This is not the first time that readers of the Kluwer Copyright Blog will have read about either the CDSM Directive or the reCreating Europe project. This post enriches the body of work in this area by summarizing the key findings of our two-phase empirical research on end-user flexibilities and end-user license agreements (EULAs) of online content-sharing service providers (OCSSPs). Our research seems to confirm that we are still a long way from saying that online service providers (especially OCSSPs) respect end-user flexibilities properly, even after the implementation deadline of the CDSM Directive.
A short summary of our two phase-mapping
First phase findings on the pre-CDSM implementation status quo
Our study on the status quo of the EULAs prior to the implementation deadline evidenced that, first, users were granted a more limited range of flexibilities with respect to the online use of intangible or service-like content than the use of works in the offline world. These flexibilities are narrowed down by the legislation itself (e.g., by the exclusion of the applicability of the doctrine of exhaustion, that is, the resale of lawfully acquired digital copies), but, more importantly, the examination of selected EULAs evidenced that platforms also tightened the grip on the potential uses of their services. For example, limitations or bans were placed on access to content on a geographical basis (“geo-blocking”) or secondary dissemination. Technical protection measures were strictly applied in many cases. EULAs remained either silent on some significant end-user flexibilities (e.g., freedom of expression-based exceptions and limitations, which might be covered by fair use in the US) or they were not clear enough on the practical application of those flexibilities (e.g., well-developed notice-and-take-down regime, but looser complaint-and-redress mechanisms). Similarly, various service providers applied misleading language, e.g., they spoke of ‘sale’, ‘purchase’ and the like, although the EULAs were purposefully limited to offer a license to the clients of the service providers. In sum, the majority of private regulatory provisions were asymmetrically in favor of the platforms.
Second, we also spotted certain effects that stemmed from the structure and business logic of the services, as well as the public laws of the European Union. For example, users of social media platforms (e.g., Facebook) as well as streaming platforms with hosting functionality (e.g., YouTube) exercised greater control – both at the upload and the access level – over the available content than users of streaming platforms without hosting functionality (the user-generated content (UGC) effect). Furthermore, service providers were limited by the existing copyright rules, but platforms that offer UGC provided an environment of greater flexibility (regulatory lock-in effect). Finally, the end-user experience was heavily affected by the fierce competition of various platforms. The horizontal (service-based, e.g., Facebook v Twitter) and vertical (company- or portfolio-based, e.g., Apple v Facebook) competition between service providers necessitates that they learn from each other, and sometimes overbid competitors’ offers. Quite a lot of end-user flexibilities stemmed from this competition, e.g., secondary dissemination, family sharing or UGC-sharing and further user benefits, e.g., subtitles (business flexibility effect).
Second phase findings on the post-CDSM implementation changes (or the lack thereof)
Our second phase empirical research on the EULAs of selected OCSSPs indicates a clear non-compliance with Article 17 in two main directions: first, by the exclusion of primary liability of platform operators and, second, by the lack of a balanced notice and takedown procedure that not only protects the legitimate interests of the right holders but also – via a complaint and redress mechanism – end-users’ rights as well.
The majority of the EULAs examined include guarantees to allow users to challenge the lawfulness of content removal, but the guarantees in Article 17 CDSM Directive do not appear expressis verbis, nor is there any specific reference to general prior content filtering mechanisms in the contractual terms. This is certainly instructive for two reasons. On the one hand, it seems that OCSSPs are sticking to well-established liability limitation clauses, shifting the liability to the end-user, thus weakening the viability of the new liability regime envisaged by the CDSM Directive. On the other hand, some platforms, such as YouTube, also actively filter uploaded content through their automated systems. They can remove content at their own discretion without notifying the right holders. But the balance between the actors concerned by the operation of the platforms – service operators, right holders and end-users – continues to tip in the direction of the first two stakeholders. At the same time, it is unclear how the platforms protect freedom of expression, freedom of creative creation and freedom of access to information, which have been among the main watchwords for criticism of the provisions of Article 17 CDSM Directive.
Although the transposition of the CDSM Directive was still underway in some Member States at the time of the completion of our studies (and, by the way, is still underway in multiple countries), the status quo seems to remain unchanged. Maintaining this status quo is clearly helped by the sporadic transposition of the CDSM Directive and by the fact that platforms with a North American background tend to align their practices to US copyright law.
We conclude that there are several policy recommendations that could support revisiting this field of law to provide due respect to end-users’ flexibilities.
The misleading terminology and asymmetric terms and conditions of the standardized EULAs of OCSSPs are worrisome not only in relation to copyright, but also from a consumer protection perspective as well. A review of whether service providers comply with the public regulatory sources regarding end-user flexibilities would be highly recommended.
The analysed OCSSPs – at the time of the review of their EULAs – diverged from the provisions of the CDSM Directive in multiple ways, especially regarding the contractual bypassing of liability under Article 17(4) CDSM Directive and the lack of introduction of end-user safeguards in line with Article 17(7) and (9) CDSM Directive. It is highly recommended that a stakeholders’ dialogue be initiated to check whether and how the private regulatory approach could be brought into conformity with the public rules.
OCSSPs’ current EULAs seem to bypass the proper application of the public regulatory sources. It is highly recommended that an analysis of whether such provisions also create any concerns from an private international law perspective (that is, whether the terms and conditions can be effectively enforced in line with the CDSM Directive’s new provisions) be carried out.
Finally, the CDSM Directive has added novel flexibilities regarding the creation and use of UGC (for example via parody, pastiche, review etc. exceptions). At the same time, EULAs are already more flexible with respect to the sharing and use of such content. It is highly recommended that there be a review of whether public regulatory sources need any further recalibration to meet the de facto online practices of end-users, as well as to move UGC into the public regulatory space rather than leaving it in the private regulatory space.
The paper on our first-phase research has been recently published: Péter Mezei and István Harkai: End-user flexibilities in digital copyright law – an empirical analysis of end-user license agreements, Interactive Entertainment Law Review, 2022, 5(1), p. 2-21. Our publication of the second-phase research is forthcoming in Public Governance, Administration and Finances Law Review (Fall 2022).