From West Wing to West Coast: What Silicon Valley's Learned from the White House

🎵 So I'm Joshua Mata, I wear many hats at Cloudflare, I'd be glad to show you them and talk to you about them after the panel, but far more interesting are the distinguished panelists to my right. Joining us is Nicole Wong, who was formerly the Deputy Chief Technology Officer of the United States. As the Deputy CTO, Nicole focused on Internet privacy and innovation policy. Prior to joining the Obama administration, she was the Legal Director for Products at Twitter and was Google's Vice President and Deputy General Counsel for Product and Regulatory Matters from 2004 to 2011. Nicole is a frequent speaker and author on a number of issues related to law and technology, including five appearances before the U.S. Congress regarding Internet policy, and we're delighted to have Nicole with us today. Also joining us is Cindy Cohen, who is the Executive Director of the Electronic Frontier Foundation, or the EFF. In her 15-year tenure at the EFF, Cindy has served as the Legal Director as well as its General Counsel, taking on numerous issues ranging from export control to electronic voting, copyright protection, and online identity. She's currently deeply involved in the issues of privacy and surveillance, both domestically and internationally, as well as reform of the Computer Fraud Abuse Act. According to the National Law Journal, and I quote, if Big Brother is watching, he better look out for Cindy Cohen. So we are, thank you, Cindy, very much for joining us here today as well. Love that quote. So, you know, to kick us off, in my role at Cloudflare, I actually spend a decent time in D.C. I won't say too much, and it strikes me that there are really two Washingtons. There's the Washington that doesn't understand technology, and there's the Washington that does understand technology. I'm very likely oversimplifying it, but do you find that accurate, and how do you think that dichotomy frames the relationship between the Beltway and Silicon Valley? So I'll kick off. First of all, happy birthday, Cloudflare. I think I could sing that without a copyright license now, so that's all great. But I'm not going to sing because we don't have enough time. So to go to the two Washingtons, I really learned everything I know about the Internet here in Silicon Valley. I can tell you, Washington, totally not like that Schoolhouse Rock cartoon about the bill sitting on Capitol Hill. And I think it has historically been true that there was like the part of Washington that got tech, or that didn't get tech, and then there was Silicon Valley. I think that's accurate, but it's changing, and for the better, for a bunch of reasons. One is, we're just bringing better people over. The line that you would see between the tech community and the political community is becoming much more porous. So starting with my former boss, Todd Park, who was the CTO at the time, he started the Presidential Innovations Fellow in 2012, and we started with 18 fellows who were taken out of the private sector and brought to make government better with technology. That is now a permanent program at the White House as of an executive order this year. So you have better people. You have the U.S. Digital Services, which is housed in D.C. to figure out how to make tech better. And by bringing in real people, for example, Megan Smith, who's formerly Google, now our new CTO, my former colleague from both Google and Twitter, Alex McGillivray, DJ Patel from LinkedIn as our first, in the history of the country, Chief Data Scientist, and a computer scientist like Ed Felton sitting at the White House who could talk circles about cryptography with anyone. That kind of firepower sitting in the West Wing is amazing, and we should take a moment to appreciate that we have that much talent in D.C. What does that mean for you as someone who talks policy in D.C.? It means that you're less often going to get blank stares from people when you describe your business, right? But what it means on the ground there is that when there's a conversation about some of the hard problems we are having about Internet governance or surveillance or encryption, there are people who speak accurately and authentically about what's at stake. That means that in the end, I hope, we will have better policy because the conversation, by definition, must be more honest when you have a technologist sitting at the table with you. I think that's right, and I think the executive branch, especially the White House, has done some great things. I think we really have to talk about Congress, though, where I don't think we have that level of technical expertise on the side that's working to preserve the liberty and the space of the Internet. And the thing that's changed from the 1990s when I started doing this kind of work more seriously is that there is actually a pretty powerful voice for turning the Internet into a surveillance state that is technically savvy. So it's not like there's just blank stares and lack of knowledge if we don't come in and make the case for freedom and understand how the technology works. There is a voice that is on the, let's take the Internet and turn it into the largest surveillance mechanism ever built, that is loud, strong, and making their case in Congress and in the executive branch. So we really do need to go. This isn't a conversation where dumb people just will kind of maybe not do the right thing but not do it because they don't understand things. This is a situation in which there is a conversation happening that if we're not a part of is going to go a very different way than the kind of the, you know, the promise of the Internet which got me interested in doing these things. And the other thing is, while, again, I'm very happy to see some people in the White House, I think that we need to think about technologists in the room, kind of the way some of the research has come around having minorities or women in the room, is if you just have one technologist in the room, they're going to get asked to fix the computers. You need two or three to be able to change the conversation in ways that are beyond kind of tokenism, beyond just stopping really horrible ideas and actually building good ideas. So we have to invest even more if you want to have an impact on the conversation. And I think the thing that has been nice to see at OSTP and the White House is that, you know, where Nicole went, it was kind of a couple people, and they're growing that. But I think we have to keep Congress in mind, and we have to, you know, maybe convince some technologists that it's okay to spend a little time on the East Coast before you come back home where the weather's better. Great. Well, so many folks in the audience are tech influencers, founders, tech executives. Nicole, having worked on both sides of the fence and talking about the shift of talent in both directions, bidirectionally, and the need, frankly, Cindy, for more people to be in the room, you know, what words of wisdom do you have for how to get involved? And maybe you can tell us a little bit about your personal story. Oh, go serve. So I had just started at Twitter and got a call, like, in January of 2013 from the White House, which I literally had to sit down, because I've never gotten a call from the White House, and asked me to come and serve. And I don't know if any of you know Todd Park. He is probably one of the most inspirational people I have ever met and talks passionately about how we can change government to make it better serve the American people. What Todd has actually done is not just talk about it, but create an infrastructure for all of us to go. The Presidential Innovation Fellows program is six months to a 12-month tour. U.S. Digital Services can be a one-year to two -year tour. There are lots of ways for you to find ways to plug in. And he and I have been talking even now, because we call ourselves the really, really West Wing now that we've come back to the West Coast, about how we get more people. Like, can we get someone who can take a three-month sabbatical but go help build a database for the Veterans Administration so it works better, right? How do we improve the immigration system so that people are not doing everything on COBOL to be able to get a green card, right? So there's lots of room for great technologists to go. And it won't be sexy work, but it will be impactful work. Let me take, for example, healthcare.gov. It's a website. The fact that we almost spent a billion dollars on a website is phenomenally crazy. And it took a SWAT team of people and technologists from the private sector to go there and fix it. And they had really low benchmarks of, like, we want to get 7 million people enrolled by the end of the enrollment date, but if you could get four to get through the process, that would be awesome. They got eight by the time March 31st rolled around. Eight million people who have never had insurance in their lives. And the letters coming in from the people were people who said, I have a preexisting condition. Neither me or my children have ever been able to have healthcare. They had it for the first time in their lives. That will change our society. So even though it's just a website, you can change people's lives by going to serve and finding a way to do it. We'd certainly love to find a way to get them on Cloudflare. We've got to fix the IT procurement system as well. But I think that when you're talking about government, there's kind of two arguments. There's two levels in which technologists need to be engaged in government. And the first one is the one Nicole's talking about, which is make government systems actually work better. That's important and it's great. It's not where I live. I live in making sure the government making the rules and setting up the processes for the rest of us don't suck and ensure space for innovation aren't just captured by the big companies who then use their power with government to try to fence out new horizons. There's a raft of issues there. So both of them are important. And one of the things I think is, I've got this idea that I haven't been able to pull together yet called Speak Tech to Power, where we really do want to take people who understand about how technology works and teach them how to be advocates in the various situations for policy changes and not just, you know, people in D.C. think technologists and they think fix my computer, right? And we need to change that so that while there are plenty of people come and fixing their computers and at healthcare.gov I'm really happy that they did, like that's not the pigeon hole that people need to stay in. People who understand technology need to be able to talk about policy in a way that helps make the rules and the limitations of those rules and the balance something that favors, you know, a continue open platform. And, you know, to me what made the Internet so awesome is that anybody anywhere in the world can have a microphone to speak to the entire west of the world and ideas can catch fire and people can have private communications and public communications and they're in control of which that they're having. We do okay on the first one I think most of the time. The second one, we got some work to do. And I think that like the work that Cindy has done for so long in EFF and so many of the other civil liberties orgs is you're totally right, which is like A, it's been a fight just to get at the table, right? And the right table, not the kids table, but the adults table to have a conversation around policy. I think that part of what we should recognize in Silicon Valley, because I think a lot of people, particularly at the startup level because you're busy building a product, but even the more mature companies for a very long time have thought that they could ignore DC. That time is closing. It is closing because they're getting smarter. It is closing because the things that we're doing with technology from Internet of things to self-driving cars head you straight into a regulatory space and you cannot ignore your regulators in that space. So you have to start exercising the muscle of wanting to be at the table. We will get the policy we deserve if we don't show up. We cannot sit and think that was a really stupid thing they passed if we haven't bothered to go and tell them why it was stupid. Right. Yeah. And you have to do both. I'm sorry. I'm going to- And I have a question though for you. Okay. So many people don't actually know that the EFF has been very instrumental to Cloudflare and actually dealing with these sorts of issues, particularly legal cases and whatnot. It'd be great to hear how the EFF can work with any company in Silicon Valley to really advance our point of view, get the seat at the grownup table, if you will. Well, I think that it goes both ways, right? Because I think if we're focused in the same direction, there's all sorts of opportunities for things to do. And in Cloudflare's case, we've actually represented Cloudflare when they got this horrible injunction because apparently you can't say groove shark on the Internet no matter who you are without getting in trouble. And we were able to help kind of narrow that back down to something a little sane. So occasionally if the issue is one that we also think is important, we will represent companies. In general, companies can afford lawyers and there's a lot of people who can't. So we focus the other direction. But if the issue is important enough, we will help people. We talk through things with companies. We try to build model policies that are things that people can do and try to consult. And then when it comes to DC, we try to work together where we can. It's not always the case that we agree with industry on certain things. But where we do, I think it's important that we try to stand together. We view ourselves as fundamentally representing the users. And users and companies often have aligned interests. They have aligned interests in intermediaries being free of liability so that people feel free to be intermediaries, for instance. And there's other instances. So we work with companies kind of directly, but we also try to work in broader coalitions. And of course, we're talking about DC a lot, but of course sometimes you just have to go sue. And so we also work with companies in terms of presenting technologically, we think technologically important cases to the courts in a way that helps the courts understand them as well. So whether it's through amicus briefs or other kinds of things. Right. So it would be great to kind of switch over to some of the issues and maybe trigger the fear impulse a little bit. Which is, as many in this audience are all too aware, around the Section 230 of the Communications Decency Act, SOPA, PIPA, numerous other legislative measures which have the potential to erode the openness of the Internet. I think it's probably not an overstatement to call these scary moments for people in our industry and probably everyone on this stage. Are there still things to fear and how fearful should we be? Well, I spend most of my time dealing with mass surveillance. And I think the fact that the NSA is sitting on the Internet at court junctures making copies of everything that goes by and then we have to trust their secret filters to make sure that we're not targeted or unnecessarily swept up in their behavior is a gigantic chilling effect on the kind of Internet that we want to build. And I think for me it's job one. Right. The Internet needs to be free of the giant all -watcher sitting at key junctures watching things. And we need to figure out how to have a system that's different than that one. And that one got built largely in secret. And now we know about it. We've got schematics in the New York Times. So I think we can't pretend anymore like we don't know what's going on. And we need to... To me that is job one. I don't think you can do a lot of other things. You know, I think every social movement, every political movement, every innovation moment happened because people were able to have a private conversation about how to make the world better. And if you can't have those private conversations without the risk of somebody listening in and having to trust that they just won't misuse that information, then you create a gigantic chilling effect. And that's about social movements, political movements. And I think it's about technological change as well. Trade secrets are important in the Valley. Absolutely. And it's not just the U.S. government, right? If you build this infrastructure, then other governments also participate and feel like it's okay too. I think it's important that we create a system that may be violated sometimes, but that there's a general sense of what's okay and what's not okay, so that right -thinking people know how they're supposed to behave. So I think that, to me, that recreating the space for a private conversation in the digital environment is job one. And we've lost that. And we lost it, I think, due to a lot of secrecy and a lot of people not paying attention, a lot of people burying their heads in the sand. So I think Internet policy to me has always been super fun, partly because it is challenging seven-dimensional chess. Because you're dealing with your government, but you're also dealing with every government that knocks on your door, and increasingly those governments are coming. There was a period in the first ten years of the Internet, so the late 90s to mid-2000s, where the Internet was largely about penetration into the U.S., the European Union, Canada, Australia, and Japan. And that was great, and we grew a lot of businesses there, but all of those countries, generally speaking, have the same approach as we do about rule of law and free expression and privacy. And what we've seen, certainly I have in the companies I've been at, since 2007 or so, the penetration into Middle Eastern countries, into Turkey, into Thailand, into China, into Vietnam, that penetration is going into systems which do not come from the same place with regard to rule of law and privacy and free expression. And we need to figure out how we're going to deal with that. So one of the huge challenges I see coming down the pike in the next several years is how are we going to make this interoperable? Because the promise of the Internet is that it is free and open and global. And the ability to moderate between all of those different values in all those different countries is becoming much more difficult. I'll add one more thing to be, this is a list of scary. So one of the things I worked on when I was at the White House, the President asked for a report on how do we think about big data? A, he was like, what is it? B, what policy things should we start to be thinking about? And I think that the masses of data that we are now collecting both in the private and the public sector and the mixing of public and private sector data poses a bunch of challenges for us. Not just from privacy, because if you talk to people about what are you worried about your data, they will often say things like I want to make sure I'm treated fairly. I want to make sure that I have the same opportunities as everybody else. That's not a privacy question. That's actually a fairness question. So there is an anti-discrimination issue that is coming up with data. And we don't fully understand or have not fleshed out what it means to ethically use data, what it means to build algorithms that are responsible. Look at VW. What are we supposed to think about how all of these companies are now implementing algorithms across data? And shouldn't you be worried about how that will be done with your financial data, your health data, your Instagram data, whatever it is. There are serious questions to be asked, I think, in the very near term that we aren't really equipped yet. We don't have the right people at the table. I think that's really important. And I think that in terms of back to your question, I think that I think that one of the biggest tools and one of the biggest things that we are spending a lot of time on now is protecting the ability for you to have encryption. And it's a very active conversation now. I believe in the White House the President is trying to formulate his position on this. And I think people in Silicon Valley need to speak up because the inability to lock your own doors will create a very different Internet than one where you get to lock your doors and you hold the keys, not somebody else or your keys aren't spread between five different, all of these strategies, right? We all know this is something this is an area where I think what technical people understand they understand it so well that they don't understand about how important it is to convince other people that you can't have a whatever door you have, back door, front door, or anything, you can't build a door that only good guys can use and bad guys cannot use. We say this over and over again in D.C. and believe me, all they say back is, well, you just got to try a little harder, geniuses, and figure that out. And I'm like, you know, I believe in innovation but I also believe in math. And math wins. And so we have to get this message through because I spent the 90s fighting the crypto wars and frankly I'm tired of it so do it as a favor to me. I don't want to do this all over again. We won the last time let's just put this thing to bed. Just because we're running kind of short on time the one final question I would have and maybe a brief response, 2016 elections are coming up candidates aren't talking a whole lot about the Internet right? What should they be talking about? What are the issues that people in this room should care about? The basis for which we should be backing which candidate? And then we'll open it up to a question or two to the audience. Have to be brief though. Well, EFF is a 501c3 so I can't get involved in election but I think the questions we need to be asking is what are you going to do about mass surveillance? Are you going to protect our ability to lock our own doors? And, you know, how are you going to preserve innovation? And it's not happening now. I think some of, you know, I do believe that the Republican primary is in a bit of an echo chamber where the only thing people are talking about is the stuff that other people are talking about. But I think the rest of us humans need to break into that conversation so we know what those candidates think because one of them is going to be the nominee. And I think we should be asking the same questions on the Democratic side. So now that in like post White House, post company, I have an amount of liberty I'm not used to. Please. Here's the thing. So in the Internet space we largely solve a lot of our own problems, right? And mostly what I hear is make sure government stays out of the way is usually what I hear. So when I think about what do I want to hear from the next presidential candidate or president about what their tech agenda looks like there are a lot of really serious problems that we face as a country and technology can solve many of them or be a part of that solution. So to me I would like to hear a candidate tell me I am going to continue to support all of the innovation happening in our country. But our talent pool is such that we should be focusing not on apps that let our dogs talk to each other but we have an infrastructure problem and we should be using technology to help us build more vibrant, sustainable and inclusive communities. We have an education problem where we are not doing the right thing by our children. We have a transportation problem particularly here in the Bay Area but across our country about how we get people and goods from one place to another. Technology can solve so many of these issues and it's kind of like building the website for healthcare.gov I actually don't think it's magical to be able to fix those solutions. It's about our talent pool being directed in the right place. So to me what I would like to see a presidential candidate say is that I am recognizing the talent that we have and I'm going to give them a platform to make sure that they are able to serve in our country. We may have time for one question Lauren has a microphone. Sorry. Right here? Sure, please. Thank you. What are some of the, what are some practical or realistic ways that you can imagine rethinking rule of law using technology? So things like machine readable law and smart contracts because I know there's a lot of techno libertarian talk about that in the Bay Area but what do you see as actually practical for the government? For a government institute rule of law? Oh, so there's like a whole constitutional process around that. There are jurisdictions which don't have strong rule of law right now for a whole variety of historical purposes. I'm not actually sure that technology, like handing them a machine readable constitution is going to help them with that. To me it's more of a grassroots issue of helping them understand the significance of what rule of law can do to enable them to move faster to society. To enable technologists to come in and then bring their technology to a society. I hope I'm answering your question. So the question is rethinking government and the way technology influences how government works. Rethinking government. So there is work being done now about rethinking government and how we serve the American people. Every agency in the U.S. government has a mission. Increasingly they are now looking at how can technology be used to advance their mission. It has less to do with rule of law per se, but there are processes that they can make easier using technology. Great. I think unfortunately we're out of time. We could probably spend all day talking about this. But please a round of applause for the panelists. Thank you. To help make the Internet faster, more secure, and more reliable meet our customer AO.com. An online retailer specializing in electrical goods from washers, dryers and refrigerators to televisions and home entertainment systems. They transact over 1 billion pounds per year. My name is Austin Davis. I'm a DevOps engineer at AO.com. We work on solutions to make development teams go faster. One of the challenges faced by AO .com was to be the best amongst their competition when it came to site performance. Faster web performance for the business is very important in our industry. In e-commerce speed is a differentiating factor. You can literally buy any product from us. You can get that from other retailers, but what you can't get from other retailers is speed. Having the best security for their business is another major differentiator for AO.com. Web security for the business is hugely important. We're seeing more attacks. We're seeing our competitors get breached. If we ever have downtime, even if it's just for 10 minutes, the cost to our business is huge. AO.com saw immediate benefits by selecting Cloudflare as their performance and security provider from quick production adoption to ease of use to competitive pricing. With Cloudflare, it seems like security is through and through in all the products, constantly in the forefront of what they do. With things like workers, with things like the security, they all seem to be miles ahead of the competition. With customers like AO .com and over 10 million other domains that trust Cloudflare with their security and performance, we're making the Internet fast, secure and reliable for everyone. Cloudflare, helping build a better Internet. music music music music music music You run a successful website that serves customers across the globe. Even after spending your hard-earned money directing customers to your site, visitors drop off before they get the chance to complete a purchase. What's happening? Modern customers are impatient. Your website load time has a tremendous effect on how long visitors will wait for your site to load. Research shows a 2-second delay can lead to more than a 4% loss in satisfaction and, most importantly, a 4.3% loss in revenue per visitor. Unhappy, impatient visitors leave before purchasing your great products, impacting your bottom line. The farther a visitor is from where your site is hosted, the longer it takes for the data to reach them, hurting website performance. That's where Cloudflare's Content Distribution Network saves the day. A CDN is a distributed network of servers designed to load your website as quickly as possible. By distributing parts of your website, such as images and JavaScript files, closer to your customers, your website can be loaded quickly, keeping your customers happy and your profits high. For more information about how a CDN works, explore the Cloudflare Learning Center. All right, excellent. Okay, so this session is the looking towards the future speculative one, which, of course, is fraught with danger. Luckily, I've got two people here whose biographies required six -point type. I'm not going to spend the next 20 minutes reading them out, but over here on stage right is Andrew McAfee from MIT. He's the co-director of the MIT Initiative on the Digital Economy. And many other things. If anybody wants this, know more about him. Or you could use the Internet, I guess. And then Larry Smarr, who is the former head of the San Diego Supercomputing Center. And by training, originally a physicist. And it's funny, in computing, there are many, many physicists, because, in fact, physics turns out to be about data a lot of the time. So we're here to talk about what the next five years are. So I was thinking about this session before we came on, and I realized that this year I've been using something close to the Internet for 30 years. So in 85, I got on a network in the UK where there's a few universities were involved. There were a few universities in the US. And looking over the 30-year span, it feels like enormous things have changed. I had anonymous FTP, it was very, very slow. But looking at the last five years, it almost feels static. It almost feels as if things haven't changed very much. Now, is this the ranting of an old man who can't see the changes? Or, you know, what's happening with the change in the Internet? With all respect, I think it's largely the ranting of an old geek. All right. Who doesn't really see some of the really profound changes. So you've been on the net for 30 years, I've been there for about 20. And as I was preparing for this panel, I thought about what mission statement we would have written for the Internet if we were being high -minded about it when we sat down and started playing with this thing. And my mission statement was something like, the Internet should connect the world's people and give them tools for education and for self-expression. That was kind of my mission statement for the Internet. And I thought, how have we been doing on that mission statement? And my one-word answer is, awesome. Just unbelievably good at executing on that mission. Particularly in the last five years, when you don't think much has been going on, the net and these tools for knowledge creation, knowledge sharing, knowledge acquisition and self -expression have flourished around the world in a way that caught most of us by surprise and is pretty darn fantastic. So the tools available around the world to anybody with a net-connected device to educate themselves, to share as much as they want in any media that they want, to contribute to our world's share of knowledge, this is honestly a big deal. And we're executing on that mission statement like crazy. The weird wrinkle for a lot of us is that, especially within the last five or ten years, a lot of the organizations that have been allowing us to execute on that mission are for-profit companies. And they're offering us a deal like, I'm going to show you ads in exchange for all these tools, all these resources that I'm giving you. And around the world, literally a billion people a day, that was Facebook's milestone a while back, said, I'll take that deal. That sounds like a pretty good deal to me. So, John, as we look ahead at the next five years, what's going to happen? My one-word answer is awesomer. We're going to get more of all that, and that's fantastic. But I think there is a looming threat out there. And I don't want to say that it's dire, and we're going to have to make the right decisions or the Internet goes away. It's not that bad. But there is a threat out there, and it's a threat from a coalition of people who don't like that bargain, who think there's something wrong with that bargain out there. And as near as I can identify, I co-wrote a book that came out in January of 2014. I've been talking about it nonstop. And I've come across these coalition members who really seem uncomfortable with that deal. And as far as I can identify, they break down into about two main groups. One is incumbents who don't like the disruption that's going on. And Exhibit A is basically the press in Europe, which has been doing a number of deeply boneheaded things when confronted with this new world of the Internet. My single favorite example was the Spanish publishing lobbying organization for newspapers in Spain succeeded in forcing Google to not link to their content. So in other words, they said, hey, on Google News, you can't link to Spanish newspaper content unless you pay us for it. And Google said, Google News is a non-revenue product. There's no revenue for us to share with you. So given that, we're going to take all your headlines off of Google News. Traffic to those sites dropped through the floor, and the publishing agency then turned around and said, no, no, please put that back. We didn't mean it. Like, what were they expecting to have happen? So we're seeing these boneheaded things, deeply protectionist, really wrongheaded things from some incumbent industries. That's to be expected. What I really was not expecting was this hostility toward the deal, toward this basic capitalist deal of I give you stuff, and in exchange, you look at my ads. To me, it seems like a really benign deal. I've been surprised over and over again at how many people I've come across who are hostile or downright enraged at how many suckers around the world are falling for that deal. And I just see the world in a fundamentally different way. So my only looming worry about the Internet is how this coalition is going to come together, and to my eyes, I'm choosing my words carefully, absolutely block the kinds of progress that we've been seeing and put us on a worse path for the next five years. All right. Larry, I mean, I flew here to the U.S. early on in the early 90s on the same vehicle I flew two days ago, which is a Boeing 747, which is the most efficient way of moving data across the Atlantic. You fill a 747 with CDs, DVDs, DAT tapes. That hasn't changed in 30 years. I think actually it was the same 747, given the look of it. That hasn't changed. At the same time, we've generated a huge amount of data from the things that Andrew is talking about. What is the challenge for the next five and so on years when you look at those two things brought together? Well, the fundamental thing that has changed is the distributing of data -generating devices. So all of your smartphones are data-generating devices. They also consume data. But in the academic research world, if you think about how many gene sequencers, mass spectrometers, microscopes, all of the things, telescopes, the things that are really discovering the universe and not just updating their Facebook page with their cat's latest brilliant achievement, that world is really kind of off the Internet because the data in the scientific world, unlike the megabyte, like if you take a picture and you put it on email or social network, it's gone instantly, right? Because the Internet's engineered for megabyte-sized objects. But science is about gigabytes. That's 1,000 megabyte-sized objects. Imagine a photograph 1,000 times more data than you're taking on your smartphone. That's a typical small object for, say, gene sequencing or anything else. And yet a terabyte, which is a million megabytes, is completely routine. And for large community efforts, petabytes, which is a billion megabytes, are in many different fields of science. But you can't put any of that through the normal Internet. And so I guess what I'm... I see everybody doing more of the same all over the world. It's a wonderful, it's a totally historic change for the human condition. But I also see all the things that it's not doing, the cloud's not doing, because it's great if you have millions of people doing small things. It can aggregate all that. But I think where we're going to see a lot of the progress is where the universities funded by federal research money are going to do what they've always done, which is bring you the world you live in. You know, where did any of this come from? Where did the Internet come from? That would be the Defense Department with DARPA and creating ARPANET. It would be the National Science Foundation in 1985 taking the results of those first few experiments like you had in England and deciding that we would build a national system, the NSF net, in 85. In 95, after it built out the regionals and the universities, networks, and it became the Internet. But I can remember it was 1990 before I saw the first .com. Always before it was .edu, .gov, or .mil for the military. And so I know you guys think you invented all this stuff, but actually it's the American citizens. This is what came out of World War II, was a plan in which the American citizens' money is taken by the real funder of early-stage innovation, the federal government, through the agencies and for the citizens invest that in the very best ideas across the universities. And the universities are a couple of training students and graduate students to that research. That model really doesn't exist at this scale anywhere in the world, and that's one reason that the United States economy is what it is compared to anywhere in the world. And I do think that model is in grave danger over the next five years. And if that happens, whether it's because of the cutting back of federal funding, the vacuuming of the top talent out of the universities into the Googles and Facebooks, et cetera, Amazons, so there's nobody left. You're eating the seed corn, people that produce, the young people coming. That is an existential threat. I agree with exactly half of that. The half that I agree with is that what makes people like Larry and me smack our heads is the fact that in about a generation, government funding for basic research, whether defense or non-defense, has declined by about a third. This is just a jaw-droppingly stupid thing to do. Because the part where I completely agree with Larry is if you trace the family tree of most big innovations back, you don't go too many generations before you find some kind of government research project. And I think it's really important not to lose sight of that. The part that I'm not on board with is this brain drain from academia. If the best minds see better opportunities to do their work and advance knowledge in the private sector than in a university, off they go. And I think our home industry of higher education is a lazy, sleepy incumbent, and it needs to have some pressure put on it. It needs disruption. Yep. Well, here it is. It will get it. Yep. All right, so we've gone back a little bit in time to the beginning of Silicon Valley with the government essentially creating it, which is kind of ironic because Silicon Valley is scared of the government, right, even though it made it. Well, the part that we should be scared of is a lot of countries around the world seem to think that their bureaucrats can pick industrial winners and losers. And the track record is incredibly clear. They cannot. So most of my economist colleagues see a really nice differentiation between funding very basic, very original research because we tend to underinvest in that otherwise. The process of commercializing that research works really well, I think better than any place else in the world in our economy. What I don't want is allegedly clever people in Washington picking technologies or industries or companies and, like, putting all the chips on the table. Terrible idea. Amen to that. Good. All right, so I want to go back... This is going to be less fun if we keep agreeing. I'm going to try and make you argue. So I'm going to go back another century. So in the 19th century, there was a technology that was created that was heralded as changing the world and everything was going to be better, which is the telegraph. And the telegraph, at the time, because Britain was a great power, we laid cables around the world and by the middle of the 1800s, you could send a message from London to Bombay in about 12 minutes, which is stunning, right? 12 minutes at that time. It would have to be relayed, but you could do it. And at the time, if you went and looked in the press, there were articles about the end of wars because people could communicate and understand each other. All the things that were kind of... We seem to sort of think about the Internet. But we've seen many countries close themselves off and perhaps even more unsurprisingly, Britain started monitoring all those cables and reading all the things that were going across them by the end of the 1800s. So is this communication medium really fundamentally changing things or are we fooling ourselves like the Victorians did? I don't think the Victorians were fooling themselves. If you look at the world over the past century and a half, one of the most striking phenomena is how much more peaceful it's become. We lose sight of that. We had a couple of nasty wars in the 20th century and it's easy to overfocus on those. Steven Pinker wrote a beautiful book called The Better Angels of Our Nature where he documented with real rigor the levels of violence, the levels of big arm conflict, the levels of smaller scale brush fire arm conflict. They are all heading in one direction and that direction is down. And part of his explanation for that is as we actually come together and communicate with each other, this is not just Pollyanna gauzy thinking, we do seem to thump on each other less. And there are way too many exceptions to that but that does appear to be the pattern. And David Brin in the previous, in two panels ago, actually referred to exactly the same book. And I'd recommend another book to all of you. How many of you have read The Victorian Internet? Yep. Okay, so the rest of you need to. Because that is the telegraph. And it was, I would argue that it was more globally, economically transformational than what we have seen since 95. Yeah, fundamentally changed everything. Okay, you guys are agreeing way too much. Okay, let's work on that. So Mark Andreessen, who's a fairly well-known investor, has said this thing about software eating the world. So why don't we talk about that? As a programmer, I'm very happy about that because it means I probably still have a job. But some people probably won't have jobs because software is going to do that. And it's also going to change the way businesses are working. Earlier on in the security panel, I said that a lot of these businesses like VW haven't realized that they're software businesses now, and that's changing things. So what does that mean, software is eating the world, and what are the consequences? I've talked with Mark about this. I came across an essay, not one that he wrote, but one that absolutely blew me away, published online, and it's called The Return of Nature. And it's written by a guy who documents these crazy phenomena around the world. I was completely unaware. Year after year, we are using less cropland across the globe to feed more people with more calories and better calories. We are actually decreasing in absolute terms the resource intensity of many, many things around the planet. One of his broad explanations for what's going on is he calls it dematerialization. You just need fewer atoms when you have more code and when you have more bits to accomplish what you want to do in the world. The evidence is striking, and I thought I kind of understood the phenomena. The evidence is absolutely striking. I think it's absolutely some of the best news on the planet. The one big caveat is I believe, and this is where Mark and I argue back and forth, I believe that one of the things we are going to need less of, one of the resources, is good old-fashioned labor. Mark thinks that entrepreneurs and innovators are going to keep coming up with things for people to do. I think they're going to keep coming up with really, really cool things. I just don't think they're going to need nearly as many people to accomplish those things, especially as we look out farther and farther into the future and our colleagues at Qualcomm develop the chips that are going to allow completely autonomous cars and vehicles and drones. Okay. In my lifetime, I think I'm going to see a staggeringly abundant economy that does not need very much human labor. More than zero, but not anything like we see today. So another way that I've talked about this is, and this is maybe because I'm a physicist, the world, you know, the physical world is made of atoms, not bits. It's a very strange thing. Take DNA. It's a molecule. Where is the A, T, Cs, and Gs of the genetic code? That's an abstraction that somehow rides on top of this. And so what we're seeing with the Internet, think of before there was a billion cell phones. Now there's a billion cell phones and there's going to be a trillion sensors in the Internet of Things. There's this plague of bits moving across the physical world. Now what software does is eats bits. And so if you think about it, sort of like the Cambrian explosion, until the bits were out there, the software would starve to death. And the more bits there are, the more the world can be read out and into a virtual version of that world, and which you now can take what we've learned about the laws of physics, which are themselves abstractions from the physical world, you can then simulate those. And so when we're talking about global warming and all that, next time you get a weather report that says the tornado is right here, that's because of incredible advances in supercomputing and simulating that physical world. So I think the software is eating the world. It ate it once before, which was before there was DNA. The world's been around a couple of billion years, but, you know, the earth. But until once self-replication became a property of molecules, it sort of built on itself, and that's the same thing we see now, which is a point that Bill Joy made when he wrote his epic blog and wired, the future doesn't need us. And that set off a global controversy, which really bears going back and thinking, because he got it, I think, very right in terms of sorting out what the implications of the software eating the world is. This is a bummer, because we're just agreeing way too much. Right. But one of the consequences of this explosion that you're talking about is that we are able to get smart about the world in ways that we never, ever could before. And this is a big deal. This is a knowledge explosion. This is like a human understanding of the world explosion. My single favorite example, how many of us use Duolingo, or have ever used Duolingo to try to learn another language? Hands way up, look around the room. This is by far the world's most popular method for learning another language. It's also the world's best, aside of having a romantic partner who doesn't share a language with you, that's first best. Second best is Duolingo, and it's a distant second, but second best is Duolingo. It's the biggest and the best platform for learning another language. The reason why that is is not because Luis Vanon, who developed it, was a scholar, or knew anything about how people acquire language. In fact, he tells a fascinating story that when he built this thing, and it started growing like a weed, and he's like, I think I've got something here. So he actually, he left academia to go start a company. I don't bemoan that fact. But he said, I thought I better go get some books about how people acquire language. So he started grabbing books and reading them about how to teach language and how people acquire language. And he said, I would go grab one book, and it would say, absolutely do this and never ever do that. And the next book I would read would flip that advice completely on its head. We had all these theories about the best way for people to acquire language, but because we didn't have enough evidence and enough data, and we couldn't run experiments, we didn't really, we were all just guessing. We were guessing for hundreds and hundreds of years. Luis, by running A-B testing on all this stuff, you know, down to the super granular level, has cracked that nut. And now we have better ways to acquire a second language than we've ever had before. I love that. All right. So we've got to our last three minutes where we're going to allow the public to ask some questions. So who wants to ask some questions of these two guys? This is Ping from Hong Kong Internet Society. So we promote Internet to get everyone online. So my question is, some areas they don't, without Internet it's hard to get online. So we aim not only to get the next billion online, but the last billions online. So what do you think about to solve this problem on how far we are from there? Thank you. I think there are two forces, actually one big force at work, and that's competition is what's going to make that happen. So if we sit around and wait for the governments and the countries at the base of the pyramid to roll out broadband to their citizens, we're going to be waiting a long, long time. What I would love for them to do instead is get the hell out of the way and let different kinds of bandwidth provider come in and fight in their countries and let Google fly their blimps and Facebook fly their planes overhead and get some connection that way. To me, that's the easiest and quickest way that we're going to solve that last billion problem. I wanted to ask you, what do you think about, you talk about the deal of accept your content with ads and stuff. What do you think right now with the ad blocking debate that's going on and also related to that the power of a company like Google or Apple to actually control that destiny? You want to go? Well, I don't worry about Google or even Apple, which is much more, I think, an inward-looking company that demands everything be a certain way because of the competition. I think they all know they could be gone in 10 years if, I mean, Apple had... Apple was almost gone. Once or two near-death experiences and now it happens to be close to the most valuable company on earth. But as long as we have this intensity of competition that we have here, I don't think anybody is safe. And my career is just long enough to remember when we were deathly afraid that Microsoft was going to choke off innovation throughout the technology sector. Anyone scared about that these days? No. Right? I have trouble getting worried about the alleged dominance of big tech companies because of the pattern Larry identifies. They do become dominant. These tend toward winner-take-all markets. These are a big deal. And then they get kneecapped from below by some little company you've never heard of that grows up into the next great big company. We should be vigilant about all big concentrations of power, but we should be a little bit confident, given the pattern in the tech industry, that today's giants might not be the giants of tomorrow, at least the movers and shakers of tomorrow. All right. Sounds like we've got time for one more question, if there is one. Sure. I've got one. You said something that resonated with me because I worked for a robotic telescope company, and we dealt with terabytes of data. I don't disagree that we're going to see innovation continue to steamroll. What I'm concerned about and what I'm wondering how we solve this is, as we build better mousetraps, how do we save ourselves from all that great technology just being used for mediocrity, like sharing those cat photos that you were talking about, instead of the scientific data that we could be advancing the human race with? Well, fortunately, the National Science Foundation has spent the last three years investing in over 130 to 150 campuses across the country and building big data cyber infrastructures based on optical fibers at which your point-to-point from the telescope to the astronomer is anywhere from 10,000 megabits a second to 100,000 megabits a second to the individual. And we can do, for instance, 9 .6 out of 10 gigabits disk-to-disk from inside of one university to inside of another university now. And I've just been funded as principal investigator to build that out across the entire West Coast as far west as Hawaii and east as Chicago as a subnational model for a separate big data freeway system. Just like in this country, you could argue that President, what Bernie Sanders calls that socialist Dwight Eisenhower, who had this big government program to do a duplicate transportation infrastructure. I mean, we had good two-lane highways, you know, Route 66, Highway 40. We had city streets. What was wrong with that? But somehow the cities decided they couldn't get from point A to point B anymore, so they built freeway systems, a redundant infrastructure made out of the same concrete and asphalt. And then Dwight Eisenhower saw that and called for hooking them together with the interstate highway system. And all of a sudden, you could go from New York, middle of Manhattan, to downtown San Francisco without seeing a stoplight. And that was radical as someone who spent 12 hours going from one state to another when I was a young person in the car. And that basically drove the economy for the next 50 years. And so the NSF, very quietly, I doubt that any of you know that this is going on, has made a massive investment of nearly $100 million in all of these universities to live in a new Internet for big data, all the while continuing to upload those very important cat videos. On that note, the most important note there, cat videos. Thank you, Larry. Thank you, Andrew. Thanks, John. Thank you.