Europe takes step towards 'censorship machines' for internet uploads

  1. All material uploaded has to go through 'content recognition technology'
  2. A separate Article, 11, would change how news is used on the internet
  3. 'Claims of censorship, upload filters, and link tax are all a total exaggeration' - Axel Voss

A key committee at the European Parliament has voted for a new provision in a legislative act that forces tech giants and other online platforms to share revenues with publishers. It is known as Article 13, and is part of an updating of the Copyright Directive.

Article 13 proposes that large websites use “content recognition technologies” to scan for copyrighted materials, though it doesn’t explain how this works in practice. This means texts, sounds and even code which get uploaded have to go through an automated filtering system, potentially threatening the creation of memes and open-source software developers.

A final parliamentary vote on enacting the change is due before the end of this year.

Two leading MEPs for and against this are Axel Voss, who leads the change for the copyright proposal, and Julia Reda, a member of the Pirate Party in Germany.

Article 13 … would mandate Internet platforms to embed an automated infrastructure for monitoring and censorship – letter of protest

Before the June 20 vote in the Legal Affairs Committee, internet pioneers criticised the proposal, claiming that it will destroy the internet in its current form. More than 70 people notable in the field signed an open letter against it, including World Wide Web inventor Tim Berners-Lee, Wikipedia co-founder Jimmy Wales, and Internet Archive founder Brewster Kahle.

“We support the consideration of measures that would improve the ability for creators to receive fair remuneration for the use of their works online,” the group stated in the letter.

“But we cannot support Article 13, which would mandate Internet platforms to embed an automated infrastructure for monitoring and censorship deep into their networks. For the sake of the Internet’s future, we urge you to vote for the deletion of this proposal.”

Article 13, a.k.a. censorship machines

Critics have described the filtering of automated systems proposed by Article 13 as “censorship machines.”

Existing systems are demonstrably bad at filtering content for copyright, and have trouble differentiating between infringement and the legal use of copyright material. The YouTube Content ID filtering system, for example, is routinely abused and has harmed legitimate content.

Automated systems just can’t distinguish between commentary, criticism, and parody and mere copying, nor could the platforms employ a workforce big enough to adjudicate each case to see if a match to a copyrighted work falls within one of copyright’s limitations and exceptions,” internet-activist Cory Doctorow told Wired.

The Tor Project on Twitter

This would break the internet. Imagine YouTube Content ID, but for everything: blog comments, tweets, Github commits, Instagram photos, replies to newspaper articles, rental listings, dating profiles. This is being proposed in the EU. #Article13

“According to the Commission, they wanted to ensure that platforms and rights-holders can negotiate fairly. Why do you need upload filters in basically the entire internet to do this? Why not trying competition law? Why not oblige them to engage in licensing agreements?” said Diego Naranjo, a senior policy advisor at EDRi, a digital-rights association.

“There are many other solutions depending on what the problem is, but definitely upload filters won’t solve any of those problems but create new ones … Furthermore, having China as the first country implementing these filters does not predict a good outcome of the idea,” he told WikiTribune.

A problem for open-source

Article 13 is not limited to video or audio, but targets any online service that allows its users to upload or share content with each other, including code-hosting platforms. As a result, developers are concerned because it affects how code is uploaded onto any platform.

This is bad news for open-source software which relies on collaborative coding platforms such as Github to flourish. If algorithms get to decide what material software developers get to share, this could result in software that’s less reliable and less resilient for everyone.

“Automated filtering of code would make software less reliable and more expensive,” wrote Github in a post. According to Github, upload filters are a form of surveillance, since content needs to be monitored,  which restricts free speech. 

Save Code Share is a campaign to raise awareness about the implications of Article 13, stating in its open-letter that the reform “fundamentally undermines the foundations upon which Free and Open Source Software is built.”

Article 11, a.k.a. link tax

The other controversial aspect of the directive that passed the committee vote of the directive is Article 11. Dubbed the the link or snippet tax by critics, Article 11 will change how news is consumed on the internet.

Automated systems just can’t distinguish between commentary, criticism, and parody, and mere copying – Cory Doctorow

Article 11 means that anyone linking to snippets of news stories would have to get a paid license. So aggregator news sites including Facebook and Twitter will be forced to buy licenses from media companies before linking to their stories. As a result, using a hyperlink to an original article could be fraught with copyright issues.

“Rights-holders and [major] publishers are the ones who asked for and are pushing for this Directive as it stands,” said Naranjo.

The EU presented Article 11 as a way for publishers to take back control, expanding what’s known as an ancillary copyright for press publishers, already established in Germany and Spain. Google News in Spain shut down its Spanish operation in 2014 due to a similar law; at the time Google News was the country’s largest news aggregation website. A year later, a study found that the law had in fact harmed Spanish publications as a result, especially smaller ones.

Making this link tax an inalienable right also conflicts with liberal licenses such as Creative Commons, under which sites such as Wikipedia operate.

“Forcing publishers who use CC to accept additional inalienable rights to be remunerated violates the letter and spirit of Creative Commons licensing and denies publishers the freedom to conduct business and share content as they wish,” wrote Creative Commons. The proposal would pose an existential threat to the over 1.3 billion CC-licensed works online, shared freely by hundreds of millions of creators from around the world.

Final vote on measure at end of year

The EU’s existing copyright law was enacted in 2001, so the reform is meant to adapt it to the digital age, or the “digital single market.”

Axel Voss, the Rapporteur of the European Parliament for the Copyright Directive, firmly believes the proposed reform will benefit both European citizens and publishers.

“I feel that the criticism hasn’t been really balanced and not based on the actual text we’ve proposed,” Voss told TNW. “That’s why all these claims of censorship, upload filters, and link tax are all a total exaggeration.”

This key vote doesn’t make Article 13 an EU Directive. The next step is a negotiation between the Parliament and EU member states to reach a common position. A final vote will take place around the end of 2018. EDRi has a breakdown of the aftermath here.

Campaigns including and will continue providing information on how citizens can make their voices heard.

Image information

  • TODO tags

      Is there a problem with this article? [Join] today to let people know and help build the news.
      • Share

      Subscribe to our newsletter

      Be the first to collaborate on our developing articles

      WikiTribune Open menu Close Search Like Back Next Open menu Close menu Play video RSS Feed Share on Facebook Share on Twitter Share on Reddit Follow us on Instagram Follow us on Youtube Connect with us on Linkedin Connect with us on Discord Email us