91

There have been lots of reports recently about how the EU are trying to ban memes. For example:

  • Trusted reviews

    The European Union (EU) is pressing ahead with efforts to ban online memes under new digital copyright laws, with the controversial rules – Article 11 and Article 13 – now having successfully passed through an EU committee.

  • Digital Trends

    The European Union’s Legal Affairs Committee, known as JURI, voted on June 20 in favor of new controversial internet legislation that is designed to protect copyright holders but may have the adverse consequence of stifling an open internet and banning memes.

  • National Review

In the last few days the EU Commission's Twitter account responded to these claims with a meme:

"How can anyone think we will ban memes" mere

I don't believe that they are trying to ban memes as their post indicates but if this new directive was passed would it be an effective ban on memes and meme sharing?

  • 9
    The most salient words here are "may have the adverse consequence of ... banning memes." It's something that could easily happen as a side-effect of the proposed law. – Mason Wheeler Jun 28 '18 at 0:31
  • @MasonWheeler well said. There's a difference between intent and result here. The result is pretty clear, the intent is far less so. – jwenting Jun 28 '18 at 4:49
  • 3
    Coming soon: To comply with new EU reg, Twitter takes down Picard meme posted by EU Commission. – reirab Jun 28 '18 at 19:25
  • 2
    Every time government (or literally anyone) tells the internet not to do something, the internet retaliates and does it on a massive scale. Even if memes were banned, people would still create memes, and more and more of those memes would be of the "memes are banned, whatever shall we do?" variety. – Chris Cirefice Jun 28 '18 at 19:55
82

BBC's coverage explains how that might happen:

Jim Killock, executive director of the UK's Open Rights Group, told the BBC: "Article 13 will create a 'Robo-copyright' regime, where machines zap anything they identify as breaking copyright rules, despite legal bans on laws that require 'general monitoring' of users to protect their privacy.

"Unfortunately, while machines can spot duplicate uploads of Beyonce songs, they can't spot parodies, understand memes that use copyright images, or make any kind of cultural judgement about what creative people are doing. We see this all too often on YouTube already.

So basically your Picard image could get zapped as a copyright violation, without human intervention. Or at least that's what the opponents of the proposal claim. I recall that something like that happened on YouTube with soundtracks, using a similarly automated technology, so the fear is perhaps not outlandish.

But as YouTube grew, so did its resources, and its need to keep business partners happy (and avoid lawsuits). Now YouTube lets anyone register their content with Content ID and hunt down infringers. The algorithm errs on the side of false positives: The site even took down computer-generated static because it sounded like an existing video of static. Rugnetta explores the chilling effect on fair use in “Automated Copywrongs,” an episode of his podcast Reasonably Sound.

The EU, in its wisdom, has made parodies expressly legal as of October 1, 2014.

"The only, and essential, characteristics of parody are, on the one hand, to evoke an existing work while being noticeably different from it and, on the other, to constitute an expression of humour or mockery," the EU rules state.

Although discriminatory parodies are not permitted:

"If a parody conveys a discriminatory message (for example, by replacing the original characters with people wearing veils and people of colour), the holders of the rights to the work parodied have, in principle, a legitimate interest in ensuring that their work is not associated with such a message."

So there's probably going to be some friction between the proposed new article 13 of robo-whatever and the existing EU legislation which does explicitly allow parodies. It remains to be seen how this conflict is going to be solved in EU directives.

Variety noted in article last year that YouTube has made it somewhat easier for such conflicts to be temporarily resolved in the favor of the disputed content:

YouTube has taken some steps to combat bogus copyright claims — but only for those flagged in its Content ID system. Last year [=2016] the video giant implemented a policy under which videos subject to disputed copyright claims are allowed to remain online and generate ad revenue while the disputes were being evaluated. However, that applies only to Content ID, not to individual copyright takedown notices submitted by third parties (as in the Bad Lip Reading situation).

Whatever you may think of YouTube's solution, scaling it to the level of the whole Internet seems much more problematic:

The experts agree: Article 13 is dreadful. Vint Cerf, creator of TCP/IP, the internet's fundamental protocol; Tim Berners-Lee, the web's inventor; Wikipedia co-founder Jimmy Wales; Internet Archive founder Brewster Kahle; cryptography expert Bruce Schneier; and dozens of other of the crème de la crème of the internet, think Article 13 is awful.

In a letter to the EU, they wrote: "As creators ourselves, we share the concern that there should be a fair distribution of revenues from the online use of copyright works, that benefits creators, publishers, and platforms alike. But Article 13 is not the right way to achieve this. By requiring internet platforms to perform automatic filtering all of the content that their users upload, Article 13 takes an unprecedented step towards the transformation of the internet from an open platform for sharing and innovation, into a tool for the automated surveillance and control of its users."

  • 41
    So basically, some politics thought AIs were fancy and tried to use them to censor some content, except AIs are dumb and kind of bad at the moment. – Hawker65 Jun 28 '18 at 9:42
  • 22
    If we can't solve The Ship of Theseus, then we can't assume we can make an AI that can do it – jrtapsell Jun 28 '18 at 10:42
  • 4
    @jrtapsell: We could if we assume the problem has a solution (that we haven't found yet). However, depending on how the AI works we may not be able to understand and thus evaluate or review the solution. That being said, The Ship of Theseus is not meant to be solved but to exemplify a limitation of our seemingly arbitrary linguistic and mental concepts of object persistence. – David Foerster Jun 28 '18 at 11:50
  • 4
    "The EU, in its wisdom, has made parodies expressly legal as of October 1, 2014.". The UK — long a shameful black hole when it comes to free speech — finally made it legal to parody other people's work risk free. They are reforming their libel laws as well. Most countries of the EU long since had parody being legal, along with truth being a defence against libel incrimination, with a "not guilty until proven otherwise" principle in effect, which the UK did not. So do not confuse the UK ad the EU on this point. Also note that the US always allowed parody under Fair Use. – MichaelK Jun 28 '18 at 12:25
  • 6
    As for examples, TorrentFreak has mocked it a couple times in the past: in the overzealous automated DMCA requests to remove infringing content from Google results based on keywords, many times companies have accidentally tried to remove pages from their own official website. – Izkata Jun 28 '18 at 19:44

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .