Professor dr juris

Jon Bing
Bøgata 7, NO-0655 Oslo,
Norway
T +47 22 67 54 00
F +47 22 67 74 78
M +47 909 67 659
[email protected]

Version 1.0
10 152 words
Copyright protection of computer programs

Jon Bing
Norwegian Research Center for Computers and Law
University of Oslo

1 Copyright and information technology

In the 1960s, computer programs were considered accessories to the very expensive computers.
There were several reasons for this, one that a program simply could not be run on any other
computer than for which it was written, high-level languages were still in the making1 and
compatibility was low. But IBM had considerable success with its 360-series announced in 1964
allowing customers to purchase a smaller system and migrate upward if their needs grew. In 1969
IBM decided – perhaps somewhat stimulated by the anti-trust suit to which it was party2 – to
unbundle hard- and software. As computer programs were separately priced, it became possible
for third parties to offer competing programs. And in such a market arose the obvious issue of
the protection of computer programs.
At this time, it was still unclear to what extent the US Copyright law applied to computer
programs. There were several court decisions, the copyright and patent systems competing for
becoming the legal framework for the intellectual property protection of computer programs.
There were also strong advocates for a third possibility, a sui generis regime for computer
programs, as it was pointed out that neither copyright nor patent was designed to accommodate
the special features of computer programs, the characteristic which occasionally is called
“industrial copyright”, programs themselves often called “soft machines”.

The author will be permitted an anecdote3 by the way of illustration. At one of the meetings of
experts4 to the WIPO in Geneva,5 there had been an unusually heavy snowfall during the
night. Struggling uphill to the WIPO building, one could see improvised tools being used to
remove the snow in order for parked cars to escape. In the meeting, the head of the delegation
of Soviet Union6 made this the basis of a metaphor pleading for a sui generis solution, “In
Geneva, where the snow rarely falls, one may allow oneself to adapt the tools at hand for the
removal of snow. If you live in Moscow, you will expect the snow to fall heavily every winter,
and you will have efficient and specialised tools. And I ask you, ladies and gentlemen, do you
think computer programs are like the snow in Moscow or in Geneva?”7

WIPO actually developed 1971-77 “Model provisions on the protection of computer software”
with the assistance of Professor Peter Seipel,8 but these were not adopted as national legislation

Kontor Office
Institutt for rettsinformatikk Norwegian research center for computer and law
Universitetet i Oslo University of Oslo
Postboks 6706, St Olavs plass 5, NO-0130 Oslo, Norway PO Box 6706, St Olavs plass 5, NO-0130 Oslo, Norway
T +47 22 85 01 01 Direct +47 22 85 00 90 T +47 22 85 01 01 Direct +47 22 85 00 90
F + 47 22 85 01 02 M+ 47 909 67 659 F +47 22 85 01 02 M +47 909 67 659
[email protected] www.jus.uio.no/iri [email protected] www.jus.uio.no/iri/nrccl.html
U:\Studieseksjonen\Master_of_laws\LLM_ICTL\Handouts og lecture outlines\JUS5660 IPLIS (tidl ICTLINTELL)\2011 IPLIS\ÅL-computer
programs-1-1.doc
Jon Bing

in any country. The model provisions were inspired by copyright, but had some elements akin to
patent protection of the content of programs. In practice the discussion of alternatives came to a
halt when the US adopted the 1980 amendments to the 1976 Copyright Act, extending copyright
protection to computer programs. A country was free under the conventions of qualifying
programs as literary works, and this made it possible nearly overnight to establish an international
scheme of protection, based on the Berne and Universal Copyright Convention.
The interest in copyright was nearly exclusively limited to computer programs. For these there
was a market, and there were strong commercial interests in protecting programs. This interest
also found different strategies for protection; one was to introduce various devices which had to
be present for the program to be executed, like an extra element for the serial plug to the printer
which then was called by the program, which failed to initiate printing if the element was not
found. This was the beginning of technical protection measures, the discussion of which later has
escalated. Another obvious measure was only to make the program available in object form,
which in turn gave rise to the doctrine of and provisions on reverse engineering in order to make
it possible to develop programs functionally interacting with another program. A characteristic of
copyright is that the protection allows anyone to access the information in the protected work,
and use this information in the creation of new and independent works. The practice of making
programs available in object form only, barred the access to the information, and reverse
engineering may be seen as a reaction to this for copyright somewhat alien aspect.
There was some interest in other aspects. A joint WIPO and UNESCO meeting9 of 1982
concerned the “problems arising from the use of computer systems for access to or creation of
works”. In the recommendation it is stated that for instance uploading of a protected work to a
computerised systems represented a reproduction in the terms of the conventions. The use of
computers to create work attracted some attention, at this time composers would use computer
programs as tools, and the recommendation10 also states that “tool” is the perspective in which to
consider such use.
However, there were considerable limitations in computerised systems at this time (1982) for a
real concern about the use of literary, musical or audiovisual works to be considered for
computerised systems. The IBM PC had been brought out the year before, the first model did
not have a hard disk, but only 5 1/4 inch floppy disks (and they really were floppy). Storage was
still expensive. Only with low storage costs the volumes of data involved for storing protected
works could be considered. In the early 1980s, the emphasis was on programs and the special
type of programs used for gaming in the first low cost specially designed consoles brought out
for the lower end of the consumer market. Also, infrastructure had to develop for the
establishment for a market for protected works. This did not happen until the early 1990. These
developments shaped the Web, and at the same time created the potential for a market in
protected works which legal policies still are unfolding, and which promise an interesting future
for the law of intellectual property related to information technology.

2 Computer programs
2.1 Source and object programs
The term “computer program” is part of our common language. As other such terms, it is used
on the basis of conventions, and consequently is somewhat vague. At the core of the concept is
the notion of a program which can be executed by a computer. The program will then have to
take the form of a set of instructions conforming to the formalism of a programming language.
The language consists of a certain predefined set of commands, which semantic is defined in detail
with respect to the actions they will cause in a computer. The language will also have a defined
syntax which has to be followed strictly for the program to be executed, or executed correctly.
The program will permit variables which may be chosen by the programmer with considerable

Page 2 of 19
Jon Bing

freedom. The program is seen as different from the data, which is what the program operates
upon, ie the relation between a word processing program and the texts processed by that
program.
In simple terms, a diagram may indicate some of the different aspects of a program:

Design material Spescification

Source code

Compilator

Feed-back
Object code Enhance-
ment

Figure 1 - Introducing computer programs

The notion of a computer program is centred on the version in source code. This is a program
written in a high level language. It may be one of the traditional languages like COBOL, but are
more likely to be a modern object oriented language like C++ or RUBY. In our context, this is
not very important. Such a programming language is often characterised as a high level language,
implying that is “high” in contrast to the programs existing at “deeper” levels in the computer –
the deepest level is where the program instructions are broken down to primitives which
correspond to the machine components, ultimately the transistors, accumulators etc which in
millions are present on the integrated circuits of the computer. The programming language
exploits these possibilities to carve out hunks with special functions. Such a hunk is then given
the name of an instruction, and using this instruction – which may be the word SORT – the
programmer invokes the whole hunk of these special functions. The high level programming
language can be read by a human who is familiar with the language, much like we learned to read
algebra. There are elements in the high level language which are similar to natural language, like
the word SORT, and its “meaning” is related to the meaning of the word in natural language. But
in contrast to natural language, the programming language has an detailed definition of its
semantic (or understanding).
Typically, the program in high level language will be run through another program to replace
the high-level instructions by the machine-related detailed instructions necessary to run the
computer. This program is called a compiler, in this the simple sentences of high-level language is
exploded into the very high number of primitive instructions implied by the definitions. The
result is the object program, which may govern the functions of a computer. The compiler also
checks the logic of the object program, and will report errors to the programmer. What may
above be implied as a one-way stream is really an interactive process, where the programmer has
the program compiled as he or she works on developing the program, using the feed-back of the
compiler to improve to program.

Page 3 of 19
Jon Bing

There is a one-to-one relationship between the source and object versions of the program in
the meaning that if two identical source programs were compiled by the same compiler, the result
would be identical object programs.

Occasionally one sees the process described as a “translation” from object to source versions.
This is not an appropriate choice of language, as translation conventionally is understood as a
situation where the translator has a certain freedom of choice with respect to picking the right
words or phrases. No such freedom exists for the compilation, it is a process governed by
strict rules. It may be better compared to a machine which converts a text of Latin characters
into Morse code, each letter being replaced by a pattern of dashes and dots.

There also is a one-to-one relationship between object and source code. Running the object code
the reverse direction through a re-compiler, a source code will be produced – and running the
same object code through the same re-compiler will each time result in an identical object
version. But there is no one-to-one version between the original object code and the re-compiled
object code. In compiling, a number of pre-defined choices will be made. Several source code
constructions may result in the identical set object code instructions. And the compiler will make
the object code optimal for the performance of the processing. Re-compiling the object code,
there will be many possibilities of generating source code structures. Originally, the programmer
had organised the source program for easy use by humans, the re-compiled program will reflect a
structure made optimal for computer processing, and therefore pose challenges for the human
sitting down to understand the re-compiled source.
In practice, the relation between source and object versions does not pose many difficulties
for copyright law.

2.2 Preparatory design work

When starting developing a program, someone must have an idea or take an initiative. There is
no lack of examples of there being the same person having a bright idea and developing this to a
successful source program. But often the situation is more mundane – an organisation has the
need for a certain function, and requests a program to be developed for satisfying this need.
Before the proper programming starts, there will be material describing the need, suggesting
solutions etc. In the diagram above, this is indicated by the boxes “design material” and
“specification”. These obviously cannot be compiled, and cannot result in a set of instructions
operating a computer. But according to the terminology of the Computer Program Directive,11
they are to be understood as “computer programs”, cf the preamble item 7:

“… wheras this term also includes preparatory design work leading to the development of a
computer program provided that the nature of the preparatory work is such that a computer
program can result from it as a later stage.”

The criterion to distinguish preparatory work which is just an ordinary literary work of non-
fiction discussing the development of a computer program, and preparatory work which is
protected as a computer program, is whether a computer program “can result from it as a later
stage”. This is certainly not a very clear criterion. It must obviously cover a situation in which a
program has been specified by a formalism – for instance quasi-coding – which leaves little
freedom for a programmer in transforming it into a source program. But how much freedom
should be allowed before there is an independent literary work and an independent program, has
to be decided in the context of a concrete case.
One should keep in mind that there may be a case of joint authorship, the programmer may
have sufficient freedom to imprint the resulting program with his or her choices, but at the same

Page 4 of 19
Jon Bing

time the preparatory work governs the programming in such a way that it becomes a derivative
work, the program is not independent of the preparatory literary work. This may be appropriate
also because there often is a functional division between the person designing the program,
resulting in a high level specification, and the person coding the program on this basis. The
designer may very well be the person with the original bright ideas, indicating functions and their
interrelation, while the programming may be of a more sweat-of-the-brow task realising the
indicated functions. In the abstract, this relation between designer and programmer,
corresponding to the issue of when a program is an adaptation of a specification, may seem
difficult. In practice, the issue will generally be solved by the rights of all those involved in the
project being collected on the hand of the employer or through contracts governing the project
(see below).
One may observe that the inclusion of preparatory design work makes it explicit that
computer programs do not have to be machine readable in order to fall within the scope of the
Computer Program Directive. In principle, a source program printed as an example in a text
book, will be a program under the Directive, though this may create some detailed problems in
practice.12

2.3 Originality
The Computer Programs Directive art 1(1) specifies that computer programs are to be protected
as “literary works within the meaning of the Berne Convention”. This implies that computer
programs are to be treated as other works within the category of “literary works”, which typical
examples are novels or text-books. This would seem appropriate, a source program has the
appearance of a text when printed out, though generally being more structured and repetitive in
the choice of terms than a piece of natural language text. However, the main difference is that the
text of a computer program through the process described above is transformed into instructions
governing a computer – it has a functional aspect not found in other literary works. Even a very
technical text relying on algebra, or a technical drawing defining in detail a machine, does not
have functions – there is no way such a work can be made to execute any process in the real
world by itself. The functional aspect is therefore a fundamental difference between computer
programs and other sub-categories of literary works, and this may have consequences for the
interpretation of the directive or national law in some situations.
The criterion for achieving protection is set out in the Computer Programs Directive art 1(3):

“A computer program shall be protected if it is original in the sense that it is the author's own
intellectual creation. No other criteria shall be applied to determine its eligibility for
protection.”

This is an attempt to harmonise – at least with respect to computer programs – how to decide
that a computer program rises above the threshold necessary to be qualified as a copyrighted
“work”. The Berne Convention does not itself have any definition or qualification of the term
“work”, though guidance may be sought in the exemplification of the Berne Convention art 2(1).
The criterion seem to require more than observing the program is not copied from an earlier
program (that it is “new”), as it has to be the creation of its author. How much should be added
to the criterion “not copies” is not very clear. But in its report on the Computer Programs
Directive, the Commission has stated that twelve member states were required to lower, and
three to increase the threshold.13
The major decision signalling that the criterion of originality was interpreted as more severe
with respect to computer programs was by the German Supreme Court.14 A decision by the same
court subsequent to the adoption of the directive, confirmed that the threshold had been

Page 5 of 19
Jon Bing

adjusted.15 One of the jurisdictions which had to raise the threshold was UK, where a “sweat-of-
the-brow” doctrine would seem to have been accepted.
In practice it may be quite difficult to appreciate the originality of a computer program. Often
a program is composed by using third-party components which are “glued” together by simple
coding, the contribution of the author’s own intellect may be the way to do it, and the result may
be similar to an anthology, to which the creative editor may have copyright. In practice, the
problem does not arise too often – and there seems to be considerable guidance in the rule of the
thumb: If a computer program has market value, it also is a copyrighted work.16
In this introduction, computer programs have been presented as instructions and the preparatory
design work which results in programs. But a “program” will also have other elements. The code
may represent graphical interfaces which may be protected as technical drawings or other
examples of art. The program may rely on feedback messages contained in separate libraries of
different languages for user communication. And there may be integrated systems for help and
assistance. To the user, this is all one “package”. Generally, such elements are not considered in
copyright terms to be part of the literary work qualified as a program (though there may be
modifications, see below under reverse engineering). And such elements may be subject to
protection under copyright law or related rights (for instance, the indexes or help files may be
qualified as protected data bases).
In general, one should be aware that a “program” is a system concept. One program may
consist of sub-programs, and interact with other programs to constitute one functional service.
To decide what is one program, will not be a technical issue alone, it also has to take into
consideration the marketing context, what is presented as “a program” to users, etc.

3 The rightholder
3.1 The original copyrightholder: Vertical, horizontal and temporal cooperation
The author of a computer program is:

“… the natural person or group of natural persons who has created the program or, where
the legislation of the Member State permits, the legal person designated as the rightholder by
that legislation. Where collective works are recognized by the legislation of a Member State,
the person considered by the legislation of the Member State to have created the work shall be
deemed to be its author.”

This is no different for computer programs than for other literary works. In many instances,
there is no problem in readily identifying the author, who is the person conceiving and coding the
program, using his or her exclusive rights as a platform from which to negotiate the exploitation
of the program.
However, computer programs are part of “industrial” copyright, and often the context will be
different from that in which the traditional literary works are created. It may be somewhat similar
to the development of text books, which may be huge projects initiated by a publishing house
involving several authors, illustrators and currently also web designers and possibly programmers.
The differences from the traditional context may be summed up:

 The initiative is usually taken by a institution or a more loosely organised “project”

 There are usually created specifications for the coding of the program
 There are usually several physical persons involved in coding the program
 Third party elements are occasionally purchased to be integrated with the program
 The program is developed for being present in the market for some time (years)

Page 6 of 19
Jon Bing

 The program will be maintained over time by making adjustments, corrections, or

adaptations over time

Project or
Designers
enterprise

R
ig
ht
s
co-operation

m
Vertical

an
ag
em
en
t

Programmers
Temporal co- Programmers for
– horizontal
operation maintenance
co-operation

Figure 2 - Cooperation in developing a program

In some jurisdictions, legal persons may be the original copyrightholder. According to the
Continental European model, the original copyrightholder must be a physical person, and the
small diagram illustrates aspects of this situation.
The coding of a program may be carried out by several, even many, persons. One may have
large projects where the program is divided into parts with well-defined interfaces. Groups may
have responsibility of one module, working in parallel with other groups under a co-ordinating
project management. The result will be a work “created by a group of natural persons jointly”,
and therefore “the exclusive rights shall be owned jointly”, cf the Computer Programs Directive
art 2(2). But this is only one of the several situations where such joint creation may take place,
indicated are also the possibility for “vertical” and “temporal” cooperation.
Here there is a “vertical cooperation” between designers, specifying the program, and
programmers, coding the program. This relationship has been briefly commented above.
A prominent feature of computer programs is their maintenance. The use of programs will
generate experiences which will be fed back to the design of the program. In any program of a
certain complexity there will in practise always be “errors” in the form of logical inconsistency.
There will always be the possibility of improving performance. Throughout the life of a program,
there will be new programs and hardware gadgets being introduced with which the program
should interoperate, which will require further development of the program. And there may be
functional enhancement. This is simplified in the diagram as a feed back loop amending the
specifications of the program, requiring further programming. Errors may be addressed by the
release of a patch to be integrated with the program. Enhancement may result in a new “version”
being released. Over time the program will be amended, perhaps even adapted. This change may
be incremental, over time the program will change – and this change will be brought about by
different persons than those involved in the first development.
The result is a “joint work”. As illustrated, this may include many physical individuals, and the
relation between the work and a person who has contributed with his or her intellectual creative
force to the program, may be rather tenuous. Some form of rights management is necessary, and
this is usually supplied by the institution, project or other entity which has taken the imitative to

Page 7 of 19
Jon Bing

the programming. There will therefore be an assignment of rights from the original
copyrightholders to the institution in question, and this institution will then face the market with
the whole bundle of rights as a platform for negotiating further contracts.
One will note this initial rights management is rather critical. If one of the original
copyrightholders fails to transfer his or her rights, there will be a defect in the title to the overall
program. In practice this may happen. Often an implicit assignment is implied by the context in
which the work is carried out. And there may be few possibilities for such an person to actually
to exploit the situation in a commercial way. These practical circumstances are probably the
reason for there being few examples of conflicts on this basis.

3.2 Employee’s assignment of copyright to employer

When organising the development of a computer program, an obvious way to organise the
project is through employment of designers, programmers and others. There will then be an
employment contract between the employer and the employee.
If a person is employed to create works – this is for traditional categories of jobs the case for
journalists, both in paper and ether media – the rights in the created work obviously have to be
shared in some way with the employer. It is understood that in this respect, the law varies
between European countries. In some jurisdiction, the employer (also when being a legal person)
may be the original copyrightholder. In other jurisdictions, the original copyrightholder is always
a physical person, the employee – and also the relationship between the employer and employee
is seen as subject to contractual arrangements and in the perspective of the labour law rather than
copyright law. In the Nordic countries, the organisations of journalists would maintain it is not
proper for the legislator to meddle with the right of the unions to enter collective agreements,
also with respect to the transfer of copyright between employee and employer.
Therefore, there is not a unified background law in this respect. In spite of this, the Directive
on Computer Programs art 2(3) harmonises the transfer of the rights of employee’s with respect
to programs:

“Where a computer program is created by an employee in the execution of his duties or

following the instructions given by his employer, the employer exclusively shall be entitled to
exercise all economic rights in the program so created, unless otherwise provided by contract.”

In this instance, European law has been harmonised. But it is a very limited co-ordination. As
indicated above, the background law on transfer between employee and employer varies between
jurisdictions. Also where there are contractual arrangements, there may be different rules for
interpretation, often in favour of the original copyrightholder (who in the traditional view is the
“weaker” part in the relationship).
The term “computer program” has to be interpreted according to the directive, but may not
be applied analogous for other types of work. On the background of a default legislation based
on the “hands off”-policy indicated above, the interpretation may be rather restrictive, for
instance not applying to design of data bases. The term “employee” is also a bit of a hurdle, in
internal law there may be different definitions of “employee” with respect to social security law,
working environment law, tax law etc, especially the qualification of “employee” with respect to a
“consultant” may be open to interpretation, as there in projects for developing a program often is
demonstrated inventive ways of associating individuals to the project. There obviously also may
be difficulties in determining whether the efforts are contained in the duties of the employment
contract, or who has taken the initiative to develop the program.
Without elaborating this further, it is obvious that national law may vary in this respect. But
these considerations on possibilities of divergences in national law should not overshadow the
principle, which is also appropriate as a guideline for managing the rights to a computer program:

Page 8 of 19
Jon Bing

The establishment of a practical regime of rights management which imply the collection of the
many rights on one hand, with appropriate upstream contractual arrangements.

4 The exclusive acts

Having decided what is the subject to the exclusive right (a computer program) and who is the
rightholders, one may approach the more succulent part of the issue – the exclusive acts reserved
for the rightholders.

4.1 The right of reproduction

4.1.1 Reproduction
The more important exclusive right is the right to reproduction, stated in the Directive art 4(a):

“ the permanent or temporary reproduction of a computer program by any means and in any
form, in part or in whole. Insofar as loading, displaying, running, transmission or storage of
the computer program necessitate such reproduction, such acts shall be subject to
authorization by the rightholder …”

“Reproduction” is interpreted according to the Berne Convention. But this does not have any
definition of reproduction. The Directive includes temporary reproductions, and this was for some
time an issue. It is believed that this issue is resolved by the Copyright Directive17 art 2, though
this does not directly apply to computer programs.18 The discussion leading to the adoption of
the Copyright Directive must be taken as the conclusion of the discussion, which may be seen as
addressing the permanence necessary fro a representation to qualify as a reproduction. It is rather
obvious that a “permanent” reproduction on paper, magnetic or optical medium is a
reproduction. It is also obvious that there are passing representations which are not qualified as
reproductions, like the reflection of a painting in a mirror.
Processing a work by a computer implies numerous instances of copying: In order to examine
a text on a computer screen, the text has to be communicated from a storage medium, typically a
magnetic disk, to the central processing unit of the computer, where the representation is used to
control the properties of the screen so that an image of the text appears. This representation in
the CPU is short-lived, the storage cache has a limited capacity (but may in modern systems be of
several gigabytes), and will be overwritten by the operating system if necessary. Also, data will by
paged in an out of the CPU using techniques for virtual memory to make the user experience that
a lager volume of data is available from the cache than in fact is the case. Similar solutions are
used in communicating data through a network, where packets may be stored intermediately in
nodes of the network waiting for forwarding capacity; the time stored varying upwards from
microseconds.
There was disagreement to whether these short-lived representations qualified as copies. The
Copyright Directive would seem to have put an end to that disagreement, and the Computer
Programs Directive uses the same terms in defining a reproduction. It must therefore be held
that all these examples qualify as reproductions. In practice, there therefore would seem to be (at
least) three degrees of permanence of a representation: Those too volatile to qualify as a
reproduction (like a mirror image), the temporary reproductions in computer based systems, and
permanent reproductions.

The author’s own view on this issue is probably visible even through the brief discussion
above. It is believed that the issue originally was caused by some concern with time-sharing
computer bureaux. Time-sharing was popular in the 1980s. Several users would share a
mainframe, accessing the computer from (dumb) terminals. A user would typically have an

Page 9 of 19
Jon Bing

account and a password, and would have access to a segment of the mass storage, where the
programs licensed by the user were available, and where data could be stored. The scenario
suggested a disloyal user accessing the mainframe, using the password of another user. In this
way, the disloyal user could process data stored on his or her segment of the mass storage
device by a program licensed by the user from whom the password was obtained. This would
then not constitute a copyright infringement as the program was not reproduced – unless one
qualified the temporary representation in the CPU of the terminal employed by the disloyal
user as a reproduction. Otherwise, one would have to refer to the criminal law on hacking,
which at that time was not very far advanced or co-ordinated between jurisdictions.

The Computer Program Directive specifies that “such reproduction” necessary for “loading,
displaying, running, transmission or storage” or the program is part of the exclusive right, which
includes typical examples of permanent reproduction (“storage”) and temporary reproductions in
the meaning indicated above (“loading, displaying, running, transmission”). The result is a very
strong exclusive right for reproduction.
In addition, the Computer Program Directive extends the exclusive right through art 4(b) to
the

“… the translation, adaptation, arrangement and any other alteration of a computer program
and the reproduction of the results thereof, without prejudice to the rights of the person who
alters the program …”

This makes it explicit that amendments or derivative works also are included in the exclusive
right. Perhaps the use of the term “translation” deserve a comment: Translation from one high
level programming language to another is not necessarily identical to translation a text from one
natural language to another. Programming languages may be related, or they may be based on
widely different design principles. If a program is written in a logic language like PROLOG, the
process of representing the rules in a language like COBOL would certainly be more than a mere
translation, and perhaps better seen as taking a detailed specification to develop a new program.
As algorithms, methods etc are not protected, one may in such a case argue that there is no
relevant derivative relationship between the programs. This is just an observation warning that
one will have to look at what actually has been done rather than the term used to describe the
process in order to decide whether it is a reproduction.

4.1.2 Delimitation for reproduction for intended use

It is obvious that the resulting exclusive right to reproduction is so strong that it has to be
modified in order to make the market work in practice. Therefore, the Computer Program
Directive art 5(1) makes a delimitation of the exclusive right for reproduction:

“In the absence of specific contractual provisions, the acts referred to in Article 4 (a) and (b)
shall not require authorization by the rightholder where they are necessary for the use of the
computer program by the lawful acquirer in accordance with its intended purpose, including
for error correction.”

One will note that the provision is secondary to a deviating contractual regulation, and therefore
presume that there is such relation between the partiers. In this context, there is no opportunity
to further discuss the practices of contracting for computer programs. Of course, these vary from
the detailed and complex documents governing the major development of a program system to
the transactions across the counter, where no negotiation takes place, and the terms will flow
from background law. It may not be unfair to maintain that the Computer Program Directive is

Page 10 of 19
Jon Bing

drafted on the understanding that the transaction will involve the acquisition of a physical
medium, typically diskettes or compact disks, on which the program is stored. At the time the
directive was drafted, it was not uncommon that a program also was executed from a diskette
inserted in the station of the computer. Development made this obsolete; the medium is only
used for transporting the program to the computer on which it is to be used – here a
reproduction takes place uploading the program to the internal memory of the computer. This
reproduction is permitted by the Computer Program directive art 5(1), as it is in accordance with
the intended purpose of the program. When executed, temporary copies will be made in the
CPU; these also are permitted as they are in accordance with the intended purpose. But without
contractual permission, only one such copy can be made of the program at the same time. It is
the original copy on the physical medium which is the copy acquired, and it is this that governs
the use of the program – the uploading to a computer is permitted because it is required for using
the original purchased copy according to its purpose. Therefore it also must be permitted to
delete an uploaded program and re-install the program on another computer, the relation
between the original copy and the reproduction remains. There may be a certain leeway found in
“intended purpose”, but the right to reproduce the program according to the Computer Program
Directive art 5(1) remains rather narrow.
As emphasised, the delimitation of the Computer Program Directive art 5 rests on the
presumption of a contractual relationship between the parties. Nevertheless, implementing the
provision in national law, at least two different strategies have been adopted. One is what would
seem to follow the presumption of the Directive, introducing a clause governing the
interpretation of licenses for computer programs – many national acts already will have clauses
for special categories of contracts, for instance the publishing contract or contracts for making a
cinematographic work. However, another strategy is to introduce the substance of the Computer
Program Directive as delimitations of the general exclusive right of a rightholder independent of
any contractual relationship, but subject to such contracts in the sense that a contract will be
given priority when the legislation does not state otherwise. The result is in practice rather similar,
but in certain special situations there may be relevant differences, see below.

4.1.3 Back-up copies

As anybody dependent on a computer system will know by their own experience, there will
occasionally be situations in which the data stored on local drives are lost. A basic principle of
data security is that there should be made back-up copies. The Computer Program Directive
permits reproduction for this purpose, cf art 5(2).

“The making of a back-up copy by a person having a right to use the computer program may
not be prevented by contract insofar as it is necessary for that use.”

The importance placed on this principle is evident partly from this being a mandatory right.
Reproduction for this purpose may be done by “a person having the right to use the computer
program”, which will include a somewhat wider circle than those having the right of reproduction
for the intended use, which is limited to “the lawful acquirer”. The right to make a back-up copy
certainly will include technical service staff routinely maintaining a computer system.
It may not be obvious why one need a back-up copy when the program is acquired on a
separate medium. If the program is lost on the computer on which it has been installed, a new
installation may be made from the original copy, as was the case in the first place – the original
copy filling the need for a back-up. But a user of a program will typically make many choices in
setting up the program to fit his or her preferences. All such features would be lost if a back-up
copy could not be made.19

Page 11 of 19
Jon Bing

The right to make a back-up copy is limited to

computer programs, a concept discussed above under
sect 2. Into what for the user is perceived as a program,
there may be elements which are not a computer
program according to the strict interpretation of the
term. A typical example would be manuals, collections of
clip-art etc. The program is also typically available to the
user in object form. There is no practical way in which
the user can qualify which part of the object code is a
computer program in the strict sense, and which
elements are accessories of a different nature. The clause
must therefore be interpreted to permit reproduction for
back-up purposes of all files presenting themselves as
part of the purchased program.
This point gains some extra interest because the
Copyright Directive seems to lack a corresponding
provision. It is in some way remarkable that an
delimitation found sufficient vital with respect to
computer programs to make it mandatory, is not even
mentioned with respect to other types of work, where it
will be just as vital if these work are in computerised
form. For such works, one therefore has to rely on the
contract to authorise the reproduction necessary for
Figure 3 - Example of files associated with
"one" program back-up routines.20
The Computer Program Directive art 5(2) uses the
phrase “a back-up copy”. This seems to imply that only one back-up copy is permitted at any
time. According to routines, back-ups are made at intervals – every day, week etc. If only one
back-up copy is permitted, this implies that earlier copies will have to be deleted when a new is
made. Again, this would be contrary to basic data security principles, which advice at any one
time to have several generations of back-up copies, preferably stored at different locations to
reduce the risk of accidental loss.
It must be justified to interpret the clause as a reference to normal back-up routines,
permitting the reproduction of computer programs to the extent this is compliant with normal
routines and loyal to the rightholder.
A special situation occurs when a licence to a program lapse. The situation may be that a
business has contracted for a program for some vital functions of its operation, perhaps as trivial
– but necessary – as accounting. There is negotiated a special license agreement which includes a
maintenance arrangement, the program being made available for an annual license fee. Variations
of such arrangements are rather common for large programs requiring user-adaptation. As time
goes on, there emerges competing programs, and the user decides to discontinue the relationship
with the first provider, and switch to another solution. The license agreement for the first
program will lapse. Without such agreement, there may be argued that no right to maintain
reproductions of the program remains. This would place the user into a very uncomfortable
situation, as the program is needed for access to historical records, such access would be required
by law in most jurisdictions for accounting records. One may look to the Computer Program
Directive art 5(2) and argue that the right to make and retain a back-up copy also extends to a
person had a right to use the program when necessary to access the data processed by the
program in the period the contractual arrangement lasted.

Page 12 of 19
Jon Bing

4.1.4 Reproduction for private use

According to the Copyright Directive art 5(2)(b), one of the possibilities of delimiting the
exclusive right for reproduction by national legislation is “reproductions … made by a natural
person for private use”. Considering the Computer Program Directive, one will not find a similar
possibility for delimiting the exclusive right to reproduction. This implies that the Directive bars
the possibility in national legislation to open for “private reproduction” of computer programs.
When the Computer Program Directive was issued, this was a rather controversial feature.
The justification obviously was the argument that computer programs were very vulnerable for
reproduction, and that the delimitation of an exclusive right to private reproduction would
represent a substantial reduction of the exclusive right enjoyed by the rightholder, and would fail
to meet the three step test of Berne art 9(2). The concern was partly that individuals would copy
programs from their work place to private portable computers or home computers, but also the
market for gaming modules, where the major market was private and the typical user (at least at
this time) juveniles.
When the Copyright Directive was passed, the provision in general on private reproduction
was made subject to a presumption of fair remuneration to the rightholder. But the regime
established for computer programs was maintained, with respect to this type of literary works,
there is no delimitation of the exclusive right for private use. In some national implementation,
one has made it explicit that this only holds for computer programs in machine readable form,
otherwise a program printed in a text book would be subject to other provisions on reproduction
than the rest of the book.

4.1.5 Analysis of programs

The Computer Program Directive emphasises the dichotomy between the expression (which is
protected by copyright law) and the ideas of a work (which are not protected). This is mentioned
in the preamble, and it is repeated in art 1(2), mentioning explicitly that “ideas and principles” are
not protected. And the Directive art 5(3) restates this in the context that a person having the right
to use a program, may examine it to “determine the ideas and principles which underlie any
element of the program”. It would seem that this is overstated in the Directive, as it is a
delimitation flowing also from basic copyright law, and would apply without any direct regulation
in the Directive.
Perhaps the reason for emphasising the point is the strong exclusive right for reproduction
given to the rightholder, and the relation to the right of decompilation, see below under sect 5.

4.2 The right of distribution

4.2.1 The exclusive right to distribution
The exclusive right of reproduction is complemented by an exclusive right of distribution in the
Computer Program Directive art 4(c):

“… any form of distribution to the public, including the rental, of the original computer
program or of copies thereof.”

The typical situation is that copies are offered for sale across the counter, and this is obviously
covered by the clause. The exclusive right requires that each copy of the program has to be
distributed under the consent of the rightholder.
Outside the consumer market, programs may be offered on the understanding that they will
be configured for the individual user – this will also be a distribution of copies. “Rental” is
explained in recital 16 as ” he making available for use, for a limited period of time and for profit-
making purposes, of a computer program or a copy thereof”. This will include the arrangement

Page 13 of 19
Jon Bing

where the license to use a program relies on periodic payments, often combined with an
“maintenance contract” for keeping the program current with respect to the developments of
hardware and relevant other programs. There is a Swedish case relating to Nintendo gaming
modules, where these were sold with an offer of repurchase of the game from the customer on
favourable terms, this were not seen as an infringement of the distribution right.21
The Computer Program Directive makes no mention of the public lending, which remains
outside the scope of the Directive, cf recital 16. The Rental and Lending Right Directive22 art 3
explicitly states that this directive is without prejudice to the Computer Program Directive art
4(c). On the other hand, it is only this provision which is excluded, the Rental and Lending Right
Directive art 5(2) there is a provision applying to the situation where the exclusive lending right is
not implemented in national legislation for computer programs, in that case, remuneration for
authors should be introduced.
The provision is cast on the basis of computer programs being made available to the market
on some sort of carrier – diskettes or compact disks being typical examples. Currently, programs
are typically acquired by downloading the program from a site. In this case, the purchase does not
presume distribution. Downloading will result in a copy being made on the computer of the
purchaser, but this copy has not been physically handed over from the rightholder. The purchase
through downloading is an example of a service (an “information society service” in the
terminology of the Electronic Commerce Directive).23 The result of the downloading obviously is
a reproduction falling within the exclusive right according to the Computer Program Directive.

4.2.2 Exhaustion
The Computer Program Directive art 4(c) make the principle of regional exhaustion of the
distribution right. It applies when a copy of a computer program is sold; the distribution right in
this copy is then exhausted.
The copy has to be subject to a sale. There may be other transactions which make a copy of a
computer program available to a party. Typically, there may be a license agreement which make
the program available for the period a license fee is being paid. This will not constitute the sale of
a copy of the program, but the qualification of the transaction may not be trivial, and will have to
take the details of the situation into consideration.
Exhaustion only takes place within the European Community, and the European Economical
Area. This is the general principle of exhaustion promoted by the EU, and a discussion of this in
general falls outside the scope of this brief introduction.
Exhaustion includes only the copy subject for sale. In the case of off-the-shelf software, it
only includes the original reproduction of the program on the carrier, which typically will be a
compact disk.
In some instances, a computer is purchased with some pre-installed programs, typically
operating system and some basic office programs. In this case, the copy has been purchased
residing on the hard drive of the computer, and the distribution right is exhausted in these copies
by the purchase, permitting the computer to be re-sold with the same programs. However, there
are details of complication, as such programs may be updated through online services, and if the
updates considered separately qualifies as works (as often will be the case for major updates), they
have not been subject to a sale as copies, and they are the result of services made available
through the net.
This is an indication of the exhaustion principle not being quite straightforward. As
mentioned, exhaustion does not apply to a program downloaded from a site. Occasionally, a hard
copy in the form of a compact disk is offered as a back-up measure, being mailed the purchaser
separately, and typically for an additional fee. The distribution right in this must be subject to
exhaustion, and may probably be sold by the purchaser to a third party without the consent of

Page 14 of 19
Jon Bing

the rightholder, while the purchaser continues to enjoy the advantages of the downloaded copy.
To avoid this, the rightholder may employ some technological protection measures.
Taking the typical example where off-the-shelf programs are purchased, the distribution right
in the original copy will be exhausted. The purchaser may therefore sell this to a third party. But
the right of reproduction is not exhausted. Therefore, the third party – who has no contractual
relationship with the rightholder – does not necessarily enjoy the delimitations of this exclusive
rights set out in the Computer Program Directive art 5. For instance, if the third party has
purchased a compact disk with a program and wants to use this on his or her computer, the
program has to be uploaded to the hard disk of the computer, which constitutes a reproduction.
Without a contract, it may be queried whether the Computer Program Directive art 5(1) applies,
as this provision may be overruled by contractual arrangement. However, in the case of resale,
the rightholder has no possibility of making such arrangements.
As briefly mentioned, some implementations have made the provision of the Computer
Program Directive part of the general delimitation of exclusive rights rather than provisions on
the interpretation of contracts. If this strategy has been used, the problem indicated above is
overcome, as a third party acquiring a copy of the program will be able to claim the same
delimitations in the right of reproduction.
But also otherwise, one should interpret the Computer Program Directive art 5 to apply also
to a third party having acquired a copy of the program in which the distribution right has been
exhausted. It may be slightly at odds with conventional copyright principle to construe the
transfer of a license for reproduction with the copy in which the distribution right has been
exhausted. But otherwise it would seem that the exhaustion to a large degree would be
meaningless, one would be permitted to transfer the copy to a third party without the consent of
the rightholder, but that third party would not be in the position of making the intended use of
the copy without such consent.

5 Decompilation
The right for decompilation is an innovation in copyright law. Its background may be found in
the dispute of the 1980s between IBM and the European Commission, which was not dissimilar
to the dispute with Microsoft which ended in 2007. In both instances there conflict was oriented
towards a dominant enterprise, and its use of market power.
Under copyright law, certain exclusive rights are granted the right holder. When the
copyrighted work is made available to the market, anyone can avail them of the content, learn
from the work and be inspired by it – and on this basis go on and create new and independent
works by themselves. The exclusive right securing the position of the rightholder is traded in the
knowledge to be gained from the protected works. For computer program, this (simplified)
justification does not work – the programs are typically marketed in object form, which in
practice cannot be read by a human.
Computer programs work in a context. When developing a program for a certain function, the
developer would like to draw upon the functions of another program. One may think of a simple
example, where a program is developed assisting the translation from Norwegian to English. By
highlighting a term, the auxiliary program will look for translations from its database. In order to
do this, the developer of the auxiliary program needs to know how to communicate with the
word processing program. This is often seen as a question of interface between a program and the
surrounding domain. With knowledge of the interface, other programs may communicate with
the program in the way defined by the program, and may obtain interoperability between
programs.
Interoperability is seen as desirable; it fosters competition and stimulates growth. Therefore it
was seen as desirable for providers of program to publish the specification of the interface of a

Page 15 of 19
Jon Bing

program. But the provider of a program may be reluctant to do so, because ignorance of the
interface may offer some competitive protection.
This is the issue which the Computer Program Directive art 6 attempts to solve. It easily was
the most controversial provision of the Directive, and the provision is rather complex. In this
context, it will not be discussed in all its details, but an attempt is made to outline its structure.
First, the provision may only be applied by someone developing an independent program. It is
the situation of the developer which may trigger the right of decompilation. The developer sees
interoperability with another program desirable. If necessary data on the interface is published or
otherwise available, there is no need24 to look further, and the right to decompilation cannot be
invoked. Only when “indispensable”, decompilation may take place.
If decompilation is indispensable, further conditions apply. The program to be analysed must
be lawfully available, the decompilation can only be performed by a licensee or another having
the right to use a copy of the program. And the decompilation can only be made of “the parts of
the original program which are necessary to achieve interoperability”. However, as the original
program only is available in object form, it may be difficult for the person doing the
decompilation to determine which parts are necessary, only what is available can be required used
for determining what to decompile.
The result of the decompilation can only be used for gaining the necessary interoperability,
and not for any other purpose. The provision emphasises that decompilation does not authorise
development, marketing etc of a program “substantially similar in its expression” or any other act
which infringes copyright. To drive this point home, the Computer Program Directive art 6(3)
refers to the Berne Convention, and emphasises that the application of decompilation cannot be
“used in a manner which unreasonably prejudices the right holder's legitimate interests or
conflicts with a normal exploitation of the computer program”.
Decompilation is therefore construed as a rather narrow right. If data on interfacing with a
program is available to developers, it will not apply. And when applying, it only applies to the
extent necessary to establish the desired interoperability with a program independently created.
In practice, decompilation result in a reproduction of the program, the object code being
transformed to a form of source code by the strict rules of the decompiling program. As noted
above (sect 2.1), it will not be identical to the original source code due to the processing by the
programs performing the compilation and decompilation. The resulting “pseudo source code”
may be quite a jumble, and require considerable skill and time to analyse. Therefore,
decompilation may be less of a solution than seen at first glance. The major effect of the
provision is that on legal policy: To encourage the publication of interface specification and
through this, exclude the application of the provision.

6 Technical protection measures

According to the Computer Program Directive art 7(c), there is established a special protection
for technical protection measures. The discussion of digital right management and technical
protection measures became very heated with respect to the Copyright Directive, which in its art
6 implements the WIPO Copyright Treaty (1996) art 1125 on technological measures. The
provision of the Computer Program Directive predates this provision, and is excluded from the
Copyright Directive, cf art 1(2)(a). Therefore, the provision on technical protection measures for
computer programs applies in parallel to the more extensive regulation in the Copyright
Directive.
At the time the Computer Program Directive was passed, such protection devices were less
sophisticated than examples currently being deployed. A typical device would be a plug (a
“dongle”)26 having to be present at the connection between the computer and printer. The
program would check for its presence, if not found, the program would fail to print. This gave

Page 16 of 19
Jon Bing

some security for the program only being unlawfully reproduced as the plug itself was difficult to
duplicate. Other measures would be key diskettes, having to be present when initiating the
program, the use of codes, etc.
The Computer Program directive applies to the possession, or putting into circulations “any
means” which facilitate “the unauthorized removal or circumvention of any technical device
which may have been applied to protect a computer program”. The act of circumvention itself is
not qualified as illegal; the provision focuses on the “means”. The term “means” is not specified,
but will probably include “devices, products or components” (cf the Copyright Directive art
6(2)), but probably not services.27 The subject for circumvention is qualified as a “technical
device”,28 it is suggested that this hardly is any limitation – any measure implemented for a
computerised system will be “technical” in some meaning of the term, but the term “device”
would seem – as suggested above – to exclude services.
There are two conditions.
First, this must be the “sole intended purpose” of the prohibited device. This is more limited
than in the later Copyright Directive. Many devices may permit circumvention, but they may also
have other and legal applications, for instance to translate formats.29
Second, the act must be for “commercial purposes”. This will exclude acts which are done by
private persons, circumvention protection devices to access a program without any commercial
purpose.

7 Term of protection
The Computer Program Directive art 8 stipulates the term of protection. This was originally fifty
years counted from – as for other works – alternative dates, the major being the death of the
author (post mortem auctoris). However, the period was amended by the Term Directive30 art 1(1)
for all literary works, including computer programs, to seventy years.
In this context, there is little reason to further discuss this provision, perhaps only with a small
note on whether such a relative long term of protection is justified for computer programs. The
argument may be that though the term of protection may be justified for other types of literary
works (and here opinions differ), it is not justified for works which is seen as rather short lived.
But the brief life-span of computer program is to some extent a myth. Computer technology
has since the beginning been subject to rapid development, which makes equipment obsolete in a
matter of years. It is fascinating to note that the programs often survive through these changes by
amendments and functional enhancement. Through incremental development, programs will
make the leap from one generation of hardware to the next. There is no lack of examples of
programs originally designed for the first personal computers in the early 1980s which still have a
place in the market or, indeed, programs designed for managing the databases made possible by
the first large magnetic discs still making their imprint on current systems in public
administration.
But to unravel the term of protection in such cases must be in practice difficult. One will be
able to determine when the term starts for the first version of a program. When the next version
comes along, one has to decide whether the amendments meets the test of originality and
represents a derivative work, or whether only changes of no copyright relevance has taken place.
In the first case, a new term of protection will be initiated for the derivative work. This will
happen time and again, the original coding may get lost in the chain of development – in
principle there may be old versions actually passing into the public domain, but in practice this
will be difficult to determine and of little practical importance.
The term of protection is, however, sufficiently long for us not to have had any direct
experience. What generally is accepted as the first electronic computer, the ENIAC, was unveiled
in 1946, was programmed by manipulating its switches and cables, only when the von Neuman

Page 17 of 19
Jon Bing

architecture was adopted for storing program in a logical notation, the first stored program for
the amended ENIAC initiated 16 September 1948. This program was written for Jon von
Neuman by Adele Goldstine, who died in 1964. Of course, other programmer would have died
before her, but it may be an appropriate note of curiosity for closing this introduction to reflect
on the fact that the term of protection of the first program will last till 2035.
1 The first version of COBOL was adopted 1968 by American National Standards Institute.
2 The complaint for the case U.S. v. IBM was filed in US District Court, Southern District of New York on 17
January 1969 by the Justice Department. The case was withdrawn on 8 January 1982.
3 This is based on my own notes from the meeting.
4 Advisory Group of Governmental Experts on the Protection of Computer Programs.
5 25 February – 1 March 1985.
6 Vitaly Troussov, Deputy Director of the Patent Examination Department, USSR State Committee for Inventions
and Discoveries
7 One will find a reference to this intervention, though stripped of the images, in the report of the meeting
paragraph 22, UNESCO/WIPO/GE/CSS/3 8 March 1985:4.
8 Swedish Law and Informatics Research Institute, Faculty of Law, University of Stockholm.
9 UNESCO is the depositary to the Universal Copyright Convention.
10 Art 14.
11 Council Directive 91/250/EEC of 14 May 1991 on the legal protection of computer programs.
12 See below under sect 4.1.4.
13 Thomas Dreier and Bernt Hugenholtz (eds) Concise European Copyright Law, Kluwer, Alphen aan den Rijn
2006:217. The express objective of harmonisation is stated in the Preamble item 4.
14 Bundesgreichthof 9 May 1985, see for instance GRUR 1985:1041, ICC 1986:681. The case is generally referred to
as “Inkasso-Programm”.
15 Bundesgreichthof 14 July 1993, see for instance GRUR 1994:39, ICC 1995:127. The case is generally referred to
as “Buchhaltungsprogramm”.
16 Formulated by Mogens Koktvedgaard, Danish professor of intellectual property law, refusing to be confused by
subtle arguments.
17 Directive 2001/29/EC of the European Parliament and of the Council of 22 May 2001 on the harmonisation of
certain aspects of copyright and related rights in the information society.
18 Cf Copyright Directive Art 1(2)(a).
19 UK courts have held that back-ups may not be necessary when the program is made available on a CD-ROM,
see the decision by the High Court of Justice, Chancery Division 23 Jan 2002 in Sony Computer Entertainment
Inc v Paul Owen and others, [2002] EWHC 45 (Ch). The decision concerns use of a device to circumvent the
geographical segmentation imposed by Sony on game modules, but the opinion of Justice Jacob on this point is
given in very general terms. A similar view is taken by Justice Laddie in Kabushiki Kaisha Sony Computer
Entertainment Inc et al v Ball et al, decision by High Court of Justice (Chancery Division) 19 July 2004 [2004]
EWHC 1738 (Ch). However, the decisions should be interpreted in the context of the disputes. A gaming
module does not offer the type of user settings which in the text is indicated as creating a need for back-up
copies also where the program is made available on a separate medium like a compact disk.
20 This point will not be pursued, but there are obvious tension between the two directives at this point.
21 Yapon AB v Ekström, Supreme Court 22 Nov 2002 {2002]ECDR (14) 155.
22 Council Directive 92/100/EEC of 19 November 1992 on rental right and lending right and on certain rights
related to copyright in the field of intellectual property. This directive has a somewhat broader definition of
“rental”, cf art 1(2).
23 Directive 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on certain legal aspects of
information society services, in particular electronic commerce, in the Internal Market ('Directive on electronic
commerce').
24 Though there will be rumours that the published data does not specify the most efficient way to interact with the
program, etc.
25 Cf also the WPPT art 18.
26 In a German case, the court held that as the program was sold with a dongle, the intended use presumed also the
use of the dongle, referring to the delimitation of the restricted acts under art 5. It might as well have been
discussed in the context of art 7(c). Cf Oberlandsgerich Karlsruhe 1 October 1996, 6 U 40/95.
27 There is a Finnish case in which written instructions of how to circumvent were available, the Supreme Court did
not find that this qualified as putting into circulation a means for circumvention according to the Finnish
implementation of the Computer Program Directive, cf Adobe Systems Inc v [A] Software distribution 3 Oct
2003 [2004] ECDR (30) 303.

Page 18 of 19
Jon Bing

28 Cf Copyright Directive art 6(3).

29 This is discussed in Kabushiki Kaisha Sony Computer Entertainment Inc et al v Ball et al, decision by High
Court of Justice (Chancery Division) 19 July 2004 [2004] EWHC 1738 (Ch) with respect to the “Messiah2” chip,
where Justice Laddie rejects that the chip has any other purpose than circumvent the copy protection of Sony
Playstation 2.
30 Council Directive 93/98/EEC of 29 October 1993 harmonizing the term of protection of copyright and certain
related rights.

Page 19 of 19