The internet can breathe a sigh of relief (for now) as the European Parliament has rejected the proposed EU Copyright Directive in its current form.

The draft Directive, which was approved by the Parliament's Legal Affairs Committee in June, proposed an overhaul of EU copyright law. While much of the Directive has been uncontroversial, two changes in particular have attracted vocal support and criticism:

- Article 11, which protects news outlets from the unauthorised republication of their news articles by other online service providers.

- Article 13, which requires online platforms to actively monitor user uploads for copyright law compliance.

Opponents of the Directive have expressed concerns that internet platforms will be "motivated to comply with the strictest version implemented by any member state" if individual states are allowed to apply their own standards for copyright infringement.

The wording of the Directive is now back up for debate and Parliament will revisit it for a second vote in September.

Article 11 - the "link tax"

Article 11 gives news outlets protection from the republication of "not insubstantial" portions of their content by online service providers such as Google or Facebook. The effect is that news outlets could charge license fees for any content that contains a link to a news article, if that link is accompanied by a snippet of the article itself (think Google's page previews). 

Those in favour say this helps news organisations, who have struggled to operate in the face of the tech giants.

Those opposed, however, have raised concerns about the potential for the law to be used as a political tool, and the uncertainty over whether a hyperlink alone would be considered insubstantial or not.

Article 13 - the "upload filter"

Article 13, the most hotly contested article in the proposed Directive, holds online platforms responsible for the publication of copyrighted material on their sites, including where this has been uploaded by their users. 

Under Article 13, an "online content sharing service provider" must obtain an authorisation from the relevant rightsholder in order to communicate or make available to the public a copyright protected work. A provider will not be liable for any unauthorised publication of a copyright work if they can prove they have effective measures in place to prevent the availability of copyright works on their platform. 

The effect of this is that online platforms could be incentivised to monitor (and filter) all image, text or other uploads by their users. Discussion around Article 13 suggests that platforms would implement automated content recognition technologies to filter uploads for copyright infringement - this would especially be the case for platforms that process large amounts of user content such as Facebook, Google, Twitter and YouTube.

Those in favour of Article 13 say this fairly protects copyright owners (a number of musicians have expressed their support for Article 13) and shifts the onus to online platforms to take an active role in enforcing copyright law. 

Opponents say this could lead to vast amounts of user content being arbitrarily blocked or removed if platforms use automated content recognition technologies to filter uploads. The types of content that could be affected by filtering technologies include, on one end of the scale, unauthorised uses of songs (such as remixes, covers, YouTube videos and so on) and videos. At the other end, it could theoretically include memes (which often utilise third party images) or incidental uses of copyright works such as Facebook pictures containing copyrighted images in the background or on items of clothing. 

Regardless of whether the Article intends to capture these kinds of uses, opponents are concerned that platforms could take the most risk averse approach and adopt policies or technologies to remove content without undertaking a thorough copyright assessment. If such a 'cart before the horse' approach is taken then the implications for freedom of speech and expression could, in the eyes of open internet advocates, be troubling.