as I have put

it to you, is it relevant to include in the program

what the inputted information is. It is the

translation of it that is the function. So that,

whether you put in from AutoCAD a series of

transitions or a particular series of transitions

or not is immaterial to the function.

MR EMMERSON:  Yes, that is correct, Your Honour, and that

would be true of most, but not all, programs, in

our submission; that, while you can get some

programs which, in a sense, do not act on incoming

Autodesk(2) 25 17/4/91

data, the ordinary sort of program would have at

least some stimulus coming in and at least some

stimulus coming out.

GAUDRON J: Well, while you are on that line and perhaps

while you are interrupted, does that mean that you

treat the receipt of information or stimulus as the
equivalent of its processing, for the purpose of

the definition?

MR EMMERSON:  The incoming digital information is what is

processed by the device according to the program,

yes.

GAUDRON J: Well, I do not know that -

MR EMMERSON:  So the information comes in and it is

processed by the device - - -

GAUDRON J:  I am asking what is comprehended by the word

"processed" in the definition. Is it satisfied by

it being received?

MR EMMERSON:	It is probably not satisfied by it merely being received with nothing more, no. Something
has to be done to that information for it to be
processed.

GAUDRON J: To that information? Not in consequence of it?

MR EMMERSON: In consequence of it, yes.

GAUDRON J: Well, there may be a difference.

MR EMMERSON:	Yes.
GAUDRON J:	And so I am asking for the purpose of the

definition, what do you say it is?

MR EMMERSON:  We would say that ordinarily information will

found against us on two aspects here: one

important, the other less so. The aspect which we

say is of importance is that the courts were

disposed to deny that in the circumstances that I

have outlined one got a reproduction in a material

form of the computer program. We would wish to

~~· challenge that and say that it is wrong.

The other aspect which is admittedly of less

general importance is this, that the court held

that it was not actually established that the use

of the respondents' locks would involve infringing

conduct of the sort that I have just described, and

it is true that there was no direct evidence that

that had happened.	We say that the inference is
blazingly clear.	The only use for these particular

devices is an infringing use, that is to say,

unauthorized running of the AutoCAD.

Accordingly, in our submission, the Full Court

was wrong on this point also and we would invite

the Court to hold that in loading a program into

Autodesk(2) 77 17/4/91

fast memory, one does thereby put it into a

material form because it is a form of storage from

which the work or adaptation or a substantial part

of it can be reproduced and that is true

notwithstanding that it may stay there for only a

short time.

Accordingly, we would invite the Court to say

that the Full Court was wrong on this point also

and that by their conduct the respondents have

authorized the reproduction of the programs. If

the Court pleases, I said that I thought that I

could finish in about half an hour and I see it is

about 35 minutes. Unless there are further matters

that the Court wishes to put to me, those are the

submissions in-chief for the appellants.

MASON CJ:  Thank you, Mr Emmerson. Mr Burnside, could we

have your outline of argument?

MR BURNSIDE: Whilst those are being handed up, Your Honour,

may I say this, that in view of the way the

submissions have developed today there would be

some preliminary matters that I would address the

Court on before starting on the outline that the

Court has just received.

MASON CJ:  What do you mean by "preliminary matters"?

MR BURNSIDE: Well, matters concerning the way in which

these devices work which, we say, is of fundamental

importance for the Court to understand because the

discussion of the legal issues involve the use of

language which can be misleading about the nature

of the notions intended to be conveyed by that

language.

MASON CJ~ Yes, well we had understood that you would give

us your own version of how it all works.

MR BURNSIDE: Yes.

MASON CJ: Thank you, Mr Burnside. We will resume at

9.45 am tomorrow.

AT 4.54 PM THE MATTER WAS ADJOURNED

UNTIL THURSDAY, 18 APRIL 1991

Autodesk(2) 78 17/4/91