United Kingdom Journals

---

You are here: BAILII >> Databases >> United Kingdom Journals >> Recent French decisions on database protection: Towards a more consistent and compliant approach with the Court of Justice’s case law? | Derclaye | European Journal of Law and Technology
URL: http://www.bailii.org/uk/other/journals/EJLT/2012/03-2/124.html
Cite as: Recent French decisions on database protection: Towards a more consistent and compliant approach with the Court of Justice’s case law? , Derclaye , European Journal of Law and Technology

---

[New search] [Printable PDF version] [Help]

---

Recent French decisions on database protection: Towards a more consistent and compliant approach with the Court of Justice’s case law?

Estelle Derclaye [1]

Cite as: Derclaye, E, ‘Recent French decisions on database protection: Towards a more consistent and compliant approach with the Court of Justice’s case law?’ European Journal for Law and Technology, Vol. 3, No. 2, 2012

Introduction

Since the official date of implementation of the Database Directive, namely 1 January 1998, 14 years have now already passed. During this decade and a half, French courts have handed down around 30 decisions on database protection. [2] Between 1998 and 2004, the year when the Court of Justice of the European Union (CJEU) handed down its first four parallel rulings on the sui generis right, the French case law on the sui generis right was quite erratic. [3] Admittedly, such state of affairs was not dissimilar in other Member States owing to the sheer novelty of the sui generis right and the vagueness of many of the Database Directive’s provisions. While the sui generis right was meant to replace the tort of slavish copying still existing in many Member States on the continent, French decisions fluctuated between cumulating the two actions and thus repairing the same damage twice, and rejecting the overlap. Since its 2004 rulings, the CJEU handed down two more decisions mainly confirming its case law on infringement. [4] One more was handed down earlier this year on the issue of originality for database copyright but also on the relationship between copyright and similar types of protection for databases. [5]

Have the French courts adjusted their rulings to follow the CJEU’s precedents since 2004? This comment compares the French case law before its British Horseracing Board and Fixtures Marketing decisions with French case law decided after them. It also checks whether these newer French decisions comply with the CJEU case law. The comment focuses on the sui generis right, including the overlap between the latter and slavish copying also known as parasitism. Copyright is left aside mainly because the French criterion of originality and other aspects of the protection by copyright before the implementation were very similar to the Database Directive’s provisions on the matter so that most decisions do not bring about a change in the law.

I. Sui generis right

The Database Directive needs little introduction for the expert intellectual property lawyer. Since the French litigation relating to the sui generis right has, not surprisingly, concentrated mainly on issues of scope and infringement, I will simply remind the reader of the relevant provisions of the Directive and paragraphs of the CJEU decisions on these issues. [6]

‘Article 1 - Scope
1. This Directive concerns the legal protection of databases in any form.
2. For the purposes of this Directive, 'database` shall mean a collection of independent works, data or other materials arranged in a systematic or methodical way and individually accessible by electronic or other means.
3. Protection under this Directive shall not apply to computer programs used in the making or operation of databases accessible by electronic means.’

‘Article 7 - Object of protection
1. Member States shall provide for a right for the maker of a database which shows that there has been qualitatively and/or quantitatively a substantial investment in either the obtaining, verification or presentation of the contents to prevent extraction and/or re-utilization of the whole or of a substantial part, evaluated qualitatively and/or quantitatively, of the contents of that database.
2. For the purposes of this Chapter:
(a) 'extraction` shall mean the permanent or temporary transfer of all or a substantial part of the contents of a database to another medium by any means or in any form;
(b) 're-utilization` shall mean any form of making available to the public all or a substantial part of the contents of a database by the distribution of copies, by renting, by on-line or other forms of transmission. The first sale of a copy of a database within the Community by the rightholder or with his consent shall exhaust the right to control resale of that copy within the Community;

Public lending is not an act of extraction or re-utilization.

(…)

5. The repeated and systematic extraction and/or re-utilization of insubstantial parts of the contents of the database implying acts which conflict with a normal exploitation of that database or which unreasonably prejudice the legitimate interests of the maker of the database shall not be permitted.’

In British Horseracing Board v William Hill, the CJEU held that:

‘30. Under the 9th, 10th and 12th recitals of the preamble to the directive, its purpose, as William Hill points out, is to promote and protect investment in data ‘storage’ and ‘processing’ systems which contribute to the development of an information market against a background of exponential growth in the amount of information generated and processed annually in all sectors of activity. It follows that the expression ‘investment in … the obtaining, verification or presentation of the contents’ of a database must be understood, generally, to refer to investment in the creation of that database as such.

31. Against that background, the expression ‘investment in … the obtaining … of the contents’ of a database must, as William Hill and the Belgian, German and Portuguese Governments point out, be understood to refer to the resources used to seek out existing independent materials and collect them in the database, and not to the resources used for the creation as such of independent materials. The purpose of the protection by the sui generis right provided for by the directive is to promote the establishment of storage and processing systems for existing information and not the creation of materials capable of being collected subsequently in a database.’ [7]

‘69. In that connection, it must be borne in mind that protection by the sui generis right covers databases whose creation required a substantial investment. Against that background, Article 7(1) of the directive prohibits extraction and/or re-utilisation not only of the whole of a database protected by the sui generis right but also of a substantial part, evaluated qualitatively or quantitatively, of its contents. According to the 42nd recital of the preamble to the directive, that provision is intended to prevent a situation in which a user ‘through his acts, causes significant detriment, evaluated qualitatively or quantitatively, to the investment’. It appears from that recital that the assessment, in qualitative terms, of whether the part at issue is substantial, must, like the assessment in quantitative terms, refer to the investment in the creation of the database and the prejudice caused to that investment by the act of extracting or re-utilising that part.

(...)

72. It must be added that, as the existence of the sui generis right does not, according to the 46th recital of the preamble to the directive, give rise to the creation of a new right in the works, data or materials themselves, the intrinsic value of the materials affected by the act of extraction and/or re-utilisation does not constitute a relevant criterion for the assessment of whether the part at issue is substantial.

(...)

86. The purpose of article 7.5 of the Directive is to prevent circumvention of the prohibition in article 7.1 of the Directive. Its objective is to prevent repeated and systematic extractions and/or re-utilisations of insubstantial parts of the contents of a database, the cumulative effect of which would be to seriously prejudice the investment made by the maker of the database just as the extractions and/or re-utilisations referred to in article 7.1 of the Directive would.

87. The provision therefore prohibits acts of extraction made by users of the database which, because of their repeated and systematic character, would lead to the reconstitution of the database as a whole or, at the very least, of a substantial part of it, without the authorisation of the maker of the database, whether those acts were carried out with a view to the creation of another database or in the exercise of an activity other than the creation of a database.

(...)

89. Under those circumstances, ‘acts which conflict with a normal exploitation of [a] database or which unreasonably prejudice the legitimate interests of the maker of the database’ refer to unauthorised actions for the purpose of reconstituting, through the cumulative effect of acts of extraction, the whole or a substantial part of the contents of a database protected by the sui generis right and/or of making available to the public, through the cumulative effect of acts of re-utilisation, the whole or a substantial part of the contents of such a database, which thus seriously prejudice the investment made by the maker of the database.’

1. Subsistence

Before the CJEU rulings in 2004, most French courts dealing with databases which were spin-offs of a main activity held they were protected by the sui generis right. [8] Courts did not make a distinction between on the one hand the investment in creating the data and the verification and presentation happening at the time of data creation and on the other hand the investment in collecting, verifying and presenting separate from the data creation. [9] They only asked the database producer to prove its investment in making the database without asking more precisely proof of the object of investment (i.e. obtaining, verification or presentation). [10] The French law which implemented the Database Directive used the term ‘constitution’ (see art. L. 341-1 of the Intellectual Property Code (IPC) rather than ‘obtaining’ (‘l'obtention’ in French) which the Directive uses instead. This vaguer term may have explained why some courts before the CJEU decisions construed the provision very generously. But this interpretation was not typical of France; many courts in other Member States handed down similar decisions. [11] Has this changed since the CJEU rulings?

Even after 2004 and up to 2007, many French courts still did not make the distinction between creation and collection of the data. [12] They were therefore not respecting the EU case law. However, the attitude, especially that of the appellate courts, started to change as early as 2005, culminating in 2009 with the French Supreme Court’s decision in the Précom case. [13] The Court of Cassation, which was for the first time asked to deal with the issue of the distinction between creating and collecting data, clearly endorsed the CJEU rulings. [14] In that case, the claimant created a web site with advertisements of properties for sale and the defendant published some of the claimant’s ads in its newspaper aimed at estate agents. The French Supreme Court held that the investment was in the creation of the ads and that the verification took place during creation and was purely formal. The Court expressly stated that ‘a company cannot claim the benefit of the protection by the sui generis right for investments linked with the creation of the content of a database’.

Nevertheless, the first instance courts decisions reported even recently, namely from 2009 to 2011, discuss the general issue of subsistence of the sui generis right in a basic or very basic way. [15] The courts hold the sui generis right infringed but do not check if the sui generis right subsists in the first place. It may have been because defendants conceded the subsistence issue. However, the decisions do not state that this was the case. Without it being conclusive, it is worth noting that the decisions originate from provincial courts. By contrast, the court of first instance of Paris seem to be more up to date with the Supreme Court case law (assuming it looked into it, which is unsure as French courts rarely refer to other decisions) as it now seems to be more rigorous with spin-off databases. For instance, in a case opposing Ryanair with Opodo, the first sued the latter for, among others, sui generis right infringement because it was possible for members of the public to book Ryanair flights on Opodo’s web sites. [16] Opodo did not contest that flight timetables are databases. The court of first instance of Paris held that Ryanair did not have a sui generis right. Even if the evidence submitted by Ryanair proved that it incurred costs to develop its web site and maintain it, [17] these costs did not prove that Ryanair made a substantial investment in the constitution (as the term goes in France), verification or presentation of the database. This is the only sentence the court utters on the matter, which is quite short. The result is probably correct as it is very difficult for spin-off database producers to prove a substantial investment separate from the investment in creating the data, but it would have been better if the court had justified its decision a little more, by stating, at least, this.

The Paris court of appeal has been recently more explicit on this very point. The case concerned web sites, exploited by Metropole Télévision, which allow users to catch up on programmes shown previously on television channels (namely ‘M6replay’ and ‘W9replay’). [18] Metropole Télévision sued SBDS Active for enabling access to the programmes on the M6replay and W9replay websites, without users being directed first to the home pages of the M6replay and W9replay websites. Metropole Télévision’s database was a web site consisting of a tool to search the television programmes classed by genre, date, title, category and links to ‘bonuses’ (e.g. interviews, games, pictures), as well as a RSS feed updating those. The court agreed with the first instance decision which said this was a database, namely a collection of audiovisual works. The court however did not think that the investments related to the obtaining, verification or presentation of the elements because the claimant’s documents did not indicate costs relating to the conception of the site, nor to the selection, indexation and classification which constitute the essence of a database. Similarly, in a case concerning a web site, the Lyon court of appeal held that the claimant did not prove its investments were made for the collection, verification or presentation of the data but instead they were for the site’s visual appearance (e.g. the invoices related to the creation of ad banners, of newsletters forms or more generally to the ‘management of the project’). [19] Likewise, the employees’ contracts (a secretary and a person dealing with communications) were too vague to prove that they worked on the creation and development of the database rather than on the company’s main activity which is the commercialisation of products.

Despite these positive developments on the verification of the investment’s object, the Court of Cassation may have created some doubt with its Lectiel v France Télécom decision. [20] In this case, the Supreme Court confirmed the Paris court of appeal’s decision that France Télécom’s list of subscribers (telephone directory) was a protected database. The defendant argued that there was no investment. But the court of appeal held that France Télécom’s telephone directory did not simply consist of its subscribers' names, addresses and numbers but had much more information than that, half of it coming from France Télécom. Whereas the appellate court noted that the database was not simply a collection of all France Télécom’s telephone directories but that there was, in relation to the copyright claim, some ‘presentation effort’, [21] it did not specifically state whether there was a substantial investment in the collection, verification or presentation of the data separate from data creation in relation to the sui generis right claim. It is strange therefore that the Court of Cassation confirmed the appellate decision. But it may be explained by procedural matters (such as the defendant not raising this point in its appeal to the Supreme Court) or by the agreement of the parties with the expert who gave an opinion as to the investment made in the database. The appellate court relied heavily on this expert statement. The Lectiel decision is therefore not entirely conclusive. Presumably therefore, the Court’s Précom ‘precedent’ still stands. As we have seen, the Lectiel decision has not had a negative impact on this subsistence point on the later appellate courts’ decisions in Metropole Télévision and TAC UP.

2. Infringement

The CJEU held in its 2004 rulings that there must be a link between the substantial part and the substantial investment. Many French decisions prior to the CJEU case law were not much detailed on the criteria to determine infringement of the sui generis right. When decisions were more precise, discussing whether a qualitative or quantitatively substantial part was taken, the courts diverged in their application of the test. Some held that the qualitatively or quantitatively substantial character of the extraction must be evaluated in function of the use that is made of it. As we now know, this is totally contrary to the CJEU case law. Thus in the PR Line v Newsinvest case, even if the extractions were not quantitatively substantial, the court held that because the extractions done by the defendant allowed it to enrich its own database, they were qualitatively substantial and thus infringing. [22] However, the court of appeal reversed the decision, holding that it was irrelevant that the extractions had enriched the defendant’s database and the extractions were not qualitatively substantial. [23] The Cadremploi v Keljob decision was more nuanced. [24] Cadremploi made a database of job offers. Keljob interrogated Cadremploi’s web site every night and selected the job offers that interested it. After this selection, Keljob extracted the essential elements of each offer, namely the title of the job, the sector, the geographical zone, the date published on Cadremploi and the URL address. Thus the content of the offers was not reproduced and the information extracted was less than 12% of the volume of the offers. The court of first instance of Paris held that while the defendant did not take a quantitative substantial part, it took a qualitative one. Even if it said that the defendant took up the value of the claimant’s database, a statement which without precision could mean the value to the user, the criterion which the Court of Justice held wrong and irrelevant, [25] court of first instance of Paris held that the extractions were of the information relating to the selection and referencing system of the database, which represent the claimant’s efforts. This reasoning seems more comparable to the link between substantial investment and substantial part that the CJEU requires.

Since 2005, similarly to the issue of subsistence, the first instance courts decisions reported briefly or barely discuss the issue of infringement. They generally simply affirm that a substantial part has been extracted or re-utilised. For instance, in Europages v Ewaycom, the first instance court of Paris simply said that the defendant infringed by taking qualitatively substantial elements of the claimant’s database. [26] In GPSPrevent v Coyote System et al, the court of first instance of Lille was even more laconic: it simply said that there was infringement of the sui generis right! [27] The commercial court of Paris, in Xooloo v Optenet Center, Optenet, France Telecom, did not think that 15.56% was a quantitatively substantial part. However, it then found infringement without discussing whether a qualitatively substantial part was taken. [28] The same section and chamber of the first instance court of Paris as in Europages held in a case opposing Optima On Line to Media Contact Israel that the latter had extracted parts of the database as proved by the fact that it reproduced three ‘trap addresses’. The defendant argued it only extracted these three addresses but the court rejected the argument by holding it was not possible as the defendant did not know that these were trap addresses. The court then simply deduces from this proof that it must have taken a substantial part but does not go further in the analysis. [29]

One decision is plainly wrong both on subsistence and on infringement: the court of first instance of Paris does not make a distinction between creation and collection of data although the database was clearly a potential spin-off (horse races fixtures along with other information such as horses’ performances and the results of the races) and holds that there was infringement because of the intrinsic value of the extracted data. For the court, because the data are indispensable for the defendant’s activity, the sui generis right is infringed. [30] Obviously, a defendant will always extract the data that is interesting to him- or herself so that if one applies this test, the sui generis right will always be infringed. The court did not check either whether the part extracted reflected a substantial investment.

Another decision seems to depart from the Keljob decision. Like in Keljob, the database was a search engine which indexed ads by way of hyperlinks leading to the claimant’s web site. [31] For the court, such indexation was not an extraction of the contents of the claimant’s database. However, the court did not check whether the elements taken are a substantial part. It assumed it. The decision may have been a policy one in favour of search engines, indexation being useful for the good functioning of the Internet. An author commented that the court restricted the notion of extraction, which seems motivated by the usefulness of search engines, and wonders whether the usefulness of search engines can allow the disrespect of third parties’ rights. [32]

Finally, one decision interprets a little loosely the French provision implementing article 7(5) of the Database Directive namely article L 342-2 IPC, which states that the database maker can prevent the systematic extraction or re-utilisation of insubstantial parts of the database when this manifestly exceeds the normal use of the database. [33] The CJEU interpreted this article by requesting that the insubstantial parts taken amount to a substantial part. In this French case, which opposed Agence France Presse (AFP) to Topix, the owner of the web site of news ‘universalpressagency.com’, the court held that the extractions were neither quantitatively nor qualitatively substantial. It thus applied article 342-2 IPC, without verifying that the insubstantial parts taken amount to a substantial part. [34] The court only cites a number of factors the addition of which seem, according to it, to lead to the application of article 342-2 and thus to infringement: the defendant’s reutilisation exceeded the normal use of the database because the universalpressagency.com news were freely accessible, the site was a competitor of AFP even if it was not a commercial site, the reception of AFP news is normally not free of charge and their reproduction is only authorised in quantities and uses defined by the terms of a subscription. It is difficult to reproach this interpretation of article 342-2 to the court as the provision leads to it. It is rare for national courts to check the national law against the Database Directive. In France, no court has applied the directive as such to my knowledge, as opposed to the courts in England and Wales for instance. [35] As to the appellate courts, nothing can be gleaned as the issue of infringement was not discussed. Likewise, the Supreme Court has yet to address this issue. So it is not possible to say yet that higher French courts now apply the EU case law on the infringement test correctly.

II. Cumulation with parasitism

Article 13 of the Database Directive allows cumulation between the sui generis right and, among others, unfair competition. [36] Indeed, as the unfair competition and infringement actions have a different cause and object (intellectual property rights are property rights while unfair competition is a tort based on a breach or wrong), they are distinct and can thus be cumulated. In France, the cumulation between the two actions is expressly provided for in articles L. 615–19 al. 2 and L.716-3 IPC and for databases, in article 341-1 al. 2 IPC.

The unfair competition action thus complements the infringement action. This complementary nature means that there can be cumulation as long as the acts of unfair competition complained of are distinct from the infringement acts. Therefore, there can only be a condemnation for unfair competition if a breach distinct from infringement can be established. If not the unfair competition action will be rejected. Courts and commentators are anonymous on this point. However, some courts and commentators disagree as to what constitute acts or facts distinct from infringement. For some courts and commentators, slavish copying, also called parasitism in France, is not a distinct act or fact. [37] Only facts without any link with the creation or innovation infringed, such as disparagement, misleading advertising, copying secret know-how or risk of confusion, are distinct facts. Others consider parasitism as an act of unfair competition and thus parasitism as a distinct fact. According to the tort of parasitism, it is unfair and thus punishable to copy another’s achievement (i.e. without making one’s own creative or innovative efforts), whether that achievement is or not protected or protectable by an intellectual property right. Therefore, in France, some courts allow an overlap between sui generis right infringement and parasitism. There can be three types of overlap: simultaneous or concurrent (while the sui generis right subsists), a posteriori or subsequent (when the term of the sui generis right has elapsed) and negative (when the sui generis right cannot subsist because the conditions were not fulfilled). In the case of the simultaneous overlap, it means that the same fact (copying the whole database or part of it) is punished and repaired twice. The three overlaps also entail that the sui generis right holder can bypass the stricter conditions of the sui generis right by using the unfair competition action instead. This also renders the sui generis right useless; there is no need to resort to it since parasitism absorbs it. For all these reasons, these overlaps lead to over-protection. Therefore, as I argued in previous publications, consistent with the principle of the freedom to copy, freedom of competition and commerce, the economics of information goods, the human rights to the respect of one’s property and to information, and the public interest, both databases which meet and do not meet the criteria of sui generis protection along those which already benefited from this protection, should not be protected by the tort of parasitism. [38]

Before 2005, courts were split on the issue of whether parasitism is a fact distinct from sui generis right infringement. [39] The majority of them (six) ruled that copying a protected database is not a fact distinct from sui generis right infringement [40] while three courts ruled in the opposite direction. [41] As there was no Supreme Court decision on the specific issue of simultaneous overlap and the lower courts were split, the issue was unresolved. As to the negative overlap, only one court condemned the parasitic copying of a database which it held unprotectable by the sui generis right. [42]

From 2005 until now, the split has remained among first instance courts. But at the higher level, the trend has gradually been to reject the overlap between the sui generis right and parasitism and the Supreme Court has rejected at least the negative overlap. Of those courts which allow the overlap (the vast majority being first instance courts), some courts allow the negative one, others, the simultaneous one. Let us first look at those courts which allow the negative overlap. One such jurisdiction is the Court of appeal of Paris which confirmed a decision of the Paris Commercial Court in its entirety. The Paris Commercial Court accepted jurisdiction when the claimant brought an unfair competition action for extraction of data from its database and the court did not look into the question whether a sui generis right subsisted and whether it was infringed. [43] The court held there was parasitism. The defendant also used the name of the claimant to sell its software. It is unclear from the decision why it found that this was an act of parasitism rather than risk of confusion.

The others are decisions from the first instance courts of Caen and Paris. The court of Caen admitted the action in unfair competition for parasitism although it rejected the claim based on the sui generis right. [44] In Adenclassifieds v Solus’immo, the Paris court found that the defendant’s search engine for real estate ads did not extract or reutilise data of the claimant‘s database of similar ads. Nevertheless, it held that it is possible in this situation to admit an action based on parasitism. In the particular case before it though, it held that parasitism was not established. [45] An ambiguous decision is that of the court of first instance of Digne in Réseau Fleuri v L’Agitateur Floral. [46] The court cumulates unfair competition and the sui generis right because the defendant created confusion. This in itself is not problematic. But then it states that there is unfair and parasitic competition. The decision is thus a little unclear on the legal principles even if in the present case, it was totally clear that there were several acts leading to confusion anyway. In conclusion, those three courts accepted parasitism as an alternative basis to sui generis right infringement.

As to the simultaneous overlap, in GpsPrevent v Coyote System et al., the court of first instance of Lille decided that the sui generis right was infringed but the court also punished the defendant for unfair competition for 25,000 € in addition to the 50,000 € for infringement of the sui generis right. The court held that the reason for the act being unfair was that the defendant benefited from an undue advantage – its copying provided a database at a cheaper cost. It is clear therefore that even if the court does not utter the word, it is condemning on the basis of parasitism, not risk of confusion. [47] In the same vein, in Xooloo v Optenet Center, Optenet, France Télécom, the commercial court of Paris found sui generis infringement and simply added that since the defendant sold the copied database at a cheaper price, there was parasitism. [48]

Other first instance courts however have rejected the simultaneous overlap. In Europages v Ewaycom, the first instance court of Paris is very clear: it states that the claimant cannot win on the basis of the unfair competition action as the act of copying is the same as that repaired under infringement of the sui generis right. [49] In AFP v Topix, the commercial court of Paris rejected the claim on the basis of unfair competition and parasitism, as there was no risk of confusion and thus no distinct facts. [50] These two decisions conflict with others of the same courts on the negative overlap. The chambers of the commercial court of Paris in Xooloo and AFP were different, which may explain the different results. However, the same chamber and same judges of the court of first instance of Paris decided the Europages and Adenclassifieds decisions, which are at odds. In conclusion, many first instance courts still allow the negative and simultaneous overlaps.

There are however more recent decisions from the Supreme Court and courts of appeal which do not cumulate the two actions. In Précom, as we saw above, the Court of Cassation held that the database in question was unprotected. It also held that there was no parasitism because the defendant indicated the source of the information and did not use the claimant’s name to capture its value. [51] The reasoning is a little cryptic though. It shows that the defendant did not create confusion. Implicitly, the Supreme Court is also saying that if there is no risk of confusion, the simple act of copying of an unprotected database is not reprehensible under the law of unfair competition. It thus rejects the negative overlap. The court of first instance of Paris decision in Adenclassifieds is even stranger, as it not only contradicts its previous decision in Europages but also the Supreme Court on this point.

In 2010, the court of appeal of Versailles held that a guide comparing different types of cars was a database protected by copyright but not simultaneously by parasitism. The court held that communicating the database to the public at a cheaper price is not a distinct fact. [52] It added very clearly that the claimant did not show in this respect a prejudice distinct from that resulting from the infringement, which is already repaired. In Metropole Télévision et al. v SBDS Active, [53] the claimant argued that the defendant’s web site reduced the number of pages viewed by the users and thus reduced its advertising revenue. But the defendant showed that when users access its web site, they are directed to the m6replay and w9replay’s relevant page including all the other functionalities of the site (link to home page and to other audiovisual works) including the advertisements. Therefore, the court of appeal of Paris concluded there was no parasitism because the defendant did not override the normal navigation process of the claimant’s web site and the advertising revenues were preserved.

In Football Dataco, the CJEU made totally clear that Member States cannot protect by copyright databases which do not fulfil the requirements of article 3 of the Database Directive (which requires the author's own intellectual creation as the criterion of originality). [54] To answer the question posed by the Court of Appeal of England and Wales in Football Dataco, the Advocate General relied on general recitals (recitals 3 and 12) which do not only apply to copyright but also to the sui generis right and stress the removal of differences in the law of Member States and the creation of a ‘uniform legal protection regime’ for databases. The CJEU followed this reasoning, also relying on recital 60 stating that article 3 of the Database Directive ‘carries out a ‘harmonization of the criteria for determining whether a database is to be protected by copyright’.’ [55] The Court of Appeal of England and Wales did not ask the same question in respect of the sui generis right as it did not arise in the case before it, but it would be strange if the CJEU would answer differently if, in future, it receives a similar question in relation to the sui generis right. By analogy, a future CJEU’s answer should be identical. If not, national laws could simply override the Database Directive and protect unprotectable databases by parasitism instead of the sui generis right although by way of the Database Directive, the sui generis right has in effect absorbed parasitism. This state of affairs would create disharmony, and thus be totally contrary to the aim of the Database Directive, in fact any directive.

Conclusion

What can we conclude from this overview of the French case law before and after to the CJEU rulings in 2004? Many first instance courts are still ‘learning the trade’ after so many years now that the sui generis right has been implemented and even 6 years after the first CJEU rulings on the topic. Many of them still do not make the distinction between creation and collection of the data and more generally fail to check the object of the investment. But with the Court of Cassation decision in Précom and the appellate courts following it, there is hope that the first instance courts will soon follow. We have the same pattern of basic or very basic decisions on infringement in the first instance courts but without higher court decisions infirming them. This makes the French case law on sui generis right infringement unsatisfactory. Regarding the overlap between the sui generis right and parasitism, first instance courts are split while the appellate courts and the Court of Cassation are rejecting the overlap. However, if one looks at the recent Supreme Court’s decisions on the relationship between parasitism and all intellectual property rights, they show that the tort of parasitism is still alive as the Court seems to constantly contradict itself. [56] It is therefore far from certain that its Précom decision in the area of database protection shows a clear trend towards rejection of the overlap between the sui generis right and parasitism. The same goes for the French Supreme Court’s case law on the negative overlap between intellectual property rights and parasitism. [57]

In conclusion, over the years, French jurisdictions seem to have become acquainted with the intricacies of the sui generis right and follow the EU case law but there is room for improvement in the French lower courts. It is not a specifically French thing however. One needs only to look across the pond and see that even appellate judges are not always good students. The Court of Appeal of England and Wales’ decision in Newspaper Licensing Agency v Meltwater is a prime example along with the even more recent Future Publishing case. [58] These cases were not dealing with the sui generis right but with copyright. Both decisions fail to apply the Infopaq ruling on originality. [59]

What to do about national courts’ failure to follow EU case law? First, is it a bad thing in the first place? It depends on the issue. On decidedly well-settled uncontroversially good EU case law, national courts should no doubt follow the EU case law. In the case of Infopaq, it is arguable that the judgment went overboard and made the law instead of interpreting it. The InfoSoc Directive [60] did not harmonise the requirement of originality across the board. But the CJEU ruled that the criterion of originality for all works is the author’s own intellectual creation. It reaffirmed this in the Painer decision. [61] Most commentators will agree that this is the, higher, continental criterion not the, lower, common law one of skill and labour. In fact, in Football Dataco, the Advocate General says it is. [62] And the CJEU confirmed that even significant skill and labour are not enough if they do not express the author’s own intellectual creation. [63] Should national courts in the United Kingdom and Ireland follow the Infopaq ruling? Arguably, they have no choice and are bound by it even if it makes the law. The only way out is for the EU legislature to reverse the judgment if it is not happy with it. However, what happens if national courts do not follow EU case law? The only sanction is for the Commission to impose fines to Member states for disobeying the CJEU. This has not yet happened in the field of intellectual property law and one wonders if it ever will... So on contentious issues, it may be that competition between national courts and between national and EU courts is a good thing. [64] It may force the CJEU to change its ‘wrong’ case law. It is arguably better to have several different precedents for some time until the best solution to the legal problem emerges than being stuck with a bad one forever. The issue in the end is how much one wants to sacrifice on the altar of harmonisation. [65] The vast majority of commentators agree that in the field of the sui generis right, the CJEU rulings in British Horseracing Board, Fixtures Marketing, Directmedia and Apis make sense and therefore, there is no reason for national judges to depart from the EU case law. Comments on Football Dataco still have to emerge but to me, the decision makes complete sense. What else to do then but organise pan-European training courses for first instance judges? They do exist in the field of trademarks and patents: OHIM organises trademarks seminars for judges [66] and there is the Biannual European Patent Judges Symposium. No similar initiative seems to exist in the field of copyright. [67] The best thing to do in the meantime then is to start putting together some specialised copyright and related rights training for European judges... Any takers? I am certainly one!