THE SUPREME COURT OF APPEAL

OF SOUTH AFRICA

 

Reportable

 

CASE NO: 118/05

 

In the matter between :

 

HAUPT, ANTON CHARL t/a SOFTCOPY Appellant

 

and

 

BREWERS MARKETING INTELLIGENCE (PTY)

LIMITED First Respondent

BREWER, CHRISTOPHER JOHN Second Respondent

COETZEE, BYRON Third Respondent

________________________________________________________________________

Before: HARMS, STREICHER, MTHIYANE, CLOETE & LEWIS JJA

Heard: 15 FEBRUARY 2006

Delivered: 29 MARCH 2006

Summary: Copyright – ‘computer generated’ need not produce correct results – improved work eligible for copyright if original – databases and database structures not computer programs – computer generated work only if no human author – creativity not a requirement of originality – control over the making of a computer program.

 

Neutral citation: This judgment may be referred to as Haupt v Brewers Marketing Intelligence (Pty) Ltd [2006] SCA 39 (RSA)

________________________________________________________________________

 

J U D G M E N T

______________________________________________________________________________

 

 

STREICHER JA

 

STREICHER JA:

[1] Anton Haupt, the appellant, applied to the Cape High Court for an order interdicting the respondents in terms of the Copyright Act 98 of 1978 from infringing his alleged copyright in a computer program known as Data Explorer and also in certain tables (or database structures) and databases. The High Court held that Haupt’s claim could not be sustained and dismissed the application. With the leave of this court Haupt now appeals against the High Court’s judgment.

 

[2] Haupt and the second respondent, Christopher Brewer, are brothers-in-law. Brewer used to be the managing director and Haupt the marketing director of Brewer’s Almanac (Pty) Ltd (‘Brewer’s Almanac’). Brewer’s Almanac did business as an advertising agency and also disseminated information of use to the advertising industry.

 

[3] During 1998 the third respondent, Coetzee, was requested by Brewer’s Almanac to write a computer program which could interrogate and manipulate what is known as AMPS (All Media Products Survey) data. AMPS data are research results produced by a media research company on behalf of the South African Advertising Research Foundation. AMPS data are based on market surveys done on a six-monthly basis by way of questionnaires and are available in binary column electronic format stored in a UFL file and captured on a compact disc. They enable one to determine, inter alia, who the readers, listeners or viewers of particular newspapers, magazines, radio stations or television programs are and who the users of various products are. The information is useful especially to advertising agencies in the planning of marketing strategies.

 

[4] Coetzee was requested to write a program that would give the user the ability to select certain questions, extract the answers from the UFL file and display those answers in a meaningful way. He used a computer program called Delphi to write the program which he called Project AMPS. To enable the selection of a question, the questions upon which the AMPS data were based had to be stored in a database. For this purpose Coetzee created a table. This was done by simply asking the Delphi program to create a new table. The program allows one to determine the number of fields required and the extent of the fields. Haupt then undertook the task of filling the table with the questions contained in the questionnaire in respect of the AMPS97b data. Coetzee, in the meantime, wrote a program that could read the data in the binary format supplied. The next step was to create the user interface or front end of the program, that is, that part of the program that determines what a user will see. Coetzee decided that the questions should be displayed in a ‘tree’ format and used the tree component of the Delphi program for this purpose. He then proceeded to write a program that could read the questions in the table (the questions database), populate the tree when the program was activated, allow the user to select a question and extract, calculate and display the data requested. By 21 June 1998 the program had developed to the extent that a tree could be populated from a questions database, a user could select one of the questions, and an answer, sometimes correct sometimes wrong, could be extracted from the UFL file and be displayed on a monitor.

 

[5] On 31 July 1998 Haupt and Brewer parted ways and from that date onwards Haupt was no longer in the employ of Brewer’s Almanac. Thereafter, Coetzee, in terms of a prior arrangement with Haupt, continued to develop the program for Haupt exclusively. In terms of the arrangement Coetzee was to receive 20 per cent of the gross sales of the program.

 

[6] The next major development of the program after 21 June 1998 was the incorporation of a ‘tree preparer’ program. Every time the program was loaded it took time to populate the tree from the questions database. To overcome this problem Coetzee wrote a program called the tree preparer program which could, from the questions database, prepare a tree populated in the format required and then save it in a tree.txt file. He took out the link to the questions database and thereafter the tree.txt file, instead of the questions database, was used to populate the tree. Haupt testified that this development took place after 31 July 1998. His evidence in this regard was not disputed. However, when Coetzee testified he suggested that the development might have taken place before 31 July 1998. According to him he established that he started working on a tree preparer program on 5 or 6 July 1998. He could not say when he completed the writing of the program but could say that it would have taken him about six hours to do so. In my view Coetzee’s belated and half-hearted suggestion that the program might have been written before 31 July 1998 is not sufficient to cast doubt on Haupt’s evidence in this regard. Coetzee also wrote a program which could search the tree view of the questions. It was not suggested that this was done before 31 July 1998.

 

[7] The data in the UFL file was not sorted and every time a question was selected a search had to be done through the whole of the file. Quick access could, therefore, not be gained to a particular sector of the data. To overcome this problem Coetzee created an answers database and changed the existing program so as to pre-sort the answers contained in the UFL file and to order and index them in a table or database structure created as in the case of the questions database structure. Once this had been done the program had to be changed again so as to look for an answer to a selected question in the answers database instead of the UFL file. This procedure had to be followed in respect of each set of AMPS data as it became available. The development took place at the beginning of 1999. At that time Coetzee also created a weightings database. This was necessary because respondents to the questionnaires have a weighting which is used to determine their actual representation in the relevant population. Subsequently various other database structures were added to the Data Explorer program.

 

[8] Every six months when new data was received the Data Explorer program had to be changed from reading the answers file to reading the UFL file and once this had been done it had to be changed back again. To obviate these changes Coetzee, in June 2000, created a new program (the converter program), separate from the Data Explorer program, to do the conversion from the UFL file to the answers file. Coetzee described the effect of the development as follows:

‘So that meant that I could leave the Project AMPS program pristine, every time a conversion came I could just do my conversion with a separate program and finished, I didn’t have to remember what I’d changed and all that kind of stuff.’

He referred to the program as the Project AMPS program although, by that time, the name of the program had been changed to Data Explorer.

 

[9] Coetzee left for the USA in October 2000. Before he did so he worked full time for a period of two months on the further development of the program. During this period Haupt paid him R20 000 per month. It was at this time that a new graphing tool was added to the program. This was done by purchasing a graphic server which was commercially available and by incorporating it into the Data Explorer program.

 

[10] The AMPS data referred to as Teen98, Child99, AMPS99a, AMPS99b and AMPS2000a were all processed by Haupt and Coetzee and marketed by Haupt together with the Data Explorer program. In respect of each set of data a tree.txt file was compiled by the tree preparer program from the questions table filled by Haupt and the data contained in a UFL file was converted to an answers database and a weightings database.

 

[11] On 26 March 2001 Coetzee, who, at that time, was still in the USA, made contact with Brewer by e-mail stating inter alia that he had heard little or nothing from Haupt. This led to the following letter from Brewer to Coetzee on 4 May 2001:

‘Dear Byron,

. . .

There’s something I’d like to discuss with you – but it MUST be kept strictly confidential and just between you and me – for obvious reasons.

I have been asked by several agencies to get busy on other database products, some of which would be impractical, but more recently SAARF asked me if I would be able to produce a “Brewers AMPS Data Program”.

Let me say straight off that I have absolutely NO intention of developing a program for this – but I WOULD be interested in coming to an arrangement with you where we would use the program you wrote to convert the SAARF data (we would then develop the front-end to make it look nice as well as looking like our other products).

I . . . also believe there is room for another player in the market (Anton is obviously doing exceptionally well – in fact he seems positively RICH) and you, yourself, would not be compromised. I’ll make that a little clearer if I can:

I need to use the program you wrote in order to import SAARF (AMPS) data into a Brewer’s database (which Hank will write). For this you will earn a “royalty” or “commission” on sales.

. . .

So here’s my question to you:

Are you interested in allowing me to use your program for converting SAARF (AMPS) data into a database (which we will write) in return for a royalty?

. . .

Whichever is the answer, please DO NOT discuss this with Anton. There is nothing devious in my saying that – it’s just that I don’t want to create any family stress or tension (especially if your answer is “no”).

There is no other reason for me not wanting Anton to know.

If we do go ahead then (if necessary) I’ll explain it all directly to him. There is absolutely nothing underhand about you and I discussing this.

. . .’

 

[12] Negotiations between Coetzee and Brewer ensued culminating in a written agreement on 2 July 2001. In terms of the agreement they agreed that Brewer’s Marketing Intelligence (Pty) Ltd, the first respondent, would develop the program, that Coetzee would provide compiled (converted) data to the first respondent to enable the first respondent to develop the program, and that Coetzee would co-operate and advise the first respondent during the development stage.

 

[13] The first respondent employed one Hank Bento to write the program which became known as the Brewer’s AMPS program. While doing so Bento was in frequent contact with Coetzee for assistance. Coetzee regularly gave advice to Bento and from time to time sent him portions of the source code of the Data Explorer program. The source code for the search function and also the source code required to incorporate the graphics server were so sent to Bento. In both instances this was done because Bento struggled to do the programming. Referring to an e-mail in terms of which he sent the code in respect of the graphics server to Bento, Coetzee testified:

‘[W]hat has happened here is Mr Bento has installed the graphing package, he’s never seen it before, and he’s tinkered with it a bit and he’s realised that there’s a lot more to this than just simply installing it, he now has to write code around it in order to use it, so he has contacted me, I’m not sure how, either with MSN or by e-mail, and said, listen, Byron, how did you use this, what, can you shortcut me on this, please, and so I said, I sent him an e-mail with the exact code that is in Project AMPS that I wrote around using the graphic server previously, and it’s about just over two pages long, the code.’

Bento conceded that he made use of these source codes.

 

[14] At that time, the latest AMPS data, namely AMPS2000b, had not been converted by Coetzee for Haupt. The first respondent acquired this data, Brewer filled a questions table with the questions in respect of the AMPS2000a data, sent the data (contained in a UFL file stored on a compact disc) together with the questions database so created to Coetzee, and Coetzee converted the data from the UFL file to an answers and weightings database. Coetzee, by using the tree preparer program, also created a tree.txt file and returned the conversions and the tree.txt file to the first respondent. In addition Coetzee supplied the Teen98, Child99, AMPS99a, AMPS99b and AMPS2000a databases which had previously been created for the Data Explorer program, as well as the relevant tree.txt files, to the first respondent.

 

[15] The first respondent marketed the Brewer’s AMPS program together with all the aforesaid converted data and the tree.txt files. The existence of the program was discovered by Haupt when he investigated why his Data Explorer program, which was being used by Nasionale Pers, was not functioning properly. He established that the malfunctioning was caused by the presence of a similar program, namely the Brewer’s AMPS program.

 

[16] Haupt thereupon applied for an Anton Piller order and for an order interdicting the respondents from infringing his copyright in the Data Explorer program and various files and folders including the aforesaid databases. By this time Coetzee had returned to South Africa. The Anton Piller order was granted and the respondents were called upon to show cause why an interdict should not be granted against them.

 

[17] On 29 May 2002 the interim interdict against the respondents was discharged and, by consent, the following order was made in respect of the application for an interdict:

‘3 The final interdict prayed for by the Applicant is to be dealt with as follows:

3. 1. Determination of the final interdict is referred to trial on 4 November 2002;

   2. The Applicant is ordered to deliver his supplementary replying affidavits if any by no later than Thursday 20 June 2002;

   3. It is ordered that the affidavits, filed and to be filed, in the present matter are to stand as pleadings;

   4. The Uniform Rules of Court pertaining to trial actions are to apply to the present matter.

4. Costs are to stand over for later determination.’

 

[18] This order should never have been made. The affidavits together with the annexures comprised 330 pages in which the issues were not properly defined. The problem could have been alleviated had the parties held a pre-trial conference as required by para 3.4 of the order but that was not done either. As a result the parties plunged into a trial without a proper appreciation of what the issues were. The trial commenced on 26 February 2003 and the respondents closed their case on 24 February 2004 after a number of postponements. No less than 33 court days were taken up. A lot of time and energy were spent at the trial on the question as to the shareholding in Brewer’s Almanac and on the question as to whether the relationship between Haupt and Brewer, while they were employees of the company, was similar to the relationship between partners. It has now been conceded by all the parties concerned that those disputes were irrelevant. An inordinate amount of time was also spent on the question whether it was Brewer or Haupt who first had the idea to write the program, and as to the relative contributions of Brewer and Haupt to the writing of the Project AMPS program. Again these issues were irrelevant as it is common cause that Haupt and Brewer were acting in their capacities as employees of Brewer’s Almanac. The parties got so entangled in irrelevant issues that the real issues, such as the nature of the works in respect of which copyright was being claimed, whether the works were original and who the author for purposes of the Act was, received scant attention.

 

[19] In terms of Uniform rule 6(5)(g) a court may, where an application cannot properly be decided on affidavit, refer the matter to trial with appropriate directions as to pleadings or definition of issues, or otherwise. In this case, as would be the case in most applications that are referred to trial, it was essential that the issues be defined.¹ Had it been done, a lot of time, energy and costs could have been saved. The order referring the matter to trial was not made by the trial judge. However, when the matter was allocated to him, he should have insisted on a proper definition of the issues.

 

[20] The evidence adduced at the trial established that nine out of ten database structures used in the Brewer’s Amps application were identical with database structures used in the Data Explorer application; that the answers and weightings databases used in the Data Explorer application were supplied to clients who used the Brewer’s AMPS program; that the tree.txt files in respect of the Teen98, Child99, AMPS99a, AMPS99b and AMPS2000a data used in the Brewer’s AMPS application corresponded 99%, 100%, 57%, 96% and 96% respectively with the corresponding tree.txt files used in the Data Explorer application; that the source code of the tree preparer and converter programs used by Coetzee for the Brewer’s AMPS program corresponded 93 per cent and 84 per cent respectively with the tree preparer and converter programs used by him in respect of the Data Explorer program; and that the source code in respect of the search and graphing functions in the Brewer’s AMPS program corresponded 83 per cent and 26 per cent respectively with the source codes in relation to these functions in the Data Explorer program.

 

[21] The High Court did not distinguish between the computer programs and the databases. It held that Brewer’s Almanac was the author of the Project AMPS program in that the company in general had overriding control of the creation of the program. In regard to the developments of the program post 31 July 1998 the High Court said that it was apparent that they were improvements and refinements of the existing Project AMPS program. It held that by expending time and effort on the improvement of the program, Haupt did not, in some way, become the holder of the copyright. According to the High Court it was not Haupt’s case that one (or more) of the developments that occurred after 31 July 1998, of which he was the author, satisfied the originality requirement of the Act and attracted copyright in its own right. In respect of the converter program the High Court held that, at best for Haupt, he could claim to be the co-author with Coetzee in respect of the changes that were made after July 1998. For these reasons the High Court dismissed Haupt’s application.

 

[22] Haupt contends that the High Court erred and that he was entitled to an order interdicting the respondents from infringing his copyright in –

1. The Data Explorer program including the search instructions, the graph instructions, the converter program and the tree preparer program;

2. Various structures of databases such as the questions.db, the answers.db and the weightings.db; and

3. The questions, answers, weightings and tree.txt compilations in respect of the Teen98, Child99, Amps99a, Amps99b and Amps2000a data.

 

[23] Haupt submitted that the High Court erred in finding that the Project AMPS program, as it existed on 31 July 1998, constituted a computer program eligible for copyright in terms of the Act. He submitted that this was the case because it produced incorrect results and because ‘sections of the raw data’ could not be read at all. I do not agree. A computer program is defined in s 1 of the Act as ‘a set of instructions fixed or stored in any manner and which, when used directly or indirectly in a computer, directs its operation to bring about a result’. It does not require the result to be correct. As at 31 July 1998 the Project AMPS program did bring about a result, sometimes correct, sometimes wrong. It did, therefore, qualify as a computer program eligible for copyright.

 

[24] The High Court was apparently of the view that Haupt could not acquire copyright in the Data Explorer program inasmuch as the program was an improvement and refinement of the Project AMPS program. It erred in this regard. If a work is eligible for copyright an improvement or refinement of that work would similarly be eligible for copyright, even if the improved work involved an infringement of copyright in the original work, if it satisfies the requirement of originality.² That will only be the case if the improvement or refinement is not superficial. The alteration to the original work must be substantial.³ See in this regard Interlego A G v Tyco Industries Inc [1989] AC 217 (PC) at 263 where the Privy Council said in respect of an alteration to an artistic work:

‘There must in addition be some element of material alteration or embellishment which suffices to make the totality of the work an original work. Of course, even a relatively small alteration or addition quantitatively may, if material, suffice to convert that which is substantially copied from an earlier work into an original work. Whether it does so or not is a question of degree having regard to the quality rather than the quantity of the addition. But copying, per se, however much skill or labour may be devoted to the process, cannot make an original work.’

 

[25] The High Court also erred in holding that it was not Haupt’s case that one or more of the developments after 31 July 1998 satisfied the originality requirement of the Act and attracted copyright in its own right. Haupt specifically claimed copyright in respect of the various databases. He may have approached the case on the basis that the converter and tree preparer program were part and parcel of the Data Explorer program but the question as to whether he was entitled to copyright in respect of these programs was canvassed to no lesser extent than in respect of the other components of the relevant software.

 

[26] In terms of s 2(1) of the Act literary works and computer programs are works that are eligible for copyright if they are original. Other works that are eligible for copyright are not presently relevant.

 

[27] Subject to certain qualifications copyright is, in terms of s 3 of the Act, conferred on the author of a work eligible for copyright. In s 1 of the Act ‘author’ is differently defined in respect of literary works, computer programs, and literary works or computer programs which are computer-generated. It follows that, in order to determine in whom the copyright in respect of the various works vested, it is necessary to first determine whether the works we are concerned with are literary works, computer programs, computer-generated literary works or computer-generated computer programs and to then determine whether they are original and who the author of the relevant work is.

 

The nature of the works

[28] As stated above a ‘computer program’ means a set of instructions fixed or stored in any manner and which, when used directly or indirectly in a computer, directs its operation to bring about a result (s 1 of the Act). It is common cause that the Data Explorer program (including the search instructions and the graph instructions) and also the converter program and the tree preparer program are computer programs. All of them consist of a code containing a set of instructions which, when used in a computer, directs its operation to bring about a result.

 

[29] A literary work includes, irrespective of literary quality and in whatever mode or form expressed, tables and compilations, including tables and compilations of data stored or embodied in a computer or a medium used in conjunction with a computer, but not a computer program (s 1 of the Act).

 

[30] In respect of the database structures Haupt contended that they were computer programs while the respondents contended that they were literary works. According to the evidence adduced at the trial a database structure does not consist of a set of instructions. It consists of a table with a certain number of columns which may vary in width. The database structures are therefore literary works for purposes of the Act.

 

[31] The Act does not contain a definition of ‘computer-generated’. In my view a work only qualifies as having been computer-generated if it was created by a computer in circumstances where there is no human author of the work. If there is a human author the work is computer assisted and not computer-generated. That is the meaning ascribed to ‘computer-generated’ in s 178 of the Copyright, Designs and Patents Act, 1988 of the United Kingdom. It also accords with the recommendation of the Australian Copyright Law Review Committee in respect of computer software protection.⁴ That is also the meaning ascribed to ‘computer-generated’ in Payen Components SA Ltd v Bovic CC and Others 1995 (4) SA 441 (A) at 450D-G although the court was not dealing with the meaning of ‘computer-generated’ in the context of the Act.

 

[32] The structures for the various databases were created by Coetzee. Although he made use of the computer program Delphi to do so they were not generated by a computer: the computer merely assisted him in creating them. It is he who decided on the number of columns, their width and the field names. The creation of such a table is no different from the creation of a document by utilising a word processor. In such a case the computer is used as a tool.⁵ The bare databases so created, therefore, did not qualify as computer-generated literary works.

 

[33] In the case of the answers, weightings and tree.txt databases it does not make sense to separate the structures of the databases from the compilations.⁶ The databases were created by Coetzee as a composite unit and their structures were significant parts of such units. They determined how the data was to be ordered in order to be utilized by the Data Explorer program. In the circumstances the answers, weightings and tree.txt databases were created by Coetzee with the assistance of the converter and tree preparer programs i.e. they were not computer-generated.

 

[34] The position is different in the case of the questions database. The structure of the database was created by Coetzee and the compilation of the questions within that structure was done by Haupt. They were literary works by definition and were not computer-generated.

 

Originality

[35] There is no definition of ‘original’ in the Act. That the work must originate from the author and not be copied from an existing source is clear but that is not to say that every work which is not copied would qualify for protection in terms of the Act. In this regard the High Court would seem to have accepted and the respondents submitted that a ‘minimal degree of creativity’ was required to satisfy the originality requirement. They relied on the judgment of the Supreme Court of the USA in Feist Publications Inc v Rural Telephone Service Co Inc 449 US 340 (1991) at 345 and 348. However, the originality requirement in the Act was also a requirement in the Copyright Act 63 of 1965, which was repealed by it, and there is no reason to believe that it was intended to have a meaning different from the meaning it had in the repealed Act. The repealed Act was based on the Copyright Act of 1956 of the United Kingdom which had a similar originality requirement. For this reason ‘original’ in the repealed Act was probably intended to have the meaning it had been held to have in the United Kingdom. There, creativity is not required to make a work original. Save where specifically provided otherwise, a work is considered to be original if it has not been copied from an existing source and if its production required a substantial (or not trivial) degree of skill, judgment or labour.⁷ In Canada ‘original’ has likewise been interpreted so as not to require creativity. In CCH Canadian Ltd v Law Society of Upper Canada⁸ it was held:

‘[A]n original work must be the product of an author’s exercise of skill and judgment. The exercise of skill and judgment required to produce the work must not be so trivial that it could be characterized as a purely mechanical exercise. While creative works will by definition be “original” and covered by copyright, creativity is not required to make a work “original”.’⁹

 

[36] There can be no question that the converter program and the tree preparer program were the products of substantial skill, judgment and labour. It took Coetzee about six months to get the converter program right while the writing of the source code of the tree preparer program would have taken him about six hours. The fact that Bento could not himself create a similar tree preparer program but had to rely on Coetzee’s assistance is in itself an indication that the writing of the program required substantial skill, judgment and labour. There can also be no doubt that the writing of the source code for the search and the graph instructions required substantial skill, judgment and labour. They were made available to the first respondent because Bento was struggling to write the code. That the incorporation of these functions into the Data Explorer program brought about a significant improvement is evidenced by the fact that both Haupt and the respondents considered it necessary that these functions be incorporated into programs such as the Data Explorer and Brewer’s AMPS programs.

 

[37] It is, in the light of what follows, not necessary to decide whether the databases, database structure and questions compilations were ‘original’. I shall, therefore, merely assume that they were original.

 

Author

[38] As stated above ‘author’ is differently defined depending on the work that is eligible for copyright. In the case of a literary work ‘author’ means the person who first makes or creates the work; in the case of a computer program ‘author’ means the person who exercises control over the making of the computer program; and in the case of any of these works having been computer-generated ‘author’ means the person by whom the arrangements necessary for the creation of the work were undertaken.

 

[39] The databases and database structures are literary works which were not made or created by Haupt. Haupt, therefore, does not own the copyright in these works. He was entitled to use them because of a tacit licence from Coetzee. The questions compilations on the other hand were created by Haupt and on the assumption that they were original he acquired the copyright in them.

 

[40] The High Court held that Brewer’s Almanac exercised control over the making of the Project AMPS program but did not consider the question of control in respect of the developments to that program after 31 July 1998. Had it done so its reasoning would have led it to conclude that Haupt controlled the making of the program after 31 July 1998. That is so because, on the evidence, the relationship between Haupt and Coetzee was the same as the relationship between Brewer’s Almanac and Coetzee in respect of the making of the program.

 

[41] In s 21 of the Act specific provision is made in respect of works made in the course of an author’s employment by another person. A distinction is thus drawn in the Act between ‘control’, in the definition of ‘author’ in respect of a computer program, and ‘control’ by virtue of a contract of employment. ‘Control’, in the definition, must therefore have been intended to have a wider meaning than ‘control’ in the employment situation, that is, a person may because of his control over the making of a computer program be the author of that program even if the creator of the program is an independent contractor.¹⁰ According to the Shorter Oxford Dictionary ‘to control’ means ‘to exercise restraint or direction upon the free action of’ and ‘control’ means ‘the fact of controlling, or of checking and directing action’. In this case Haupt instructed Coetzee as to the end result that was to be achieved, Coetzee then did the technical work required to achieve that end result and from time to time effected improvements. However, Coetzee was all along in constant contact with Haupt and he accepted and executed detailed instructions from Haupt. As he progressed he submitted his work to Haupt for it to be checked and approved by him. In the properties section of the Data Explorer program Coetzee indicated that the copyright was owned by Softcopy, the name under which Haupt was trading. The allegation by Haupt in his founding affidavit that it was always agreed between the parties that Haupt was the owner of the program was not disputed. This being the understanding between the parties, Haupt could at any time direct in which direction the development of the program should proceed or could terminate further development if he wished to do so. Haupt was, therefore, in a position of authority over Coetzee insofar as the development of the program was concerned. He was in command and Coetzee subjected himself to such command. It is true that Haupt was in no position to instruct Coetzee as to how, technically, to achieve his requirements but I agree with the High Court that one does not need to be a computer programmer to be able to control the writing of a computer program. For these reasons I am of the view that Haupt controlled the writing of the computer programs written after 31 July 1998.

 

[42] It follows that the copyright in the Data Explorer program, which included the search instructions and the graph instructions and also the copyright in the converter and the tree preparer program, vested in Haupt.

 

Infringement

[43] The respondents contended that, in the event of it being held that the copyright vested in Haupt, they did not infringe such copyright. Section 23(1) provides as follows:

‘Copyright shall be infringed by any person, not being the owner of the copyright, who, without the licence of such owner, does or causes any other person to do, in the Republic, any act which the owner has the exclusive right to do or to authorize.’

 

[44] Section 11B provides that copyright in a computer program vests the exclusive right to reproduce or to authorize the reproduction of the computer program and to do certain other acts not presently relevant in respect of the computer program in the Republic. It is, however, not only the reproduction of the entire work that would constitute an infringement. In terms of s 1(2A) the reproduction of a work is, unless the context otherwise indicates, to be construed as a reference to the reproduction of a substantial part of the work.

 

[45] The first respondent reproduced part of the Data Explorer program in its Brewer’s AMPS program and Brewer caused it to do so. The respondents submitted that such reproduction was not substantial. However, where a part of a work is reproduced the question whether a substantial part had been reproduced depends much more on the quality than the quantity of what had been taken.¹¹ It is true that in this case only 26 per cent of the graphic component and 83 per cent of the search component, consisting of 63 lines of several thousand lines of source code, had been copied but then those lines were copied because Bento found it too difficult to write them himself. These components were clearly considered to be a valuable ingredient of the program by both Haupt and the respondents. For these reasons, the copying referred to, in my view, constituted the reproduction of a substantial part of the Data Explorer program. It follows that the first and the second respondents infringed Haupt’s copyright in the Data Explorer program.

 

[46] The respondents agreed that the Brewer’s AMPS program would be marketed in South Africa and that Coetzee would receive a royalty of 15 per cent of the full sales price of each sale. Coetzee assisted the first and second respondents in infringing the appellant’s copyright in the Data Explorer program by providing the source code in respect of the search function and also the source code required to incorporate the graphics server. In the premises Coetzee made common cause with the first and second respondents and co-operated with them in so far as the infringement by them of the appellant’s copyright in the Data Explorer program in South Africa is concerned. By doing so Coetzee himself infringed the appellant’s copyright in the Data Explorer program in South Africa.¹²

 

[47] The tree preparer program and the converter program do not form part of the Brewers AMPS program. They were in possession of Coetzee and were used by him to create tree.txt, answers and weightings databases in respect of the AMPS2000b data. In doing so neither he nor any of the other respondents did, or caused any other person to do, any act as contemplated in s 11B which Haupt as owner of the copyright had the exclusive right to do. Haupt was unable to advance any basis for his contention that his copyright in respect of these programs had been infringed.

 

[48] Haupt’s case is that his copyright in the questions compilations done by him was infringed in that each one of them ‘was reproduced in first respondent’s program in the exact same manner and form and thereafter sold by first respondent to its clients’. The case is not borne out by the evidence. As I understand the evidence the compilations were no longer required after the tree.txt files had been created.

 

[49] The order in terms of which the matter was referred to trial was made by consent. All the parties are, therefore, to blame for the fact that the trial lasted 33 court days as a result of various irrelevant issues having been canvassed. In the circumstances I consider it just and equitable that Haupt, who should have succeeded in the High Court, should have been deprived of 50 per cent of his costs.

 

[50] Haupt also appealed against the costs order made by the High Court in respect of two interlocutory applications, one on 11 September 2002 and the other on 11 November 2002, relating to discovery. In respect of the first of these applications the High Court ordered that the costs would be costs in the cause. In respect of the latter application the High Court considered it reasonable that the parties bear their own costs. Haupt could not point to any irregularity or misdirection and could not submit that a court, acting reasonably, could not have made these costs orders. In the circumstances this court cannot interfere with them.¹³

 

[51] In the result the following order is made:

 

1. The appeal is upheld with costs, including the costs of two counsel.

2. The order by the High Court is set aside and replaced with the following order:

‘1 The respondents are interdicted from infringing the applicant’s copyright in its computer program known as “Data Explorer”.

1. 2. The respondents are ordered to deliver up within 7 days all infringing copies of the work to the applicant.

   3. The respondents are ordered, jointly and severally, to pay 50 per cent of the costs of suit, including :

3. 1. The qualifying fees of Mr Marius Bosman;

   2. The costs of the Anton Piller application, including the costs of the supervising attorney and experts used therein.

   3. The costs of the application on 11 September 2002.

 

 

__________________

P E STREICHER

JUDGE OF APPEAL

 

 

 

CONCUR:

HARMS JA)

MTHIYANE JA)

CLOETE JA)

LEWIS JA)