 Good morning, everyone. We know that there's a few of you still getting lunch, but we'll get started so we can get through most of the questions and then move to audience questions. Welcome today to the Great Net Neutrality Debate being put on by both Berkman Klein and Jolt. This talk is going to be mostly about the January 4th order. And our two experts, Matt Wood and Professor Christopher You, are going to be debating the implications of it, the policy, the legal background of it. Professor Christopher You is the John H Chestnut Professor of Law Communication and Computer and Information Science at the University of Pennsylvania and the director of its Center for Technology Innovation and Competition. Professor You is a graduate from Northwestern and clerked for Anthony Kennedy before joining the University of Pennsylvania. Matt Wood is the director of the policy director at Free Press, which is one of the country's leading net neutrality advocacy groups, which has been very involved in the state attorney's general lawsuit recently. He'll be taking the other side of the debate. And a HLS grad and former Jolt subsider, I am told, in this file. You got to play to the crowd. So the rules for today are going to be simple. We're going to give each presenter two minutes to speak to their point and then give two minutes to the other for rebuttal. We're going to start with just an intro from each of them, telling them a little bit about themselves, and then we'll get right into the debate. So if Matt, you'd like to lead off, and then we'll move to you, Professor. Sure. So briefly, I assume a lot of people have heard about net neutrality. It's been in the news. It's been much discussed. And here you are. For those who haven't been as much into the weeds and in the docket, the FCC's order was voted on last December, as I was saying. It was released just a few weeks ago here in January. It took away the net neutrality rules that the previous Federal Communications Commission put in place in 2015. And these really aren't new concepts or new rules. I would start by saying net neutrality is basically a fundamental nondiscrimination law for internet access or for internet users. That's really the debate, is not just the rules themselves which did things like, and have now been repealed, but did things like prevent internet service providers from blocking or throttling or charging more to reach certain websites or use certain apps. Those rules are important, very much so, and we care about those. But we also care at my organization about the legal foundation for the rules. And they were based. The repeal is not effective yet. But the rules that have been repealed were based on what's called Title II of the Communications Act, and specifically the land provisions in that Act that prohibit internet service providers or at least prohibit telecom providers from unreasonably discriminating against their customers. So that's really the foundation is, do we need that law in the first place? We think yes, and do we need rules that then make that more specific and build on top of that nondiscrimination mandate? Well, I'm delighted to be here. One, because it's hosted by the Journal of Law and Technology. Actually, I've published twice there. I wrote an article called Beyond Net Neutrality, which was dated 2005, showing my total inability to predict the future. And for many years, was the most cited article by the Journal. I don't know if it's still true, but it's been a hot topic, but it was most downloaded on the website, and that's how it shows. I also feel I bear a certain responsibility for being here, which is, I actually wrote what many people regard as the second article on network neutrality. And it's sort of funny. Tim Wu was credited with creating the word because he's the first one to use it in print in 2003. In a journal, they actually, the next year, the conference next year asked me to write a response, and he wrote a rebuttal to that. And so actually, in some ways, I was the first to engage him on this, although the issues go back much farther. So I bear some responsibility for starting this fight. I do not take responsibility for how long it's gone on. I've been predicting its deaths, and it's at least 2007, and it just refuses to go away. I think it's interesting. Matt, the most recent order is the 2016 order, depending on the 17 order or 2018, depending on whether it's voted and when it's released, it depends on how you think about it, is actually the last in a long series. And it's actually, I think, a little bit more complicated than Matt suggests, which is network neutrality has been a bit of a moving target. In 2005, in the policy statement, it really was about non-blocking. And in fact, Michael Copps in a series of speeches really says we need to make this about non-discrimination because at that point it wasn't. In 2010, it did become about non-discrimination, and then in the end it became about, we added in 2015, interconnection. And so we have different aspects of this order that have moved along. The two things I'll say just to preface it is from a historical standpoint, since the beginning of IPv4, there has been something called the type of service flag which was designed for prioritization. It was retained in IPv6, and they added a field called the flow label. This has been part of the design since the beginning, has not always been used much, but it's an interesting exploration. Do we have five or two? It's five in the beginning, isn't it? Or do you want two? Okay, it's fine. The other thing, so first, as a historical matter, it is going longer, I'll leave the last second. The second thing I will say is, no technical architecture does everything well. They all involve trade-offs. And we had an architecture that was designed for a context in the 70s and really blew up in the 1990s around web and email, which was about a PC connected to a phone line doing file transfer programs before we had smartphones and wireless. The world we're in and have had to talk about this in conversation is very different now. And the interesting question to me is as what we demand from the network becomes more intense and more heterogeneous, when and under what circumstances should we expect that architecture to change? And that to me is sort of how I think about this problem. All right, thank you both very much. We're gonna get into the questions. We're gonna start with you, Professor Yoon. We would like, so most of us in this room have heard of net neutrality, we've heard about it, talked about it, but it's not clear to a lot of us as students what exactly the terms of the old open internet order were, what the January 4th order did. Can you explain to us really quickly what that order did in your own words? And then we'll go to Mr. Wood. So the first, if you think about the evolution of this, it pulled back from the 2015 order. The 2010 order we can talk about, but I'll set that to one side, it's the third of three orders. The 2015 order created three rules and a general conduct rule. Non-blocking, transparency, if you will, I'm gonna paraphrase, non-blocking and non-discrimination. Now they're not gonna use the word non-discrimination because of the reason the 2010 order was set, struck down because they used that word, but effectively, no disadvantage or these different aspects. The 2017 order kept transparency but modified it, got rid of the blocking rule, got rid of the no-pay prioritization rule, and got rid of the general conduct standard. And so we still have and changed the legal basis on which they acted from what the DCC did in 2005 was they did not fold, we're the only country in the world that did not fold the internet in the traditional regulatory regime that governs the telephone network as developed. In 2015, we changed that, 2017, 18 changed that back. So the legal basis changed and the scope of the orders changed was just transparency now. Yeah, I mean, maybe I'll take a step back and go back to what we were talking about at the beginning too and say, obviously this is a long debate. It actually probably started in the 60s if we don't wanna keep going back to the beginning of time and talk about when computers first started being used to send information over phone networks. So if we throw a lot of dates at you, 2005, 2010, 2014, 2015, 2000, I mean, it gets confusing but basically as I started by saying, I think it's the same debate throughout even though as professor you mentioned it's changed over time as to exactly what's encompassed. So as he said, the 2017 order repealed the 2015 order and that order had these conduct rules that prevented things like blocking, throttling, discrimination by internet service providers. It did retain some kind of transparency rules. So we have in the hopper now transparency rules that say if the ISP tells you it's going to block, it's okay, they still have to tell you. They have to tell you what they're gonna do in advance much like we see with privacy policies today and the federal trade commissions enforcement of those. So there's some debate about whether or not that transparency alone framework will be sufficient having stripped away all of the conduct rules. And I guess I would say that we don't think it will be put it mildly because for one thing we don't think that's enough to protect internet users. But for two, the major thing this last order did was once again shift the legal ground that we're standing on. And so even for things like transparency requirements they've changed the portions of the statute they're pointing to and in some cases not pointing to if we really want to get into the weeds and talk about administrative law and some of the failures in the order on that ground. So it stripped away the rules but it also changed the legal framework and I'll sound like a broken record here but to us that's very important. The legal framework and preserving internet users' rights to non-discriminatory telecommunications is exactly what we think the FCC should be doing to go back to what Professor Yu said at the beginning. Yes, there are ways in the network to prioritize certain kinds of content. Really our question and I think everybody's question should be who gets to make those decisions? Is it the content provider and the internet user on the edges of the network or does the internet service provider in the middle have a role to play in deciding what you can do and what you can say online? Thank you. To follow up on a question or sorry on a point you brought up Professor Yu could you explain for everyone in the room those four concepts that you brought up blocking transparency, discrimination can you flesh out a little more exactly what those concepts mean? So transparency is a bit, it has changed over time but it's basically explaining to people what service you're providing them. And in general we all understand networks get congested and right now Netflix is about a third of the prime of the internet in terms of peak times which is what matters and what causes a congestion. YouTube is about another between the two of them is a little over half and so it's largely video. And so what we see is these enormous things of video have really changed the way we do things. It is very common for networks in an application neutral way to when you use too much bandwidth cut you back. So there are bandwidth hogs you use a lot of it and the reality is if you do that if you hit your bandwidth cap they would say there are appropriate congestion management techniques as long as you disclose what they are and everyone knows what they're getting and they buy the tier that they should get there's probably a way to do that that complies with transparency. Non-blocking is you can't stop someone from reaching a particular website. So suppose that you didn't want them to reach I mean one of the standard accusations is there was a labor dispute in a Canadian ISP blocked people from reaching the union website because they didn't like the content and the non-blocking rule says said you can't do that or you can't slow it down so much that it's effectively blocking you're not literally blocking it but yeah it really are. The no pay prioritization rule was you can't pay for play you can't pay for a higher quality service and get better service. And the last reasonable conduct was the backup was sort of a general reasonable requirement about what you should do. All of these were subject to exceptions one for reasonable network management one for security illegal activity and the like and there was another category called specialized services the classic examples point in the orders. Amazon used to use dedicated bandwidth to send updates of your to your Econ your Kindle reader for books and they paid sprint for it and many people use for example dedicated bandwidth for voice calls because voice calls are very sensitive to delay so what they do is they have a special channel to do that that's what they called specialized services and in the order as defined those were things that were application specific in other words you're not doing a general service to the internet sub-priority service you're selling a very specific one for something that needs it like voice or the Kindle and as we said the transparency rule still survives in a modified form all the other things I discussed have now been repealed. Mr. Wood would you like to respond? Yeah Alex was asking if we'd agree on anything so I can say we do probably agree largely on those contours I think a few things to add I'll get logical for a second all networks can be congested some networks are congested not all networks are congested at all times and so that's really where at least one of the touch zones of this debate was when Comcast was in 2007 throttling back or effectively blocking all BitTorrent streams for some of the users because BitTorrent might be congesting the network this is a copyright debate as well people were concerned about pirate video or illegal video, uncopyrighted video so Comcast was basically dropping and blocking all BitTorrent streams and to me that's wildly over inclusive by them and eventually Comcast after a hearing right here hosted by the Berkman Center then in 2008 changed its policies and stopped blocking all BitTorrent streams instead decided to throttle people back when and where congestion occurred in the network so that's important and you know bandwidth hugs are sometimes also referred to as paying customers so people demand video and sometimes it gets congested but often times it's not and that's why we think that there is in any number of reasons to have reasonableness tests for what the ISPs are doing and do you think the final solution Comcast came through is appropriate? Yeah it was, it was time and place specific if you're gonna address congestion you should address where and when it occurs you shouldn't have a traffic cop on a corner that's not congested the second thing I'll say and final thing on this is paid prioritization is thrown around a lot and a lot of people phrase it exactly as the professor you did talk about pay for play really what the internet service providers mean when they talk about paid prioritization is not you paying more for faster speed because people do that today you can pay for a higher service tier what they mean is having the edge provider the content provider the person on the other end of the line also paying them to get their content to you faster so sometimes this gets very couched in economics debates terms when people talk about two-sided markets you know we could or one could debate the benefits of those but it's different than just simply saying you can't pay more for more speed because you and I can question is does AT&T or Comcast get to charge Netflix extra as well Thank you very much Mr. Wood this is a point of question to you you are part of a lawsuit of 22 state attorneys general to block the FCC's order this past year can you explain your theory of the case and why you think you're likely to win Sure in two minutes No First of all I am proud to have been appointed an attorney general but that's not actually the case it is a lawsuit that 22 states have joined in but they will be a separate party you know Free Press, Public Knowledge, Open Technology Institute few of us groups that are based in DC and other places are also challengers or petitioners on the FCC rule just adopted last December also in that suit right now are Mozilla and Santa Clara County which has taken an interest because obviously Silicon Valley is very near and dear to their hearts so I'm gonna stick with procedure for a second and then we'll talk more about the case on the merits I'm sure as we go those lawsuits were probably filed early I mean if the substantive intricacies here weren't enough for you there also are a lot of different triggers for filing deadlines and people taking Fed courts might be interested in this and nobody else everyone else can take in that but you know the FCC's finality of their rules is actually based on when they're published in the federal register that has not even happened yet here some five weeks after the FCC voted on December 14th to repeal these rules and to change legal framework once again so the reason I dwell on that is yes we went to court but those suits were filed early we explained that to the FCC they probably agree they wouldn't tell us they did for sure and that's the reason we did it is that back in 2015 when as there always is there was litigation a few folks filed early and they actually had the lottery run at that point and the appellate court that heard the case was picked in what you might say was a jurisdictionally premature time period for that to happen so when we do go back we'll argue about all these things I've been talking about legal classification the need for and the structure of the content rules the transparency rules and whether those are effective and whether all of this was properly noticed speaking of the attorneys general it's really been the New York attorney general who's led on this but talking about the unprecedented nature of this docket at the FCC I'm sure there were tens of millions of legitimate comments in the docket and that blows away every previous record but there also I think are undeniably lots of fake and fraudulent comments in that record where senators names have been used and people who passed away their names are in the docket and so it's kind of a mess and that's why there are procedural aspects to our challenge as well as substantive ones on the legal definitions and on the conduct rules and the content of those rules so I agree that it's an interesting phenomenon it will take a while for it to be published in the federal register actually the 2015 order issued under the Obama and the investigation I think took six months, eight months no, no, that was 2010 that was 2010 2015 was relatively tidy the rules were voted at the end of February and they were in the federal register by April 12th so it takes, but even that's a couple it's not unusual for it to take time it was released on January 4th and there's a certain amount of this has to happen I give a lot of credit to the current chairman he's adopted a number of transparency policies which we've never had before orders that they have voted on get released to the public three weeks in advance and that had never happened before I think those are all really positive there's an aspect of this challenge which I think is unlikely to succeed the Supreme Court in 2005 in Band-Ex held that this is governed by that the agencies have discretion it's ambiguous, for those who, FCC gets Chevron deference for those of you who are lawyers but the agencies have a fair amount of discretion to do things as long as it's reasonable and the reclassification that they just did again in January was basically that reclassification and the question then will be as Matt says is did they turn nice quarters in doing that did they explain the record and all that and agencies generally get deference on this stuff they win these more they don't I think that they're likely to win but we shall see I mean that's predicting litigation is always risky what Matt didn't talk about that's really invoked by the state issues which is something that's coming up in the news in the cycle right now is among other things the FCC preempted additional state action not only did they withdraw the rules but they preempted the ability for states to act in the void and that is a step that has really never been in play before and it's very different there's a lot of interesting law we've the courts of DC circuits I mean I'm sorry the FCC has done this before with something called the computer inquiries which was the before the big was the big regulatory regime goes back and forth in this and we'll have some interesting discussions about that and some states are looking for interesting ways and creative ways to try to take action in the space and the most interesting part of this suit will be that because that's something we've never really seen before let's do two more or one more round on this because I think we're starting to get into a flash point here so we'll give you Mr. Wood two minutes to respond and then you again Professor you yeah I mean those are those are two interesting aspects where you finished and I'll pick up there the preemption question is very interesting I can't claim to have exhaustively researched this but pretty exhaustively and I think the FCC is only one on this theory once in recent times and that is citing no express statutory preemption authority they have said basically we're going to occupy the field broadband is an interstate service and so the states can't act there even though we the FCC have decided the federal government has no jurisdiction either so that's interesting I mean the current chairman has done a lot of things differently than his predecessors one of the things he did when he was in the minority and in dissenting quite often in the Obama administration was he said you can't preempt state laws that prohibit local broadband systems by municipalities unless you cite specific statutory authority for that and he also said the specific statute cited by that FCC was not enough so now we have this much broader field preemption claim coming out from this FCC saying we don't need to cite a specific statute we preempt because the service is interstate and we think that's a better regulatory policy now I think it's an interstate service too don't get me wrong I just think everything else the FCC has done is problematic and that's why the states can step in and as they've always done there are some local aspects to interstate services and we've always had in the phone system that kind of interstate versus interstate issue that goes all the way back to commerce clause decisions so that's interesting and we can talk more about that on the Chevron deference point I agree this has been a Chevron line of cases up till now Chevron deference is not infinite and it's funny because in 2005 when the FCC went the wrong way in my view they were upheld by the Supreme Court and brand X on Chevron deference in dissent was noted liberal fire brand Justice Scalia saying no that's wrong this is clearly based on the statute a telecommunication service meaning something that transmits information for us and is not itself some kind of information service so that line between transmission and content which is complicated but we think clear and it has to be maintained so in the era of Justice Gorsuch now when people saying maybe we shouldn't have Chevron deference or major questions I think coming at this as an academic and an advocate is interesting because as you said so well trying to predict what they'll do this time around just based on the last few rounds is pretty hard the specific preemption case he's talking about which went the other way involves a slightly different issues for those of you who don't follow this closely many states have prohibited cities within the same state from building their own municipal broadband networks and the FCC in the last administration put an order saying we preempt that rule and there is a line of Supreme Court authority tracing back to Gregory versus Ashcroft that's limit the federal government's ability to interfere with the state's relationship with its own cities that aspect really isn't here that federalism overlay doesn't apply here in the sense that there's not overturning a state law that has a state-wide law that affects relationship with the cities here you get to an interesting set of cases from the computer inquiries interesting to lawyers and geeks like me it's preemption so for anyone who's a lawyer knows this is God awful law I mean this is not gonna be neat tidy law but there's something called North Carolina Utilities Commission out of the Fourth Circuit California Three out of the Ninth Circuit there's a whole line of cases and I will say that they are complicated I mean it's not like you'll see uniform wins or losses but the FCC won in the fourth on computer two they lost in the ninth and computer three and we're gonna have to have a very interesting fight about that particular preemption aspect of it but what's interesting is I agree with it's actually I we're surprisingly agreeing on a lot of stuff I think this should be interstate the difference if you want to sharpen the dispute a little bit I don't think discrimination non-discrimination is always the right attitude so to use a student audience to give you an example one beneficiary that you almost all benefit from of classic discrimination is the classic student discount which is you see you're getting charged less for the same meal that I get charged and the answer is there are times that you want to do that particularly because if you charge one uniform price the students who are have no money can't go and so what you can do is actually get create a discounted price for people that's still above cost and still good for everybody so they can go and shop and do well and at that point you're actually if you had to do a uniform price you wouldn't be able to do it but the ability to differentiate that makes it possible and I will say from a service standpoint I would pay for a better connection from my home to my office and my email server and a handful locations I do well because I use them so intensively and they're just much more important to me those sorts of interesting services in some of them that I'm happy to talk about will start to show some of the so we're not in complete agreement here this is supposed to be a debate some of the differences between me and Matt Thank you Mr. Wood a question for you if our concern here is unfairness and censorship the blocking and the discrimination principles we raised earlier what stops the Federal Trade Commission from simply coming in being our main regulator when there's cases of unfairness and anti-competitive discriminatory acts and bringing enforcement actions what prevents us or why should we not just go that route? Yeah, there are several things that we think of as problematic with the FTC just we had a whole other huge category of law into two minutes I think I would try to narrow it down here and say the Federal Trade Commission does have a similar statute to the Federal Communications Commission in some respects now the FTC doesn't have rule-making authority as a general rule so they couldn't have the kinds of rules against blocking and throttling in advance and it would be all after the fact enforcement and I can't help but pause on that point and note that our friends at the internet service provider community change their tune on that quite a bit they'll say Federal Communications Commission we need bright line rules we need to have clarity in advance you'll say what about the FTC and they'll say yes, yes we love the FTC because we do after the fact adjudication and everything's post hoc so I've learned this trick in DC if you don't like a standard you call it vague and if you like it you call it flexible and really the only difference is the adjective but the reason the FTC probably can't even do anything about blocking if they do have any power if the standard is a good way to come at it is that here we get to squeezing a whole area of law in the two minutes the FTC statute says unfair and deceptive practices are prohibited now the Federal Communications Commission that draws the ire of the cable and phone company says unjust and unreasonable so you wanna talk about a great fight for lawyers like unjust and unreasonable is too hard on us but unfair and deceptive we understand perfectly I don't believe it but I also don't think the FTC can stop blocking because they have essentially mothballed the unfair part of their statute they do a lot of antitrust enforcement at the FTC and they also do a fair amount of privacy policy work and again that goes back to basically if the company tells you what they're going to do and then keeps that promise they're okay so if the company says you know what we are gonna block some websites I don't see how the FTC could go back at them unless they were to unearth and revive this unfairness doctrine that they have been lost themselves in court on but more we're talking about the 70s now rather than the 90s and the aughts and more recent cases when the FTC really withdrew from that part of its statutory mandate. So it's a bit of a knit but lawyers do this actually the FTC does have rulemaking authority and if you go to the webpage I was surprised you see Actions Actions Actions They have about 20 or 30 rules there now the interesting thing is they have a higher level of procedural requirements in the hoops they have to jump through I don't I think that the point is well taken is it's harder for them to make rules I think that's true but it's often stated in the category of they lack real that's just not true it is they're held to a higher procedural level I don't think we really disagree on that but it's just whether that's good or bad is a different question. What's interesting to me is the question about rules in advance versus ex-post adjudications is a really interesting one and from a lawyer's standpoint what always struck me was I always think about three different forms of literature one is the rules versus standards literature which were taught in law school I'm also thinking about the difference between what we call per se illegality that you can't do it absolutely rules in advance versus rule of reason and antitrust law and the last thing was the codification movement that happened in the late 1900s where people said we need rules in advance and what's interesting to me is the rules in advance are as never as clear as the people who want them think they are and the ex-post adjudication probably are never as vague contracts, torts, property almost everything we teach in the first year was all done in common law and basically we understand how to do most of those things. What's striking to me is the internals of the rules versus standards or the per se illegality rule of reason framework says when do we do rules? It's when we have a lot of experience with something we can say with great confidence when there's no baby in the bathwater we don't need to do these costs but there's a reason why we still do tort liability through reasonable some of the totality of the circumstances when we could have clearly bright line rules and conduct that's very common and it's the reason the codification movement ultimately failed which is when you have a dynamic multivariate phenomenon and I would say internet access right now is one it actually there's a real danger that by creating a rule you may lose something the best example is what we call zero rating for those of you who have T-Mobile they will allow you to write T-Mobile allow you to use a product called Binjohn which will let you do unlimited video without accounting extra data cap the last administration did an order report was very skeptical of it the current administration withdrew that report and ended that investigation that's the kind of emerging practice that sort of hangs in the balance between these two interpretations Thank you Professor an economic question for you we know that half second delays even as small as that can cause significant drop offs in subscribers and users of any app or service with that in mind how are we not concerned that paying for extra speed by big players like Pokemon Go like Netflix is going to cause competitive problems for small startup companies that want to get into data heavy services that won't be able to front the same amount of cash as them it actually is not quite true that half second delays are bad for all apps actually the key idea is that apps vary in their tolerance for delay so if it's a file transfer program if an email shows up half a second later most of us won't notice if you're doing a phone call and there's a half a second lag on it you'll be talking over your other person constantly we've all had that experience and if you look back in the original design of the internet they had it optimized essentially for email and file transfers and they started to change the architecture from the very beginning because voice needs a different set of services in other words if it drops a packet we need to get something that's fast even if it's not as reliable it's basically the trade off they made and we're starting to see this with more and more applications in particularly video real time interactive applications that are particularly image heavy are very sensitive to delay pre-recorded video if you don't mind it buffering for a second once it starts it'll actually work flawlessly and so one of the interesting things about prioritization is if you are on your phone call and you're in a dead spot most networks are engineered that if you're on your phone and you're having trouble getting in the low bandwidth they will hold your data and give you your voice they will discriminate between two applications generally for your benefit because you care about the lag and if you're on your phone it's more important that it comes through and you're probably not checking your email anyway and so this is an interesting way that taking advantage of the growing heterogeneity of the way people are using the internet to actually vary the way the internet provides you services and in some ways I think of what we're seeing in terms of the development of the different services is a response to the natural response of a provider trying to satisfy customers let me give you another example financial services have exited the internet completely why? they need millisecond latencies markets move prices and in fact one of the hottest retail market real estate markets in Manhattan is the real estate around the network interconnection point because they need to get closer and closer to them and it's gotten to the point where they've actually introduced lag, uniform lag so people don't do this race to the bottom trying to buy real estate that's another example of an app that needs something very different that the current best efforts internet can't support and so we start to see are they and because they can't get it from the network they're paying for a private line connection which is their alternative they're still paying just not in a shared infrastructure that would support the way we've gotten the internet okay so there's a lot there as what Professor Yew has just said shows us is that we do have private lines still possible even with the net neutrality rules when they were in place and we've always had that in the telecom network so that is not really something new yes remote surgery and medical applications often comes up don't you want people to have good access to medical care on the internet well I probably don't want surgery done on the best efforts internet you know that probably would be something where you'd want to buy a dedicated line so that has happened in the past that still happened under the rules that have just been struck down it's probably something that will be allowed in the future as well no matter what regime we come to even if the current we would say radically deregulatory attempt by this administration doesn't win out so there are ways to solve for this I think we're doing a lot of legal noodling I don't want to get into engineering and pretend I know enough about that stuff but for a lot of the kinds of delays and latency issues that Professor Yu is talking about capacity solves them or does a lot to solve them I guess I would say this to be somewhat more careful and qualified so yes as I said earlier some networks get congested and some get congested a lot most of the current landline broadband networks we have coming into your home are not getting congested at least not in most parts of the day and so there are questions about economic incentives that we want to talk economics should we be incentivizing building out more capacity as we see with things like Verizon Fios and Google Fiber when that was a thing or should we be allowing providers to use that common infrastructure and to limit capacity or at least to respond to limited capacity by charging more for a bigger portion of that pipe so I'll go back again to this fact that we actually can and do by more speed today as internet users you said you would pay more for a better connection to certain servers you use the question is should you be allowed to do that I would say yes should then the internet service providers say you've bought more capacity here's what you're allowed to use it on only on certain servers only on certain websites only on certain applications frankly I don't need Comcast or Verizon telling me which services I can use with the capacity I purchased I would rather have that capacity and make that decision for myself and that's why we think the non-discrimination rights that we had in place under the law and under the rules are so important it leaves the user in control of that rather than the edge provider and the ISP negotiating between themselves for who gets there first this is our last question before we'll go to live audience questions and this can be for either of you to lead off with and have the other respond we had we saw recently in the past couple years Brazil implementing their civil rights framework for the internet a bit of a departure from what we're doing here in the United States to what extent to the broadband policies of other countries affect how we approach net neutrality obviously internet services are cross country servers are in one place clients can be in another are we expecting other people to follow in the footsteps of the recent order or do we think that that's going to be a problem compatibility-wise going forward I don't know what to say exactly I mean yes it matters and especially because people do talk to each other in other countries and the internet is a network of networks as we all know or should know so it definitely matters I think that this tends to get perhaps spun a little bit too much in debates like if we do this then they'll do that over there and that'll be the end of the world I don't know that that's really the case and again I see advocates doing this all the time they'll point to Europe and say see it's terrible because the speeds are too slow but then some people used to say oh it's too fast there was an advocate in DC who used to talk about we don't need what did he say Maserati's speeds Toyotas are fine for most people so you just get this kind of understandable but sometimes silly attempt to claim either that the laws in other countries have an extra territorial effect on us here quite literally or that the precedent set will be bad and I wouldn't deny that the precedents are bad I mean I'll point to the Obama administration official for once and not the current administration you know one of them got in trouble for saying we don't want China blocking why would we want AT&T blocking I think that's an interesting question I wouldn't say that those two are exactly the same thing or despite what there's another former Republican commissioner McDowell who used to say or another one too Harold Furch got off you know if when they see us regulating the internet here which is debatable and I think they're framing it wrong that will embolden China and Russia to block and I think people do these things for their own reasons you know global politics around communications rights are not dictated or swung strictly by the FCC even though of course it matters how the network operates across international borders so if you're asking about how international developments will affect US law whether you believe that I believe that's always been somewhat limited frankly but in this current administration probably even more so I mean it's just the reality we're in I don't think it matters they're all you know I do think US policy has tremendous influence internationally most regulators don't have the capacity to do studies and so what they generally do and I went to I was at the Mobile World Congress which is the biggest gathering they have a ministerial there when the 2015 order dropped and the basic thing they say is if the EU and the US are on the same page and at that point the EU basically was the presumption of most regulators is that's what they're gonna do or the burden of proof is clear on someone else to pull the other argument to go the other way and what you've probably seen is the EU had a very strong proposal out of the parliament they've retrenched some we've retrenched some and I'm sure I'm actually going in the end of next month and we'll be interested to see how they react my guess is they'll be a bit puzzled by trying to make the sense of it because it's easier for them when there's a clear consensus between the US and the EU and they sort of follow that lead as we go but I would say probably less of the world will follow the US lead under the current administration than has been true in the past I realized I didn't fully answer one question because you asked it in terms of innovation your last one to me I want to boil it down because I got into the examples what I would take away there are innovation interests on both sides of this issue to me there are innovators like financial services and other things who want something different from the network they can't get it from the current architecture and even among people who are currently using the current architecture bloggers who are sent text stuff does not need big fancy speeds video needs big fancy speeds and so what we're seeing is if we have a single uniform internet with one class of service and everyone has to pay for it what you're gonna end up having is people paying for stuff that they don't necessarily need a blog actually is not this doesn't need to speed latency it doesn't really matter that much a half second delay on text is going across and we're starting to see as differences come out that way as well so to me there's innovation on both sides and getting more complex into it's much more interesting to me can I just answer on that one briefly I want to get to the I think the blogger who doesn't need video speeds is paying for their own upstream capacity and doesn't need to overpay most users are not going to say I only want text the user might pay more to have a more robust line that can handle video but I just I don't think this is a much longer question we get into now but I don't think it's fair to say the blogger is paying for my videos it's just not really the way network economics work if you ask me but that's obviously not just a debate but a symposium maybe we can have some day all right we've got about 15 minutes left we're going to turn to live audience questions here you can take this one we're going to so we'll take off the time limit for this one if there's any questions that you submitted that you didn't hear this is the time to ask them and then we'll let you all we'll let you all head out at about 1250 because we know a lot of you have one o'clock classes I guess one question I have is what would you say to the consumer who pays for 10 megabytes per second who actually prefers the new rules because Comcast allows them to get 100 megabytes per second if they're accessing some approved video service like Google or Amazon to that consumer is that actually a benefit I don't think that's happening I mean I guess it's in my mind it's similar to zero rating is that well no but see under the old rules too this you're really talking about capacity and so if Comcast charges me for up to 10 megabits per second sometimes it's gonna be less than that if the network is congested if instead they're providing faster speeds to me that's good for me it might be bad business for them although I would say it isn't even that because it doesn't actually charge that it doesn't they don't have a big marginal cost for doing that nothing about the old rules prohibited them providing more raw capacity to you you know it was done neutrally so I have 100 megabits per second now accidentally even though I'm paying for 10 I can use it for anything video anything else like I just don't I kind of reject the premise of the question I guess without being too argumentative because I don't know if I'm missing something right maybe I'm missing so under zero rating it wouldn't it would be limited to that specific service yeah but we didn't see zero rating you know you're talking about speed rather than the amount I guess that's what that's what I'm missing so you're saying you know you're allowed to go over your cap so yes you know the speed is how much I'm getting at a certain time the cap is how much I'm allowed to use per month and while professor you is right there was this debate about zero rating funnily enough because they see it as a better business most of the mobile carriers have now gone back to an unlimited approach as well so that's another fascinating area of debate you know are we better off if bandwidth is metered out and I'm only allowed to use up to a certain cap per month and then I pay for over that some people would say that's better for consumers and others would say it's worse but the new rules you know do allow for that for that kind of zero rating because there are no rules preventing it they've taken away all the rules I would say under the old rules that were just struck down that behavior might have been allowable too it's just that the FCC was able to and was in fact looking into whether it was actually pro user and pro edge provider rather than being something that was strictly pro cable company I think you hit the nail right on the head so if you took binge on away from my kids they would be pissed you know they're getting a real benefit from doing this now there's a couple really interesting things about this first T-Mobile is the number four player and if they try to compete based on network quality and bandwidth AT&T and Verizon have them right now now they're trying to make it better it opens up the number of dimensions they can compete because they're appealing to a small group of subscribers and they're doing a niche play basically and they're finding people who are probably more tolerant of packet loss and all these other things and I actually think it's a really good competitive move that allows diversification to make those things happen interestingly AT&T with the direct TV acquisition is experimenting with different new business models video distribution Comcast is making a video product available to its subscribers it's complicated because it's part of their cable serve I mean we get into these interesting definitional problems but what we're seeing is these new business models really emerge that I think are quite exciting and if you have to bear in mind binge on was pioneered by Metro PCS when it was the number eight provider in a 3% market share I'm sorry I'm often providing LTE on 1G spectrum these tiny slivers I'm sorry what they do should be per se legal and they were subject to the first network neutrality complaint under the 2010 order there's no complaint much broader we do as Matt points out we do have private line service but there's a new thing going on that's sometimes called network virtualization or network slicing which is part of the key 5G business model which is the problem with the private line solution that Nat mentions is you have to buy it 24, 7, 365 you buy a line and it's yours what they're experimenting now is basically temporary private line service on a stand up and take down basis which is I need this connection not all the time but I need it between this hour and this hour so I'm doing something and they're gonna allow you to reserve resources in a short term private line service and it's gonna give you a better faster connection and you're gonna have to pay for it and many people see that as the version of 5G and many people believe and there's a whole bunch of writing coming out of Europe and other places that think that certain implementations of network neutrality because this is pay for play you're paying for better service on a temporary basis they think that strong form implementations of non-discrimination potentially stand in the way now maybe we don't and we'll get into arguments we never really got to that part of the implementation of the order but the question is will the ambiguity around this harm people from experimenting with different business models my position would be is expose case by case I've been called in successive speakers in a panel a network neutrality regulations and deregulationist because I believe in expose case by case enforcement which means I do believe in some enforcement but I think you wait to let someone try the practice and if the consumer harm is shown stop them but the problem is by reversing it the other way you never find out you put the burden of proof on innovation and you lose the breathing room that innovation needs to experiment with different solutions I do think we need some form of backstop here to make sure that it's fair but absolutely to prevent people who haven't tried something from trying it in the first instance so we can find out what the impact is I think it's bad for innovation I will briefly fill time as you go to the next question binge on has gone away and it's in that form because they've gone to unlimited so I think I just have to say if you say to people you can have free video over your limit they'll say yay but if you say instead actually there is no limit people will prefer that over having something parceled out to them on a non-neutral basis even though they're not sitting there thinking about it in that neutrality terms and we can argue with the carriers about whether it's a better business model for them or for the entire internet if we have metered plans instead but you know it is not just the it's not a fact of life there has to be a limit and only by allowing this kind of innovation can we get past those limits we've actually gotten past them today with most carriers well but the other thing that's interesting to me about binge on I do think well we have a factual question I think it still exists because unlimited plans get I do think that if you have to buy above a certain plan you can get it even if you have a cap the most fascinating thing to me about it is it's a video specific solution which is it takes the result it mostly succeeds by taking the resolution from high definition down to standard definition to reduce the bandwidth it's requiring now this is a very creative solution that actually is targeted at the problem which is video it cannot be done in an application neutral way you cannot reduce the resolution of other things because video is a unique product in this way and it addresses the problem that is causing congestion on the network today and so what's interesting to me is on a technical side if we say everything has to be done in a pure application neutral way you will lose some efficient technical solutions which we would otherwise have so my question is in the 2015 rulemaking order the FCC only cited one violation of what would become the net neutrality rules and that violation they were able to correct without the expanded powers of the 2015 rulemaking so I guess my question is what's to say that this becomes a bigger problem in 2017 than it was pre-2015 and what really is the necessity of keeping expanded rules that weren't necessary in the first place which one are you thinking of because I think they cited more than just one I thought they only cited one maybe two and in both cases I don't remember the exact which exactly the case was but in both cases I know that they were able to solve the all right so the legal framework has shifted over time but we have had some kind of rules or principles in place basically continuously so I mentioned earlier that Comcast throttling BitTorrent and what happened there was that Kevin Martin the Republican FCC chairman said that's not a thing we want we don't want Comcast blocking BitTorrent so he actually took an enforcement action against Comcast Comcast sued now they changed their behavior after the enforcement action so that's a good thing but they sued and they struck down the legal framework on which that enforcement action was based so again I think I would not entirely agree with the notion that it was solved without these rules it was solved with a version of these principles later found to be legally unsound by an appellate court there are others and the reason I say there's more than one is you had telephone companies blocking VoIP you had providers blocking access to things like FaceTime and Skype on the mobile network you've had providers blocking payment applications that Google owned in favor of their own payment applications this is funny the ISP payment application was called ISIS so bad branding but how could they have known in advance so there have been actually several circumstances there are some a little bit outside of the ISP realm where Verizon was blocking text messages sent by NARAL and saying we have the ability and authority to block unsavory content so I think all of those problems were addressed with versions of these rules and principles that were later found to be unsound by an appellate court based on the legal foundation and the 2015 rules are largely the same they've evolved over time as Professor you notes largely the same as those principles that were struck down but now once again based on solid legal authority to address these and future problems that we just might not be able to contemplate right now I guess I agree with the question basically you don't see a lot of these examples in fact the last time I had one of these debates was last week and the person I was debating says we don't actually see a lot of these I think I can think of two examples but one of the I guess I should be saying this as a professor actually wrote a paper looking at all examples cited some of the ones that actually Matt mentions are often cited in my opinion miscited in the debate so the NARAL was about text messaging actually that's outside of network neutrality have the order don't have the order won't affect it and prohibited by title two but my point is that it's not a solution in search of a problem that we found but it's not this wasn't really about the networking in a network neutrality order Google wallet and the payment system uses near field communications which also isn't really touched by this now people would say we don't really care it's all about communications and to some extent the problem we have is an old statute that's really built around technology so we have to care I mean this we're fighting over just definitions written decades ago which were not didn't have the internet in mind but there's an interesting question hanging in the balance here about how many examples you need before you act the flip side is to remember the other side is how many examples in the other side of innovations we're going to lose and we have to take this all into account to really make them go and so what I think that was interesting is I do find that there was a dearth of concrete examples as you mentioned but I do think that what Matt says is the alternative is to build more capacity instead of managing the network the problem with that is it's expensive and what we see is a lot of people in rural communities and minority communities actually filed an opposition because if you can only build more bigger pipes your breakeven number goes up and so there's actually a concern that there's a deep interaction between this and some of the digital divide stuff not just an innovation standpoint but in terms of the service we provide people and so the question is if we can do an OPEC solution and operational expenses instead of capital expenses why will we tie one hand behind our back because there's different things that are appropriate at different times we have time for one last question I'm going to keep talking annoyingly the rural and communities of color solution is a little more complicated than was just said and there actually were a lot of people who were on our side in that and I do want to go back to this notion that we can discriminate our way to innovation and I think that's the genius of the internet is that the innovation can happen in a lot of places so I would just ask people to question whether or not we need to discriminate in the network in order to innovate or if in fact the absence of discrimination is what allows for innovation to flourish So I guess my question is you previously made the distinction between discriminating speed for price against consumers which we both recognize has already kind of gone on with these ISP discriminating against content providers in terms of speed for the price they charge them but doesn't that already happen in some ways with larger companies having either duplicate servers in various areas or CDNs so I guess really what changes under the new system? So CDNs and other methods to increase I mean a lot of this debate stems from 2005 Ed Whitaker who was then CEO of SBC which is now AT&T said Google shouldn't be allowed to use my pipes for free and so what he was claiming was for the first time I think the phone company's right to charge not just their customer but the person making the call as well you know there's a whole history here of telecom services being regulated as to price and so they could exchange money in with each other and they could make that connection work or they could each just charge their own customers and each phone company making itself whole based on resort only to its own customers but I think what you're pointing to is the notion that yes I can improve my speeds if I take technological steps to get my content closer to the user that's just the fact of physics and of networks the question is should the ISP be charging for that privilege or be monetizing it in some way or should others be allowed to take those steps to improve their speeds and not have that kind of gatekeeper in the middle in terms of what the cable or phone company does to their traffic? So to step back about what this issue is about if you want to get better service you can build your own data center close to your customer you'll get lower latency you'll get better responses and it belongs to you and other companies call content distribution networks the biggest of which is Akamai based here in Cambridge do this on a third party basis you don't have to build your own data center we'll build a data center and we'll sell it out to third parties as a service and we'll give you better service but you're gonna have to pay for it this is a great example of substituting storage for networking in other words we don't really rely on the network we'll just build more storage facilities and reduce that that's a great move when you can't get priority service from the network it's your only move you can only do it in the technically build more facility solution the interesting question is can we open up the space where you don't have to pay that person instead we'll find a network based solution that actually may allow different kinds of sharing and different kinds of aspects because what really makes us all go the reason the internet works so well is we're all sharing capacity that used to be dedicated to one person we can make people buy their own CDNs and do this but this actually unlocks the potential and that's the whole idea behind network slicing and we've seen it in cloud instead of having your own computer you share it with other people so you don't need to pay for it when you're not using it this is all the notion of virtualization in many ways there are business models that would take this to the next step in the network that are actually in extreme tension with traditional network neutrality and the worry is that those business models will be stifled with two restrictive limits on what the providers can do I think it's just a part of note that increasing speed is not the same thing as prioritizing one packet over another though so ISPs were not stifled from providing CDN services they in fact did that under the every version of the rules we've had and yeah that is the question should we be allowing for new business models that prioritize certain packets and essentially slow down others or is it better to have a neutral framework where you can pay for more speed you just can't pay for your speed to be faster than the other guys thank you very much it is 1251 so we're gonna wrap up there but thank you both for a phenomenal discussion on behalf of Berkman Klein and Jolt and we'll see you hopefully in the future you're on the same topic