SAS Institute Inc v World Programming Ltd

It is a cliché of copyright law that copyright does not protect ideas: it protects the expression of ideas. But the utility of the cliché depends on how ideas are defined. This dichotomy has made its way into international treaties and European legislation. The international treaties include the Agreement on Trade-Related Aspects of Intellectual Property Rights (“TRIPS”), article 9 (2) of which provides:

“Copyright protection shall extend to expressions and not to ideas, procedures, methods of operation or mathematical concepts as such.”

Article 2 of the World Intellectual Property Organisation Copyright Treaty (“WIPO”) is to the same effect. Both these treaties are part of the international legal order of the European Union.

Council Directive 91/250/EEC (“the Software Directive”) contains the following recitals (numbered as in the judge’s first judgment):

“[13] Whereas, for the avoidance of doubt, it has to be made clear that only the expression of a computer program is protected and that ideas and principles which underlie any element of a program, including those which underlie its interfaces, are not protected by copyright under this Directive;

[14] Whereas, in accordance with this principle of copyright, to the extent that logic, algorithms and programming languages comprise ideas and principles, those ideas and principles are not protected under this Directive;

[15] Whereas, in accordance with the legislation and jurisprudence of the Member States and the international copyright conventions, the expression of those ideas and principles is to be protected by copyright;”

This dichotomy is carried forward into article 1(2):

“Protection in accordance with this Directive shall apply to the expression in any form of a computer program. Ideas and principles which underlie any element of a computer program, including those which underlie its interfaces, are not protected by copyright under this Directive.”

Article 1 (3) provides that:

“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.”

The Software Directive has since been replaced by Council Directive 2009/24/EC but without relevant alterations of substance. I will continue to refer to the original Software Directive.

Running alongside the Software Directive is Council Directive 2001/29/EC (“The Information Society Directive”). Its recitals proclaim that:

“(20)

This Directive is based on principles and rules already laid down in the Directives currently in force in this area, in particular Directives 91/250/EEC [i.e. the Software Directive], 92/100/EEC, 93/83/EEC, 93/98/EEC and 96/9/EC, and it develops those principles and rules and places them in the context of the information society. The provisions of this Directive should be without prejudice to the provisions of those Directives, unless otherwise provided in this Directive.”

Thus the principles underlying the Software Directive also underlie the Information Society Directive. This is also confirmed by recital (50) which says:

“Such a harmonised legal protection does not affect the specific provisions on protection provided for by Directive 91/250/EEC. In particular, it should not apply to the protection of technological measures used in connection with computer programs, which is exclusively addressed in that Directive. It should neither inhibit nor prevent the development or use of any means of circumventing a technological measure that is necessary to enable acts to be undertaken in accordance with the terms of Article 5(3) or Article 6 of Directive 91/250/EEC. Articles 5 and 6 of that Directive exclusively determine exceptions to the exclusive rights applicable to computer programs.”

Article 2 provides that:

“Member States shall provide for the exclusive right to authorise or prohibit direct or indirect, temporary or permanent reproduction by any means and in any form, in whole or in part:

(a)

for authors, of their works …”

As Mr Howe QC pointed out, unlike the Software Directive the Information Society Directive does not expressly deal with the scope of copyright protection (i.e. what is capable of being protected by copyright). Nevertheless the ECJ (and latterly the CJEU) has supplied the omission. What is protected is the expression of an author’s “intellectual creation”. The phrase “intellectual creation” does not appear in the Information Society Directive. The point emerges most clearly from the decision of the court in Case C-5/08 Infopaq International A/S v Danske Dagblades Forening [2009] ECR I-6569. At [33] to [37] the court said:

“33 Article 2(a) of Directive 2001/29 provides that authors have the exclusive right to authorise or prohibit reproduction, in whole or in part, of their works. It follows that protection of the author’s right to authorise or prohibit reproduction is intended to cover ‘work’.

34 It is, moreover, apparent from the general scheme of the Berne Convention, in particular Article 2(5) and (8), that the protection of certain subject-matters as artistic or literary works presupposes that they are intellectual creations.

35 Similarly, under Articles 1(3) of Directive 91/250, 3(1) of Directive 96/9 and 6 of Directive 2006/116, works such as computer programs, databases or photographs are protected by copyright only if they are original in the sense that they are their author’s own intellectual creation.

36 In establishing a harmonised legal framework for copyright, Directive 2001/29 is based on the same principle, as evidenced by recitals 4, 9 to 11 and 20 in the preamble thereto.

37 In those circumstances, copyright within the meaning of Article 2(a) of Directive 2001/29 is liable to apply only in relation to a subject-matter which is original in the sense that it is its author’s own intellectual creation.”

Thus the court is taking an expression used in one Directive and applying it to another. In fact the origin of the expression is in the international legal order represented by the Berne Convention. In that way the court interprets Directives dealing with intellectual property consistently. It does so in order to establish “a harmonised legal framework for copyright”.

The court has also considered what amounts to an “intellectual creation” on a number of occasions. The essence of the term is that the person in question has exercised expressive and creative choices in producing the work. The more restricted the choices, the less likely it is that the product will be the intellectual creation (or the expression of the intellectual creation) of the person who produced it. In Case C-393/09 Bezpečnostní softwarová asociace – Svaz softwarové ochrany v Ministerstvo kultury [2011] FSR 18 (“BSA”) what was in issue was a graphic user interface which enabled a computer user to communicate with a computer program. The question referred was whether it was protected by the Software Directive. The answer was “no”. The reason was that it was not “a form of expression of a computer program”: see [42]. However, both Advocate-General Bot and the court itself also considered whether it could be protected by the Information Society Directive. The Advocate-General began at [73] and [74] by recognising that the development of an interface required considerable intellectual effort on the part of its developer. That intellectual effort included using a programming language to create a complex structure. But that did not necessarily mean that the product of that intellectual effort qualified for copyright protection. He continued at [75] and [76]:

“75 The difficulty as regards determination of the originality of the graphic user interface lies in the fact that the majority of the elements which comprise it have a functional purpose, since they are intended to facilitate the use of the computer program. Accordingly, the manner in which those elements are expressed can be only limited since, as the Commission stated in its written submissions, the expression is dictated by the technical function which those elements fulfil. Such is the case, for example, of the mouse which moves the cursor across the screen, pointing at the command button in order to make it operate or of the drop-down menu which appears when a text file is open.

76 In such cases, it seems to me that the criterion of originality is not met, since the different methods of implementing an idea are so limited that the idea and the expression become indissociable. If such a possibility was offered, it would have the consequence of conferring a monopoly on certain companies on the computer program market, thus significantly hampering creation and innovation on that market, which would run contrary to the objective of Directive 2001/29.”

The court specifically approved this reasoning. It said at [48] to [50]:

“48.

When making that assessment [i.e. the assessment whether the GUI is the author’s own intellectual creation], the national court must take account, inter alia, of the specific arrangement or configuration of all the components which form part of the graphic user interface in order to determine which meet the criterion of originality. In that regard, that criterion cannot be met by components of the graphic user interface which are differentiated only by their technical function.

49 As the Advocate General states in points 75 and 76 of his Opinion, where the expression of those components is dictated by their technical function, the criterion of originality is not met, since the different methods of implementing an idea are so limited that the idea and the expression become indissociable.

50 In such a situation, the components of a graphic user interface do not permit the author to express his creativity in an original manner and achieve a result which is an intellectual creation of that author.”

What seems to me to be clear from this passage is (a) that if expression is dictated by technical function then the criterion of originality is not satisfied; and (b) that, where that is the case, the product is not an intellectual creation of the author at all. It is of importance to note that this emphasis on questions of function applies to the Information Society Directive and not just to the Software Directive. What is equally important is that the court approved the Advocate-General’s description of the policy of the Information Society Directive which is, like that of the Software Directive, to avoid conferring a monopoly on certain companies in the computer program market, thus significantly hampering creation and innovation in that market. One of Mr Hobbs’ major points was that what counted in the context of the Information Society Directive was “features” rather than “functions”. But in my judgment that argument is inconsistent with what both the Advocate-General and the court said in BSA.

In Joined Cases C-403/08 and C-429/08 Football Association Premier League Ltd v QC Leisure [2012] All ER (EC) 629 the court held that a football match was not entitled to copyright protection under the Information Society Directive. One reason for that, as the court explained at [98], was that “football matches … are subject to rules of the game, leaving no room for creative freedom for the purposes of copyright.” In Case C-145/10 Painer v Standard Verlags GmbH [2012] ECDR 6, again in considering the Information Society Directive, the CJEU said at [88] that an intellectual creation is the author’s own “if it reflects the author’s personality” and went on to say at [89]:

“That is the case if the author was able to express his creative abilities in the production of the work by making free and creative choices.”

The CJEU made a similar point in relation to database right in Case C-604/10 Football Dataco Ltd v Yahoo! UK Ltd [2013] FSR 1 at [37] to [39]:

“37 Secondly, as is apparent from recital 16 of Directive 96/9 , the notion of the author’s own intellectual creation refers to the criterion of originality (see, to that effect, Infopaq International A/S v Danske Dagblades Forening (C-5/08) [2009] E.C.R. I-6569; [2010] F.S.R. 20 at [35], [37] and [38]; Bezpečnostní softwarová asociace - Svaz softwarové ochrany v Ministerstvo Kultury (C-393/09) [2010] E.C.R. I-13971; [2011] F.S.R. 18 at [45]; Football Association Premier League Ltd v QC Leisure (C-403/08 & C-429/08) [2012] F.S.R. 1 at [97]; and Painer v Standard Verlags GmbH (C-145/10) [2012] E.C.D.R. 6 at [87]).

38 As regards the setting up of a database, that criterion of originality is satisfied when, through the selection or arrangement of the data which it contains, its author expresses his creative ability in an original manner by making free and creative choices (see, by analogy, Infopaq International [2010] F.S.R. 20 at [45]; Bezpečnostní softwarová asociace [2011] F.S.R. 18 at [50]; and Painer [2012] E.C.D.R. 6 at [89]) and thus stamps his “personal touch” (Painer [2012] E.C.D.R. 6 at [92]).

39 By contrast, that criterion is not satisfied when the setting up of the database is dictated by technical considerations, rules or constraints which leave no room for creative freedom (see, by analogy, Bezpečnostní softwarová asociace [2011] F.S.R. 18 at [48] and [49], and Football Association [2012] F.S.R. 1 at [98]).”

This test may not be quite the same as the traditional test in English law. As Advocate-General Mengozzi explained in Football Dataco Ltd v Yahoo! UK Ltd:

“36 It is common knowledge that, within the European Union, various standards apply as regards the level of originality generally required for copyright protection to be granted. In particular, in some EU countries which have common law traditions, the decisive criterion is traditionally the application of “labour, skills or effort”. For that reason, in the United Kingdom for example, databases were generally protected by copyright before the entry into force of the Directive. A database was protected by copyright if its creator had had to expend a certain effort, or employ a certain skill, in order to create it. On the other hand, in countries of the continental tradition, for a work to be protected by copyright it must generally possess a creative element, or in some way express its creator’s personality, even though any assessment as to the quality or the “artistic” nature of the work is always excluded.

37 Now, on this point there is no doubt that, as regards copyright protection, the Directive espouses a concept of originality which requires more than the mere “mechanical” effort needed to collect the data and enter them in the database. To be protected by the copyright, a database must—as art.3 of the Directive explicitly states—be the “intellectual creation” of the person who has set it up. That expression leaves no room for doubt, and echoes a formula which is typical of the continental copyright tradition.”

If the Information Society Directive has changed the traditional domestic test, it seems to me that it has raised rather than lowered the hurdle to obtaining copyright protection.

Our domestic legislation confines the doing of a restricted act (e.g. copying) to doing that act in relation to the work as a whole or any “substantial part of it”: Copyright Designs and Patents Act 1988 s. 16 (1), s. 16 (3), Nova Productions Ltd v Mazooma Games Ltd [2007] EWCA Civ 219 [2007] RPC 25 at [29]. It has long been the position in domestic law that what is substantial is a question to be answered qualitatively rather than quantitatively. In Infopaq the court said that parts of a work are entitled to the same protection as the work as a whole. But the parts in question must “contain elements which are the expression of the intellectual creation of the author of the work”: [39]. This is now the test for determining whether a restricted act has been done in relation to a substantial part of a work. Both counsel agreed that to interpret section 16 (3) in this way was consistent with the court’s duty to interpret domestic legislation, so far as possible, so as to conform with European Directives. I do not think that anything in the decisions of this court in Nova Productions Ltd v Mazooma Games Ltd and The Newspaper Licensing Agency v Meltwater Holding BV [2011] EWCA Civ 890 [2012] RPC 1 casts doubt on that proposition.

In the course of the reference in our case both the Advocate-General and the court discussed the distinction between ideas and the expression of ideas. The Advocate-General began his discussion at [42] to [44] concluding at that point that the originality “of a work” lies not in an idea, but in the expression of an idea. At this point the Advocate-General was dealing with works generally, not limited to computer programs. However, at [47] to [50] he recognised that elements of creativity, skill and inventiveness manifest themselves in the way in which a program is put together; and that copyright protection for a program is conceivable from the point at which the selection and compilation of its elements are indicative of the creativity and skill of the author. He concluded that the protection of a computer program was not confined to the source code and object code but extended to any other element expressing the creativity of its author.

The Advocate-General then turned to consider what counts as an idea, rather than the expression of an idea; in particular the functionality of a computer program. He defined that expression at [52] as follows:

“The functionality of a computer program can be defined as the set of possibilities offered by a computer system, the actions specific to that program. In other words, the functionality of a computer program is the service which the user expects from it.”

He then gave an example taken from the facts in Navitaire Inc v easyJet Airline Co Ltd [2004] EWHC 1725 (Ch) [2006] RPC 3. In short he said that the functionalities of a computer program are dictated by a specific and limited purpose:

“In this, therefore, they are similar to an idea. It is therefore legitimate for computer programs to exist which offer the same functionalities.”

But he added at [55] that:

“There are, however, many means of achieving the concrete expression of those functionalities and it is those means which will be eligible for copyright protection under [the Software Directive]. As we have seen, creativity, skill and inventiveness manifest themselves in the way in which the program is drawn up, in its writing. The programmer uses formulae, algorithms which, as such, are excluded from copyright protection because they are the equivalent of the words by which the poet or the novelist creates his work of literature. However, the way in which all of these elements are arranged, like the style in which the computer program is written, will be likely to reflect the author's own intellectual creation and therefore be eligible for protection.”

He pointed out at [57] that to allow the functionality of a computer program to be protected as such by copyright would amount to making it possible to monopolise ideas. At [60] he referred to Infopaq and the Information Society Directive and said that given that a computer program must be regarded as a literary work, the same approach should apply.

After further discussion he said at [63]:

“Whatever its nature and scope may be, it is my view that the functionality, or indeed the combination of several functionalities, continues to be comparable to an idea and cannot therefore be protected, as such, by copyright.”

The important point here is that even a combination of several functionalities (which entails choice by the programmer or developer) falls on the ideas side of the line. Nor does the Advocate-General restrict himself to protection under the Software Directive: he speaks of copyright generally. He added in the following paragraph:

“64 Similarly, it is my opinion that the foregoing analysis cannot be called in question by the nature and extent of the skill, judgment and labour expended in devising the functionality of a computer program.”

Thus the nature of the skill and judgment expended in devising the functionality of a computer program (which will inevitably involve making choices) still falls on the ideas side of the line.

I do not consider that the court disagreed with that analysis. Thus at [39] the court said that:

“On the basis of those considerations, it must be stated that, with regard to the elements of a computer program which are the subject of Questions 1 to 5, neither the functionality of a computer program nor the programming language and the format of data files used in a computer program in order to exploit certain of its functions constitute a form of expression of that program for the purposes of Art. 1(2) of [the Software Directive].”

The court did not itself define what it meant by “functionality”; and in those circumstances there is no reason to suppose that it meant anything different from the way in which the Advocate-General defined the word. Moreover, at [40] it specifically approved what the Advocate-General had said at [57] of his opinion:

“… to accept that the functionality of a computer program can be protected by copyright would amount to making it possible to monopolises ideas, to the detriment of technological progress and industrial development.”

As Mr Howe QC pointed out, in that paragraph of its judgment the court expressed the policy in very general terms (not limited to protection under the Software Directive). Since it is exactly the same policy that, in BSA at [49], the court held applied to the Information Society Directive, the expression of the policy in general terms is readily understandable.

Mr Hobbs QC drew attention to the court’s statement at [45]:

“The Court also points out that the finding made in para. 39 of the present judgment cannot affect the possibility that the SAS language and the format of SAS Institute's data files might be protected, as works, by copyright under [the Information Society Directive] if they are their author's own intellectual creation.”

It will be recalled that in [39] the court had discussed (a) the functionality of a computer program, (b) the programming language and (c) the format of data files used in a computer program. By contrast in [45] it considered only (b) and (c). One must infer, therefore, that the court considered that the functionality of a computer program could not be protected under the Information Society Directive. So I do not think that this carries SAS Institute’s argument further.

It is also the case that the court’s discussion about protecting the functionality of a computer program under the Software Directive was at root a decision about where to draw the line between ideas and expression of ideas. Since the same division between ideas and expression of ideas and the same underlying policy apply with equal force to the Information Society Directive, it would be entirely unsurprising for the line to be drawn in the same place.

The Advocate-General dealt with protection of the SAS Manuals under the Information Society Directive at [102] to [122] of his opinion. At [109] he made the same distinction (with a cross-reference) between ideas and the expression of ideas that he had made in relation to the Software Directive. Ideas, procedures, methods of operation and mathematical concepts as such are not protected by copyright because they fall on the ideas side of the line. At [110] he said:

“In this case, the referring court states that WPL has, in particular, taken the keywords, syntax, commands and combinations of commands, options, defaults and iterations from the SAS Manuals in order to reproduce them in its program, as well as in the WPL manual.”

It is thus clear that the Advocate-General (a) appreciated that WPL had taken combinations of commands and (b) had taken them from their descriptions in the SAS Manuals. Nevertheless he said at [111]:

“In my opinion, these elements, as such, do not qualify for the protection conferred by copyright.”

He then considered the selection of statistical operations described in the SAS Manuals; and pointed out that WPS offered “the same selection of statistical operations”. Nevertheless his pithy conclusion at [115] was that:

“The WPL System does not reproduce the description [in the SAS Manuals] of those statistical operations but simply executes them.”

He dealt with mathematical formulae at [116] and a specific statistical operation at [117]. He concluded at [118]:

“In my view, it follows from the foregoing considerations that those various components correspond to ideas, procedures, methods of operation or mathematical concepts. Consequently, they are not, as such, eligible for the copyright protection conferred by Art. 2 (a) of Directive 2001/29 [i.e. the Information Society Directive].”

The notion of “ideas, procedures, methods of operation and mathematical concepts” is not expressly found in the Information Society Directive. On the contrary, it is a read-across from the Software Directive. In other words, what counts as an idea, for the purposes of a computer program, also counts as an idea for the purposes of a manual. If something counts as an idea it is not “eligible” for protection i.e. not capable of protection. It is also abundantly clear that the Advocate-General regarded all the detailed elements specified in the question referred (which included the combination of commands and the selection of statistical operations implemented in the SAS System) as falling within the scope of ideas rather than expression. Both combinations and selections are choices of a kind, but plainly in the Advocate-General’s opinion they were choices of the wrong kind to be capable of protection by copyright.

The court itself dealt with this part of the case as follows:

“66.

In the present case, the keywords, syntax, commands and combinations of commands, options, defaults and iterations consist of words, figures or mathematical concepts which, considered in isolation, are not, as such, an intellectual creation of the author of the computer program.

67.

It is only through the choice, sequence and combination of those words, figures or mathematical concepts that the author may express his creativity in an original manner and achieve a result, namely the user manual for the computer program, which is an intellectual creation…

68.

It is for the national court to ascertain whether the reproduction of those elements constitutes the reproduction of the expression of the intellectual creation of the author of the user manual for the computer program at issue in the main proceedings.”

In my judgment a fair reading of the court’s judgment, and especially of [66], shows that the court agreed with the Advocate-General on this point. Paragraph [66] is not simply a repetition of paragraph [110] of the Advocate-General’s opinion. It includes all the features mentioned in the latter paragraph but adds more. The additions are those which the Advocate-General discusses at [115] to [117]. Moreover as WPL point out, in [66] the court is stating that these elements are not even the intellectual creation of the author of the program. If they are not the intellectual creation of the author of the program it is hard to see how they can be the intellectual creation of the author of the manual which describes the program. I note also that the question referred to the CJEU asked specifically whether the level of detail with which the functionality of the program had been reproduced made any difference. The Advocate-General was clear (at [62]) that it did not; and the court gives no hint of disagreement. In saying that these features are not the intellectual creation of the author of the program the court is echoing the Advocate-General’s view that to the extent that there are choices involved, they are choices of the wrong kind to be capable of protection by copyright. This, in my judgment, is the answer to Mr Hobbs’ submission that the developers of WPS fettered their own choices and took the choices made by the designers of the SAS System. These choices fall on the ideas side of the line, and are not protected by copyright.

Mr Hobbs emphasised the court’s statement that the functions (or features) mentioned in [66] were not an intellectual creation “in isolation … as such”. From that he drew the proposition that those features, in combination, would attract copyright protection. There are two points to make about that. First, in [66] the court was dealing with the intellectual creation of the author of the program; not the intellectual creation of the author of the manual. Second, what the court was discussing was the question of intellectual creation, rather than the expression of an intellectual creation. We know from BSA that what is critical is not the intellectual creation, but the expression of the intellectual creation.

Thus when we come to [68] the question that the court is leaving to the national court is not whether there has been a reproduction of the intellectual creation of the author of the manual, but whether there has been a reproduction of the expression of the intellectual creation of the author of the manual. That is a significantly narrower question.

The essence of this claim is that in writing WPS in the Java programming language (and subsequently in C++) WPL has copied the SAS Manuals. This argument seems at first sight to be counter-intuitive, because the SAS Manuals themselves do not contain any programming language. They describe the functions that the program is to perform.

Following the judgment of the CJEU the question remaining for the judge was whether the reproduction in WPS of elements described in the SAS Manuals constituted the reproduction of the expression of the intellectual creation of the author of the user manual. It is important to stress that it is only the intellectual creation of the author of the user manual that counts. The intellectual creation of the author of a different work (e.g. the computer program itself) is not relevant to this question. In addition, and perhaps more importantly, for the purposes of copyright what is relevant is not the intellectual creation itself, but the expression of the intellectual creation of the author of the manual: Infopaq at [39]; CJEU in our case at [65] and [68]. The functionality of a computer program is, quite simply, not a form of expression at all.

In his first judgment the judge dealt with the Manual to Program Claim as follows:

“[255] Assuming that the reasoning which led Pumfrey J in Navitaire to conclude that it is not an infringement of the copyright in a computer program to replicate its functions without copying its source code or design is correct, I consider that by parity of reasoning it is not an infringement of the copyright in a manual describing those functions to use the manual as a specification of the functions that are to be replicated and, to that extent, to reproduce the manual in the source code of the new program. Once again, it is a question of the kind of skill, judgement and labour involved. Copyright in a literary work, be it a computer program or a manual, does not protect skill, judgment and labour in creating ideas, procedures, methods of operation or mathematical formulae. It follows that it is not an infringement to reproduce such things either from a computer program or a manual….

[259] … counsel for SAS Institute submitted that WPL had reproduced the form of expression of the mathematical formulae set out in the SAS Manuals in the WPS source code. In principle, this argument seems to me to address the issue from the correct perspective. As I have said, however, the evidence does not establish that SAS Institute's employees expended any particular skill, judgement or labour in devising the form of expression of those formulae as distinct from other mathematically equivalent formulae. Furthermore, it is not possible to translate such formulae directly into source code and WPL has not done so. As I have said, there is some resemblance of form between the WPS source code and the formulae in the SAS Manuals, but it should not be overstated. Having regard to the quality as well as the quantity of what has been taken, and provided that it was permissible for WPL to reproduce the ideas etc. described in the SAS Manuals, I am not persuaded that WPL has reproduced a substantial part of the expression of the SAS Manuals.

[260] … counsel for SAS Institute did not dispute that a single mathematical formula was unprotectable by copyright. He submitted, however, that the SAS Manuals contained a compilation of statistical methods and that WPL had reproduced that compilation. Thus in the case of example 14, he argued there were at least 10 possible methods of calculating percentiles in the statistics literature, SAS had exercised skill, judgement and labour in selecting five of those 10 while omitting the remainder and WPL had reproduced that selection in WPS. In my judgment, however, this argument involves an intellectual sleight of hand. If there is a compilation of statistical methods, the compilation is to be found in the SAS System. The authors of the SAS Manuals did not create that compilation, they simply reproduced it. Although the authors of the SAS Manuals undoubtedly exercised skill, judgement and labour in devising the form of expression of the Manuals, and in particular their wording, they did not exercise any skill, judgement or labour in making the compilation of statistical methods. It follows that, to use the language of Infopaq, what WPL reproduced from the SAS Manuals in this respect does not “express the intellectual creation” of the authors of the SAS Manuals.

[261] I would add that I do not accept that the SAS System does contain a compilation of statistical methods. It is clear from the evidence that the collection of statistical methods contained in the current version of the SAS System has grown steadily over a period of around 35 years. The collection has grown by accretion as a result of both customer feedback and competitive pressure as well as SAS Institute's own initiatives. While there may be individual procedures or functions in respect of which it could be said that an author or group of authors consciously planned and implemented a particular selection of methods, in general the collection was not planned and there was no overall design. It was therefore not the intellectual creation of an author or group of authors. In this respect it stands in the same position as the collection of commands in Navitaire: see the judgment of Pumfrey J at [92].”

In this passage the judge has combined the traditional test and the Infopaq test. But that combination is, if anything, more favourable to the authors of the SAS Manuals than a strict application of the Infopaq test.

In his third judgment the judge answered the question that the CJEU left to him in the following way:

“In so far as counsel for SAS Institute argued that WPL had reproduced compilations of (i) formulae, (ii) keywords, (iii) default values, (iv) comments and (v) optimisations from the SAS Manuals, I would repeat the answers I gave at [260]-[261], and in particular the first one. The authors of the SAS Manuals did not create such compilations, the authors of the SAS System did.”

SAS Institute first criticises the judge for having carried forward his reasoning on the scope of the Software Directive into his interpretation of article 2 (a) of the Information Society Directive. They say that the two directives are separate and self-contained pieces of legislation; and that the protection given by the latter is broader than the protection given by the former.

In the light of recitals (20) and (50) of the Information Society Directive I do not consider that the first step in this argument can be sustained. Nor, as I have shown, is it supported by the Advocate-General’s opinion in our case or the judgment of the CJEU both in our case and in others.

In addition when considering the concept of “intellectual creation” the CJEU has built on decisions on different directives, as can be seen most clearly from the quoted extract from Football Dataco Ltd v Yahoo! UK Ltd (see [35] above) in which the court reasons by analogy from one directive to another. Indeed in Infopaq itself the notion of the author’s own intellectual creation was carried across from the Berne Convention and the Software Directive into the Information Society Directive, even though the latter Directive does not use the phrase. The conclusion must be that the same concept of what is capable of protection (as a form of expression rather than as an idea) applies to both the Software Directive and the Information Society Directive. Since the court’s avowed intention was to establish a harmonised legal framework for copyright, this is not surprising.

It is also the case that the European legislation was enacted against the background of many international treaties and well-established principles of copyright law, and that the relevant directives must be interpreted against that background: Infopaq at [32]. One of these principles is that for an infringement of copyright to exist, the defendant’s work must represent the claimant’s work in some real sense. This is an embedded feature of copyright law. In Designers Guild Ltd v Russell Williams (Textiles) Ltd [2000] 1 WLR 2416 Lord Hoffmann explained at [25] that:

“… a copyright work may express certain ideas which are not protected because they have no connection with the literary, dramatic, musical or artistic nature of the work. It is on this ground that, for example, a literary work which describes a system or invention does not entitle the author to claim protection for his system or invention as such.”

To similar effect is the observation of Jacob LJ in Nova Productions Ltd v Mazooma Games Ltd at [51]:

“Similarly and more generally, a written work consisting of a specification of the functions of an intended computer program will attract protection as a literary work. But the functions themselves do not.”

Copinger & Skone James on Copyright (17^th ed) explain the principle thus at 7-37:

“It has already been pointed out that in general terms a work is not reproduced unless what has been produced represents the work in some real sense. A description in a novel of a scene from nature is thus not infringed by a drawing made to depict that scene. So, in the context of a literary work, the copyright in a book which described a method of teaching mathematics was not infringed by making a series of coloured rods which demonstrated that method, the copyright in written instructions for the making of a garment was not infringed by making the garment, and the copyright in the words and numerals in knitting guides was not infringed by making garments to those instructions. Again the copyright in a book of recipes would not be infringed by making a dish according to one of the recipes.”

I do not consider that either the Software Directive or the Information Society Directive undermines this basic principle; nor did the decision of the CJEU in this case.

It is, however, fair to say that the judge does not seem to have appreciated that the judgment of the CJEU had, to some extent, changed the question. What is protected is the form of expression of an intellectual creation. The intellectual creation itself is not protected; and the functionality of a computer program does not count as a form of expression. The functionality of a computer program (in the sense of what it does and how it responds to particular inputs) falls on the ideas side of the line. It falls on that side of the line whether one is considering the Software Directive or the Information Society Directive. Accordingly, in my judgment the judge was wrong at [260] to concentrate on who did what, and in his third judgment to rely on that paragraph of his first judgment as containing his conclusion. It would have been preferable simply to say that the copying that SAS Institute alleged was not the copying of the form of expression of an intellectual creation, which is close to how the judge put it in [255] of his first judgment. In short, what WPL took was not capable of protection by copyright. Thus although I do not agree with the way in which the judge formulated his conclusion in his third judgment, I consider that he came to the right answer.

I should also say that for that reason in my judgment it does not matter whether the creation of the computer program preceded the writing of the manual, proceeded alongside the writing of the manual, or followed the writing of the manual. In Nova Productions Ltd v Mazooma Games Ltd Jacob LJ was dealing with the example of an “intended” program; and Lord Hoffmann’s statement of principle was entirely neutral on the question of timing. The examples given in Copinger & Skone James exhibit a variety of different sequences of events. In the light of the judgment of the CJEU the common thread underlying these cases is that what is taken is an idea rather than the expression of an idea.

There is also the general policy of the directives to be considered. I have already quoted what the CJEU said in our case at [40] and what it said in BSA at [49]. The policy underlying both the Software Directive and the Information Society Directive is identical. It would be contrary to that policy if SAS Institute could achieve copyright protection for the functionality of its program indirectly via its manual which simply explains that functionality. Moreover, as the judge pointed out at [253] of his first judgment, the European Commission took the view in its amended proposal for the Software Directive that it was permissible for a third party to use manuals published by the creator of an original program as a source of information about interfaces in order to create an interoperable program and, to that extent, to reproduce the original program.

Mr Hobbs submitted that the judge’s conclusion was inconsistent with WPL’s admission that copyright subsisted in the SAS Manuals. I do not agree. No one doubts that copyright in the manuals subsists. The question is: what does that copyright protect?

In short, although we are not bound by it, I agree with the Advocate-General’s pithy conclusion that:

“The WPL System does not reproduce the description [in the SAS Manuals] of those statistical operations but simply executes them.”

In my judgment, therefore, the judge was right in his conclusion that the elements of the program that WPL obtained from the SAS Manuals in creating WPS were not the form of expression of the intellectual creation of the authors of the manuals.

At [261] of his first judgment the judge appears to have been saying that the SAS System was not the intellectual creation of its authors because it had grown by accretion without an overall design. He repeated this reason in his third judgment. In my judgment he was wrong to do so. First, the subsistence of copyright in the SAS System was admitted on the pleadings, and the judge’s reasoning seems to me to go behind that admission. Second, I can see no reason in principle why something that has grown by accretion should, for that reason alone, be deprived of copyright protection. But my disagreement with the judge on this point does not affect the outcome of the Manual to Program Claim.

I would therefore dismiss the appeal in relation to the Manual to Program Claim.

Subject to a procedural point the appeal in relation to the Manual to Manual Claim stands or falls with the Manual to Program Claim. This follows from the judgment of the CJEU at [69] in which the court said:

“In this respect, the examination, in the light of Directive 2001/29, of the reproduction of those elements of the user manual for a computer program must be the same with respect to the creation of the user manual for a second program as it is with respect to the creation of that second program.”

If, therefore, the Manual to Program Claim fails, so must the Manual to Manual Claim, in so far as it concerns the underlying elements of the program itself. Linguistic reproduction of the precise terms of the SAS Manuals is a different matter, as the judge rightly recognised. He held that in some limited respects WPL had infringed the copyright in the SAS Manuals; and WPL do not challenge that conclusion.

The judge’s conclusion is also supported by the reasoning of Pumfrey J in Navitaire Inc v easyJet Airline Co Ltd. He gave the following example at [127]:

“Take the example of a chef who invents a new pudding. After a lot of work he gets a satisfactory result, and thereafter his puddings are always made using his written recipe, undoubtedly a literary work. Along comes a competitor who likes the pudding and resolves to make it himself. Ultimately, after much culinary labour, he succeeds in emulating the earlier result, and he records his recipe. Is the later recipe an infringement of the earlier, as the end result, the plot and purpose of both (the pudding) is the same? I believe the answer is no.”

This part of his judgment was approved by this court in Nova Productions Ltd v Mazooma Games Ltd. I think that in his opinion in our case the Advocate-General also approved its reasoning, given that he based his examples on the facts of that case. In so far as the text of the WPS manual was copied from the SAS Manuals, which the judge found established to some extent, the claim for copyright infringement succeeded, and there is no appeal against that. But in so far as the WPS manual described the WPS program, which had been created from observation of the functionality of SAS and its description in the SAS Manuals, it is exactly analogous to the writing by the second chef of his successful recipe, which does not infringe copyright in the first chef’s recipe.

This attack on the judge’s third judgment therefore fails.

However, Mr Hobbs had another, different attack. He pointed to the judge’s conclusions at [317] and [319] as follows:

“The only issue here is whether WPL has reproduced a substantial part of the relevant SAS Manuals. WPL contends that it has gone no further than indirectly copying the content of the SAS Manuals via the WPS source code, and that this is not an infringement because all it has done is to copy the ideas, procedures, methods of operation and mathematical formulae described in the SAS Manuals rather than the expression of the SAS Manuals.”

“I have concluded that WPL went further than merely copying the ideas etc. described in the SAS Manuals. For the reasons given in paragraph 148 above, I consider that WPL substantially reproduced the language of the SAS Manuals even though its policy was not to do so. Applying the test laid down in Infopaq, namely whether the parts which have been reproduced express the intellectual creation of the authors of the SAS Manuals, I conclude they do.”

Mr Hobbs’ point is that, having regard to the case management order which stated that the pleaded similarities were to be treated as representative, once the judge had found that there had been infringement that was the end of the Manual to Manual claim. He should simply have entered judgment for SAS Institute on that claim.

I have already set out the terms of the case management order. What was before the judge was a selection of alleged similarities, which were to stand as representative samples of the manuals as a whole. I do not, however, consider that the terms of the order compelled the conclusion that if a single instance of infringement were to be established that meant that the whole of the WPS Manual should be treated as infringing. Within the selection the judge was free to decide that some instances of copying had been proved, and others had not. As I have said the underlying purpose of these orders was that whatever proportion of copying was found in the representative samples could be extrapolated across the manuals as a whole.

This attack on the judge’s partial dismissal of the Manual to Manual claim also fails. I should, perhaps, record that Mr Howe told us that in fact the WPS Manual that the judge considered had been withdrawn and a new one will be written from scratch.

The use of the Learning Edition was, in principle, governed by the terms of the contractual licence. There were four versions of the licence in issue. WPL bought two copies of version 1.0 in October 2003, one copy of version 2.0 in March 2005, two copies of version 4.1 (a) in July 2007 and seven copies of version 4.1 (b) in March 2009. The relevant terms were these:

“Important: Please carefully read the terms and conditions of this License Agreement ('Agreement') before clicking on the 'Yes' button. By clicking on the 'Yes' button, the individual licensing the Software ('Customer') agrees to these terms, and SAS Institute Inc ('SAS') will authorize Customer to use the SAS Learning Edition Software ('Software') in accordance with the terms and conditions of this Agreement. If Customer does not agree to all of the terms of this Agreement, click on the 'No' button and return the Software to the supplier from whom Customer obtained it. Customer will receive a refund of licence fees applicable to and paid by Customer for this Software, if any, provided Customer returns the full Software package with proof of purchase within thirty (30) days of date of purchase or the supplier's return period, whichever is the longer.

1.

Licence Grant

1.1

In exchange for Customer's payment of all applicable fees and compliance with all of the terms and conditions of this Agreement, SAS hereby grants Customer a[n individual] nonassignable, nontransferable and nonexclusive license to use the Software on one (1) workstation at a time, for Customer's [own self-training] non-production purposes only. Concurrent usage or use on a network is not authorized.

1.2

Customer may not use the Software in any outsourcing, facilities management or service bureau arrangement or any data or information technology management operation by or for third parties.

1.3

… Customer may terminate its license to use the Software by either returning all copies of the Software and associated documentation to SAS or by destroying copies….

2.

Intellectual Property Rights

The Software is copyrighted. ... Customer may not reverse assemble, reverse engineer, or decompile the Software or otherwise attempt to recreate the Source Code, except to the extent applicable laws specifically prohibit such restriction…. Customer may not distribute the Software to any third party in modified or unmodified form.

4.

Export Restrictions

… By accepting this Agreement, Customer affirms that it is located in and is a lawfully admitted permanent resident of a country to which the United States permits SAS to send the Controlled Material. Customer further affirms it is a party to whom the United States allows SAS to provide the Controlled Material…

…

9.

Complete Agreement

This Agreement and any invoices relating to the Software set forth the entire agreement between Customer and SAS related to the Software and supercede any purchase order, communications or representations regarding the Software…. ”

The words in italics were contained in versions 1.0 and 2.0 but not in versions 4.1 (a) or (b). The licences were governed by the laws of the state of North Carolina and the USA. However, the trial was conducted on the basis that there was no difference between those laws and English law. So was the appeal.

The judge found that WPL was in breach of the terms of the licence in two different ways. First, it had allowed use of the Learning Edition by employees other than those who had “clicked through” the licence and hence fell outside the definition of “Customer”. Second he held that WPL had used the Learning Edition for purposes that were not “non-production” purposes. However, he referred to the CJEU the question whether WPL’s use of the Learning Edition for a purpose that was not permitted by the licence was nevertheless permitted by article 5 (3) of the Software Directive, in which case the terms of the licence prohibiting that use would be void to that extent.

Article 4 of the Software Directive is headed “Restricted Acts”. It provides so far as relevant:

“Subject to the provisions of Articles 5 and 6, the exclusive rights of the rightholder within the meaning of Article 2, shall include the right to do or to authorize:

(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;

(b)

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…”

Article 5 is headed “Exceptions to the restricted acts”. It provides so far as relevant:

“1.

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.

…

3.

The person having a right to use a copy of a computer program shall be entitled, without the authorization of the rightholder, to observe, study or test the functioning of the program in order to determine the ideas and principles which underlie any element of the program if he does so while performing any of the acts of loading, displaying, running, transmitting or storing the program which he is entitled to do.”

Article 9 (1) states that:

“Any contractual provisions contrary to Article 6 or to the exceptions provided for in Article 5 (2) and (3) shall be null and void.”

The basis on which the court answered the questions under this head was set out at [47] and [48] of its judgment. The court noted that the question referred related to a person who had obtained a copy of a computer program under licence and carried out acts “with a purpose that [went] beyond the framework established by the licence.” It noted that the licence in this case restricted the licence to non-production purposes, but that WPL “used the various copies of the Learning Edition of SAS Institute’s program to perform acts which fall outside the scope of the licence.”

Having set out the question and the factual framework in that way, I am bound to say that, at least on first reading, the CJEU gave a clear answer to the question referred. It concluded at [61]:

“It must therefore be held that the copyright in a computer program cannot be infringed where, as in the present case, the lawful acquirer of the licence did not have access to the source code of the computer program to which that licence relates, but merely studied, observed and tested that program in order to reproduce its functionality in a second program.” (Emphasis added)

When the case came back to the judge SAS took the point that whereas the CJEU’s judgment had been given on the basis that WPL was “a lawful acquirer of the licence”, WPL never did acquire the right to use the program, which was restricted to the individual employee who clicked through the licence. The judge rejected that argument in his third judgment as follows:

“60.

As counsel for SAS Institute accepted, WPL is a legal person. As he also accepted, it obtained copies of the Learning Edition. It did so under licences. It is true that the licences were entered into by particular employees acting as agents for WPL. It is also true that the licences restricted the use of those copies of the Learning Edition to those specific employees. It also true that other employees of WPL used those copies in contravention of that restriction and that WPL is vicariously liable for such use. None of that detracts from the fact that WPL obtained copies of the Learning Edition under licence.

61.

It follows that, by virtue of Article 5(3), WPL was entitled without SAS Institute's consent to observe, study and test the functioning of the Learning Edition in order to determine its underlying ideas and principles. In my judgment it does not matter whether it exercised that right by the "licensed" employees or the "unlicensed" employees, leaving aside for the moment the fact that the "licensed" employees committed acts extending beyond the scope of the licence. Article 9(1) renders null and void any contractual restrictions contrary to Article 5(3). In my view this includes a contractual restriction on the employees by whom a legal person in the position of WPL can exercise the right under Article 5(3).

62.

To put what is much the same point in a slightly different way, I consider that it makes no difference to the issue with regard to Article 5(3) whether WPL's acts fell outside the scope of the licence because (i) the copies of the Learning Edition it purchased were used by "unlicensed" employees or (ii) the "licensed" employees used the copies of the Learning Edition for purposes which were not permitted by the terms of the licences. Either way, WPL acting by its employees used the Learning Edition in ways which fell outside the scope of the licences.”

In other words, what the judge was saying was that WPL lawfully acquired the right to use the program, but the manner in which that right was to be put into effect was through the medium of the particular employee who clicked through and accepted the licence terms. It was on that basis that a term of the licence restricting the manner in which the right could be put into effect was invalidated by article 5 (3).

I will take the points arising under this part of the claim in reverse order. I deal first with the question whether the fact that WPL used the Learning Edition for a purpose not permitted by the terms of the licence means that it cannot rely on article 5 (3). In my judgment both the Advocate-General and the court have answered that question with an unequivocal “No”. Despite Mr Hobbs’ extremely skilful advocacy it is in my judgment clear that the court distinguished between acts permitted by the licence, and the purpose for which those permitted acts were carried out. Once you have crossed the threshold of being entitled to perform acts for any purpose specified in the licence, article 5 (3) permits you to perform those same acts for a purpose that falls within article 5 (3). To my mind that is the only reading of the court’s judgment that is consistent with [61].

It follows, in my judgment, that if the judge was right about the meaning of “non-production” purposes, then that contractual restriction is invalid to the extent that it prohibits the observation, study or testing of the functioning of the program in order to determine the ideas and principles underlying it. Mr Howe also argued that the judge was wrong in his interpretation of that phrase. In view of my conclusion about article 5 (3) it is unnecessary to deal with that argument; and I express no view about whether the judge was right or wrong.

The second question of interpretation boils down to: who is “the Customer”?

The Re-Re-Amended Particulars of Claim do not, I think, allege that WPL was not entitled to use the Learning Edition at all. Rather, it is said that the activities of its employees were not permitted by clause 1.1 of the licence; and it is also asserted that WPL had accepted the terms of the licence. In those circumstances, I do not think that it is open to SAS, on the basis of its pleaded case, to assert that WPL was not a lawful acquirer of the licence.

It was not disputed that the licences must be interpreted in the same way as any commercial contract. The general principles of contractual interpretation are well-known and I do not need to rehearse them. There were, however, some points of disagreement on the details, with which I need to deal. The first concerns the relevance of and weight to be given to factual background. In considering the questions of interpretation with which he was concerned the judge took into account what he called “the background knowledge” which he set out at [277]. The judge took into account among other things what was on the packaging in which the Learning Edition was supplied. He also appears to have taken into account what could have been learned from the SAS website, even though it was not established that WPL had visited the site. In this latter respect I consider that the judge pushed the concept of background knowledge too far. Almost anything is available on the internet these days, and simply because something is available on the internet does not mean that it is relevant background: The Movie Network Channel Pty Ltd v Optus Vision Pty Ltd [2010] NSWCA 111 at [97] to [100]; Toth v Emirates [2012] EWHC 517 (Ch) [2012] FSR 26 at [44]; Hamid (t/a Hamid Properties) v Francis Bradshaw Partnership [2013] EWCA Civ 470 at [49]. To be fair, Mr Hobbs did not suggest that it was.

I begin with the language of the licence itself. In my judgment the only real indication that the licence is granted to a single human being is the rubric right at the beginning of the licence: viz “By clicking on the 'Yes' button, the individual licensing the Software ('Customer') agrees to these terms”. It seems to me to be obvious that only a human being can click on the “Yes” button. But that, as it seems to me, is not in itself a strong pointer. The word “individual” can be read as meaning “a single person”, which would include a legal person such as a company. In addition a human who clicks on the “Yes” button may do so as agent for the person (in this case a company) who bought the program and is, in the ordinary sense of the word, a customer. There are other indications elsewhere in the licence that point to the conclusion that the Customer may be a company. First, in both clause 1.3 and clause 4 “Customer” is referred to as “it”, which does not suggest a human being. Second, the prohibition on concurrent usage would be unnecessary if the licence only extended to the single human being who had clicked on the “Yes” button. Third, it is difficult to envisage how a single human being would use the Learning Edition on more than one workstation at a time. Fourth, the licence, and in particular clause 1.1, is drafted on the basis that the person who pays for the program is the “Customer”. Where, as in this case, it is a company that pays the purchase price, the assumption must be that the company is the Customer. Fifth, the fact that the chosen defined term is “Customer” is itself an indication that the licensee is the same person as the person who bought the product: Chartbrook Ltd v Persimmon Homes Ltd [2009] UKHL 38 [2009] 1 AC 1101 at [17]; Cattles plc v Welcome Financial Services Ltd [2010] EWCA Civ 599 [2010] 2 BCLC 712 at [35]. Sixth, SAS positively avers that the licence agreement subsists between it and WPL. It is difficult to see how there could be a licence agreement with a contracting party under which the contracting party acquired no licence.

I do not consider that the background material detracts from these points. The judge referred to the fact that the packaging calls the Learning Edition a “personal learning” version. But that is consistent with the view that the person in question is a company. The packaging also says that “students” and “business professionals” will learn from the edition. But clearly the actual operation of the program will be carried out by a human being whoever the “Customer” is; so that, too, carries the argument no further. What is of equal, if not more, significance is that SAS Institute describes the SAS System as “the world’s leading business intelligence and analytical software”. Since it promotes the software as business software it is natural to suppose that businesses want to learn how to use it; although obviously corporate businesses will do so through the agency of human beings.

Finally, one cannot forget that the licence agreement is offered on a take-it-or-leave-it basis. The purchaser has only two choices: click on the “Yes” button or click on the “No” button. There is no room for negotiation. If there were any doubt about the meaning of the licence at this stage, in my judgment the application of the contra proferentem principle would tip the balance in WPL’s favour. In my judgment the judge was wrong to rule out the principle at an early stage in his analysis. Accordingly, in disagreement with the judge, I would hold that WPL had the right to use the Learning Edition. Once one arrives at that position, there is no restriction on the number of employees whom WPL may authorise to observe, study and test the program, provided that they do so one at a time and at a single workstation at a time. It follows, therefore that WPL were not in breach of the licence by authorising multiple employees to use the program for the purposes of observation, testing and study. Thus I reach the same conclusion as the judge, but for a different reason.

This means that it is unnecessary to consider whether the judge was correct in deciding that article 9 also invalidated the restriction to a single employee that he found to exist. Again, I express no view about whether he was right or wrong in that respect. Nor is it necessary to consider Mr Howe’s argument that it is now too late to raise the point.

In my judgment, however, for the reasons I have given this attack on the judge’s third judgment also fails.

I would dismiss the appeal.