This paper briefly discusses the recent Football DataCo v Yahoo! C-604/10 decision by the Court of Justice for the European Union on the interpretation of Database Directive 96/9EC which concerns the harmonised copyright scheme for original information collections provided by the Directive.
Cite as: P Virtanen, “Football DataCo v Yahoo! The ECJ interprets the Database Directive”, (2012) 9:2 SCRIPTed 258 http://script-ed.org/?p=468
DOI: 10.2966/ scrip.090112.258
© Perrtu Virtanen 2012.
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 2.5 UK: Scotland License.
1. Introduction: The National Procedures and the Case before the CJEU
The Court of Justice for the European Union (CJEU) recently delivered its judgment on the boundaries of the exclusive rights granted by the EC Directive 96/9EC on Legal protection of databases concerning soccer fixture lists in Football DataCo v Yahoo! and Others. This time, however, the reference for a preliminary ruling did not touch upon the sui generis database right aspect but rather another tier also found in the Database Directive. This other tier creates harmonised copyright protection for the arrangement and selection of the materials in databases.
All six claimants in the case were involved with the creation of football fixture lists in the English and Scottish football leagues, while the respondents, a media enterprise and betting companies, used the fixture lists without licenses from the claimants. According to the referral by the England and Wales Court of Appeal, the procedure for drawing up a fixture list required a significant amount of labour and skill to satisfy a multitude of competing requirements while respecting the applicable golden rules for pairing the teams and sequencing the matches accordingly, as found by the judge at first instance. The trial judge held that the lists created were eligible for copyright protection. By contrast, the appellate court held doubts about whether the skill and judgment expended was of the right kind for the purpose of attaining copyright protection for the lists. It referred the case to the CJEU for a preliminary ruling. The CJEU posed the following questions:
(1) Whether the intellectual effort and skill of creating data should be excluded in connection with the application of art 3(1) Directive 96/9/EC; Whether the “selection or arrangement” of the contents, within the meaning of that provision, includes adding important significance to a pre‑existing item of data, and; Whether the notion of “author’s own intellectual creation” within the meaning of that provision requires more than significant labour and skill from the author and, if so, what that additional requirement is.
(2) Does the Directive preclude national rights in the nature of copyright in databases other than those provided for by article 3(1) Directive 96/9/EC?
For the first question, the CJEU held that the resources deployed for the purpose of determining the time and identity of teams corresponding to each fixture of the leagues related to the creation of the data in question and were of no relevance in assessing eligibility for copyright protection where the protection resides in the selection and arrangement of the data giving the database its structure. Accordingly, the intellectual effort and skill in creating the data were not relevant in determining eligibility for copyright protection.
The notion of an author’s intellectual creation refers to the criterion of originality which is satisfied when, through the selection or arrangement of the data contained in a database, its author expresses his or her creative ability in an original manner by making free and creative choices. By contrast, the 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. 
Further, no other criteria than that of originality can be applied in order to determine the eligibility of a database for the copyright protection provided by the Database Directive. Therefore, provided that the selection or arrangement of data is an original expression of creativity by the author it is irrelevant whether or not the selection or arrangement includes “adding important significance” to that data. On the other hand, the fact that the setting up of the database required significant labour and skill of its author cannot as such justify the protection of it by copyright under Directive 96/9, if that labour and skill did not express any originality in the selection or arrangement of that data.  It is for the referring court to assess whether the football fixture lists in question satisfy the above mentioned criteria for copyright protection.
As for the second question, the Court opined that Directive 96/9 aimed according to recitals 1-4 to remove the differences which existed between national legislation on the legal protection of databases, particularly regarding the scope and conditions of copyright protection which adversely affected the functioning of the internal market, the free movement of goods or services within the EU and the development of an information market therein. In that context, and set forth in recital 60, the Database Directive carries out a “harmonization of the criteria for determining whether a database is to be protected by copyright”. Accordingly, subject only to transitional provisions, Directive 96/9 precludes national legislation which grants databases, as defined in Article 1(2) of the Directive, copyright protection under conditions which are different to that of originality laid down in Article 3(1) of the Database Directive.
2.1. Initial Remarks
Since its adoption in spring 1996, the Database Directive has generated abundant national litigation in the Member States leading to at least seven decided referrals to the CJEU on the proper interpretation of the Directive thus far. Of those, the current Football DataCo v Yahoo!, not to be confused with still pending case Football DataCo v Sportradar at the time of writing, is remarkably the first finally decided referral at the CJEU concerning the less famous part of the Database Directive on the harmonised copyright protection for original compilations of information.
The CJEU held that the resources deployed for the purpose of determining the time and identity of teams corresponding to each fixture of the leagues relate to the creation of the data in question and are of no relevance when assessing eligibility for copyright protection when the protection resides in the selection and arrangement of the data that give the database its structure. At first glance the conclusion may seem perplexing, perhaps even harsh regarding both the rigidity of division into two, exclusive categories and the subsequent placement of the work completed into one falling out of copyright protection entirely. While the creation of data – as opposed to selection and arrangement thereof – criteria appear as straightforward dividing concepts at level of law, drawing the distinction between them in an actual case may be harder.
To start with the particulars of the case, perhaps selecting the relevant team pairs nevertheless returns eventually to the context of arrangement since all the teams are supposed to face each other a given number of times. Hence, it can be suggested that arranging the pairs in sequential order when the relevant applicable parameters make pairing and timing difficult amounts to giving structure to the database in question.
Further, the CJEU held that no other criterion than that of originality is be applied and that criterion is not satisfied when the setting up of database is dictated by technical considerations, rules or constraints. In contrast, pursuant to the findings of fact made by the domestic trial judge, the process of preparing the football fixture lists is not purely mechanistic or deterministic. Instead, it requires very significant labour and skill to satisfy a multitude of competing requirements while respecting the applicable rules as far as possible. Consequently, the work needed is not merely an application of rigid criteria but rather requires judgment and skill at each stage and especially so when the computer programme does not find a solution for pairing the teams, and sequencing to achieve an optimal home-away match sequence for every club also takes into account the organisational constraints and wishes of each club. As a result, the sequencing of matches is, as evidenced by fact-finding of the trial court, a significant skill that cannot currently be done by a computer programme.
Although limited – as opposed to dictated – by technical considerations, this organization of the data nevertheless may well stamp the maker’s personal touch to the scheme provided the factual findings are correct. While the difficult task of finding a working match schedule is successfully accomplished, there are also other alternatives for completing the fixture list. The result is thus not predetermined and limited to only one, resultant fully correct list while others are not. Accordingly, it is suggested that the conclusions by the CJEU deserve further consideration and analysis.
2.2. Further Discourse
The following attempts to assimilate these somewhat deviating considerations strive to promote understanding the underlying principles and rationales of copyright database law. They are not all found expressis verbis in the actual judgments unless otherwise indicated and therefore serve only as modest propositions for discourse and supplementary arguments to elucidate and develop the law on the field further.
First, while the sequencing of the teams is undeniably difficult, the resultant football fixture list may nevertheless appear as a simple grid or list presenting the paired teams and dates for games: the arrangement of the contents of the list in a grid or list can therefore be very basic. It was therefore putting the scheduling into a working timetable, as opposed to creating the expressive format present in the list that required the very significant skill. The skill required in the first task, the matchmaking, however original and creative, is not necessarily present in the latter which shows its results in a list. In other words, the skill and judgment present in crafting a match scheme in the case is not the same as required from the creative structure of a copyright database.
The division between the two, at times dubbed between “preparatory works” and the actual result in the form of claimed work is neither new nor unproblematic, though. Perhaps not surprisingly, it has actually arisen before in Britain concerning copyright “compilations” and remarkably in football betting context, exempli gratia in the House of Lords Ladbroke Football Ltd v William Hill Football Ltd case of 1964. The opinions of the Law Lords attest that in the case decided pursuant to then valid UK copyright law in some cases the preparatory work is separable from resultant list, in others not and much appeared to depend on the circumstances of the individual given case.
Even in there, only the skill and labour relevant for the purposes of originality was that mattered, though. Furthermore, as for arrangement, the then applicable skill, labour and judgment criteria were required in the way which the wagers were expressed and presented to the eye of the customer, including particularly the arrangement of the document and of its headings and the way in which such headings were described and coloured, etc. By the same token, considering selection, the same applied to selecting the wagers to be offered out of the total of matches.
As mentioned above, the resultant list in the recent case, it can be argued, barely conveys the exigencies and dexterity present at preparation stage to the final, mundane lists which are the decisive starting point as both for content and form for the appraisal of a copyright work pursuant to the relevant Database copyright scheme.
The absence of the information present in the database to give indication on the “personal intellectual creation” in the selection or arrangement residing only in background work has remarkable practical ramifications. How does one then establish whether the database in question is copyright protected or not? The distinction between the copyright approach present here before the CJEU and the relatively new sui generis regime of the Directive is obvious. The missing link between the “preparatory work” in obtaining the materials and the relevant database, not presenting the existence of a substantial investment and making an accordingly informed decision impossible as to whether the database in question qualifies for sui generis protection at all is actually one of the peculiar methodical weaknesses of the current ancillary rights system.
The possible remaining point of confusion is whether it is apposite to classify the skill and labour expended in the original game scheme planning as “creating the data”. Rather, the sequencing of the relevant games does not easily translate into creation of the data i.e. the same terminology is used in the judicial appraisal of sui generis databases without the risk of creating confusion.
It is suggested that what constitutes the contents in the database is mainly the names of teams and game dates. Further, given the particulars of the case, these are “created” by the claimants. The domestic courts had already decided in the case, based on earlier CJEU decisions in three football fixtures cases discusses above and the British Horseracing case that the database contents did not qualify for sui generis protection since the creation of their contents could not be considered as obtaining, verifying or presenting it as required by Directive 96/9, article 7(1). The creation of data has accordingly become standard in the sui generis case law of the CJEU to differentiate article 7(1) activities from those not meriting sui generis protection and is likely to be understood in that sense.
This consideration may well render the latter deliberation on the existence of creative choices in sequencing the matches superfluous. Nevertheless, the game scheme patterning is arguably neither entirely predetermined nor dictated to be mechanistic without choices or any creative input. Rather, the concurrent and colliding natures of given parameters that factually limit the number of available options leave room for several solutions. Putting them in order of preference while picking out the final arrangement leaves room for discretion. This perhaps renders the composition of the schedule as exactly the demanding task that the domestic courts recognised during the trial. And yet the resulting expression of the football fixture list is not necessarily creative or original. As the CJEU stated, the procedures for creating the lists, as described by the referring court, if they are not supplemented by elements reflecting originality in the selection or arrangement of the data in those lists, do not suffice for database in question be protected by the copyright as provided by Article 3(1) of the Database Directive.
As for the second main question concerning the co-existence of parallel copyright schemes in relevant fixture lists, the European Court adopted a purposeful construction of the Database Directive. Article 3(1) says: “In accordance with this Directive, databases which, by reason of the selection or arrangement of their contents, constitute the author’s own intellectual creation shall be protected as such by copyright. No other criteria shall be applied to determine their eligibility for that protection”. What the article clearly states is that the “personal intellectual creation” criterion is the sole, applicable standard while others are excluded: the present judgment unequivocally confirms this. The same criterion concerning computer programmes was among others already present in the so-called software Directive during the drafting of the Database Directive.
Further, recitals 1 to 4 emphasise harmonisation while recital 60 provides for transitional existence of overlapping national rights with those provided by the directive. Accordingly, read together with Article 3(1), the Court took the view that the Directive precludes national databases as defined in Article 1(2) of the Directive which gives additional, domestic copyright protection under conditions different from those of the Directive.
Skill and labour are not available alternative copyright criteria for other copyright schemes concerning databases that meet the somewhat technical definition for the database of the Directive. The conclusions on both accounts are not surprising in the light of the Database Directive but they help reduce authoritatively the uncertainty and speculation about the coexistence of a national, parallel copyright regime. For some, this of course removed the possibility of a complementary copyright scheme strengthening the protection domestically.
Yet one may ask if those information collections that do not fit the criteria of “database-ness” by the directive are eligible for a different domestic copyright regime. The CJEU rulings, in 2009 Infopaq International and recently Painer have arguably rendered author’s own intellectual creation an e-universal copyright criterion for several categories, if not “all sorts” of works. The stance currently taken by the CJEU may also lessen the likelihood of these eventualities.
Formerly there may well have existed a deeper doctrinal and practical schism between the concepts used for copyright criteria in different EU countries at more general level. Creativity, even in the guise of personal intellectual creation referring to the notion of originality as interpreted by the CJEU, is different from originality as understood e.g. traditionally in the UK where copyright law suggests merely that the work is not copied or derived from another work, together with criteria such as skill, judgment and labour.
Provided the Infopaq I-Painer standard is understood exclusively to determine the application of copyright criteria across the whole spectrum of works, the rulings will bring the national copyright threshold rulings to heel. The “approximation” of domestic regimes may well have taken place by judicial fiat in lieu of Union wide legislation covering all copyright works, but that is not the discussion here. In the context of copyright databases in the current case, the requirement given derives straightforwardly from the Directive.
The CJEU pioneered the field of copyright databases and gave its first preliminary ruling on the interpretation of several aspects of copyright databases embraced in Chapter 2 of the Database Directive. The arrangement and selection of data take precedence over its creation. The exclusivity of the personal intellectual creation threshold and rebuttal of the existence for an alternative domestic copyright regime for databases were the main points and of the argument employed by the CJEU. The decision can be seen as a long-awaited, complementary clarification in the field of database copyright. It remains for the national courts to decide finally on the particular facts of the case.
Notably, besides the copyright issue and the referral to the CJEU, in this case both sui generis and copyright were present at national level in the actual domestic litigation, although the former was given short shrift. But the football saga does not end here. The Football DataCo v Sportradar case before the CJEU will touch on database contents based issues in respect of Internet information distribution and jurisdiction rooted on sui generis right. The issues therein touch on many potential chokepoints regarding Internet distribution of data that the case deserves its own treatment in a separate paper.
Further, there is a fresh British domestic High Court case where football game results and other in-game information was collected and this gathering was held as obtaining, thus qualifying for sui generis protection, giving rise to a subsequent finding of an infringement. In addition, the currently hotly debated open data initiatives do not concern only the opening up of public authority data and its implications on data protection. Arguably those proposals and resultant legislation together with related licensing schemes will trigger issues on IP rights residing in databases at several levels and this may well show that the sequence of questions concerning the authoritative interpretation of the EC Database Directive regime is not yet finished. The football cases discussed here, with their limited factual circumstances, have served the purpose of opening the game in both database layers of protection regulated in the Directive. I will now pass the ball to other players in the field of copyright protection for further refinement.
* Post-doctoral Researcher, Helsinki Institute for Information Technology (HIIT), Aalto University. Warm thanks to Professor Jukka Kemppinen and Dr Olli Pitkänen for their observations. All comments and corrections are welcome. The hyperlinks were last accessed on 16 June 2012 unless otherwise indicated.
 Football DataCo Ltd, Football Association Premier League Ltd, Football League Ltd, Scottish Premier League Ltd, Scottish Football League, PA Sport UK Ltd v Yahoo! UK Ltd, Stan James (Abingdon) Ltd, Stan James plc, Enetpulse ApS,  CJEU C-604/10 (hereafter Football DataCo v Yahoo!).
 Directive 96/9/EC of the European Parliament and of the Council of 11 March 1996 on the Legal Protection of Databases,  OJ L77, at 20-28.
 Football DataCo and Others v Yahoo! and Others,  EWCA Civ 1380, at 4 and 5.
 Football DataCo and Others v Britten Pools Ltd and Others,  EWHC 841(Ch): “The process of preparing the Fixture Lists…involves very significant labour and skill in satisfying the multitude of often competing requirements of those involved.”, at 41; “This work is not mere ‘sweat of the brow’.”, at 43.
 Football DataCo v Yahoo!, paras 33 and 35.
 Ibid, paras 38-39.
 Ibid, paras 37, 40- 41.
 Ibid, para 42.
 Ibid, para 43.
 Ibid, paras 48- 52.
 The British Horseracing Board Ltd and Others v William Hill Organization Ltd ,  C-203/02; Fixtures Marketing Ltd v Organismos prognostikon agonon podosfairou AE (OPAP),  C-444/02; Fixtures Marketing Ltd v Svenska Spel AB,  C-338/02; Fixtures Marketing Ltd v Oy Veikkaus Ab,  C-46/02; Directmedia Publishing GmbH v Albert-Ludwigs-Universitt Freiburg,  C-304/07; Apis-Hristovich EOOD v Lakorda AD,  C-545/07. On domestic case-law on copyright cases, see P Virtanen, “Evolution, Practice and Theory of European Database IP Law” (2008) 303 Acta Universitatis Lappeenrantaensis, 138-141. On CJEU case-law reviewed until 2008 see E Derclaye, The Legal Protection of Databases: A Comparative Analysis (Cheltenham: Edward Elgar Publishing, 2008). On recent French cases, see by the same author, “Recent French Decisions on Database Protection: Towards a More Consistent and Compliant Approach with the Court of Justice’s Case Law” available at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1989031.
 Football DataCo Ltd, Scottish Premier League Ltd, Scottish Football League, PA Sport UK Ltd v Sportradar GmbH (German) and Sportradar (Swiss),  Case C-173/11.
 Football DataCo v Yahoo!, at paras 32 and 36; Pursuant to Article 3 (1) of the Directive: “In accordance with this Directive, databases which, by reason of the selection or arrangement of their contents, constitute the author’s own intellectual creation shall be protected as such by copyright. No other criteria shall be applied to determine their eligibility for that protection”. (emphasis added)
 The arrangement of games and the rules employed in organising the game sequence differs in the English and Scottish football leagues. It is given in Football DataCo and Others v Britten Pools and Others, at paras 9-40.
 The division into selection and arrangement proves problematic since the task was and can be presented as selecting the team for every time slot which then imports “selection” into play; this shows how formulating the presentation of the relevant processes easily becomes the hostage of legally relevant language.
 As described in Football DataCo v Yahoo!, this is usual. The case also divides the tasks of the domestic and EU courts and provides the requisite margin of discretion concerning factual circumstances as to whether there is sufficient originality in the particular case: “it is for the referring court to assess, in the light of the factors set out above, whether the football fixture lists in question in the main proceedings are databases which satisfy the conditions of eligibility for copyright protection”., at para 43.
 There is no reproduction of the relevant fixture lists either in the domestic or the CJEU judgments. Instead, the method of preparing them and the rules employed are expounded in detail.
 See para 42 of Football DataCo v Yahoo!, “…the fact that the setting up of the database required, irrespective of the creation of the data which it contains, significant labour and skill of its author, as mentioned in section (c) of that same question, cannot as such justify the protection of it by copyright under Directive 96/9, if that labour and that skill do not express any originality in the selection or arrangement of that data”.
 Ladbroke Football Ltd v William Hill Football Ltd,  1 All ER 465 where reference was made by Lord Pearce to an earlier High Court case, Football League Ltd v Littlewoods Pools Ltd,  1 Ch 637.
 See e.g. H Laddie et al, The Law of Modern Copyright and Designs (London: Butterworths, 2000), vol 1, at 206-209 and the cases referred to therein.
 See the judgment in Football DataCo v Yahoo!, opinion of Lord Evershed. On uncertainty as to what exactly constitutes the relevant criteria for originality in the UK copyright law for various works together with the harmonsing effect of European copyright law, see L Bently and B Sherman, Intellectual Property Law (Oxford: OUP, 2001), 80-82.
 The “what is worth copying is prima facie worth protecting” adage has been refuted on several grounds. That discourse falls, as regards the harmonised EC regime, outside the scope of this paper. See e.g. Laddie et al, note 20 above, at 203.
 P Virtanen, “Database Rights in Safe European Home: The Path to more Rigorous Protection of Information” (2005) 202 Acta Universitatis Lappeenrantaensis 299.
 Football DataCo v Yahoo!, para 42.
 Council Directive 91/250/EEC of 14 May 1991 on the Legal Protection of Computer Programs,  OJ L122, at 42-44, art 3(1). It is also present in article 6 concerning photographs as set out by Directive 2006/116/EC of the European Parliament and of the Council of 12 December 2006 (codified version) on the Term of Protection of Copyright and Certain Related Rights,  OJ L372, at 12-18, (subsequently amended as for extension of term for performers and sound recordings by Directive 2011/77/EU). The original Term Directive 93/98 (1993) also employed the author’s own intellectual creation criterion concerning photographs in art 6.
 Infopaq International C-5/08  ECR I-6569, paras 35, 37-38; Painer C-145/10, para 87; Bezpeĉnostni softwarová asociace C-393/09, para 45; Football Association Premier League and Others, C-403/08 and C-429/08, para 97.
 See e.g. Laddie et al, see note 20 above, at 82-83 and 203-205.
 See e.g. G Karnell, “European Originality: A Copyright Chimera”, in Intellectual Property and Information Law: Essays in Honour of Herman Cohen Jehoram (The Hague: Kluwer 1998) 201-209; Newspaper Licensing Agency et al v Meltwater BV et al ,  EWCA Civ 890, at 20.
 See the High Court Decision in the case by Justice Floyd, in  EWHC 841 (Ch), at 92.
 Football DataCo Ltd et al v Sportradar and Football DataCo Ltd t et al v Stan James et al, [2012 ] EWHC 1185 (Ch). The case involves largely the same parties as the Football DataCo v Sportradar Case referred to the CJEU (C-173/11). The current case on copyright databases, the interplay between the different layers of database protection and the compound effect of all the cases after both preliminary rulings and final domestic judgments remains to be seen.
 The underlying idea of open data can be roughly characterised in that certain, usually previously public authority held data should be freely available for the public to use and republish, both for commercial and non-commercial purposes. The purposes of open data movement complement and run partly parallel to open source, open content and open access schemes. See in the current context of the case e.g. Data.gov.uk from the UK and in the EU the Directive 2003/98/EC of the European Parliament and of the Council of 17 November 2003 on the Re-use of Public Sector Information,  OJ L345, at 90-96, and its revision initiatives, e.g. in EC Commission Digital Agenda for Europe 2010-20, I Pillar, “Action 3: Open up Public Data Resources for Re-use”.
See e.g. S Kulk and B Van Loenen, “Brave New Open Data World?” (2012) available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2039305.
 See e.g. E Derclaye, “Does the Directive on the Re-use of Public Sector Information affect the State’s Database Sui Generis Right?” in J Gaster et al (eds), Knowledge Rights: Legal, Societal and Related Technological Aspects (Vienna: Austrian Computer Society, 2008), at 137-169.