CULT Opinion on the Copyright in the Digital Single Market Directive: bad on filtering, press publishers’ rights and TDM, but putting users back in the picture!

In the process of the adoption of the Directive on Copyright in the Digital Single Market (the copyright review), several European Parliament (EP) Committees will draft Opinions that will then need to be ‘taken into account’ in the Report by the lead Committee of the EP on this matter, namely the Legal Affairs (JURI) Committee.

The draft Opinion [PDF] on the copyright review from the Culture and Education (CULT) Committee has been published last week, making CULT the first one to have a go at rethinking the un-ambitious proposal by the European Commission. The Rapporteur of the draft Opinion is MEP Marc Joulaud (EPP, France) and FINALLY, someone has decided to talk about the users in a positive and proactive manner when looking at the copyright review. That alone, in itself, deserves to be applauded and to highlight our appreciation, we have decided to embed several of these user generated content illustration in this blog post.

However, while a few elements of the Opinion seem to indicate some willingness to introduce positive elements, they are sadly outweighed by changes that are clearly bad. So let us briefly outline the Fifty shades of grey in this first attempt at redrafting the European Commission’s initial proposal. The proposed amendments will be referred to hereunder with the acronym AM.

Eyes wide shut / welcome to total darkness: Opinion must be rejected on content filters (Article 13), press publishers’ rights (Article 11) and text and data mining (Article 3)

Sadly, the positive attempts in this Opinion are largely offset by the negative proposals in the areas of content filters, press publishers’ rights and text and data mining (in no particular order of ‘really not good’).

Filtering of user uploaded content (Article 13)

Where the Opinion totally fails to deliver, is on its proposed redraft of Article 13. First, C4C considers that content filtering is a matter of legal enforcement that is totally out of place in the framework of this copyright review, especially as the initial proposal by the European Commission goes in a direct collision course with the Ecommerce Directive and established case law by the Court of Justice of the European Union (CJEU).

But aside from that principle-based approach, the redraft proposed by MEP Joulaud considerably worsens the original text (who would have thought that was possible?) at various levels:

The text is amended to apply to ‘digital content platforms‘ (AM28 & 71-72) instead of ‘information society service proiders’ and shifts from services that ‘store and provide’ to ‘content uploaded or displayed’ by users. By removing the condition of ‘storage’, the Opinion brings in all the content that is (live) streamed, embedded, etc. hence dramatically increasing the scope of the content filtering provision, which is already absurd as originally proposed. It also makes it technically even more unrealistic, as filtering before upload of a live stream is bound to be challenging, to say the least.
The scope is also increased in terms of covered content, as the draft Opinion replaces ‘large amounts of copyright protected works or subject matter uploaded by their users’ with ‘significant amounts of user-generated content, copyright protected works or other subject-matter uploaded or displayed by their users’ (AM28).
At the same time, the rightholders able to claim the benefit of a licence from platforms is limited to the larger ones (AM 72), as digital content platform providers are mandated to ‘enter into fair licensing agreement with any requesting rightholder, provided that he category of works covered by the rightholder represents a significant amount of the content displayed on the platform’. In other words, collecting societies and major labels can come to the table but individual authors or small rightholders are excluded.
At some point, the text dwells on ‘unfair competition issues’ (AM27) which seem a bit out of place in the context of an EU copyright review.
And the draft Opinion feels the need to clarify that this content filtering provision does not apply to search engines (AM28), which makes one wonder if anyone thought the original actually did.

The draft Opinion also tries to clarify the redress mechanism (AM30, 73 & 74) by notably requiring rightholders to process complaints in a reasonable time.

Press publishers rights aka ancillary copyright (Article 11)

Here again, let us start from the premise that the proposal to include a neighbouring right for press publishers should be purely and simply deleted and that the Member States that have adopted such a right should be asked to remove it.

Looking at the proposed draft Opinion from that perspective, whilst one can consider that decreasing the duration of the protection from 20 years to 3 years (AM69) is a good step, it is still a step in a direction that should not even be considered. The same is true when looking at the effort to limit the scope to ‘professional’ publications (AM24) and to activities for commercial purposes (AM68).

Some of the proposed amendments even seem to make the whole debate even muddier, the references to hyperlinking now stating that ‘This protection does not extend to acts of hyperlinking, or to the text fixating the hyperlink, where such acts do not constitute communication to the public under Directive 2001/29/EC’ (AM24) and ‘Where content is automatically generated by an act of hyperlinking related to a press publication, such content should be covered by the protection granted to press publications under this Directive’ (AM25). The latter is specifically puzzling: is it intended to refer to the type of automatic embeds generated on social media platforms where, if a user adds a link to an article in his status update, that links automatically transforms into a news snippet?

Text and Data Mining – TDM (Article 3)

Whilst the Rapporteur did spot that ‘science’ is a term that means everything and nothing and hence tried to accommodate both the soft and hard sciences disciples (‘The term ‘scientific research’ used in this Directive is to be understood as referring both to the natural sciences and the human sciences’ – AM1) and tried to somehow close the loophole in Article 3 (3) (AM45), his take on the text and data mining (TDM) exception seems to be one that tries to replicate the ‘French’ approach to it, which is currently the most restrictive in place in the EU. This includes:

the obligation to destroy content used for TDM ‘once the all the acts necessary for the research have been performed’ (AM3 & 44) – which does not really fit the need of researchers, as pointed out by the Max Planck Institute: ‘In the field of scientific research there is a particularly strong interest in the verifiability of research results. This may determine that the relevant datasets have to be retained and made available for the purpose of verification’.
the obligation for rightholders to be compensated for TDM uses (AM5 & 47), hence completely negating the fact that the TDM “exception only applies to content for which one has legal access, or the fact that not all rightholders even claim compensation for that legal access as such (e.g. in case of open access). The claimed harm that needs to be compensated relates to ‘ the consequent investments that will be required by rightholders to make technically possible and facilitate the wide use of text and data mining techniques’ (AM5), a claim that is not substantiated by anything in the European Commission Impact Assessment or by any current practices, especially User as no obligation has been put on rightholders that would justify such investments (and at any rate, many of them intend to set up their own TDM platforms, so they would do it for their own purposes).

On the ‘lighter’ / positive side of the spectrum: Opinion should be improved on users and creators (more) in the spotlight & (timid) freedom of panorama

Freedom of panorama (AM11 & 55): limiting to non-commercial is more restrictive than the Infosoc Directive

MEP Joulaud displays a courage which was absent in the European Commission proposal, by acknowledging the ‘popular’ request from European citizens to get rid of the absurdity of an absence of an harmonized Freedom of panorama exception in the EU. However, his proposal (AM11 & 55) fails to truly deliver a satisfying solution, as it limits the proposed exception to non-commercial uses, hence making this flavour of freedom of panorama more restrictive than the current text of the Infosoc Directive 2001/29 authorizes (see Article 5 (3) sub h).

Whilst some people may think that this ‘non-commercial’ uses limitation is not a big issue, it is worth pointing out that the CJEU has in several cases shown a willingness to interpret the notion of ‘profit making’ quite broadly, considering that the fact that a website displays advertisements can be enough to consider that there is a commercial dimension at play (and honestly, is there still much left on the Internet that does not display advertising?).

User Generated Content ‘exception’ (AM12, 13 & 56)

Sometimes, things look too good to be true. When spotting MEP Joulaud’s willingness to integrate the user into the debate around copyright by proposing a form of user-generated content (UGC) ‘exception, we could only cheer with enthusiasm. And, to a certain extent, even after a weekend reading the Opinion, our enthusiasm is still present, although mitigated by some concerns. Indeed, the definition of the content covered by UGC seems extremely broad as it does not seem to be purely limited to content created or remixed by a user. But all in all, this is certainly a worthy effort by MEP Joulaud.

Increasing the possibilities for creators to be fairly remunerated (Article 13 to 15)

The draft Opinion also looks at the (contractual) situation of creators and tries to make the provision proposed by the European Commission slightly more effective by notably permitting claims by creators against rightholders to be brought before a court or other competent authority through an organisation acting on the creator’s behalf (AM34 & 35).

One can wonder however if the addition of ‘demonstrated’ in AM34 does not set a higher threshold for creators to bring such claims forward.

A general approach to avoiding contractual bypasses of exceptions & limitations: but what about technical protection measures bypasses?

Where MEP Joulaud did fully grasp the current loophole of the Infosoc Directive regime, is in his consistency at specifying that the exceptions he proposes or adds on cannot be bypassed by contractual provisions, as he makes such provisions ‘unenforceable’.

One may however regret that this principle is not established as applying to ALL existing and future exceptions and limitations to copyright, and why it is not extended to technical protection measures, which often also limit the ability of citizens or institutions to benefit from exceptions and limitations granted by the law. In other words: be even bolder!

In the darker shades of grey: better organized licensing (but still licensing) for education (Article 4) and a mandatory Reprobel regime for Member States (Article 12)

Education exception for digital (Article 4)

Whilst AM8 & 9 acknowledge that it is not easy for an educational establishment to be aware of the existence of licences and to find them, and hence obliges Member States to ‘use or develop appropriate tools, such as a single portal or database’, it still remains stuck in the tunnel vision of licences will solve all, a premise that is not necessarily shared by all parties involved nor confirmed by practice.

MEP Joulaud also seems to consider across his draft Opinion that any exception to copyright automatically triggers harm for the rightsholder, hence requiring that compensation mechanisms be established (AM52 in the case of education).

The so-called Reprobel provision (Article 12)

The originally proposed Article 12 seems aimed at contradicting the judgement of the CJEU in the Reprobel case (C-572/13). The CJEU confirmed that the rationale of the fair compensation requirement is intended to compensate for the harm suffered by right holders, and concluded that publishers are not subject to any harm by, in this case, the reprography and private copying exception. Article 12 aims to reverse this decision by giving a right to all publishers, ‘including those of press publications, books or scientific publications’ to claim a share of the compensation for uses under exceptions.

MEP Joulaud’s proposal goes further (AM 57 & 70) by making it mandatory for Member States to ‘provide that where an author has transferred or licensed a right to a publisher, such a transfer or a licence constitutes a sufficient legal basis for the publisher to claim a share of the compensation for the use of the work’.

So to conclude on a French note in acknowledgement to the effort made by MEP Joulaud to seemingly table all his amendments in English, our summary of this Opinion is: ‘Bel effort mais peut mieux faire’ (Nice effort but can do better). Or, in other words: you set yourself a laudable objective (see below): now act upon it!

‘It is the Rapporteur’s view that the proposal does not acknowledge the position consumers, as service users, now occupy in the digital environment. No longer playing a mere passive role, they have become active contributors and are now both a source and recipient of content in the digital ecosystem. Indeed, digital content platforms base the entire design, business model and optimisation of their services around the dual role of their users. From a legal standpoint, it is also the opinion of the Rapporteur that digital practices of users do not benefit from legal certainty under the current copyright rules, in particular the exceptions and limitations, and therefore require a specific approach, a fourth pillar within this Directive.’