Hi, I'm Danny O'Brien -- I work at EFF on Article 13. So we talk a little a bit about this earlier this week: https://www.eff.org/deeplinks/2018/11/yes-eus-new

Hi, I'm Danny O'Brien -- I work at EFF on Article 13. So we talk a little a bit about this earlier this week: https://www.eff.org/deeplinks/2018/11/yes-eus-new-copyrightdirective-all-about-filters https://www.eff.org/deeplinks/2018/11/yes-eus-new-copyrightd... I've mentioned this in Hacker News threads before, but here are the few things to bare in mind. Firstly, the SME exemption (as well as the specific exceptions for Wikipedia and Github) was put in at the last moment in an attempt to win enough votes to get the Directive through the plenary following widespread opposition. It only exists in the European Parliament text, which is currently being negotiated with the original Council text. The news from the trilogues is that there's lots of lobbying to get it removed. The exemptions sound good for Github and Wikipedia on paper: but Github and Wikipedia are the services that currently exist, and have been effectively grandparented in. It doesn't speak for all the other potential services that we can't describe, because they don't exist yet -- and won't exist if there's a liability regime that would have stopped Github and Wikipedia in their tracks if it had written before. Thirdly, we know from experience that these exemptions don't work. In the European Parliament text, we have a blanket liability regime, which means that rightsholders can sue you, or your provider, by default. You can argue back, "oh it's okay, we're covered by this exemption", but you have to prove that you're really covered by the exemption -- and the expression of that exemption in each of the 28 implementations of the Directive in the member states. It creates a default of liability, and then fences of a small section of the Internet where you may be protected. In the mean time, you -- as a person who hasn't paid up for a licensing arrangement with the major rightsholders -- will be on the receiving end of repeated orders to take down content, with the understanding if you don't successfully argue against that exemption, or the moment you cross that SME line, you'll be liable to an unbounded extent. Why would you take that risk? How could you protect yourself against that risk? Also note that many of these exemptions are expressed in the Recitals, which merely indicate the spirit of the law, the specific rules for transposition in the Articles themselves. Generally speaking, if you are threat-modelling new law, you might as well ignore the Recitals, because there is a substantial lobbying and legal community who have a big financial incentive to guide lawmakers, and deploy lawsuits in such a way as to sideline those non-binding commitments. The exemption language was an attempt by lawmakers taken aback by the force of the opposition to Article 13 and 11, to both win over votes in the Parliament, and stop GitHub, Wikimedia and its users from complaining. The reason why Github, and Wikimedia and their users continue to complain is that they don't believe they can act in the future as though these exemptions will really work for them — and they have an interest in maintaining the rest of the Internet ecosystem, which will be still left out in the cold. Remember too, that almost every platform we use to communicate online has been accused (and sued) as a method to encourage infringement. The music industry attempted to sue the MP3 player out of existence as an infringing device (https://www.zdnet.com/article/stop-the-music-diamond-multimedia-faces-lawsuit-over-its-rio-player/ https://www.zdnet.com/article/stop-the-music-diamond-multime... ), the movie industry sued YouTube claiming that it was primarily a device for piracy. https://en.wikipedia.org/wiki/Viacom_International_Inc._v._YouTube,_Inc https://en.wikipedia.org/wiki/Viacom_International_Inc._v._Y....