In Apis-Hristovich v Lakorda (Case C-545/07) the European Court of Justice has given further guidance on the database right and, in particular, on what factors to take into account when assessing whether an extraction is a substantial part of a database.
The Facts
Apis brought proceedings before the Sofia City Court for cessation of the allegedly unlawful extraction and re-utilisation by Lakorda of substantial parts of its modules ‘Apis pravo’ (‘Apis law’) and ‘Apis praktika’ (‘Apis case-law’), which form part of a general legal information system. Apis also sought compensation for the damage it had suffered.
Apis claimed to have made a substantial investment in the compilation, verification, systemisation and updating of the databases of the product modules, mainly by digitalisation, conversion, correction, technological processing and consolidation of the texts of legislative measures, and legal editing. It claimed that persons who previously worked in its software department before founding Lakorda unlawfully extracted substantial parts of its modules, which permitted the latter to produce and market, in September 2006, its own modules, called ‘Balgarsko pravo’ (‘Bulgarian law’) and ‘Sadebna praktika’ (‘Case-law of the courts’), which form part of the general legal information system ‘Lakorda legis’.
Lakorda was said to have extracted:
• the texts, in their consolidated version, of more than 19,700 documents, comprising normative measures that were then in force, measures amending or repealing earlier measures, and non-normative measures
• more than 2,500 documents, which were earlier versions of legislative measures from the period 2001 to 2006
• 2,516 unpublished judicial decisions, obtained by Apis with the permission of the relevant courts and which it had collected in the ‘Apis praktika’ module
• data connected with those documents, such as references between the documents and the legal definitions of certain terms and concepts.
This included 82.5% of the total number of documents contained in one product module. It was claimed that the unpublished decisions had a special value.
The fact that this unauthorised conduct took place was demonstrated, according to Apis, by the presence in Lakorda’s modules of features identical to those in the Apis modules, such as editor’s notes, references to translations of the documents into English, commands, fields, hyperlinks and the chronology of legislative measures.
Lakorda denied any unlawful extraction and re-utilisation of Apis’s modules. It contended that its ‘Lakorda legis’ system is the fruit of a substantial, independent investment. Establishment of the system involved a team of software specialists, lawyers and managers and is based on original computer programmes for the establishment, updating and visualisation of databases, allowing data to be processed and information to be accessed in a faster and more efficient way than other legal information systems. It claimed that its modules have a fundamentally different structure than those of Apis.
Lakorda also contended that, in setting up its project, it relied on its contacts with various national and European authorities and used publicly accessible sources, such as the Darzahven vestnik (Official Journal of the Republic of Bulgaria) and the official websites of national institutions and courts. That reliance explained the great similarity of the contents of its modules and those of Apis and the presence, although limited, of features similar to Apis’s modules regarding, in particular, references to translations and commands. Moreover, under the relevant Bulgarian law, official measures adopted by State bodies are not covered by the copyright rules. Lakorda also pointed out many differences between the systems and suggested that some similarities result from the generally applicable punctuation rules of the Bulgarian language.
The Referred Questions
The Sofia City Court referred the following questions for a preliminary ruling:
‘1. How are the terms “permanent transfer” and “temporary transfer” to be interpreted and to be delimited in relation to each other for the purpose of:
– determining whether extraction within the meaning of Article 7(2)(a) of Directive 96/9 … from a database accessible by electronic means has taken place[?]
– at what point in time is it to be assumed that extraction within the meaning of Article 7(2)(a) of Directive 96/9 … from a database accessible by electronic means has taken place[?]
– what is the significance, for the assessment of extraction, of the fact that the content of a database extracted in this way has served to create a new and amended database?
2. Which criterion is to be applied in interpreting the concept “extraction of a substantial part, evaluated quantitatively” if the databases are divided into separate subgroups and are used in these subgroups, which are independent commercial products? Is the size of the databases in the entire commercial product or the size of the databases in the relevant subgroup to be used as the criterion?
3. In interpreting the concept “a substantial part, evaluated qualitatively”, is the fact that a certain type of data allegedly extracted was obtained by the database maker from a source which is not generally accessible, so that it was possible to procure the data only by extracting them from the databases of that very database maker, to be used as a criterion?
4. What criteria are to be applied when determining whether extraction from a database accessible by electronic means has taken place? Can it be regarded as an indication that extraction has taken place if the maker’s database has a particular structure, notes, references, commands, fields, hyperlinks and editorial text and these elements are also found in the database of the person who has committed the alleged infringement? In the carrying out of this assessment, are the various original organisational structures of the two opposing databases relevant?
5. When determining whether extraction has taken place, is the computer program/the system for database management material if it is not part of the database?
6. Since, according to Directive 96/9 … and the case-law of the Court of Justice of the European Communities, “a substantial part of the database from a quantitative and qualitative point of view” is linked to substantial investment in the obtaining, verification or presentation of a database: how are these concepts to be interpreted in relation to legislative measures, and measures having individual application, which have been adopted by executive State bodies and are publicly accessible, to their official translations and to case-law?’
The judgment includes discussion of the admissibility of the questions which need not concern us here.
Extraction Generally
The topic of chief interest is the further guidance given by the court as to what constitutes a substantial part of a database. It is hard to see how the considerable attention paid to the difference between a temporary transfer and a permanent transfer is justified because it seems to have little bearing on the end result, although theoretically relevant to the level of damages. However, the judgment does seek to clarify a few points that most would already have thought clear.
The ECJ noted that it had already decided (see the British Horseracing Board at [51] and [67] and Directmedia Publishing at [34]-[38]) that extraction must be given a broad interpretation as referring to any unauthorised act of appropriation of the whole or a part of the contents of a database, the nature and form of the process used being immaterial. The objective pursued by the act of transfer is immaterial for the purpose of assessing whether there has been an extraction: ‘it is of little importance that the act of transfer in question is for the purpose of creating another database, whether in competition with the original database or not, or that the act is part of an activity, whether commercial or not, other than the creation of a database’. (This factor might however be relevant to damages.) A different arrangement or organisation of the extracted material is also irrelevant, especially bearing in mind the technical possibilities of reorganisation associated with electronic databases.
Dealing with the referred questions that were essentially evidential, the ECJ found as follows:
‘The delimitation of the concepts of ‘permanent transfer’ and ‘temporary transfer’ in Article 7 of Directive 96/9 is based on the criterion of the length of time during which materials extracted from a protected database are stored in a medium other than that database. The time at which there is an extraction, within the meaning of Article 7, from a protected database, accessible electronically, is when the materials which are the subject of the act of transfer are stored in a medium other than that database. The concept of extraction is independent of the objective pursued by the perpetrator of the act at issue, of any modifications he may make to the contents of the materials thus transferred, and of any differences in the structural organisation of the databases concerned.
– The fact that the physical and technical characteristics present in the contents of a protected database made by a particular person also appear in the contents of a database made by another person may be interpreted as evidence of extraction within the meaning of Article 7 of Directive 96/9, unless that coincidence can be explained by factors other than a transfer between the two databases concerned. The fact that materials obtained by the maker of a database from sources not accessible to the public also appear in a database made by another person is not sufficient, in itself, to prove the existence of such extraction but can constitute circumstantial evidence thereof.
– The nature of the computer program used to manage two electronic databases is not a factor in assessing the existence of extraction within the meaning of Article 7 of Directive 96/9.
Substantial Part
The judgment on this point is worth reading in full (click here) but the following headline points can be made.
- What is a substantial part must be assessed in relation to the size of the database from which the extraction is made, not the size of the database in which extracted material ends up.
- The level of investment in the creation of the materials has to be constantly borne in mind, for example (on these facts) that might make otherwise unpublished material of more significance when assessing the level of extraction because the acquisition of those materials may have required greater investment.
- The fact that a database is compiled from official ‘open access’ resources does not rob it of protection in terms of the sui generic database right.
