When dry-sounding pieces of legislation quietly make their way through the European Parliament, it's hardly surprising that they get little attention from the wider world.
The Directive on Copyright in the Digital Single Market, for instance, which was voted through last week, intends to tackle the problem of EU citizens uploading copyright-infringing material to the internet.
Understanding Article 17
But the internet is a global entity, and laws passed in any major territory can reverberate way beyond its borders. The directive's most controversial component, Article 17, has been described as a threat to the internet's democratic freedoms; not just for those living in Europe, but for all of us.
The crux of Article 17 is the way it places responsibility for the copyright problem on internet platforms such as Google, Facebook, YouTube and many smaller ones. As the directive makes its way into law, the platforms will become liable for any copyright-infringing content that European users choose to upload. As a concept, it seems fair enough: the people and organisations who create the video, audio and text upon which the internet thrives, have never reaped the same rewards as the platforms that deliver it. But how would this infringing content be detected?
Why automation is not enough
Critics of the directive say the only way of doing so is by automated means. Artificial intelligence would scan every piece of media for potential infringement, and positive matches would be blocked. But there are doubts about the capability of machines to make such decisions correctly. Non-infringing content, whether satirical memes, photos containing corporate logos or wedding videos with pop music playing in the background could all be blocked as a result.
Annmarie Bridy, professor of law at the University of Idaho, says the use of such blunt instruments to satisfy the new laws is inevitable. "Anyone who says we shouldn't worry about over-blocking because filtering technology is smart, doesn't understand the technology. It's almost inevitable that these measures will affect access to content for millions outside the EU."
Digital activist Cory Doctorow believes it will also affect creators outside the EU. "If you're running a platform, which rules do you apply? You take the most restrictive rule and apply it to everyone," he says. "For example, if you want to sell a car in California, the emissions standards are very high. That means most production lines in the world produce vehicles that adhere to those standards, and California effectively becomes an exporter of those standards. I believe the EU is about to become an exporter of terrible internet policy."
The directive was first proposed back in September 2016, and there have been many drafts, amendments and concessions along the way. Non-profit platforms such as Wikipedia were exempted, as were smaller companies that could never even hope to afford to implement it.
Fair use exemptions for memes and satirical works were also incorporated – but would automated filtering even be able to tell the difference? Fears of the use of such automated systems were addressed in the legislation, but Doctorow mocks its drafting. "They said, OK, we'll add a clause that says 'don't use filters if at all possible'. But imagine I said 'I don't want an elephant, but I legally require you to produce a large grey land mammal with four legs and a trunk – and let me reiterate, this must not be an elephant'."
'An absolute love note for trolls'
Given the sheer quantity of material internet users upload every day, automated filtering is inevitable, most experts say, regardless of the law. As a result, Article 17 wades into the most fundamental argument regarding censorship: do you stop it before it happens, or remove it afterwards?
Ada Palmer, associate professor of history at the University of Chicago, recently drew parallels with similar arguments that took place as long ago as the 17th century, and noted English poet John Milton's point that "censorship before reaching the public has a vast impact on what is said, and even what is thought".
Two of the internet’s father figures, Vint Cerf and Tim Berners-Lee, both issued statements stressing that Article 17 would affect freedom of speech, but the EU reiterated that this was not its aim, and that it merely seeks to clip the wings of the biggest internet companies and rebalance power towards creative industries.
Doctorow’s concerns lie beyond the clumsy filters that could block innocent uploads. He also believes that the databases used to check for copyrighted material could be easily gamed and abused – after all, if anyone can create a copyrighted work, anyone can submit that work to a database and claim it as theirs.
“You could use a bot to submit the entire works of Shakespeare every 10 seconds,” he says, noting that companies banning people from making such uploads could end up being sued. “It’s an absolute love note for trolls,” he continues, adding that if databases can be abused for mischievous purposes, they can also be manipulated for political ends to suppress information.
Conveying the potential loss of freedom
Before the vote, major platforms such as Reddit, Patreon, YouTube and The Internet used the hashtag #SaveYourInternet to raise awareness. The games streaming site Twitch was particularly vocal; online gaming communities thrive on videos of people playing copyrighted games that feature copyrighted music and graphics. Games studios have traditionally chosen not to enforce their rights, but Article 17 adds new complexity to the debate.
Bridy cautions against policies that emerge from moral panic and sees strong elements of such panic here. "The online speech codes and filtering mandates that are coming out of the EU are a serious global threat to human rights of free expression," she says. "They should be recognised for what they are, no matter how well-intentioned they seem to be."
Doctorow says it has been hard to convey to the world the potential loss of freedoms we've hitherto taken for granted. "One problem is that the subject is complicated and technical," he says, "and the other is that the negative impacts are somewhat speculative and a long way away".
What we do know is that the directive has passed. The irony is that a number of MEPs have since claimed they accidentally pressed the wrong button during the crucial vote. The implications of their mistake may be felt far beyond the chamber.
