On 12 November 2020, Dr João Pedro Quintais explained the problems with Article 17 at the CIPIL Evening Webinar: ‘Article 17 and the New EU Rules on Content-Sharing Platforms’. With the recent developments on the topic, this webinar was much needed.
First, Quintais draw attention to how Article 17 is the tip of the upcoming Digital Services Act iceberg, due to tackle illegal content online. After explaining where we are today with Article 17, he delved into the complex concepts embedded in Article 17, like the notion of Online Content Sharing Service Providers (OCSSPs) and the mitigation measures of Article 17(5)-(9).
This presentation was rounded off with a reference to the implementation recommendation written by Quintais, Frosio, van Gompel, Hugenholtz, Husovec, Jütte and Senftleben to minimise the risks of broad filtering and over-blocking: filtering and blocking should be limited to the files which are identical or equivalent to the “relevant and necessary information” previously provided by the rightholders to OCSSPs. The absence of this match should result in leaving the content available to the public on the platform pending a decision regarding its unlawful nature. In the case of partial matches, platforms should offer a user declaration or flagging scheme, so that users can declare that their upload is covered by a copyright exception or limitation. Most importantly for the partial match scenarios, where the match does not meet the prima facie infringement threshold, the upload should not be automatically blocked. This would provide a safeguard for the protection of freedom of expression and information online.
The webinar on ‘Opening Copyright Exceptions Through Fundamental Rights’ hosted by Sean Flynn and the Program on Information Justice and Intellectual Property at Washington College of Law of American University took place on 20 November 2020. Professor Christophe Geiger gave a thought-provoking presentation on the interesting subject of “Fair Use” through Fundamental Rights in Europe indicating that the concept of fair use already exists within the case law of the ECtHR, as well as the national case law of some Member States. First, he described the current situation of copyright exceptions within the EU law by emphasising the case law of Spiegel Online, Pelham and Funke Medien. He noted that in Funke Medien, AG Szpunar took a favourable approach towards freedom of expression as an external limitation, although unfortunately the CJEU took a more restrictive approach. That being said, Professor Geiger highlighted that the recent trend in European jurisprudence of interpretation in light of fundamental rights is a departure from the traditional categorical immunity of copyright law from any external freedom of expression evaluation.
The conclusion was that a fair use limitation is already in place with the balancing test introduced through Article 10(2) of the ECHR and Article 52(1) of the Charter. There is rich case law, especially from the ECtHR, for examining limitations on the exercise of the rights and freedoms. As a big fan of the fair use exception, I was happy to hear that EU copyright law does not particularly need a ground-shaking new exception; we can use the Charter to craft one. I am hoping that Member States take this into account when implementing the CDSM Directive. At least they should read Geiger’s ‘“Fair Use” Through Fundamental Rights In Europe: When Freedom Of Artistic Expression Allows Creative Appropriations And Opens Up Statutory Copyright Limitations’.
Modern Poland Foundation’s CopyCamp 2020 was full of interesting presentations. On 25 November 2020, the topic of the day was ‘Content Filtering and other Internet Disputes’. Julia Reda gave a presentation on Article 17 and asked the question on everybody’s mind: What is it Really Good for? First, Reda underlined the fact that Article 17 has already been challenged before the CJEU with case C-401/19 Poland (see here). She then provided a piece of insider information: apparently, during the hearing, the AG asked parties “if you are required by the law to go from Luxemburg to Brussels in two hours, do you really have a choice whether to use a car or go by foot?”
Reda examined the main concepts regarding the problems with Article 17. In relation to its impact on freedom of expression and information, she underlined the issue of false positives, especially when it comes to creative commons and public domain. She then explained the contradictory nature of Article 17, which incorporates a ‘best efforts’ obligation and an obligation to guarantee the unavailability of illegal content. Reda noted that during the Poland case hearing, the EU Commission argued that only manifestly illegal content should be removed under Article 17, since otherwise the obligation to guarantee the availability of legal content would be violated. Yet, Spain and France argued that everything that automated filters found illegal should be blocked. This is a highly problematic statement which completely ignores the relationship between time and online freedom of expression. The suggestion that redress mechanisms can restore the damage to end-user rights is arguably naïve.
The insufficiency of Article 17’s safeguards was underlined. Reda stated that there are no mechanisms described in the Directive for how to achieve the requirement “to guarantee the availability of legal content”, unlike for the best efforts obligation, i.e. the removal of the illegal content. There was then a moment of silence for the safe harbours; they are long gone, according to Reda, and Article 17’s impact on smaller platforms will be dramatic, since they will be forced to invest in costly filtering systems. Finally, with regard to national implementations, Reda pointed out that the CJEU will not deliver its decision in the Polish challenge before the implementation deadline, so even if the CJEU decides that Article 17 violates the Charter, it will be too late, as legislation will not be automatically invalid on the national level.
Following Julia Reda’s presentation, Dr Bernd Justin Jütte and Giulia Priora also gave a must-watch presentation titled ‘Teaching Exceptions for Exceptional Circumstances – Learning from the pandemic for a proper implementation of Article 5 DSM Directive’. As they highlighted, the new teaching exception obliged Member States to offer protection for digital uses. This is particularly crucial at the moment, when teaching has moved online because of the pandemic.
Lastly, Edyta Bednarczyk talked about the German draft implementation of Article 17. Bednarczyk noted that, in addition to the statutory copyright exceptions and limitations, the German proposal also contains a second copyright exception regarding the non-commercial uses of the copyrighted work. This new exception for “mechanically verifiable uses authorized by law” covers the non-commercial use of third-party material below a quantitative threshold (20 seconds of video or audio, 1000 characters of text, or individual images up to a file size of 250 KB.) This “de minimis exception” presents potential problems, since even that minimum use may constitute a valuable part of the work.
The draft’s 90 per cent match rule (i.e. the assumption of false flagging where the content matches the rightholder information by 90% or more) is not always applicable (consider e.g. creative commons licenses or works in the public domain). Moreover, there are no safeguards within the German proposal regarding the prevention of faulty blocking of such content. These faults aside, the overall conclusion was that, from the perspective of controlling over-blocking, the German draft is satisfactory.
3 December 2020 was the day of the SIPLR conference on ‘IP in the Digital Environment’. The conference started with the hot topic of website and streaming blocking injunctions, presented by Lord Justice Richard Arnold. The presentation started with a brief explanation of primary and accessory liability. It was stressed that blocking injunctions result from neither, but rather represent a special type of liability, as platforms are targeted because they are identified as “best placed” to end an infringement. Arnold LJ then gave examples from UK case law regarding platform injunctions, most importantly the Cartier case. He stated that the criteria developed in that decision were derived from Article 3(2) of the Enforcement Directive, with the exception of the fair balance requirement, which was based on CJEU case law. He pointed out the difficulties caused by the different national interpretations and implementations of Article 8(3) InfoSoc and Article 11 Enforcement Directive, that have been only partially bridged by the SABAM and UPC Telekabel decisions. Inevitably, the problems with the notion of communication to the public were underlined. Arnold LJ explained how national courts have different views on efficacy and proportionality, as they don’t have extensive knowledge on website blocking injunctions, which only feeds the lack of harmonisation. The presentation was rounded off by stressing the need for greater consistency and a harmonised legal framework for online injunctions to solve this international problem.
A safe harbour presentation with the smart title of ‘Roasting the Host’ was delivered by Eleonora Rosati. First, Rosati took us down memory lane with a timeline of developments regarding intermediary liability starting from Svensson and GS Media, through Filmspeler and Ziggo, followed by legislative events, such as the CDSM Directive and the US Section 521 Study, all the way to the IP Action Plan and the Digital Services Act. Recently, the European Commission underlined the need for clarifying and updating the responsibilities of platforms, so it is a good time to re-examine intermediary liability.
Rosati pointed out the changed status of online platforms: they are no longer just a messenger. This “Hermes to Hades” change feeds the need for stricter rules when it comes to platform liability. Rosati also explained how the CJEU has tried to make sense of the notion of communication to the public with new concepts like “indispensable intervention”, “new public”, “profit-making intention” and “knowledge.” As the platforms evolved, their actions and the types of intervention they perform also changed, resulting in different decisions regarding their liability.
Regarding the upcoming YouTube decision, Rosati asked the important question of “how far can the Pirate Bay decision be stretched?” So far only Advocate General Saugmandsgaard Øe’s opinion has been released. Rosati underlined the fact that, according to the AG, platforms are not directly liable under the InfoSoc Directive and, following the AG’s approach to the concept of communication to the public, they can only be held liable when the platform is a piracy-focused one or when the platform presents the content as its own. However, according to Rosati, this approach requires ignoring a big portion of the CJEU case law. We shall see what the CJEU thinks about the AG’s opinion next summer.
It will also be interesting to see what will happen to platforms and the safe harbour protection as the implementation deadline for the CDSM Directive approaches (7 June 2021) and the debate on the new proposal for a Digital Services Act (15 December 2021) takes off. Here’s to 2021 and its webinars, seminars and ideas!