How to Write Bad Copyright Laws – Leaked Proposal for a New EU Copyright Directive

The EU Copyright regime has been subject to calls for reform for at least since 2008 (Green Paper). The reform process began in earnest in 2014 with the publication of thepublic consultation. The 2015 Digital Single Market Strategy put copyright high on the list of future interventions, promising a draft proposal by the end of 2016. The 2015 Copyright White Paper outlined the pursued aims in more detail, quoting wider access to content, more adapted exceptions, a well-functioning marketplace and an effective enforcement system. Few things in either the Strategy or the White Paper indicated a very radical approach except, perhaps, calls for a ‘follow-the-money’ mechanisms, which was neither defined nor properly explained.

A draft of the promised new Copyright Directive had been leaked a couple of days ago and so has the impact assessment that led to it. The final text of the proposal (which may be different from documents produced here) will be subject to co-decision procedure and intense lobbying both in the Parliament and the Council and is likely to get changed. Nevertheless, assuming that this is the starting position, what do these documents show about the proposed vision?

The proposed directive is meant to complement the Database Directive, Copyright Directive, Rental and Lending Rights Directive, Software Directive, Orphan Works Directive and the Collective Management Directive and not replace them.

It contains five titles, three of which bring new concepts in. Title II, adapting some exceptions and limitations to the digital cross-border environment, introduces new exceptions for text and data mining, use of works for teaching activities and preservation of cultural heritage. Proposed Article 3 allows text and data mining by research organisations of works to which there already is lawful access. Any contractual provision that says otherwise shall be ineffective. Text and data mining are defined as automated activities that generate patterns, trends and correlations. A typical text and data mining context is copying small amounts of material from journals and reports and assembling them for further use. A similar exception had already been introduced into UK law in 2014, although that one applies to non-commercial uses only whereas the Draft is silent on the commercial side of mining. Article 4 allows use of digital works for non-commercial teaching purposes only. Member States are allowed flexibility in implementing this provision. Finally, Article 5 allows cultural heritage institutions to make copies of works in their permanent collections for purposes of preservation only. Overall, the exceptions offered are welcome but exceptionally modest. Even a very cursory look at the literature will reveal a need for a far wider reform of the exceptions regime which needs to be reformed not only in terms of its largely voluntary nature but also its scope and depth.

Title III deals with licensing practices and wider access to content. It contains three relevant provisions. Articles 7 – 9 deal with uses of out-of-commerce works by cultural heritage institutions, specifically with licenses for non-commercial purposes that collective rights management organisations give to cultural heritage institutions for out-of-commerce works. These are relatively uncontroversial measures but are also relevant in a relatively narrow context. Article 10 attempts to cover agreements for making audiovisual works available to on-demand platforms (only partially covered by the above-mentioned Collective Rights Management Directive), but only says that such parties “may rely on the assistance of an impartial body”. This is a bizarre addition. Nothing prevents the parties from relying on such assistance at present and it is not clear how such a reminder is supposed to help anyone.

Title IV relates to well-functioning copyright marketplace. Article 11 introduces what the media (justifiably) calls the “link tax”. The article obliges Member States to provide publishers of news publications (defined very broadly in the Draft) with the rights contained in Articles 2 (reproduction right) and 3(2) (right of communication of the public) of the Copyright Directive for the duration of 20 years. This is essentially the creation of a completely new, previously non-existent neighbouring right. This measure is aimed at 3rd party aggregators (mainly but not exclusively Google) who show snippets of news and/or link to them. The heavily criticised idea would crete strains between USA-based aggregators and the EU and has potential negative freedom-of-speech implications. It seems to protect, for no apparent economic reason, the diminishing revenues at news companies at the expense of arguably more innovative models and is almost certainly going to be difficult if not entirely impossible to implement.

Most surprising, and most damaging, is Article 13 which covers uses of protected content by information society services (ISSs). It demands ISSs which store large amounts of works uploaded by users (i.e. user-generated sites) to take “appropriate and proportionate measures” to ensure that agreements with the rightholders are functioning, and to prevent the availability of works not covered by these agreements – including through use of “effective content identification technologies”. ISSs must provide rightholders information on such technologies. The provision, which in its present form is nothing other than a demand for monitoring, is in direct contravention of Article 15 of the E-Commerce Directive which explicitly prohibits monitoring. The drafters of this proposal are either not familiar with the 15-year-old largely uncontroversial article (nor, for that matter, with the content of the CJEU Sabam case) or are chosing to deliberately ignore it. Either is very alarming. The mechanism proposed here, even if the discrepancy is completely ignored, is unjustifiably tilted in favour of the rightholders. Further to that, only very large operators (such as YouTube) would have the financial and organisational power to install “effective content identification technologies”.

The proposed text has already been subject to heavy criticism (see here and here). In my own opinion, such criticism is entirely justified. The produced text is badly written, is ill-conceived and based on poorly tested ideas. Moreover, it contradicts the already established and well-functioning EU law (Copyright Directive, E-Commerce Directive, CJEU case law). The sooner it is changed (or abandoned altogether), the better.


BEREC Guidelines on Net Neutrality – Part II (Final Version)

On August 30, the Body of European Regulators for Electronic Communications (BEREC) published their final Guidelines on net neutrality, as mandated by the Connected Continent Regulation (Reg. 2015/2120). The Guidelines do not crate new rules but interpret the ones contained in the Regulation. Their importance lies in the fact that national regulatory authorities (NRAs) have been left in charge of the implementation and monitoring and this is intended to help them apply the Regulation and ensure a degree of uniformity.

A draft version had been made available on April 6 (I commented on it in a previous post) and the final text is a result of an extensive public consultation. I emphasised earlier that, although the Draft supports net neutrality in principle, it also allows zero rating, traffic management and specialised services under controlled conditions, which is a good starting point. The final version remains on the same position.

Public debate so far has been roughly divided into two camps: Internet Service Providers (ISPs) have argued for weak regulation while general public has been leaning towards a strong one. Although the final version does reinforce the open nature of the Internet, it does leave a lot of manoeuvring space for ISPs, essentially crating a relatively weak and rather bureaucratised set of rules that depends on NRA scrutiny.

Although, some comments seem to be claiming that “loopholes” left in the Regulation have been tightened, this is, in fact, not the case. The Guidelines allow for flexibility: zero rating, traffic management and specialised services, all remain allowed. The ISPs rely on all three as zero rating and specialised services are increasingly seen as lifelines in an industry where investing in new technology is prohibitively expensive but very risky and subject to competition from unregulated over-the-top (OTT) providers.

In respect of zero rating, which is not specifically mentioned in the Regulation, BEREC recognises that end users’ choice may be materially restricted. Commenting on Article 3(2) of the Regulation, however, the Guidelines do not call for their outright prohibition. Recitals 44 and 45 simply say that providers’ respective market positions are to be taken into account as well as the extent to which the users’ choice is actually been restricted. This is a regime which leaves balancing to the NRAs. The only situation in which zero rating is prohibited (Recital 55) is where, having reached the data cap, all applications but the zero rated ones are throttled.

The BEREC position on traffic management is dependent on the rather specific and relatively clearly worded limitations already imposed in the Regulation. The basic position here is that traffic management is allowed under controlled conditions. The main impression gained is that the Guidelines are descriptive, giving examples of what might constitute typical attacks or allowed measures.

Finally, specialised services are also allowed in Regulation Article 3(5) under NRA-controlled conditions and the Guidelines do not change the position in that article dramatically. Paragraph 111 comments that it is necessary for a specialised service to actually require a “level of quality that cannot be assured over a IAS”, that is to say, specialised service must be the only option for delivering the electronic product adequately. Paragraph 112 emphasises that defining a fixed list of specialised services is not necessary considering the fast pace at which technology develops.

In summary, the present Guidelines offer a basic net neutrality protection, preventing the ISPs from discriminating between apps and services flowing through the basic pipe. At the same time, they allow reasonable traffic management and specialised services, which is a positive development. More worrying is the fact that both are subject to far too many administrative hurdles and the uncertainty of regional NRA scrutiny. In my own view, net neutrality regulation was not necessary as potential problems could easily be addressed through the existing laws and adequate application of competition rules. At best, the present rules provide only a basic shield against (at present still largely hypothetical) violations. In the worst case, their overzealous application by NRAs would act as a burden to developement of new services.

An Overview of EU Cyberlaw Literature

I often get asked, by students and colleagues alike, what books to get and how to use them. Since Cyberlaw is, in the memorable words of Frank Easterbrook, “the law of the horse” – in other words, a collection of disparate legal disciplines, working with EU Internet Law does, unfortunately, mean working with many different disciplines. The situation is not so dire, however, as decent works exist in all of them and there are also some truly excellent ones. In this post, I will aim to give an overview of the books that I find particularly useful and work with relatively often. It is probably worth saying that, although Internet law changes relatively rapidly, books do not get out of date so easily. This is for a simple reason that EU Directives have a relatively long shelf life (E-Commerce and InfoSoc directives, for example, date to 2001) and the European Court produces a ´small number of judgments in each of the areas listed below.

Policy. No general books exist that cover EU Internet policy (although some references to policy and governance in individual areas are to be found in the works below). The following two books are not specifically EU oriented but they give a good overview of the main issues  involved. Ian Brown (ed.)’s Research Handbook on Governance of the Internet (Elgar 2013) is a good introduction to governance in cyberspace while Reed’s Making Laws for Cyberspace (OUP 2012) throws light on the law making process.

General literature. Few general works exist, although the situation is better now than it used to be. My own EU Internet Law (Edward Elgar 2013) is coming out in the 2nd edition later in 2016. It is complemented by the Research Handbook of EU Internet Law (Savin/Trzaskowski, Edward Elgar 2014). Neither is meant as a textbook and those looking for one should probably look at the Introduction to EU Internet Law (Trzaskowski/Savin/Lundqvist/Lindskug ExTuto 2015). A compendium of texts, cases and materials has been published in 2015 (Koščík et al. eds., available online) and is a useful collection of primary sources.

Those interested in the historical development of the main EU directives have a couple of volumes to consider. Lodder and Kaspersen’s eDirectives: Guide to European Union Law on E-Commerce (Kluwer 2002) remains the only article-by-article comment on the major EU Internet directives including the E-commerce, the Copyright and the Data Protection directives. Similar effect is achieved in Lilian Edwards’ (ed.) The New Legal Framework for E-Commerce in Europe (Hart 2005), which is an indispensable work for understanding policy decisions behind E-Commerce Directive.

Surprisingly, no up-to-date books covering electronic commerce only exist today and by far the best ones are the two volumes quoted above. Elgar’s Research Handbook on Electronic Commerce Law is due in September 2016 and will cover this gap to some extent.

Copyright is one of the areas where there is abundance of excellent books. A general work covering all EU IP Rights is Trevor Cook’s EU Intellectual Property Law (OUP 2010). Although no longer up to date in terms of case law, it gives a good overview of the basic directives. The Stamatoudi/Torremans detailed commentary on EU Copyright Law (Elgar 2014) is the most comprehensive article-by-article reference on all EU copyright-related directives including a section on EU policies. An earlier version with a similar scope (OUP 2010) is the Walter/von Lewinski’s European Copyright Law: A Commentary. An overview of the Member States’ implementation of the Copyright Directive is the Lindner/Shapiro Copyright in the Information Society: A Guide to National Implementation of the European Directive (Elgar 2011). Kur/Dreier European Intellectual Property Law: Texts, Cases & Materials (Elgar 2013) is excellent as a textbook in this area as is Pila/Torremans book with the same title (OUP 2016).

Trademark. Botis/Maniatis/Mühlendahl/Wiseman’s Trademark Law in Europe (3rd edition, OUP 2016) is a comprehensive overview of the area. Kur/Senftleben’s European Trademark Law is out in February 2017.

A review of the new EU patent system can be found in Pila/Wadlos (eds) The Unitary EU Patent System (Bloombsury 2015).

The most recent summary of cybercrime issues is Gillespie’s Cybercrime: Key Issues and Debates (Routledge 2015). A view of the EU cybersecurity policy issues is to be found in Christou’s Cybersecurity in the European Union: Resilience and Adaptability in Governance Policy (AIAA 2016)

Consumer issues are covered relatively comprehensively. Up-to-date general works exist, including Weatherill’s EU Consumer Law and Policy (2nd edition, Elgar 2014) and Micklitz/Reich EU Consumer Law (2nd edition, Intersentia 2014). Policy issues are covered in Leczykiewicz/Weatherill The Images of the Consumer in EU Law: Legislation, Free Movement and Competition Law (Bloomsbury 2016). Djurovic’s European Law on Unfair Commercial Practices and Contract Law (Bloomsbury 2016) covers the UCP directive. Twigg-Flesner’s (ed.) Research Handbook on EU Consumer and Contract Law is due in September 2016.

Private International law, while being subject to a multitude of general titles, has only a few specific ones that apply to the Internet only. An example is Svantesson’s Private International Law and the Internet (3rd edition, Kluwer 2013). Specifically consumer-oriented is Tang’s Electronic Consumer Contracts in the Conflict of Laws (2nd edition, Bloomsbury 2015).

Comparative law: Knowing what is happening in the United States is an important part of understanding where the EU laws come from and why they are written in the way they. A good general overview (including cases and materials) of American Cyberlaw up to 2010 is Bellia/Berman/Frischmann/Post’s Cyberlaw (4th edition, West 2011). Grimmelmann’s Internet Law: Cases & Problems (Semaphore Press, 2015) provides a more up-to-date look.

Books on privacy have traditionally concentrated on national implementation of EU laws. The current situation is made more complicated by the recent adoption of the General Data Protection Regulation, which fundamentally changes the current EU law and has yet to be subject to a comprehensive coverage in a book format. The best overview of the current law is to be found in EU’s own Handbook on European Data Protection Law (EU 2013). An up-to-date discussion of policy and constitutional issues are discussed in Hijmans’ The European Union as Guardian of Internet Privacy: The Story of Art 16 TFEU (Springer 2016). Further to this, there is Lynskey’s The Foundations of EU Data Protection Law (OUP 2015) which does cover some aspects of the initial GDPR proposal.

In addition to the above, the following three are also of interest.

Telecommunications law (or, as the EU itself puts it – electronic communications) is, strictly speaking, a separate branch of law that applies to carrier only (the wires) and not content on the Internet. Nevertheless, since convergence between content and carrier has been blurring this boundary for a considerable time now, some knowledge of the area is needed. Three comprehensive books exists on the subject. Scherer’s (ed.) Telecommunication Laws in Europe (Bloomsbury 2016) is an overview of national laws but it does begin with an introductory section on EU telecoms laws. The most comment-like treatment of the 2009 telecoms package is Nihoul/Rodford’s EU Electronic Communications Law (OUP 2011). A comprehensive analysis of the ex ante approach to telecommunications is Hou’s Competition law and Regulation of the EU Electronic Communications Sector (Kluwer 2012). Finally, my own EU Telecommunications law is due with Elgar later in 2017.

BEREC Guidelines on Net Neutrality – Part I

On June 6, the Body of European Regulators for Electronic Communication (BEREC) published a draft set of guidelines concerning Net Neutrality. The authority to draft them came from Regulation (EU) 2015/2120  –  the “Connected Continent Regulation” – whose Article 5(3) obliges BEREC to issue guidelines for the implementation of the Regulation by August 30, 2016, in cooperation with the Commission and after consulting the stakeholders. Since it is the national regulatory authorities (NRAs) in each Member State who are in charge of implementation, the guidelines direct NRAs on how to implement the Regualtion.This makes them rather significant in practice.

On its website, BEREC claimed that it received “almost half a million contributions” in response to its call. Considering the fact that net neutrality has been in the public eye for a while now, it is not surprising that BEREC Guidelines should attract attention, but it is curious that the number of responses (most of which were from ordinary citizens) should be demonstrated in this manner. The drafting process itself is not particularly transparent, with concerns that BEREC drafting process happens behind closed doors and under potential influence of lobbying.

The Connected Continent Regulation has as its main purpose the removal of roaming and the introduction of minimum rules on net neutrality and it is not a full revision of the 2009 EU Telecoms regime (which is promissed for December 2016). The Regulation introduces a couple of basic safeguards. First, the Regulation does not define net neutrality. The introduction of a proper definition was a hotly debated point in previous drafts with proponents of net neutrality arguing for a clear definition and the opponents against. The resulting regulation simply says, in Article 3(1) that

End-users shall have the right to access and distribute information and content, use and provide applications and services, and use terminal equipment of their choice, irrespective of the end-user’s or provider’s location or the location, origin or destination of the information, content, application or service, via their internet access service.

This statement, which roughly corresponds to usual definitions of net neutrality, simply means that internet traffic must be provided without discrimination based on origin or content of that traffic. At the same time, this does not affect agreements between providers and users on “price, data volumes or speed”. The main net neutrality point is then repeated in Article 3(3):

Providers of internet access services shall treat all traffic equally, when providing internet access services, without discrimination, restriction or interference, and irrespective of the sender and receiver, the content accessed or distributed, the applications or services used or provided, or the terminal equipment used.

The same paragraph allows “reasonable traffic management”. Reasonable are only those measures which are “transparent, non-discriminatory and proportionate” and which are not based on commercial considerations. Such measures cannot result in monitoring and must be temporary. Providers are in particular not allowed to:

block, slow down, alter, restrict, interfere with, degrade or discriminate between specific content, applications or services, or specific categories thereof

In principle, traffic management is allowed only when EU laws demand it, to preserve the integrity and security of the network or to prevent congestion.

Finally, the Regulation allows the providers to offer specialized services, which are not defined but which are “optimised for specific content, applications or services, or a combination thereof” where such optimization is necessary. Such specialised services must not be offered as a replacement for regular Internet services and must not result in the degraded quality of the former.

The draft guidelines comment on each of the articles. Some of the comments are crucial. Particularly important are those on zero rating, traffic management and specizlied services.

In response to bundling of free applications and services with the Internet access (such as, for example, offering free Spotify or HBO subscriptions), the Guidelines paragraph 33 indicates that these ought to remain legal as long as they are not priced differently than the rest of the traffic. Paragraph 35 specifically emphasizes that ISPs are not allowed to contractually ban specific content (e.g. VoIP). In response to zero-rating practices (where the provider does not count traffic associated with a particular application towards data cap, thus making them essentially free), the Guidelines differentiate between several situations. If all other services are slowed down except the zero rated ones, once the cap is reached, such practices would infringe Article 3(1) (as per paragraph 38). In respect of all other practices, the Guidelines call for a complex assessment exercise (paragraphs 43  – 45) the purpose of which is to determine to what extent the end-users’ choice if actually restricted. At present, there is little evidence that zero-rating is harmful and some evidence that it is beneficial.

In response to traffic management, BEREC simply clarifies the conditions in Article 3(3) without adding anything particularly controversial. It acknowledges, at the same time, the legality of traffic management. The guidelines do not insists on application-neutral technical measures, possibly leaving space for controversy. On the other hand, “permanent or recurring” measures may not be legal (paragraph 70) showing, at least, that traffic management measures cannot be used as a backdoor for openly violating neutrality.

In response to specialized services, BEREC emphasizes the fact that such services need to be carried at a specific level of quality and cannot be assured by standard best effort (paragraph 95). The job of verifying whether optimized delivery is objectively necessary is left to NRAs . If the level of quality can be adequately assured through a regular connection, the specialized service would be illegal. BEREC is careful to emphasize that what constitutes a specialized service today may not be so in the future. Creating black and white lists, therefore, makes little sense. As a way of example, paragraph 109 quotes Voice-over-LTE (VoLTE), linear broadcasting IPTV services and real-time health services, as those which may required specizlized access today.

There are two important points to note from the Draft Guidelines. First, they follow the spirit of the Regulation by prohibiting opportunistic discrimination. This, in itself, is not necessarily a novelty since current laws can be interpreted to give the same effect. Additionally, this intervention may be irrelevant as there is little evidence of actual net neutrality violations in Europe. Second, they are permissive and flexible towards issues that matter to the industry – traffic management and specialized services. This is a positive development, as restrictive rules can and do have a negative impac on investment which, in telecommunications industry, is highly risky.

The comments this author has seen can roughly be divided into three groups. The first are individual citizens who invoke net neutrality as a fundamental right, often with little or no understanding of what it actually implies and how problems that arise from net neutrality issues differ from those happening on the backbone. The second are proponents of net neutrality on both sides of the Atlantic, who argue for stronger rules and criticize both the Regulation and what they perceive as BEREC’s permissive position. The third are industry representative who are against strong regulation. They are against overly prescriptive ex ante regulation, they fear future litigation over ambiguities left in the Regulation and worry about a potential slow-down in investment. A manifesto co-signed by BT Group, Deutsche Telekom, Ericsson, Hutchison Whampoa Europe, Inmarsat, Nokia, Orange, Proximus, Royal KPN, SES, Tele2 AB, TIM – Telecom Italia, Telefonica, Telekom Austria Group, Telenor Group, Telia Company and Vodafone, and claiming support from Ahlers, Airbus Defence & Space, Royal Philips, Siemens and Thales Alenia Space recently warned that no invetsment in 5G networks will be forthcoming unless uncertainties introduced in the Guidelines are removed.

In the view of this author, net neutrality legislation is unnecessary, counterproductive and technically difficult to implement. The final version of Guidelines may, if drafted carefully, eliminate some of the problems. I will comment on these when they are published later this month.





The Commission’s New Proposals on the Digital Single Market, May 2016

In May 2015, the European Commission published its Digital Single Market Strategy. The document outlined a number of targeted actions which the Commission was to undertake in 2015 and 2016. The Strategy consists of three pillars entitled: I) Better access for consumers and businesses to digital goods and services across Europe, II) Creating the right conditions and a level playing field for digital networks and innovative services to flourish and III) Maximising the growth potential of the digital economy. Behind the somewhat non-descript names hides a desire to address a number of relatively serious problems. In the first pillar, this includes simplifying and increasing the volume of e-commerce across EU borders, dealing with the problem of geo-blocking,  reforming copyright rules and reviewing the Satellite and Cable Directive. The second pillar has the task of reforming the telecoms rules, the audiovisual media framework (AVMS Directive) and the role of online platforms. In the third pillar, the Commission looked at free flow of data and standards and interoperability.

At the time of publication the Commission specifically underlined the removal of geo-blocking and the creation of level-playing field as being important. The first was, at that time, understood in its broad sense, as including both geo-blocking in sale of goods and services and geo-blocking that stems from differences of copyright law but very soon the attention concentrated on the former only (considering the difficulties that would arise form attempting to remove the latter). The level-playing field, on the other hand, refers to a complicated issue of the relation of OTT players and the incumbent telecommunications companies.

The first proposals resulting from the 2015 DSM Strategy came already in December 2015. They included a successor to the failed Common European Sales Law (CESL) – in the form of the Proposal for a Directive on certain aspects concerning contracts for the supply of digital content, coupled with a Proposal for a Directive on certain aspects concerning contracts for the online and other distance sales of goods. In addition to this, and delivering partially on the promise to remove geo-blocking, a Proposal on ensuring cross-border portability of online content followed.

Today, partially following on the promises made in the 2015 Strategy, the Commission made five new proposals.

  1. In the area of geoblocking, it proposed a Regulation on addressing geoblocking and other forms of discrimination based on customers’ nationality, place of residence or place of establishment. The gist of this proposal, which does not address copyright or AVMS issues, is found in Articles 3 and 4. Article 3 prohibits  measures which limit access to traders’ online interfaces for reasons related to nationality, place of establishment or place or residence of the customer. Article 4 outlines 3 situation in which geo-blocking is illegal: a) when the goods are not delivered cross-border by the trader or on his or her behalf; b) where the trader provides electronic services (other than matter subject to copyright), c) where the goods are physically supplied on traders’ premises but not in the customers’ place of residence. In any case, the regime does not apply to certain SME enterprises.
  2. In the area of cross-border parcel delivery, it proposed a Regulation on cross-border parcel delivery services. The proposal increases price transparency and regulatory oversight of cross-border parcel delivery.
  3. In the area of consumer protection, a Regulation on cooperation between national authorities responsible for the enforcement of consumer protection laws has been proposed. Among other things, it allows for immediate take-down of websites hosting scams.
  4. In the area of audio-video media services (AVMS Directive), a new Directive is being proposed. The new regime has better rules for protecting minors from harmful content and for protecting all citizens from incitement to hatred. The proposal also aims to improve the definition of the country of origin and introduces a minimum 20% limit on European content in broadcasts.
  5. In the area of online platforms, the Commission will maintain the existing intermediary liability regime, found in Article 12-15 of the E-Commerce Directive. Instead, in a Communication, it proposes to address the issues of intermediary liability through the updated AVMS Directive, in the coming copyright reform package and through  encouraging self-regulatory efforts by online platforms.

There is no doubt that at at least the geo-blocking and AVMS proposals bring significant changes to the present picture. I will address these proposals in more detail in the weeks to come.

Interconnecting content on the Internet lawfully – CJEU and hyperlinking of copyrighted material

On April 9, an important CJEU case reached the Advocate General Opinion stage. The case involves Sanoma, publishers of the Playboy magazine, who have commissioned nude photographs of a certain Ms Dekker. GS Media is the publisher of a blog which carried a link to a file containing the photos hosted on the Australian file depository system Filefactory removed the photos on demand from Sonoma but GS Media refused to remove the link on their website.

The plaintiffs brought an action, claiming that the publication of the links constituted copyright infringement. The Supreme Court of the Netherlands referred several questions to the Court of Justice.

‘1      (a)            If anyone other than the copyright holder refers by means of a hyperlink on a website controlled by him to a website which is managed by a third party and is accessible to the general internet public, on which the work has been made available without the consent of the right holder, does that constitute a “communication to the public” within the meaning of Article 3(1) of Directive 2001/29?

(b)      Does it make any difference if the work has also not previously been communicated, with the rightholder’s consent, to the public in some other way?

(c)      Is it important whether the “hyperlinker” is or ought to be aware of the lack of consent by the rightholder for the placement of the work on the third party’s website mentioned in 1(a) above and, as the case may be, of the fact that the work has also not previously been communicated, with the rightholder’s consent, to the public in some other way?

2      (a)   If the answer to question 1(a) is in the negative: in that case, is there, or could there be deemed to be, a communication to the public if the website to which the hyperlink refers, and thus the work, is indeed findable for the general internet public, but not easily so, with the result that the publication of the hyperlink greatly facilitates the finding of the work?

(b)      In answering question 2(a), is it important whether the “hyperlinker” is or ought to be aware of the fact that the website to which the hyperlink refers is not easily findable by the general internet public?

These, essentially, boil down to a question of whether posting a link to a website amounts to a communication to the public in terms of Article 3(1) of the Copyright Directive.

These questions have already been addressed in two previous CJEU cases, decided in 2014.

In Svensson and Others (C‑466/12, EU:C:2014:76) the Court ruled that “the provision on a website of clickable links to works freely available on another website does not constitute an ‘act of communication to the public’, as referred to in that provision.” This was because the work had not been communicated to a new public nor had technical means that differ from the original ones been employed. Importantly, the Court  concluded that Member States are not at a discretion to give Article 3(1) a wider interpretation than it already has.

Likewise, in BestWater International (C‑348/13, EU:C:2014:2315), the Court said that “framing” of one website (YouTube) inside of another does not constitute communication to the public in terms of Article 3(1). In this case, the initial communication had not been authorized but the Court did not give further clarification as to how that fact might change the decision.  Like in Svensson, the Court concluded that no new public had been granted access to the work and no new technical means had been used.

Neither case had really clarified the situation concerning linking to a potentially unlawful source nor whether the hyperlinker’s awareness of the illegality played a role. The AG is acutely aware of this in paragraph 43.

The present case thus gives the Court an opportunity to determine whether the authorisation by the copyright holder of the making available of the protected work on the other website (‘initial’ communication) is essential to a finding that there is no communication to the public within the meaning of Article 3(1) of Directive 2001/29.

In the view of the AG, in order to be considered a “communication to the public” as per Article 3(1), the intervention of the hyperlinker must be “vital or indispensable”. In comparison to Premier League case, where no football matches would have been shown had the public house proprietors not taken the steps they took, the hyperlinkers here merely facilitate access to what is, essentially, already there. In the view of the Advocate General, the intervention of GS Media is not indispensable for getting hold of the photographs which are already stored, legally or not, on Filefactory’s servers. For that reason, no “act of communication” had taken place, as required by Article 3(1).

The second part of Advocate General’s analysis concerns the analysis of the “public” to which the work had been communicated. Following the “new public” line of reasoning, the AG concludes that, since no initial authorisation happened here, the “new public” criterion is of no relevance. Although the website’s audience is wide, the photographs have already been made available on other websites. The importance of that becomes apparent in paragraphs 73 and 74:

  1. In any event, it is clear from the judgment in Svensson and Others(C‑466/12, EU:C:2014:76, paragraph 31) that if a hyperlink makes it possible for users of the site on which it appears to ‘circumvent restrictions’ (43) put in place on third-party websites to limit access to protected works, (44) the hyperlink in question constitutes an indispensable intervention without which those users could not enjoy the works. Accordingly, that intervention makes the works in question available to visitors to the website in question, in this case visitors to the GeenStijl website, and therefore constitutes an act of communication to a public which must be authorised by the copyright holder pursuant to Article 3(1) of Directive 2001/29.

  2. On the other hand, it is clear from paragraph 31 of the judgment in Svensson and Others(C‑466/12, EU:C:2014:76) that it is not sufficient that the hyperlink facilitates or simplifies users’ access to the work in question. […]

In analysing other circumstances which may be relevant, the AG  says that “[…]as a general rule, internet users are not aware and do not have the means to check whether the initial communication to the public of a protected work freely accessible on the internet was effected with or without the copyright holder’s consent.”

The final conclusion is mostly in agreement with the Svensson line of cases. The conclusion is that posting a link to unauthorized content is not communication to the public, that it is irrelevant whether posters were aware that authorization had not been given and that hyperlinking which “simplifies” access rather than find new audience is not communication to the public.

In our own opinion, this approach leaves some important questions unanswered. The AG indicated that the link poster’s awareness of the illegality of the content linked does not play a role since either the conditions in Article 3(1) are fulfilled or they are not, in which case awareness cannot play a role (paragraph 63). This may be so but such an interpretation does not distinguish between two types of situations in which hyperlinking can be damaging. The first is the situation in this case where the primary infringement is taking place on a server not connected with the hyperlinker and where the hyperlinker is referring to something that is very likely already in distribution in some form or another. The hyperlinker here is, if not entirely neutral, certainly not primarily a promoter of the infringement.

The second type of situation is entirely different and refers to websites the sole purpose of which is to distribute illegal copies of audio-video files otherwise in some form of commercial distribution. In this scenario, an operator of a forum or blog is posting links to file deposit sites, such as the one in question here with full knowledge and awareness that these files are distributed in violation of copyright law. Most of these files may, indeed, also be linked from other forums. Applying the Svensson line of cases, a conclusion could be reached that a) since there is (largely) no new audience and no new technical means and, since this is just a simplification of (an illegal) distribution already happening, there is no making available; b) the complete lack of good faith on the hyperlinker’s part is not relevant and c) if the right holder desires to commence legal action for infringement or obtain an injunction, they may do so against the primary infringer which is, presumably, the person or persons posting on the file depository which, in its turn, is subject to general ISP liability protection.

It appears that the suggestion in paragraph 73 quoted above, which refers to Svensson paragraph 31, does not resolve the tensions here. The Svensson paragroah 31 says:

‘where a clickable link makes it possible for users of the site on which that link appears to circumvent restrictions put in place by the site on which the protected work appears in order to restrict public access to that work to the latter site’s subscribers only, and the link accordingly constitutes an intervention without which those users would not be able to access the works transmitted, all those users must be deemed to be a new public, which was not taken into account by the copyright holders when they authorised the initial communication, and accordingly the holders’ authorisation is required for such a communication to the public’. (emphasis added)

What is promoted here is “indispensable” hyperlinking as opposed to “facilitating” hyperlinking. The former is an act of communication to the public, the latter is not. The problem arises from the insisting on the existence of copy restrictions. If such restrictions (as per Article 6 of the Copyrigth Directive) had been put in place, the hyperlinking is a communication to the public. If not, it will fall under the non-essential “facilitating” kind.

Such a conclusion cannot possibly be right for at least two reasons.

First, why would the right holder only be granted protection where they have availed themselves of the possibilities Article 6 offers? In that scenario, a right holder who distributed through, say, iTunes, would be protected, whereas the one who broadcasted on TV or streamed very likely would not.

Second, copyright violations on the Internet are usually facilitated through multiple fora which all link to the same files on the same depository sites (often multiple ones). No distribution would here be possible at all, if it was not for these hyperlinks since depository files do not themselves offer search facilities. In the same manner, the reason why PirateBay and similar sites are the usual targets of copyright infringement lawsuits is precisely because the essential element of illegal file distribution is their facilitator function.  In other words, the facilitator function is often the essential element in the copyright violation chain irrespective of whether there is a “circumvention restriction” or not.

All the previous cases are essentially about the extent to which content on the Internet can be interconnected. The AG’s conclusion in this case preservers the functionality and importance of hyperlinking as an important tool through which modern Internet functions but it fails to address a host of other cases where hyperlinking has no other purpose than to distribute illegal files. It remains to be seen to what extent the Court accepts his opnion.

Do network-wide ad-blocks really violate Net Neutrality? – An EU perspective

Ad blocking has been on the rise for a number of years. The 2015 Ad Blocking Report claims that ad blocking is estimated to cost publishers $22 billion in 2015, the year in which it has grown 41% compared to the previous year. At the same time, ad blocking penetration stood at 15% in the USA and 25% in Germany. And it keeps growing.

While ad blocking as a problem has so far concentrated on the users and their deployment of blocking software on the perriphery, several sources (see e.g. FT article) have recently reported that EU mobile operators intend to introduce network-wide advertisement blocking. One of the first credible threats came in February 2016 from the mobile operator Three. They rely on a technology produced by an Israeli company called Shine.

Blocking ads on mobile networks seems to be a result of increasing frustration of mobile operators who see ad networks enjoy increasing profits without the need to invest in the infrastructure. Mobile internet ad spending is set to stand at about $100 billion in 2016 with a projected increase in the years to come and network operators seem to want a share of that cake. Platforms which rely on ads, such as Google, on the other hand, argue that blocking makes no sense since content production which they engage in also requires investment and relies heavily on income generated by advertising.

Do network-wide ad-blocks violate the principle of net neutrality?

First, net neutrality is a complex set of problems that tends to get oversimplified and politicized. In reality, few understand it properly and even fewer have a grasp of the relationship that law has to technology. To put it in very simple terms, net neutrality is a principle which commands Internet Service Providers to treat all traffic equally, irrespective of the type, content, platform, application, etc. It can be understood as a principle that describes how Internet operates at present (it possesesnon-discriminatory properties) but it can also be understood as a regulatory principle (legislation is needed to protect net neutrality). Often the two come together in a (rather political) claim that, since the Internet is “neutral”, legislation is needed to addressed real or perceived threats to that neutrality. Importantly, the regulation of interconnection and peering is, by definition, not covered by net neutrality. In other words, net neutrality only controls what is happening on the “last mile“, between an ISP and the user. It says nothing about how ISPs regulate their own interconnection to other providers. Such relationships are only subject to contract law and competition law.

There are, at present, two opposing camps in relation to how net neutrality ought to be regulated. One is claiming that net neutrality is necessary to preserve the free and open Internet such as we know it while the other is saying that introducing legislation to protect net neutrality hampers investment and is a solution in search of a problem. This author tends to agree with the latter view.

The US and the EU have different views on net neutrality. While the US introduced a relatively strong form of net neutraluty in 2014, the EU opted for the weak one in its 2015 telecoms reform (effective from 30 April 2016). The EU version allows for fast lanes, as long as there is no discrimination in the main internet pipe.

How would EU rules react to the ad blocking problem? The key provision is Article 3(3) in the new Regulation:

Providers of internet access services shall treat all traffic equally, when providing internet access services, without discrimination, restriction or interference, and irrespective of the sender and receiver, the content accessed or distributed, the applications or services used or provided, or the terminal equipment used.

The first subparagraph shall not prevent providers of internet access services from implementing reasonable traffic management measures. In order to be deemed to be reasonable, such measures shall be transparent, non-discriminatory and proportionate, and shall not be based on commercial considerations but on objectively different technical quality of service requirements of specific categories of traffic. Such measures shall not monitor the specific content and shall not be maintained for longer than necessary.

Providers of internet access services shall not engage in traffic management measures going beyond those set out in the second subparagraph, and in particular shall not block, slow down, alter, restrict, interfere with, degrade or discriminate between specific content, applications or services, or specific categories thereof, except as necessary, and only for as long as necessary, in order to:

  1. comply with Union legislative acts, or national legislation that complies with Union law, to which the provider of internet access services is subject, or with measures that comply with Union law giving effect to such Union legislative acts or national legislation, including with orders by courts or public authorities vested with relevant powers;

  2. preserve the integrity and security of the network, of services provided via that network, and of the terminal equipment of end-users;

  3. prevent impending network congestion and mitigate the effects of exceptional or temporary network congestion, provided that equivalent categories of traffic are treated equally.

This provision makes it obvious that “discrimination, restriction or interference” is not allowed but also that “reasonable traffic management rules” is. Such traffic management must be “transparent, non-discriminatory and proportionate” and cannot be based on “commercial considerations”.

There is mounting evidence that mobile ads do gobble data and slow traffic (see here and here). Ender Analysis released a report in March 2016, claiming that between 18% and 79% of traffic is attributed to mobile ads. All that is required, in other words, is for the operator to show that ad traffic is causing disturbances in network traffic. This would almost certainly be allowed to stand even if the ultimate motive is commercial, i.e. if the real drive is an increase share of mobile ad revenues.

The key to understanding this problem is in the nature of the ads in question. Mobile ads are demanding on traffic and integrated with the content. They bring data congestion, security and privacy risks. As such, they would fit perfectly within letter b) (integrity and security of network). The Ender report further says “On the basis of this investigation, an estimate that says advertising accounts for half of all data used by publisher pages on iPhones does not look unreasonable. Publisher mobile pages are bloated, and advertising is an enormous part of that.”

It is entirely possible that network-wide ad blocking may have implications that arise out of contract law, advertising law or unfair competition law.  In the view of this author (although there are also opposing views) blocking ads on a network basis is not a net neutrality issue and is easily justified by reasons of traffic management.  A preliminary answer will be provided by Body of European Regulators for Electronic Communications (BEREC), who is supposed to published their draft net neutrality guidelines at some point in June 2016. The final answer will almost certainly have to wait for a Court of Justice reference.