We've got less than a day until the key vote on the wording of the new EU Copyright Directive, when members of the EU's legislative committee will vote on whether to include controversial mass censorship language in the proposal that the parliament will vote on.
These two proposals are the "link tax" (Article 11), which forbids linking to news stories without a license (but doesn't define "news story" or "linking," leaving it up to 28 member states to make 28 contradictory rules for this); and "copyright filters" (Article 13), requiring that everything Europeans post to the internet be automatically checked for copyright violations first, in a kind of vast, unprecedented version of YouTube's Content ID, but for everything — your Tinder profile, the code you check into Github, the skins on your Minecraft avatars.
Article 13 is every bit as defective as Article 11: it insists that companies filter their users' posts, but does not penalize people who abuse the system by falsely claiming copyright over works that don't belong to them, whether tactically (a political party using false copyright claims to suppress embarrassing footage before a key vote), or trollishly (claiming all of Shakespeare just for the pleasure of banning anyone from quoting him).
Sarah Jeong covered this today for The Verge: "legislators everywhere seem to have accepted the status quo of a digital landscape that's dominated by a handful of giants with self-perpetuating silos."
The vote will be early in the day (Central European Time) on the 20th. There is still time to contact your MEP.
For anyone who's been on the internet long enough, the problem with Article 13 is pretty clear. It's YouTube Content ID but for the entire internet. Axel Voss, the member of European Parliament who is taking the lead on the copyright bill, has argued that the actual proposed language never mentions a filter, although that just raises the question of what using "effective technologies" to prevent copyright infringement means, other than filtering. Although the most recently revised language exempts sites such as "online encyclopaedias," clearly aiming to exclude the likes of Wikipedia, Voss has said in the past that he cannot predict which platforms will be affected.
Remember the time YouTube Content ID took down a video with birds chirping in the background because an avant-garde song in its copyright database also had birds chirping in the background? Remember the time NASA's videos of a Mars landing got taken down by a news agency? Remember the time a live stream got cut off because people started singing "Happy Birthday"? And all this happened despite the fact that Google is really good at what it does.
Opponents of the proposal argue that if smaller platforms are also required to implement upload filtering, it'll not only be a significant burden to them, but they're likely to do a much worse job. Placing content-filtering obligations on platforms encourages them to block as much as possible, and it gives them little incentive to let innocent content through. FOSTA in the United States is a good example of how these incentives work. After the passage of the law, which was intended to combat sex trafficking on the internet, Cloudflare dropped Switter, a sex worker-friendly Mastodon instance.
New EU copyright filtering law threatens the internet as we knew it [Sarah Jeong/The Verge]
