Copyright law trapped in the web of hyperlinking: the AG’s Opinion in the VG Bild-Kunst case (Part II)

Kluwer Copyright Blog
October 26, 2020

Please refer to this post as: , ‘Copyright law trapped in the web of hyperlinking: the AG’s Opinion in the VG Bild-Kunst case (Part II)’, Kluwer Copyright Blog, October 26 2020, http://copyrightblog.kluweriplaw.com/2020/10/26/copyright-law-trapped-in-the-web-of-hyperlinking-the-ags-opinion-in-the-vg-bild-kunst-case-part-ii/


The first part of this blogpost analysed the main theoretical foundations of the AG’s Opinion in the VG Bild-Kunst case (C‑392/19). The second part focuses on the application of this theoretical background to frame links and inline (automatic) links. As will be shown, the AG has proposed a methodological model of distinction between hyperlinks on the basis of technological and functional characteristics, while a decisive factor behind this differentiation is the intensity of the appropriation of a third party’s content.

a)  Access, but how? The modalities of access as a balancing factor

Probably the most innovative reasoning of the Opinion is the AG’s advancement of a theory on access, which is based on the substantial differences between linking techniques. By doing so, the AG proposes a limitation of the vague right of controlling access to content on the basis of the specificities and the modalities of providing access. In this context, the AG recommends a differential treatment for clickable and non-clickable links. Clickable links (both simple hyperlinks and framing) are a different but usual way of accessing content by the same authorised public. On the contrary, automatic, non-clickable links  are seen as a more intrusive intervention by the linker who plays a decisive role in communicating the linked work to a new public which was not taken into account by the copyright holder when the work was initially made available (par. 98 of the Opinion). For the AG, the automatic embedding of the work in the linker’s webpage ends up in a de facto appropriation of that work, since the disconnection of the work from its initial source is a priori hidden and brutal, while its result is the substantial enjoyment of the economic value of the work in a way that the rightholder cannot be presumed to have implicitly authorised.

The functional or economic equivalence of this act with acts falling within the rightholder’s control is also decisive. For a user, there is no difference between an image embedded in a webpage from the same server and one embedded from another website (par. 93 of the Opinion). Therefore, in so far as the effect of an automatic link is the same as that of a reproduction made available to the public independently there is no reason to treat them differently. Par consequent, it cannot be presumed that the copyright holder took that public into account when authorising the initial making available of the work (par. 95 of the Opinion).  Furthermore, and even more restrictively, this finding should also apply irrespective of whether the work is embedded in the form of a thumbnail or whether the embedded object is a miniature of the original work (par. 120 of the Opinion).

Accordingly, the AG excludes technological protection measures against framing from the protection of Article 6 of the Directive 2001/29, by clearly distinguishing these measures from the access restrictions to which the Svensson decision refers. This delicate exclusion is based on two interconnected arguments. First, there is a fundamental difference between the two restrictions. Access restrictions limit the circle of persons capable of having access to the work. Persons who gain access to it by circumventing those measures therefore constitute a new public (par. 128 of the Opinion). However, in the view of the AG, protection measures against framing restrict neither access to a work nor even a means of accessing it, but only a manner of displaying it on a screen. In other words, they are not a fence, but some closed windows in a house that can still be seen by the street. Second, and more substantially, for protection measures against framing there can be no question of a new public, because the public is always the same: the public of the website targeted by the link (par. 129 of the Opinion). Consequently, protection measures against framing, though lawful, are not eligible for protection under Article 6 of Directive 2001/29 (par. 136 of the Opinion). On the contrary, protection measures against the embedding of works from other websites are technologies designed to prevent or restrict acts which are not authorised by copyright holders because through a process of transformation of the work, that is to say of the code of the webpage containing that work, they give the copyright holder control over use of the work in the form of its embedding in another website (par. 135 of the Opinion).

The difference in the legal assessments for framing and automatic links is not technologically neutral and clear-cut. Neither is the distinction between access restrictions and protection measures against framing. However, it leads to a graduated and more calibrated approach on the legality of unconventional links, which is more compatible with fundamental rights. As noted by the AG, apart from the technical and functional differences between those two types of links, that distinction enables the best possible achievement of one of the objectives of Directive 2001/29, that is to say the objective of ensuring a fair balance between the interests of copyright holders and the interests of users. (par. 115 of the Opinion).

In a digital copyright law ecosystem where access to content is broadly being controlled, there is a need for the emergence of certain criteria able to delimit this control. Certainly, the focus on the technological and functional modalities of providing access (clickable and non-clickable links) is a straightforward solution, which can in practice serve as a safe guideline for linkers who are often lost when trying to understand the complex copyright law principles and their interpretations. On the other hand, this approach is tied to technical specificities, while in fact the most substantial argument supporting this thesis is a legal one, i.e. whether providing access to content results in the appropriation of the work or not. Considering the latter argument, the distinction between framing and inline linking appears a priori artificial. The AG admits this ambiguity, but at the same time proposes to clearly affirm the rightholder’s control only in the most flagrant cases of reaping the linked copyright protected content.

b)  Will the CJEU use the AG’s “Ariadne’s string”?

It remains to be seen whether the AG’s propositions will be followed by the CJEU to find a way out from the labyrinthine overlap between copyright law and hyperlinking. Svensson’s main rule of safeguarding linking activities when a work has been made available lawfully has become an emblematic principle of  EU digital copyright law, while a priori its application does not depend on linking technicalities, as the CJEU implicitly confirmed in BestWater (C-348/13).

The significance of linking that the CJEU has recognised for the sound operation of the Internet under a fundamental rights approach might count in favour of an overall exception for all kinds of links from the application of the right of communication to the public where the linked content has been first made available lawfully. If, however, the CJEU chooses to deepen the discussion, as the AG did, and to differentiate between linking techniques on the basis of the intensity of their interference with the copyright holder’s monopoly, it is also possible that both framing and inline linking will be treated similarly, even though this will result in a more profound reversal of Svensson’s “safe harbor” for linking activities.

Moreover, the AG’s rich Opinion opens the debate for a series of other fundamental, but also controversial, questions, such as whether it is possible to invoke technological protection measures as a sufficient proof of the rightholder’s intent to delimit the accessibility of his work (rightholders do not always control content protection policy or the use of the technological protection measures and collective management organisations may require the use of such measures without being expressly mandated by their members, see par. 125 of the Opinion). Likewise, the question of the application of contractual or licence restrictions as a solid proof of the author’s intent to exclude hyperlinking (even clickable links) remains unanswered. The AG’s focus on the concept of the author’s consent and state of mind regarding the public targeted by each act of making available of a work appears to favour the treatment of contractual restrictions in  a way similar to technical access restrictions. However, it might not be clear whether a disclaimer forbidding linking is imposed by the rightholder or the licensee. Furthermore, even if it is accepted that the rightholder’s  will to restrict access via some or all kinds of links is explicit and clearly indicated on the website containing the copyright protected content, the axiom of importance of hyperlinking, which forms part of the EU copyright acquis, might prevail. Otherwise, the extended use of hyperlinks by Internet users would be significantly hampered if they have to consult or interpret the website’s disclaimer before linking.