Law and Information

At the Berkman Center, we released today a paper on the current state of implementation of the EU Copyright Directive (EUCD), with emphasis on the transposition of the provisions on the legal protection of technological measures sections (such as encryption, digital watermarking, copy-control technologies, and the like.) In this study, we have taken a closer look at the relevant definitions, exemptions, sanctions and remedies associated with the national anti-circumvention laws. What are the key findings?

First, our analysis reveals that uncertainty over the scope of provisions aimed at protecting technological measures as well as the definition of crucial terms (such as ‘effective measures’) persists – even at a rather basic level. The question, for instance, as to what extent access control mechanisms fall under the definition of technological protection measures and, as a consequence, are protected by the anti-circumvention provisions has been contested.

Second, the study explores different ways in which national implementations have addressed the problem of privately applied technological protection measures vis-�-vis the traditional exceptions to copyright within the framework as laid down in the EUCD. As demonstrated in the paper, incumbent member states have not made broad use of the possibility to take measures ensuring that private copying exceptions will survive technological protection measures, and have gone different paths as far as the implementation of the public policy exception as set forth by the EUCD are concerned.

Third, a brief analysis of some approaches to sanctions and remedies taken by EU member states suggests that member states have interpreted the relevant provisions of the EUCD – calling for “appropriate sanctions and remedies” – in different ways. While all countries impose civil sanctions in the case of a violation of anti-circumvention provisions, differences remain with regard to criminal sanctions. The regimes range from significant criminal sentences for both acts of circumvention and trafficking in circumvention devices and services to copyright laws that stipulate modest fines, but no imprisonment in the case of a violation of the anti-circumvention provisions.

The project website also provides an interactive chart with a resource page containing international and national legislation on technological protection measures with focus on the relevant laws of EU member states for further research. We intend to update the site as soon as new anti-circumvention provisions have been enacted in a EU member state.

I’ve just finished reading four important judgments by the European Court of Justice (ECJ) concerning the EU-Directive on the Legal Protection of Databases (“Database Directive”.) The judgments of the Court in the cases Fixtures Marketing Ltd v. Oy Veikkaus Ab, The British Horseracing Board Ltd and Others v. William Hill Organisation Ltd, Fixtures Marketing Ltd v. Svenska Spel AB, and Fixtures Marketing Ltd v. Organismos prognostikon agonon podosfairou (OPAP) relate to similar factual circumstances (for a summary, click here), i.e., databases of sporting information such as horse racing information in the British Horseracing Board case and football fixtures in the other three cases. Certain pieces of information from these databases were used by third parties for commercial gambling operations. In proceedings before the relevant national courts, the claimants alleged that these uses by the gambling operators were an infringement of the claimants’ sui generis database rights under the Database Directive. In each case, the national courts referred a set of question to the European Court of Justice. The questions raised by the national courts concern, inter alia, the definition of the term “database” in the Directive, the scope of protection (especially the “substantial investment”-requirement), and the infringement of the sui generis right through extraction or re-utilization. In today’s post, I’d like to focus on the definition of the term “database”.

What is a “database”?

In order to determine the scope of protection of databases under the Database Directive, one first has to consider the definition of the term ‘database’ as used in the Directive. According to Article 1(2) of the Database 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.” In Fixtures Marketing Ltd v. OPAP, the ECJ confirms that the term has a wide scope, “unencumbered by considerations of formal, technical or material nature.” (para. 20) In this context, the Court points out that electronic and non-electronic databases are protected under the Directive. Moreover, the ECJ confirms that the definition does not include quantitative requirements, i.e., that the definition does not depend on the question as to whether a significant number of data or materials is involved or not. (para. 24) Consequently, the definition as set forth by the Database Directive is broader than the one suggested in Section 2(5)(A) of the U.S. Database and Collections of Information Misappropriation Bill, where the term ‘database’ is defined as “a collection of a large number of discrete items of information produced for the purpose of bringing such discrete items of information together in one place or through one source so that persons may access them.” (Emphasis added.)

The ECJ further clarifies that it is not required that a database is its maker’s own intellectual creation to be classified as such. However, the criterion “originality” plays an important role in the assessment whether a database enjoys copyright protection under Article 3 of the Database Directive. (para. 26) In this regard, the standard laid down by the Database Directive is similar to Feist.

In sum, the classification of a database depends on three (analytically distinct, although related) elements:

(1) Collection of ‘independent materials.’ Article 1(2) requires, inter alia, that the database consists of a collection of independent works, data or other materials. This criterion, according to the ECJ, requires that the materials are separable from one another “without their informative, literary, artistic, musical or other value being affected.” (para. 29; emphasis added.) It is an open question whether or not this interpretation is more restrictive than the one suggested by the Advocate-General, according to which the criterion should be understood “as meaning that the data or materials must not be linked or must at least be capable of being separated without loosing their informative content.” (para. 39 of the Opinion, emphasis added.) The “independence”-requirement is particularly relevant in the context of musical recordings and with regard to movies. While it seems clear that the sound of a musical recording as such or the pictures of a movie do not fall within the scope of the definition (see also recital 17 of the Database Directive) because they are note separable without affecting the artistic value of the work, it remains unclear whether the compilation of musical recordings on a CD or the compilation of video clips on a DVD, for instance, is a database or not. The Advocate-General seems to suggest that a compilation of musical recordings does not qualify as a database; para. 29 of the Opinion.) I would argue, by contrast, that different musical recordings on a CD (think about a “Best Of” album as the prime example), by and large, are independent, systematically or methodologically arranged, and individually accessible works. Arguably, exception to the rule might exist, for instance in cases of interrelated “variations” on a theme, where the separately stored and individually accessible musical works cannot be separated without affecting the musical value of the works. In any event, the qualification of a compilation as a database is only a necessary, but not sufficient requirement for sui generis protection. In fact, the Database Directive seems to suggest that compilations of different musical works are databases, but are not protected due to lack of substantial investment as required by Article 7 (see recital 19).

(2) Systematic or methodical arrangement. The independent materials must be systematically or methodically arranged. Recital 21 of the Database Directive states that it is not necessary “for those materials to have been physically stored in an organized manner.” In essence, this criterion makes sure that randomly accumulated information does not fall within the scope of the definition. The ECJ clarifies that this second condition “implies that the collection should be contained in a fixed base, of some sort, and include technical means such as electronic, electromagnetic or electro-optical processes … , or other means such as an index, a table of contents, or a particular plan or method of classification, to allow the retrieval of any independent material contained within it.” (para. 30) According to the Advocate-General, it is sufficient “if a structure is established for the data and they are organised only following application of the appropriate search programme, and thus essentially through sorting and, possibly, indexation.” (para. 40 of the Opinion.) As a result, the threshold established by the “arrangement”-criterion is rather low. This conclusion is confirmed by previous case law in EU member states (for an overview, see the Database Right File by Bernt Hugenholtz.) In C-Villas, for instance, the Austrian Supreme Court held that a homepage with information on eight holiday houses located on the Caribbean island had been systematically arranged, since the websites were individually accessible and the villas described per Island, village, and based on the housing equipment. In the German case Babynet, a District Court held that a website with 251 alphabetically (sic!) arranged links is a systematically arranged collection of independent (and individually accessible) materials.

(3) Individually accessible. The third element of the definition set forth in Article 1(2) of the Database Directive requires that the independent materials making up the collection are individually accessible by electronic or other means. Against this backdrop, the Advocate-General argued that the mere storage of data is not covered by the term database. (para. 41 of the Opinion.) However, scholars have argued that it is unlikely that independent works, data and materials can be stored systematically or methodologically without being individually accessible. In the context of the present cases, the ECJ did not have reason to further explore this question. Instead, the ECJ discussed the second and third requirement in one paragraph and concluded that, together, they make it possible “to distinguish a database within the meaning of the directive, characterised by a means of retrieving each of its constituent materials, from a collection of materials providing information without means of processing the individual materials which make it up.” (para. 31)

In sum, the ECJ confirms the broad meaning of the term ‘database’ in Article 1(2) of the Database Directive. Thus, it is not surprising that football fixture lists have been considered to be a database within the meaning of the Directive. However, it is important to note that this is a necessary, but not a sufficient condition for the grant of the sui generis right set forth by Article 7 of the Database Directive.

In next posts, I will take a closer look at the sui generis protection in general and the “substantial investment”-requirement in particular, and discuss the infringement of the sui generis right through extraction or re-utilization.

Comments, as always, most welcome and much appreciated.

We are pleased to announce the next installment of the 2004/05 edition of the Harvard-Yale Cyberscholar Workshop Series. We invite you to our second session to be held on Thursday, November 11 from 6:00-8:00pm in room 111 at Yale Law School. We are delighted to host a session with two distinguished speakers:

Caio Pereira, JSD candidate, Yale Law School, Information Society Project Fellow, on “Universal Access Policies in Developing Countries“, and

Ivan Reidel, SJD candidate, Harvard Law School, Berkman Center’s Digital Media Team, on “Competition and Deregulation in the Music Industry“.

RSVP required.

Over the past few months, I’ve been working on a couple of Berkman papers on developments in international and national copyright laws as far as digital media is concerned. I’ve been looking at U.S. law, at the legislative and regulatory developments in Europe (both EU legislation and national implementations of EU law), and at selected jurisdictions in the Asia-Pacific region such as Australia, Singapore, Malaysia, China, Japan, and South Korea. The papers provide a rough overview of the copyright laws and regulations in these countries, and analyze the different paths and stages of evolution of the international copyright ecosystem. The reports also describe the current state of digital media law in action.

Currently, I’m working on one of the reports’ conclusion section. As always, it’s quite a challenge to come up with “crispy” takeaway points. Well, here’s the way I intend to frame it; any feedback is very much appreciated.

It seems to me that one might roughly distinguish between three stages of development of copyright protection across the world:

1) Copyright laws at a very nascent stage. Copyright laws of countries in this category do either not exist at all, or have not incorporated the relevant protection levels set forth by the Berne Convention or TRIPS.

2) Copyright (IP) laws that are TRIPS compliant.

3) Copyright laws that have incorporated the WIPO treaties or are otherwise in compliance with WCT/WPPT.

These categories, in turn, say something important about the driving forces of copyright legislation across the globe. By and large, TRIPS has become the major force aimed at creating a level playing field of IP protection in many parts of the world. It strikes me that TRIPS, today, is particularly important in the transition from “nascent” to “well-developed” copyright regimes. The implementation of the WIPO treaties, in contrast, can be understood as the “fine-tuning” of copyright legislation – from “well-developed” to “advanced”, so to speak. One other important driving force hasn’t been mentioned here, but is discussed in the papers: bilateral free trade agreements (such as the AUSFTA and the USSFTA.)

The second question is: What are the effects of the international treaties and bilateral trade agreements on copyright legislation? Well, it’s quite obvious: The international treaty system works as a leveler and harmonizes the fundamental issues and corner stones of a copyright system. However, the Berkman papers will also demonstrate that significant differences among national laws remain. Even if we look at countries that are in compliance with the WIPO treaties, we find different approaches, especially with regard to definitions (e.g. of technological protection measures), exceptions (e.g. private copying), and sanctions/remedies against infringements. Another addendum is necessary: The digital media landscape across the globe – even the most “advanced” – also varies significantly when it comes to the “law in action.” The reasons are manifold and include huge differences in civil and criminal procedure laws as well as, of course, economic, cultural, historical… differences in law enforcement practices. However, the papers will also illustrate that law enforcement – as far as online piracy is concerned – is increasingly an internationally orchestrated multi-actor (rightholders, rights organizations, governmental task-forces, etc.) effort.

The third question I’d like to touch upon is: what are the effects of international copyright frameworks and corresponding national legislation on any given information environment? Here, one might want to distinguish between four stakeholders: Users/consumers, rightholders, businesses, and policy-makers. Unfortunately, I do not have much empirical data to support potential claims about the effects of legislative developments on each category. However, anecdotal evidence suggests certain trends. In essence, one might argue that the “most advanced” copyright systems have a bias towards protection of rightholders and businesses at the cost of users. Especially the review of current case law in European jurisdictions suggests that this trend is not only a U.S. phenomenon, but structural in nature. (To be sure: the interactions among contract and copyright law, technology, and business models are rather complicated. You’ll find an analysis of the interplay between these elements in our iTunes case study.) One final point with regard to policy-makers: The papers will demonstrate that policy-makers – at the national level – still have some leeway in the way they design their copyright ecosystem, despite international harmonization.

Against this backdrop, my research interest moves towards “best practice models” in the digital age. Stay tuned. (BTW, I’ll link to the above mentioned papers as soon as they become available.)