 And my introduction as well. I'm a legal fellow at Wikimedia Foundation. My name's Kabir. And there are some, in case there's an emergency, look for the other hosts here, which are the other legal fellows. That's Sarah. Just raise your hands, guys. These are guys who have done the actual work. So they would help you in case of an emergency. And there are stairs there in case you have to take them. OK. Now coming to the panel. Basic introduction for all the panelists here. We have Abby Walmer. She's a senior policy manager at GitHub. And we have Eric Simon. She's the senior product counsel at Google. We have Heather Meeker. She's the partner at OML, Wendy, and Myers. We have Professor Mark Lemley. He's William Edge Neocombe Professor at Stanford Law School. And my former video game law professor as well. And we have Mishie Choudhury. She's the legal director at Software Freedom Law Center in New York. And so now coming to the panel's topic, EU copyright in EU, you guys have come here for that. So essentially, after five years of intensive discussions, multi-stakeholder negotiations, and need I say, aggressive lobbying, the European Union finally passed these reforms. And their title copyright in the digital single market. The EU Commission says that these reforms aim to unify Europe's digital market and product and protect the online creativity. Just to give you some context, the last EU reforms were introduced in 2001. A lot of things have changed since then. We have Facebook. We have Netflix. We have memes. And there's a lot of internet penetration. And now you have computers in your pockets. So the discourse has essentially changed. And naturally, we have seen a lot of support for these reforms, particularly coming from music companies and some publishers who always felt that their copyrighted material was uploaded online. And they were not fairly remunerated for it. There's been some opposition for this as well, coming from digital rights activists, some tank companies, and a very nice free and open encyclopedia online. It's called Wikipedia, just in case. And essentially, their argument is that this will kill the free flow of information on the internet, as we know it, and also kill the digital creativity in the European Union. Now, there are a lot of new provisions which these laws have introduced. But there are two key provisions which we are going to focus on, which are Article 15 and Article 17, formerly Article 1 and 13, respectively. I'll just give a brief overview of what these provisions are. So Article 15, it's the link tax provision. People call it that. And essentially, it seems to be targeted at search engines and news curators like Google News. And essentially, it obligates platforms to be producing more than quote unquote, single words, or very short extracts, to get a license from the publishers. So essentially, these platforms will have to pay to use the links from news websites. A model like this has already failed in Germany and Spain. And potentially, it will also impact people's right to freedom of information. So if you're not even writing journalistic content, still be worried about it. Coming to the next provision, which is Article 17, it's the upload filters provision. It requires commercial sites with over 5 million users to make quote unquote best efforts, to preemptively buy licenses for anything the user may possibly upload, which is all the copyrighted content. So essentially, it's just changing the entire ecosystem as we know it. Currently, if there is some sort of a violation, a user just red flags it and lets the platform know. And the platform would be the one to act upon it, either remove it or just tell the person why they're not removing it. Now it just changes the discourse. Now they have to preemptively do it. Just to give you sort of a context, 400 hours of video is uploaded on YouTube every minute. So humanly, it's not possible to filter this content. That is why we're going to have automated filters, which are criticized so much for being error-prone. But this provision has also found a lot of support. People think it is fair and appropriate. And it finally gives these creators like a right to set terms and conditions for others to reuse their content commercially. There are critiques. They think it will kill the startup industry. It will restrict the freedom of speech and expression and the right to information as well. And it will also, there's a narrative that it will solidify the dominant players who now can afford these filters. No guesses for guessing the names. So what really happens now? There are a lot of unanswered questions. But all the members of the European Union have about two years to incorporate these provisions into their national laws. You'll have some questions, whether it is the end of the free and open internet as we know it, or whether the fears are really over-exaggerated. Does it really give content developers a level playing field? These are some of the questions which all of us have been looking for answers to, and I don't have them. So let's go to our panel. And before that, I'll just introduce our moderator, Alison Devonport. She's also a prominent member of the Wikimedia's legal team. So over to you, Alison. Am I on? Can people hear me? Great. Thank you, Kabir, for that wonderful introduction. I don't know that I would describe myself as a prominent member of the legal team, but I am a member. I am, just for context, I'm the Wikimedia Foundation's Tech Law and Policy Fellow. It is a long-term fellowship that is focused primarily on policy with a little bit of law thrown in there for fun. And as you might expect, my last year has been largely focused on e-copyright, which is why the summer fellows asked me to moderate this panel. But you don't want to hear me talk. You want to hear these very smart people talk. So I'm going to ask Abby to kick us off. You and I have sort of talked about e-copyright. We've been in the throes of it together. And one thing that was really great out of GitHub was that you were able to negotiate a carve out in the copyright reform for open source software repositories. And I was wondering if you could share a little bit of that experience and sort of introduce yourself as well. Sure. Yeah, thanks, Allison. So I'm Abby Volmar. I do policy at GitHub. GitHub is a software development platform. It's where people build software. So when you think about what that means from a policy perspective, we're representing software developers on the broader ecosystem on issues that affect them in policymaking. So when we learned about this proposal from the European Commission in September of 2016, it was that basically platforms where people are uploading content would need to think about copyright infringement and use upload filters to prevent it. And so for us, we are a platform where users are uploading content and software code is copyrightable. So that definitely meant us. And at the same time, the public content that users of GitHub are uploading and sharing with the public is not something that they're doing to make money. And we, as a platform, are not making money off of it. So in terms of what the motivation was behind the objectives of this law, whether you thought it was a good idea or not, just didn't apply at all to what GitHub was doing. So we didn't think it was a good idea. And we wrote a blog post and tried to get that idea across. But in terms of how we were representing developers on it, we had this very specific argument we could make about open source software and how software development was really at risk for this. And so the Free Software Foundation in Europe and Open Forum Europe, which is also a nonprofit in Brussels, launched a coalition or a petition and got community support called Save Code Share. And we joined that. And I went to Brussels, talked to people there, talked with Wikimedia, Mozilla, others who were really concerned about Article 13, which was what it was at the time. It's now 17. And we realized there was a lot more we could do there. And Wikimedia was already making great inroads in explaining why online and psychopedias should be carved out. And along those lines, we were able to convince policymakers that, hey, open source software also doesn't make sense here. So yeah, I guess maybe I can leave it at that. I could get into the whole process, but I want to make sure everybody else has time to chat. So is that sort of a good show? Yeah, we'll circle it back around for that later. Thanks. So I actually wanted to just send it right down the line to Mishi and ask what, if any, involvement was there with the SFLC in the EU copyright reform? And was it on the greater radar of the free software community? She's already said my name is Mishi Chaudhry. I'm the legal director of Software Freedom Law Center in New York. I also am the founder president of SFLC.in, which is the Indian equivalent of Electronic Frontier Foundation. How many of you actually are familiar with the EU copyright director? This is a very smart group. So SFLC was not an official organizer of the big campaign, which FSFE and Open Forum Europe launched. But a lot of our clients did join the campaign. And mostly that time, our role was them asking, do you think this is right to join? What do you think is your advice? I'm a large project. I am Debian, or I'm a patchy software foundation, or I am XYZ. And what is your analysis? What is the problem here? Or should we join it and what we should be doing? My ask the question about the familiarity is I think that all of us in this room do realize that, obviously, internet was the information superhighway. It should have been the universal education system, which is what you guys have proved by not surveilling anybody and letting everyone read. But I think the companies also realize that the jig is up. The era of unregulated behavior is not going to continue any longer. But because there's social harms, there's a lot of anti-tech sentiment. But there's also this other side of the story where there's redistribution of revenues, where Hollywood and the music industry behaves in a very different way. So no matter how much anti-tech sentiment might be there, but there's also this thing when that determines what your policies would be, then what is it that we get is this mishmash of Article 15, 17, and how it operates, and without taking into account what the technical realities are or are not, and only targeting one major company or two large company. So the analysis becomes a little confusing. And also it depends, who are you advising that time? Most of the projects are headquartered in the United States. This is the EU, but these are open source projects. They're going to be actually distributing everywhere, making code everywhere. People are distributed everywhere. But just keeping in mind that what the EU directive was trying to do, which is having the German publishers, the French broadcasters on one side, trying to say that YouTube is making so much money, we need a much larger share of it. And that changing the copyright law itself, and that fed into the analysis quite a bit. Thank you. And again, moving down the line, you sort of mentioned that a lot of your involvement was through clients. And my question for you, Heather, is what was the sort of general sense among your clients? Were they panicking about this? Were they happy about it? Did they have no awareness it was going on? I would say they were mostly not panicking about it. But one thing that may be happening as a response, and this probably needs to be borne out over time, is that when emerging companies happen and they have some business model that involves using electronic content somehow, having this kind of, I don't know if prior restraint is the right word, but it's kind of a law that says, you have to stop this before it happens instead of, we're going to apply liability after it happens. And when you have that, what you're taking away is the ability of emerging companies to do an ask forgiveness, not permission kind of business model. And if you look around at the really successful electronic businesses that at least some of us love, not with standing tech hatred, a lot of those were ask forgiveness, not permission models. People would come out with a model. Nobody would really understand whether it was lawful or not under copyright law. And then if you're successful enough in the marketplace, you at least had a war chest to fight the battles. But if you have regulations that say, you can't even engage in this activity, it's not that we're going to apply liability to you, but you can engage in it at all, then you're actually significantly curbing innovation, innovative business models. And that's probably where I see it the most, because clients come and say, we want to do this business model, and we as technology lawyers have to say, well, you're never going to be able to do that business model that's totally unlawful, or that happens, or well, it might be. And so if you make a big splash in the marketplace and you get out in front of it, you might be able to basically have the model and then become a company that is fighting the legal battles to keep content available. So that's where I see it most. Yeah, and I think that's a really interesting answer, given the sort of distinctions that ended up being in the law at least around Article 13 between small businesses and larger platforms. And we'll talk later about whether those were well written or not. But I want to turn to Erin next to sort of talk about this from Google's perspective. There was at least a little bit of a narrative that this was sort of meant to bite the big tech companies. And so I'm interested in hearing about what the process at Google was like around this reform. Yeah, so hi, everybody. I am Erin. I have a slight head cold. And I'm very happy to be here. I am a Product Counsel on Search at Google. And for the last many months, I have spent a lot of time in Europe focused very intensely on this because one of the things I spend the most time on in my product counseling is Google News. I think it is totally fair to say that this is a law targeted at us. There's what used to be called the link tax and then civil society organized and made hyperlinks radioactive and then became the snippet tax and the law that was the content ID law. And I think it's really kind of funny because content ID is, and I don't think this is a brag, I think it is the most advanced copyright management system on the earth. I think some enormous percentage of YouTube's copyright management happens through content ID. And something like 85% of copyright owners choose to leave their work up and monetize it. They don't actually want it taken down. They don't think there's a problem with unauthorized material appearing on YouTube. The problem is they want a better negotiating position when they sit down with YouTube and hash out how much royalty they're going to get from it. That is my cynical real politic view of this, but I dare you to disagree with me. And I don't personally work on YouTube so much, but I work on news. And I have to say, our mission is to make the world's information universally accessible and useful. It sounds corny. It is a little corny, but we actually believe it. And when we can show you a small preview of the thing that you will get if you click a link, you are more likely to click that link. We have ample data to show that. But so I kind of see the copyright directive as there are a couple of really great things in it. We're not going to talk about them here, but making the public domain really the public domain by denying protection to works whose copyright has expired is fantastic. Legitimizing the kind of extended collective licensing that's normal in the Netherlands and the Nordics I'm for you. But I think that the two articles that we're talking about really do reflect kind of a failure of Google's political influence relative to a small coalition of large news publishers and media organizations. And I'm not going to tell you that I want Google to have more political power, but I will tell you that I started working there because I believe in a free and open internet. And that was the best way to make that a reality. And that is the thing that is most in jeopardy at this moment. Great. Thank you. And then to round off our panel, I want to ask you, Professor Limli, and this might be a big question to situate this copyright reform in maybe the context of US law, how it differs, how it might influence it. Yeah, all right. I'm Mark Limli. I teach IP law at Stanford Law School. I'm also a partner at Dury Tongre here in San Francisco. How do we situate this in the context? So let me think about three ways to think about that. So one would be, how do we think of it as kind of copyright policy, copyright theory? I am, if anything, more cynical than Erin on this. Copyright battles, political battles in the United States and in Europe have long been about a vision of copyright law that's just, are we trying to promote the interests of copyright owners or are we trying to promote the interests of users and readers? And I come down very strongly on the side that copyright is about not just producing content but making content available to the world and that that's why we want to balance. Content owners sometimes come down on the no, no, give us more money and we'll produce more stuff and that's the point of copyright law. I actually think the EU directive goes further and is targeted at a third vision, which is a mercantilist vision. I'm not even sure that the answer is they want to promote more copyright owners. I think the answer is they want to promote more European folks at the expense of US folks. And that's been a kind of long running theme in European tech policy, unfortunately. You see it in the antitrust policies, you see it in the trademark policies and I think it's a little ironic because it's quite likely that the people who are most likely to benefit from this are Hollywood and the people who are most likely to be hurt by this are European consumers and European startups. I think Google actually is gonna do fine in part because they have the content ID system. But I think it seems pretty clearly from a kind of philosophical perspective, it's not just targeted at let's promote content owners, it's targeted at let's promote our national interests rather than the interests of readers. So I think that's at the level of kind of theory. I mean, at the level of copyright policy, I think this is a very fundamental shift in the way we think about the internet. The idea that I can't actually link to or provide a single small headline in the course of a link without paying, without permission, right? The idea that I can't host content provided by someone else on the site without some sort of advanced clearance is a pretty fundamental change in the way the internet works. I don't think it's a good change, but I think it reflects a pretty different vision of what we want the internet to be. I'm old enough that I remember the information superhighway from the early 1990s. We were gonna build a big, broad band platform, and what we were gonna get out of that was the existing TV stations were gonna send lots of channels of pre-programmed content over it. This was the internet we were going to build. It was a kind of program channel internet with just more stuff given to you by the people whose job it is to produce stuff and deliver it to you. And the internet happened sort of while no one was looking, and around that, and we saw this amazing outgrowth of creativity from sources we'd never expected to see creativity from, and everyone could use it and benefit it. That is not the EU copyright directive model of the internet. The EU copyright directive model of the internet is the information superhighway circa 1993. There will be an internet, there will be stuff on the internet, but by God, we're gonna control who puts that stuff on the internet, what it looks like, and the existing corporate players are the ones who are gonna deliver it, and that's it. And that I think is part of a kind of broader and to me very worrisome fracturing of the fundamental principles of the internet. You add to that the kind of Chinese approach to the internet with a different set of software companies with a different set of values and increasingly a different set of hardware platforms on which it is built, right? And I think we're seeing a move towards the world in which there isn't gonna be an internet. There are gonna be several different and incompatible models of how one delivers data online. Hey, thank you. And I think we're gonna get back to that sort of idea very soon of who are the actual winners here and who's gonna suffer. But first, I still wanna keep us on the topic of actual sort of tactics during this reform because I think we have a good group of people here to give different perspectives on that. And one thing I wanted to ask about was grassroots campaigns. So Wikimedia ran a very small grassroots campaign with our community members. We actually ended up blacking out the front page of Wikipedia for those who are unaware or were not in Europe at the time. And sending people to call their representatives. There were a couple other organizations and websites that did this, Save Your Internet, Copyright for Creativity. And what I'm interested in is sort of the panel's impressions and anyone who can take this of how those grassroots campaigns worked. Were they, they obviously weren't effective in stopping the law, but were they effective in raising awareness? Do you think that they had other benefits or do you think that they may have harmed the cause? I can jump in again giving the story of software developers. So for us, I guess I left off saying that I went to Brussels, met with organizations like Wikimedia, Mozilla and some of the companies with open source programs that were really concerned about how this might affect them and also members of the open source community, just individual developers. But at the same time went to Parliament and met with people like Julia Reda who is a member of Parliament that was one of the lead negotiators and very, very vocal about the problems with Article 13. So she totally got it, but she said, hey, you know, great to hear what GitHub thinks. I wanna hear, are the members of Parliament need to hear what individual developers think? And so we, you know, wrote blog posts, tweeted about it, got the word out to developers and said, hey look, this is happening in the EU, it affects you. They wanna know, the policy makers they wanna hear from you. And we got people writing into us because we had said, you know, feel free to contact us if you want some more background or like help with figuring out who to contact and that sort of thing. So we got plenty of people writing to us saying like, you know, open source is my life, the EU is funding my research, like I don't understand how this could be happening, like I really need to make sure that they understand like what they're thinking about doing here and why it's a bad idea. And you know, I think if you'd look at what happened, it was pretty soon after that, that the European Parliament and the European Council, which were at the time thinking about amendments to the initial proposal, both tried to figure out a way to protect open source software, code sharing platforms. The language was a little bit different, one was about open source, one was about code sharing, but basically the idea there was what we were getting at, what these developers were telling them, like hey, you need to enable us to collaborate, you need to enable us to share information on platforms because that's how we innovate, that's how we create stuff. So I think that was a really cool illustration of how grassroots, you know, really needing to hear from individuals was much more powerful, a company can say what they want, but in the end, the policy makers I think really were interested in hearing from individuals. And do you think sort of the specificity of the ask, there was helpful in ensuring that that was successful? I do, and it's sort of going back to what I was saying earlier, that for us, you know, when we intervene from a policy perspective as GitHub, we're always thinking like pretty laser focused about software development. And when we did first write to the world about this, we didn't say, you know, exactly what the whole thing should look like. We just sort of said from the perspective of software development, here's why this is a bad idea. And then the response we got, you know, especially after the developers all wrote was to try to focus on excluding us, which honestly wasn't, that's not the best outcome from our perspective. I mean, the fact that the whole thing passed is like not great, which I'm sure we'll get into. But to your point, I think maybe part of why we were able to get at least that much was that we were using this pretty specific illustration of why, hey, for this sector, it's a really bad idea. And they were like, okay, sort of like with Wikimedia. Okay, well, we'll take online encyclopedias out of it. Okay, we'll take open source development platforms out of it. So I just to comment on that, I mean, I do think this is something you see a lot in the legislative process around copyright in the US as well as in the EU, right? Which is we're gonna do a thing and a bunch of people show up and say, oh my God, this is gonna mess with my life, this is gonna mess with my life, this is gonna mess with my life. And the response of the legislators is fine. We'll carve out you and you and you. And then we'll do a bad thing to everybody else who didn't have the political ability to show up and that's I think a little bit what happened here. I mean, we should be clear that the process, the grassroots process worked in the sense that article 17 is not as bad as original article 13. It's not as bad because it carves out GitHub and Wikimedia and various other folks. It does some things that are sort of efforts to try to kind of meet halfway or be compromises. Many of those I think are frankly just, I don't know if they were intended as window dressing but they end up as window dressing. So no general obligation to monitor. You don't have to monitor your website. You do however have to make sure that nothing that anyone has ever complained about appears ever again posted by a different person in a different form on your website. So feel free to do that without generally monitoring your website if you can think of a way to do it. People said, oh my God, there's gonna be, you know, you're gonna eliminate free speech, you're gonna eliminate fair use, you're gonna eliminate all of the sort of things that people use with copyrighted content that's not illegal. No problem says article 17. We just make it so that stuff that isn't illegal isn't excluded, right? Nevermind how we're gonna try to figure that out, right? Just automate upfront the process, not only of finding whether this is copyrighted content but figuring out whether it's fair dealing or fair use under the different laws of each of the member states of the European Union. So I think a lot of those promises are illusory but they reflected an effort, I think, to respond to each of the individual constituencies except Google who came in and said, hey, these are the bad things that are gonna happen if you do this law by carving out that constituency. Yeah, and I just wanna echo that and expound on it a tiny bit. The Save the Link campaign, I don't know if anyone in this room was part of it, worked. Hyperlinks got excluded from article 15. The Save the Meme campaign kinda worked. The anti-filtering, the civil society showing up and saying filters, harm-free speech. Computers can't tell what is legitimate for uses of copyrighted content. Worked to the extent that regulators said, okay, we won't use the word filter but we still expect you to do a thing that has the same effect. But I think it's that kind of, like there was a coalition that then kind of, certain pieces got solved just for them and I know that wasn't the intent but once GitHub has their exception, legislators stop meeting with GitHub because they say this is solved. And the problem is, not just like, everybody else who's not GitHub who wants to maybe start something similar but that we don't know what the next thing is gonna be. If Wikipedia didn't exist and hadn't gone to Brussels then the law wouldn't have a carve out for them and then they could never be formed. And so it's that door closing for all of these beneficial novel future uses that is, I think, pretty dangerous. I think that was sort of our philosophy when we decided, because we did have long talks about whether we wanted to get involved in this, especially since we had that carve out. And I think part of our philosophy there was we, especially Wikipedia, which depends on citations, we depend on the entire internet. And so to sort of turn our backs once we got that carve out didn't feel quite right. So now, I think for GitHub, it might appear that way and maybe policy makers didn't listen to us as much after, I don't know, but we had a similar view where even if the platform GitHub itself was carved out that doesn't help all the developers that are building stuff on our platform. They are still needing to figure out what they need to do. We're continuing to make this message known to policy makers, but I mean, to your point, maybe they sort of stopped like, oh, we took care of you, I don't know. But I think Wikimedia also did a really nice job there. Even though you had the online encyclopedia language, you made this point about the broader internet and why that's important. Yeah, but yeah, I agree that ultimately it was a hard point to make to lawmakers in particular. So I think now we should move on to the substance. Oh, goodness. Sorry, everyone. And I think we've been talking a lot about Article 13, 17 and Article 11, 15. So I wanna start with Article 13 slash 17. Who do we think is going to benefit most from this? And who do we think is going to be harmed by it? And this is for everybody. I do think audible magic is going to really benefit our people. I mean, before I start about this, I do have to say one thing is that this internet until now has been very great in the sense of that everybody who was not even in the United States assumed that they had First Amendment rights that happened to me all the time. If I were in India, where is the other place I practice? And they would say, I have First Amendment rights. And I'm like, India's First Amendment actually restricts free speech. So it's a different constitution. But the reality is that the large platform companies are no longer the same companies as they were in the 1990s. Whether it is social media platform companies who didn't even exist that time, things have really metamorphosized into a very different era we live in. We also live in a place where the companies behave very differently based on the jurisdiction they are in. They still play the Lexlokai server games when they're in another country saying, all my servers are in California. I can't give you any data. That's why I can't take off, take down any content. Here are all the problems which I have, which means we are going to see far more regulatory structures coming out of different countries which is going to try and use law to break the net as we know it. And it's in the making and it's coming. And like the CDA 230 and the copyright DMC 512, the content is obviously distinguished. Copyright is used many a times. And but other content also where moderations are starting, we are seeing a lot more regulation which is very similar to what we've seen coming out of EU or sometimes far more restrictive. So that's going to happen. That is a reality. It's not going to be the same regulations. And also the same arguments of the great free and open internet is not going to work anymore. People don't want to break democracies. People don't want elections to be run via WhatsApp or interference through social media platforms. Society now sees real harm. And then when regulation comes, then everything gets muddled up. And that's one of the big problems of the EU directive also that when the conversation happens, the sentiments are, oh, we have to protect our own people, our own German publishers, our own French broadcasters and YouTube because it is so big, that's the enemy. Let's go target them. And then everybody tries to attack it from that point of view. So more regulation is coming. Sorry for the bearer of bad news but I'm not the person who's done anything. But, and who's going to benefit? I do think Audible Magic will benefit. One of the things, a question which was sent to us was to think about whether there is going to be an open source alternative to something like Content ID. And I'm like, have we seen the patent portfolio of Audible Magic and the fingerprinting tech? What goes in it? Of course, the money which gets invested in it is very different. Google has that kind of money and of course, the reference files, it's also about, it's a lot about data. It's not just about that you can build a tool and then it will continue to check it. So they are going to benefit. I do think that when a regulation is written, which looks like it's only targeting one particular company, then obviously it's not going to cover a large swath which you want it to cover and these kind of small car routes will appear. So, small and medium enterprises are not going to benefit. The Europeans are not going to have suddenly another one because suddenly the cost of entry of the market is really, really high now. So, and other than Google, major players do license tech from Audible Magic. I don't see many people coming up with another new platform or alternative to all of that. So they are definitely going to benefit. They were also there lobbying and saying, well, we know how you can have, we are the gold standard, gold industry standard about compliance with Article 13. So here it is, use our tech. That's how you do automated content regulation. So they are benefiting for sure. Hopefully I do not know how the redistribution of the wealth from the various people who think they can get a better deal by negotiating with YouTube works out. And if they think that they can get a better deal, we will see how that really does because Content ID also did give them a way to monetize what was already their content. Without that, the only game in town that time was takedown. It made it a system that they could monetize it. So those I'm not so sure, but at least one particular entity I can identify. Go work there then. I'd like to make another comment about, first of all, make sure everybody understands what Audible Magic is. It's like a music fingerprinting sort of technology so that if you say upload something, you can very quickly check whether it is something that is a known copyrighted work. But I would like to point out something about that. Those systems really work for the types of copyrightable works that have canonical forms, right? And so Audible Magic or some kind of content ID system can tell whether you've uploaded the latest hit that comes out of a record company produced and released. But it isn't good at all at figuring out every other kind of content that doesn't have canonical forms. Open source software is a really good example of that. It's just all over the place. You can do content checking on binaries of open source, but checking source code is such an unbelievable information problem. A lot of you probably know this, but there are companies that do this. They do code audits. Those systems are slow, expensive, error-ridden, and who's ever looked here at a Black Duck report? I mean, it is not a pleasant thing to do. So if you're looking for copyrightable content in something that isn't just like a little pill of content that's put together like a song or a movie, it completely breaks down. So in answer to the question and drafting on what you've said, the content checking systems for everything besides like your record company song, your movie studio movie, they're not there, right? And I don't know, maybe some people will come up with them, but those are a really different technology problem from the kind of content checking that we have already. I'm sure people are working on it, but it's not there right now. So that's something very different, and I'm sure Audible Magic will benefit, but there's maybe another generation of tools that will benefit as well. Can I just pick up on that, because I think it then becomes really important what for each of the tech companies the default is when there isn't an Audible Magic code, right? So this looks like a music file, right? It does not look like something for which we have an existing authorized music file from Audible Magic. What should we do with it? How? One important difference between the old and the new version is the old version just said, nope, that's entirely off limits, keeping me sure it stays off the internet. The new version doesn't say that. It says that you don't have to do that until you get some complaint about it, but that doesn't mean that tech companies won't decide that the easiest and quickest thing to do with something that's probably illegal but hasn't been caught yet is just not allow it on the system until it's been vetted through Audible Magic. And that's really important because there's a whole bunch of stuff that's valuable to people other than tech companies on the internet that will not be on the internet if that turns out to be right. So if you like actually watching Twitch streams or figuring out how somebody can get through this video game level, if you're 11 years old and you like putting your videos on TikTok, if you're a new musician who hasn't made it into the mainstream and doesn't have a mainstream publisher and wants to upload your music, all that stuff isn't necessarily going away from the internet, but there's a very strong structural incentive that says, hey, I'm outside the system and we're probably going to get in trouble if we allow this stuff, so maybe the easiest thing to do is just not allow it. Now, we'll see what happens and different companies may take different approaches, but the default, I think, for those and for a bunch of new things that no one's even envisioned yet, right, the sort of whatever the next version of the sort of video meme is, is probably going to be, it never goes up. Just saying, there was a fact which after the director was passed, which was also put up, which exactly said this, that they're not using the term filters, but the best efforts provision, they rely that something innovative will come out of the industry. So they're already assuming that something innovative is going to help them identify and then be able to take all that content down. And I read that fact and I re-read it and I was like, so they're already prepared and they think this is what's going to create jobs in Europe? But that's the thinking behind all that's going on where the technical requirement is de facto what the best efforts will turn out to be. Yeah, and by the way, I don't think the open source developers are going to be rushing out to develop these things. That is probably the last thing on their list to develop. I guess it's remotely possible, but I'm not betting on that as a robust and popular open source project. And I'll just add, because we've been talking about the technical difficulty in recognizing when files are similar, when a work is included in another file, don't forget that knowing who owns what and who has given permission to whom is incredibly difficult. So difficult that our professional music licensing often gives us like 125% of ownership, 50% of ownership. The music labels themselves don't know what they own and haven't regularized it enough to be able to tell anyone else. And yet, we are supposed to magically intuit that is secretly the PR arm of Viacom and is allowed, but this user isn't. In each of the countries of the EU separately. Right, copyright is territorial. And that's even before we get to what is inevitably going to happen, which is the effort to enforce this globally beyond the borders of the EU. Oh yeah, I was going to save this for later, but Mark blew up my spot, so I'm going to say it now. The very next fight we will have is once we implement whatever filtering, I'm not committing to this. I'm saying as soon as filters exist to detect infringement under European law, right, holders will say that is red flag knowledge under the DMCA, boom, no more safe harbor. They will be wrong, and you will all tell them that they are wrong for the simple reason that European law is much more restrictive than United States law. Europe does not have fair use, United States does, much is not infringing here that is infringing in Europe, and these filters will be so badly calibrated that there will not be a good signal even of European infringement. Okay, thank you. Yes, and we definitely will circle back around to the greater global implications of this later. But since we're talking about, you know, kind of winners and I don't want to say losers, but I did. Let's talk about, let's shift it over to this sort of so-called snippet tax, Article 11. Who comes out on top there? Does anybody? Okay, so I got to get on my rant about this one, right? So as Kabir mentioned, this has been tried before in other countries in Europe. It even succeeded in Belgium for a while. So in Belgium, the news publishers sued Google News and they said, hey, you're linking to our news sites. That's copyright infringement. And the Belgian court said, yep, it is. So this is perhaps the dumbest shakedown in the history of dumb shakedowns. So just to set some context here, right? Google News does not in fact run ads. Google News, of course, does not charge money, right? Google News is a money-losing venture that makes news available. How does it make news available? Well, it doesn't copy news sites. It links to information voluntarily placed on the web by the news sites themselves. So when in Belgium, the court said, yep, putting a headline with a link to the actual news site as an act of copyright infringement, Google News said, okay, I guess we won't do that anymore. Sorry, we violated your rights. The news publishers were aghast because they get more than 50% of their link traffic and therefore more than 50% of their advertising revenue from Google News driving people through their links. So they said, no, no, no, we don't want you to take our pages, our links down. We want you to pay us money for providing for free half of our revenue. And the problem is that you're not paying us to generate money when you're doing it yourself for free. That's what the snippet tax does. That's the new rule we've got. Now, the question is, what happens? And it may be that Google is a big enough company and it's committed enough to the sort of public interest story that we want to make this information free and accessible that they will pay some amount of money out of the goodness of their hearts. But that's the only reason, honestly, why you should pay that amount of money. I mean, the logical hard-nosed business thing for Google to do in this situation is to say, okay, no European news unless you come into a deal with us, right? Because we now have a copyright obligation. And let's talk about the terms of that deal. Let's see. We'll give you half of the money we're making. That's zero from this link. Why don't you give us half of the money you're making from running ads on the news that comes from this link? Any logical snippet tax would involve money being paid from the news industry to Google. And I don't know that that's where we're going to end up. But the idea that there's a value gap because the people who are making money from the links on the other end are not being paid enough by the people who are making no money is pretty ridiculous. Erin, would you like to speak for Google? Yeah, I have a lot to say on this topic. I attempt not to rant, but Mark just did it so eloquently. I do think the market has set the price of a link from Google's homepage to a website. And the price is that websites pay us for those. They're called search ads. So I agree that the idea of Google paying to send traffic to a publisher is not economically rational, but I think a lot of what drives European copyright is not economic rationality. I think there is a philosophical difference between EU and the US, which is in the US we see disruptive innovation as a good thing. We have made new and better stuff. Maybe it's not actually better. That disruption can be harmful. In Europe it is, oh no, disruption. We must protect our legacy industries. We care about them. So I have spent a lot of time in Europe. I have talked to a lot of European folks about this. And there is a real kind of core belief that if you make money in any way connected to a piece of copyrighted material, you owe some of it back to the copyright owner. I think in the US we are much more comfortable with this idea that you might expand the pie and that you might create value that didn't exist before that you were entitled to. And maybe you do that in a way that infringes the copyright and so you get a license, and maybe you do that in a way that doesn't infringe the copyright, but creates benefits for everyone. I think that that's how Google works. Users benefit in being able to access information. Publications benefit in having readers find them and become revenue-generating for them. We benefit because like Google, we make a lot of money. But like all of that together does not mean that the balance was unfair before. And without commenting on how the balance might change going forward under the new laws, I think we should be attentive to this uneasy way that European copyright in particular doesn't mesh with the way the Internet kind of fundamentally works as this free exchange of data. You know, this thing that grew up in DARPA in academic contexts with this norm of just like it's there, it moves. We're now seeing like the CJEU's line of cases on linking is phenomenally complex and makes almost no sense except as a uncomfortable tension between this idea that copyright is kind of firm and brittle and the Internet is going to flow around it as best it can. I think I'll stop there because I could rent for hours on this. Thank you. Anybody else or? Cool. Oops. Your nail polish caught his eye. Yeah, so I think I do want to even though you just said that you don't want to talk about it, but I do want to move on to sort of what this means for the future. Like what does this look like going forward? So in the case of Google, I'm sorry to make you talk twice in a row. How much is this going to change your current systems? Is that change going to sort of echo out beyond the EU? And how much of that? Sorry, go ahead. This also reminds me of something I didn't work into my last answer, but I think is important. The drive behind the EU copyright is not economically rational vis-a-vis Google except that we have money and they want some of that money, but it is economically rational for a handful of powerful publishers to suppress the ability of their competition to be found online. If you're in Spain and you don't know where to go for news, you go to El País. And so in some ways an effort that makes it harder for people to know that there is a regional publication, that there's a hometown paper or somebody close to a story who's reporting on it, like that's also part of what's going on here. As far as what we are going to do about it, I cannot share a great number of details. France is voting on its bill in about five days and I am not going to upset that apple cart for all of the plaudits from all of the very smart copyright people. But we are committed to the free and open web. That's what search was built on. That's how search works. And we are doing our damnedest to make sure that we can come up with an approach that works for everyone, not just a handful of already successful well-known publications. It will depend on the details of each particular country's implementation. This is a directive, not a regulation, which means it's not self-enforcing. Each country has to pass its own law and the directive leaves quite a number of important things open to be decided at the national level if they want to. Hooray for the digital single market and uniformity across Europe. So a lot of this is to be determined, or at least to be announced. Thank you. And I sort of want to ask the same question to Abby and Michi about the sort of open source software community. I know that there is a carve out, but is anything in the reform going to change how things are done or create different levels of risk than there were before? So you mean in the EU itself, you're not asking globally? So just with respect to that, I alluded to this earlier that for us, even though the language pretty clearly is GitHub where you're looking at that exclusion, developers are building things on GitHub and those might be subject to Article 17. So there's still a lot of concern within the open source community about how that will play out and also making sure that when these national governments are looking at the directive and putting it into their system, their local national systems, that the exclusion that is there remains intact. And so I think that's kind of where things stand right now in the EU. So everyone said there are a lot of uncertainties, but one thing which at least I am hearing a little bit, not a lot, but in 2010 I was in Beijing and I gave a talk which nobody paid any attention to because it was about trade and why should you pay any attention about trade with China in 2010. But the idea was that sharing and collaborative innovation is something which is so important and such an integral part of how the work, how the infrastructure of the world is going to be built, that there has to be a recognition of it formally and the trade system should also recognize it. When I say sharing, that's free culture and free and open source software licensing as well. I think at least the lobbying efforts did result in something positive. The scour out language may not have been perfect, but there is a recognition, at least at the commission level there is, they understand what open source software is, that none of these companies, none of anything would have actually been possible if the underlying open source software systems did not work. So or something like Wikipedia or all of these people, I think now that one has a realization that there is some understanding, it needs to be pushed a little further so that if there is a general recognition of open source and how software or how culture is being made, then it also provides a good context for any of the carve outs like these. If you know that this is how people are building software, this will never be included and push much further about the importance of it, about the licenses and how people do this work everywhere so that whatever other TBDs we are seeing, they will already be very much cognizant of the fact that this is not something somebody has to come again to see for it. But other than in future, I do have to say that the US does push a lot on IPR rights and some of the developing countries where I have seen is that they have now imbibed it into it that their copyright regimes have to be at the highest strictest level, no matter what. So that's why the little bit of legislation it comes in interacts with intermediary liability of safe harbor provisions. We are seeing a much stricter regime coming either through courts or even at the or even at the level of regulation. And again, I have to say that unfortunately a lot of sensible regulation is also being driven by a very anti-tech sentiment. So that means that nuanced understanding or discussions are not happening. Can I make a comment too? And Abby, I'm curious what you might think about this. But one of the issues that I see going forward is that the assumption that software is software and other stuff is other stuff and never the twain shall meet is a failing assumption and it fails more every day. And as an example, the anti-996 thing on GitHub, that was really interesting because that was actually a political movement started in a tool for a software repository. I found that really interesting because it wasn't soft. Well, there was software involved, but a lot of what was there if you look at the page, it's really interesting. It was about labor laws and so forth. And so anything that carves out, say, software is obviously too narrow, not only because these tools are used for all sorts of things, but because software is converging with many other things. What about data? What about the next big company that's the GitHub of data? I mean, that and the protection of data by copyright, which is a very strange and scary topic in itself means that all sorts of stuff could be limited that we wouldn't even expect it to apply to based on our models for what's going on today. Yeah, sure. That's a really good point and kind of ties into something that I did want to get a point I wanted to get across before the end, so maybe I'll just do it now, but kind of an opportunity where everybody in this room can be thinking about getting involved in these kinds of discussions going forward. So unfortunately, a lot of the people who are making these laws don't have a really great grasp on technology, and this is a really good example. Even something like software, how many things that we do, how much of what is on our phones, on our laptops, is powered by software developers might be somewhat understandable, but then there's a bunch of other stuff in the world that people don't even realize is basically powered because somebody is building software to make it run. So I think just the wide application of what developers are doing and how much software code is important to the world was something that people didn't get, but also how is software even built? Today it's really built usually when people are working together, and whatever creative endeavors you engage on, you might realize that often it's helpful to have more than one person thinking about it, more than one set of eyes on it, and for software development that's certainly true, and you can have thousands of contributors to a project. So even just getting people who are making these laws to understand why collaboration is so important, not only for software development, Wikipedia is a great example too, where just having that functionality where people can be interacting like that is so important, and it's not just about somebody, I guess I'll do my reference to something that maybe not everybody remembers, but the movie Hackers. It's not just people sitting there like a hoodie and programming. Today it's a lot more pervasive, and I don't think that's obvious. So for everyone in this room, I'm sure there's something that you could inform policymakers about related to the interface between technology and policy in one of these discussions that they might not know and that might really help them understand better what to do. Thanks. So we are almost out of time. I think I just want to finish off with one sort of rapid-fire question, which is about implementation. So this is a directive that needs to be implemented in all countries within the EU over the next two years, and I'm interested to hear from each of you what you think the most important part of that implementation will be. You can either name a country, or a sort of strategy, or a aspect of the reform, but just what do you think the focus should be moving forward? If anyone wants to volunteer, I'll call. You said rapid-fire because... Yeah. So there is actually a kind of council forming to come up with authoritative guidance to member states on how they should be transposing the directive, and so there's still actually a lot of opportunity to influence how these kind of open questions get resolved. I can take off several for Article 15. You know, who counts as a press publisher? Like, what is a very short extract that one is permitted to have that is not subject to the law is, you know, France once again going to go kind of wild over territorial applicability and try to assert jurisdiction over the whole world? They haven't yet, in their lie that I've seen. Okay, rather than just list off other things, I'm going to say that the thing I want us to pay the most attention to is not just in the copyright directive, but in the future of Internet regulation, the delegation of judicial power to private organizations. Because, you know, it sounds great to say tech companies should be more responsible and should, you know, be suppressing illegal content. But when we do that without a court telling us what's illegal, we're the ones deciding what's illegal. I personally have been the decider for the right to be forgotten in Europe. And I try to do a good job. Like, we all do our best. But at the time I was making decisions about what European citizens had a right to see and not to see, I had not visited most of the countries in Europe. I don't think that a good law delegates that power to a 30-year-old in California. You know, much as it makes for a good chatter at cocktail parties. So I actually, like, as we are having these conversations about the balance of responsibility and openness about how to limit the real harms that arise from open platforms, I think we need to think about who we want making these decisions. All right, I'm going to dodge outside the question a little bit and think about sort of the scope question. So this is the first thing that has ever happened that's made me think that Brexit is a good idea. But so one question is what happens in the UK, right? I don't think anybody knows. But I think the implementation decision that I think ultimately, or issue that ultimately will be most important, is not the implementation decision in the EU member states. It is what is coming, which is the effort of the content industries to bring the same thing here. To say, hey, look, they've adopted notice and stay down, whatever they're going to call it. We have a value gap here, and that is going to merge with the tech lash that's already been mentioned. And, you know, we fought this off with SOPA and PIPA eight years ago, but that was a very different political environment and I think there's a substantial risk, not that this law, but that something like this law that abrogates 512 is on its way. Maybe I'll just jump in on that too and kind of take a similar idea of, like, not exactly the implementation in the member states, but what else is sort of tangential to that? In the EU, there's the major intermediary liability protection is in the e-commerce directive. And just a few days ago, there was a leaked version of what the commission is thinking about doing around that and how to undo it in certain ways. And so, if you think about the EU copyright directive and what it's doing to intermediary liability with respect to copyright, that is now sort of setting a precedent in some ways for the EU, and it will be elsewhere too, but we're already seeing in the EU efforts to think about how they can take similar approaches to chipping away at these protections that, as Mark so eloquently mentioned in the beginning, enabled us to have an information superhighway that was quite different than what that early vision of it, the internet that we know is nothing like that information superhighway model that would have been much, much worse for creativity, innovation, et cetera. Unfortunately, I think that we're watching the same reality TV, whether you're in U.S. or India. India is just two seasons ahead. In terms of right-wing and whatever, go back to wherever. If you're in India and you ask the government a question, you worked whole, go to Pakistan. Today here also, when you ask some questions, you're told to go somewhere. But why I say that is especially in this intermediary regulation system, we've already seen some versions of that coming in India. What scares me is that it's all very well to say we don't want censorship by proxy. And I've done that in courts and I've done that at policy level also. But there's not a really good alternative answer that we are also providing. Courts in most places are not going to be fast. I'll be able to dispose these cases very, very fast. And the amount and the scale of content is such. And then the harms are real that whether it's the society or our regulators or the legislators who actually really don't understand what this is all about are just going to make demands. And this is definitely giving them at least some kind of a blueprint to say this perhaps will work or this is what we are going to demand. I am curious, of course, about what Germany does, mostly because a lot of open-source licensing litigation does happen there. But not for that reason, but because legacy industries definitely were there. And in the past, there has been some kind of legislation which has come from there which is interesting. UK, as Mark already said, when it leaves we will learn different things. But we have to... It won't work the same way to say, oh, save the internet, internet is dying. We'll have to come up with some really concrete proposals to say, okay, we don't want censorship by proxy, but this is what we want. Otherwise, that's what we would look at. Yeah, I would just say that capital is a coward and every time I hear about some interesting thing, particularly coming out of Europe, I kind of sit and wonder, at what point will my technology clients decide they don't want to do business there? And it hasn't happened yet in most cases, but one of these days it's going to happen. And then just looking from the point of view of how private capital works to develop technology companies, investors are going to start saying, no, we don't want you to do business there. It's too expensive, too difficult, too much legal exposure. And so in a way, that'll happen independently of the laws, right? People will vote with their feet. People, investors do that. And that worries me along with just sort of a more general question about our regulations, right or wrong, that people will just say, it's too much. We don't want to do it anymore. And I think that would be to the detriment of everybody on either side of the pond. A bunch of doomsday scenarios. I love it. So I think we are going to take some questions from the audience now. There should be a microphone over there. Thank you. This is great. I'm Andrew Bridges from Fenwick and West. Can I just sneak two questions in with one? Number one, I understand that Poland is launching a legal challenge to the directive. How are people handicapping that? Will others join? That's the first question. Second, I think Aaron made a really good point and I think maybe Heather touched on it as well. Will people in Europe be able to figure out what they're missing by this regulation? And can others help create an awareness of what they're missing in this fashion in a way that might put pressure years down the road on undoing it. I always thought it would be nice to have a chamber of stifled commerce of companies that had to go out of business because they couldn't navigate the regulation. And they need a seat at the table at OECD in places like that. I'd love to know your thoughts on those. So let me take the second one. One of the problems with promoting innovation is that you quite literally never know what you're missing. You can see the companies that are shut down, although you rarely see them turn into the things they would be. If we'd shut down YouTube three months after it started, which is entirely plausible, no one would think, oh, we've missed out on this extraordinary explosion of creativity by a bunch of people we never expected to be video producers. That's not how we would have told the story of YouTube being shut down. The one good news is if we can keep this confined to Europe, if we can fight off the efforts to do it in the United States and elsewhere, then we might actually get a useful test case along the lines Heather suggests. So there's already evidence that a more restrictive European copyright basically drove the cloud service industry out of Europe and to the United States. There's a very good paper by Josh Lerner at the Harvard Business School that does a study on this. People didn't develop new startups in that space in Europe because it didn't look like a very good environment. I think if you can tell that story again and again, if you see where the new musicians come from, if you see where the new content comes from, you see where the new tech companies come from, and it's not Europe, that starts to become a persuasive story. In fact, one might go farther and say that we have already seen that experiment, and there's a reason why all of the big internet and tech companies are American. We might also see this natural experiment because Germany has a very active startup scene pushed very hard for an exception in Article 17 that allows small and startup businesses not to have to do copyright filtering until they hit. Yeah, there are a couple of kind of cliffs that they can fall off of. One is the number of users, one is the amount of total turnover, I think, and the other is just three years. So there is a clock on every startup now that deals with user-generated content that either it becomes massively successful enough to afford audible magic, build its own whatever, or it dies. I want to see those business plans that say I'm going to be in business for three years or until I make X amount of money and then I'm just going to close up. Those will be real successful. But the other thing that we saw that wasn't powerful enough to avert the directive, but I think did have an impact, was that when there was a campaign about memes and about YouTube creators, people losing the ability to share silly videos with their friends, you MEP's children, teenage children were telling them, what are you doing? And the lawmakers are like, this is the first time my own kid has ever paid any attention to what I've done. And so I think there are those ways to reach them. I also wanted to come back to the Poland question, because I think this is kind of hilarious. Poland is fairly well known as a repressive regime accused of infringing human rights itself and has seized on the opportunity to say, I know, wait, I know you are, but what am I, I know I am, but what are you? I'm sorry, I'm a little sick. But Poland has mounted a challenge to the directive as any member state can before the court of justice of the European Union, saying that it infringes the charter and European's fundamental right to receive an impart information. This is a way of Poland being like, well, you say we're bad, you're bad, you guys do it, everybody does it. So I handicapped the success of that effort with exactly the same handicap of who else joins it. I think if it is only Poland, it is probably not going to go very far. But if other member states can be persuaded to join in and maybe it's getting a little late for that, there is a real concern about automated filters, inability to recognize something as a protected parody and that could be, you know, important political speech or otherwise. So we'll just sort of have to see how that goes. Any quick responses over here? We'll go to the next. Okay, go ahead. Hi, Fred Jennings here from GitHub, but very officially taking off my GitHub hat for this question. I've been debating internally about whether to ask an obnoxiously specific implementation question or a perniciously broad theory question. And the theory question dovetails well with what we've been talking about, so I'm going to go with that. One thing that's been on my mind with copyright that we've talked about here and also like more broadly with other sort of internet regulation laws being passed both in the EU here in various other countries is the territoriality question that you all have touched on in one way or another. And I'm curious, you know, sort of casting forward into the future of this trend, whether it's sort of tech-lash driven or copyright markets driven. How do you all see this coming to a head with the concept of territoriality in law and sovereignty? At what point do, you know, US-based companies start saying, hey, this is something that is sort of a, you know, trade protectionism by other means? And at what point do these internet regulating laws begin to sort of challenge or break down the fundamentally non-territorial space of the internet as an informational space? I'm just curious to hear your thoughts on, you know, how we kind of harmonize this open sort of borderless space with notions of legal sovereignty. That's already happening actually. You have to say 1998. That's the point at which we'll start. USTR is in India right now. In the last, for the last 10 days, this negotiation has been going on. It's in the context of data sovereignty, actually. So, because India doesn't have an omnibus legislation, just like the US. It does have some fractional very little protection for data, but there has been a draft about data protection. And because WhatsApp's largest market is also in India, and there has been a discussion about WhatsApp pay, and we now know Libre is coming. So there has been a lot of, the equivalent of Federal Reserve came up with a policy saying all the data of India, and Indian citizens should stay only in India, which is like not even a copy. That's not how cloud computing works. That's not how anything works. But that's there. And the idea and the push behind is, and this is a lot coming from Indian large telecom players to say, we've lost the bus, we missed the bus. Now the American companies are monetizing all this data, they collect all this data, they make tons of money, we've missed the bus. Now the government should come to our rescue just like the Chinese did, and you have to do something. And so much so that there has been a policy hijack exactly by the local companies to say you need to do something. So there has been assertion now in the draft legislation in the regulations which is coming to say Indian's data belongs to India. And they're like, when during the trade thing there was just a loss of $6 billion about when India was said, okay you cannot send any of these materials. The Indian government actually said $6 billion is nothing. The money which we will make monetizing 1.25 billion Indian citizens, that's worth a fight with the United States. That's the direction we're already going. Those talks are already on right now. And the data sovereignty discussions and the calls are so loud that the largest telecom provider in India wants to vertically integrate all that they are doing and the push against e-commerce giants like Amazon is apparent. We're already seeing that and it's starting from data protection but it's not stopping just there. I think we've been doing geolocation for 20 years now. And we've been doing it imperfectly and more importantly we've been doing it on individual kind of one-off pieces. Turkey says there's something that offends us on YouTube so we're going to shut down all of YouTube and it's shut down for a day or two and then it comes back in Turkey with that video removed in Turkey but not in the rest of the world. France says no Nazi memorabilia can be sold here so we have different rules for what auctions show up on eBay in France and eBay not in France. And that's been true for a while. I think what's increasingly happening is that we're now making it part of the mainstream internet that everybody's experience will be increasingly different on a daily basis depending on where you are. Now, that's not going to be perfect. People will tunnel around it. People use VPNs to get out of China but you can make it harder, you can make it inconvenient, you can make it illegal. And I think that's the direction we're going to go and that's why when I mentioned earlier I think we are kind of seeing the subalconization of the internet. I think that's going to continue. Okay, I think from the activity of the fellows so take it away. Hi, Mitch Stoltz from the Electronic Frontier Foundation. A couple of you talked about this dynamic in copyright legislation where something very broad and sweeping is proposed and the folks that show up to resist it get a carve out and the folks who don't have the means to do that or the people just to do that get screwed since this is sadly probably not the last piece of bad copyright legislation we're going to see. I wonder if you have thoughts on strategy for avoiding those sort of carve ups next time? Yeah, I would say show up. They can't carve everybody out if everybody shows up. But no, I think part of the problem is there is a cynicism about the people who are advocating. If you go to Brussels and you say this is going to hurt the internet that we love, you only say that because you're Google to me. They say to GitHub you only say that because you want an exception. Well, I don't know what they say to you. You can tell. You just say that because Google told you to. Oh, yeah. This is also a danger actually. But there is definitely a business interest and I'm not going to argue that Google has no business interest in better copyright law. But there's also just a genuine, let's not break everything just to target us. It's, I think, important for people who don't have an axe to grind or who don't have an ostensible axe to find, to get involved to be unaffiliated and say this is just important for everybody or to find constituencies that seem legitimate enough but also widespread like people who want to write software, people who want to put their books online or share their photography. And like we have this problem in all of politics and there's a word for it that someone will remind me that I read in high school but where interests are widespread but diffuse they lose out to interests that are focused in a small group of people because those small group of people care enough to show up. I'll hand it off to Mark who knows the word. Public choice theory, right. So I will say I agree show up is part of it and one of the things to notice that one of the great things the internet has done is it has meant that show up doesn't mean show up with bags of cash and well paid lobbyists that is one way you can show up but as the meme example indicates one way you can show up is get enough teenagers upset that they're going to lose their memes right and have them talk to their parents right and that means so those people show it up and they got a carve out they got an exception that said I mean the problem is how do you mobilize how do you find those people and that's an issue and I so the one thing I will note with apologies to Aaron right we did this extraordinarily well in SOPA and PIPA right so if you remember the SOPA PIPA thing right it blacked out the internet on a day in January Google a bunch of other folks blacked out their sites 10 million people contacted their congressman more people have called their congress people about copyright law than any other issue in modern history that's astounding and that's because a bunch of people were got the got the message but they got the message in part because they were signaled by other people hey this is a problem and for what it's worth right I will note that SOPA and PIPA Google and Wikimedia blacked out their websites when it came to the EU copyright directive Wikimedia blacked out their websites Google didn't I do think that makes a difference and if we had the story would have been Google's fake grassroots protest blah blah blah like no I'm serious like we I hear you right but I think that was the story in SOPA and PIPA too but when 10 million people called it made a difference it works different over there we could not behave the same we have to be actually careful to distance ourselves from some of the you know fellow travelers because we don't want to pollute them with like oh well you're just Google Stooges but I will say also 20,000 people marched in the streets of Germany against the copyright directive did you see it on the front page of any German newspaper absolutely not and that's because the newspapers wanted it to pass so they were not reporting fairly on what happened and I mean I really wasn't kidding and you just brought this up this Google Stooges idea I mean it was really sad you could see basically anybody who had the same opinion as the big tech company that the policy makers were really upset about and trying to write this legislation around all of a sudden gets labeled as oh you're taking money from Google or Google told you to do this or whatever and it's like really hard to make your point when people are just making assumptions about you but I think you know setting that challenge aside if you can go through and just sort of make your point and make it earnestly and in this case about the carve out tactic explain to policy makers that hey that's never going to be a good strategy to kind of come up with this like laundry list possible exceptions when especially in technology you can't possibly anticipate the things that are going to come up later so why not take a step back and try to actually like make the legislation itself more tailored instead of using carve outs to try to make up for mistakes that you're making you can just say I'm Microsoft now okay so I think that ends our Q&A period but we do have a reception going on after this I see there's still food still drinks so if you have burning questions still I think our panelists are sticking around but everybody give it up for wonderful panel and there are some more fellows for arranging it and yes we have thank you guests for all of you