Reema Moussa  0:00  
From the Internet Law and Policy Foundry, this is the Tech Policy Grind Podcast. Every week, our fellows chat with leaders in the technology and internet law and policy space on recent developments and exciting topics such as privacy, internet governance, cybersecurity, tech legislation, and more. I'm your host, Reema Moussa, and I'm a member of the fourth cohort of Foundry Fellows. The Foundry is a collaborative organization for internet law and policy professionals who are passionate about disruptive innovation.

Joe Catapano  0:46  
All right. Good afternoon, everyone. Thank you for being here on a beautiful mid August afternoon in DC. It's the we don't usually get weather this nice here at this time of year so so it cooperated with us and we're very happy and we're very happy you're all with us, in person as well as our virtual guests. I am very honored to be on this stage with with these three folks here. So on behalf of the Internet Law and Policy Foundry, I would like to welcome you all to our live podcast event, The Internet Past, Present and Future: A Conversation with Internet Pioneers Drs. Vint Cerf and Steve Crocker. My name is Joe Catapano and I am part of the fourth class of Foundry Fellows. In my day job I manage stakeholder engagement for the North America region for the Internet Corporation for Assigned Names and Numbers or ICANN, as it so happens two gentleman to my left are both past chairs of that organization. The Foundry is a collaborative organization for internet law and policy professionals who are passionate about disruptive innovation. The Foundry offers members a platform for professional development, constructive debate, and network building within a cohort of skilled attorneys and policy analysts eager to help shape the development of internet law and policy. So our two distinguished guests here today, in many ways do not need an introduction, but I will attempt to do a brief one. So Vint Cerf is Vice President and Chief Internet Evangelist for Google. He contributes to global policy development, and the continued spread of the internet. And he is the co designer of the TCP IP protocols, and the architecture of the Internet. Steve Crocker is President and CEO of Edgemoor Research Institute, a nonprofit developing policy tools and concepts for collection and access to registration data. In the early days of network development, he organized the network working group, which was the forerunner of the modern Internet Engineering Task Force, and initiated the Request for Comments series, or RFC of notes through which protocol designs are documented and shared. My co host today is Reema Moussa, also part of the fourth class of Foundry Fellows. And I will turn it over to her to introduce herself and go over some of our guidelines for the event.

Reema Moussa  3:29  
Thanks, Joe. And thanks, everyone for being here today. My name is Reema Moussa. I am currently a law student at the University of Southern California, Gould School of Law down in Los Angeles. So I'm really happy to be here in DC with you all today. I'm also a fellow, as Joe mentioned, with the Internet Law and Policy Foundry where I am the producer of the Tech Policy Grind Podcast, and also the West Coast Regional Chair. So we'll start with some sort of moderated questions that Joe and I came up with and then we'll get into the submitted audience Q&A. All right, now I'll pass it back off to Joe to get started.

Joe Catapano  4:17  
Thank you very much. So, you know, for me, part of the really interesting kind of story of, of this whole thing called the internet is that, you know, not only did you both work on the project, but you've also known each other for a very long time. I mean, you've been great friends, since childhood and Van Nuys High School and the whole nine yards. So before we dive into some of the more kind of, you know, policy topics and things like that, maybe giving the audience a sense of, you know, how did you to become friends and how did that friendship kind of evolved into the work itself?

Vint Cerf  5:00  
So I met Steve around 1959 at Van Nuys High School, and the first thing we discovered was a joint interest in mathematics. And then Steve, notice we didn't have a math club. And so of course, that was high on his agenda. So a math club got put together, and we ended up competing in the LA area for various and sundry prizes. We also discovered an interest in computing, which way back in 1960, was sort of early relatively early days. Steve got permission to use the computers at UCLA and I tagged along. There's some stories to go along with that, but I'll let Steve elaborate on those and maybe a little bit about how it was that you came to Van Nuys because you had also had high school in Evanston, Illinois.

Steve Crocker  6:02  
Well, there's a couple of hooks to follow there. Yeah, so in my in my case, I was my high school years were a bit turbulent, my parents had split up. My mother had moved out of Los Angeles back to Chicago, where she was from, and in a very unplanned and unscheduled way I wound up essentially alternating years between Evanston Township High School and Van Nuys High School. At Evanston, I managed to get introduced to computing over at Northwestern University on an IBM 650, one of the first commercially produced computers that you have to look it up in the history books. And then I was abruptly back at Van Nuys where I had met Vint earlier and we continued our friendship. But before I continue with the computer part, let me let me talk about that math club. So when I arrived at the at the high school for 10th grade and discovered, to my horror, that there was no math club, how can there not be a math club in high school? We checked what the rules were, and the answer was, yeah, we could, we could go start a math club. So five of us, all boys, of course, decided to start this math club. And we learned one of the least I learned one of the most important lessons that has endured, because one of the boys decided that we needed a constitution. We spent the entire year arguing about a constitution, we got no mathematics done. And it was perhaps one of those situations where even though that was a disaster, in a sense, it was a super cheap lesson that paid off many, many times in the future. So then skipping forward, the latter part a high school, as Vint and I got access to some computers at UCLA and you'll enjoy this part. We were we were playing a some silly equations, I wanted to map out the these equations. And there was a machine in the engineering department, a Bendix G15, which was about the size of a Coke machine. And I didn't have any keys or anything, but I've been given permission to use it. And in those days, there was very little security that was not like today. And the and, and nobody was using it on the weekend. So on one Saturday, Vint, and I trekked all the way over from Van Nuys over the hill, the mountains, arrived at UCLA and we get to the building and the building is locked. And Vint, observes that the second floor window is open. I think you know, we're not next thing I know he's on my shoulders, climbing through the window. It's one of these crank windows. Then he comes by and opens the door. It's one of the standard push bar doors. And we tape the door so that we can come in and out and get lunch at the cafeteria. And so we spent the day working on the computer. All the rooms were open. There wasn't any issue there. And we cleaned up properly afterwards. This was Spring 1961. The spring 61. Over the next decade, we graduated high school, and went to college, Vint went off to Stanford, I hung around at UCLA. Vint finished in a reasonable amount of time, I didn't. Kept hanging around UCLA, but eventually I wound up in grad school. And then I got a job at ARPA, now called DARPA same agency. 1972 I'm working there, I've got a top secret clearance, Vietnam War is in high gear. I've got super long hair and a beard, which I had assumed might cause a problem, but it didn't turn out to cause a problem. And then there was this little incident at the Watergate. And the Watergate burglars got discovered. And how did they get discovered? They got discovered because they had taped the door to the institution. And the guard came along a plainclothes guard and discovered called the cops and the rest is history. And, and this shiver that went down my spine was something. But, but we survived all of that.

Vint Cerf  10:42  
Small little addition to this story. Many years later, around 1999, Bill Clinton and Hillary had these Millennium evenings. This is honor the past and imagine the future. So Eric Lander and I are the featured speakers. We're supposed to talk about genetics and informatics. I wanted to say DNA meets DNS, but they said that was too geeky. So Hillary introduces us and she tells this story about breaking into the second floor window. And you know, this is like 30 years later. And I still don't know to this day where she got that information, but you talk about little shivers, running up and down your spine.

Well, there was there was a situation where our common thesis advisor Jerry Estrin, was retiring, and he was being honored by the department, and all of his graduate students, or as many of them could, could make it came. And there was a thing Friday night, and then Saturday morning a brunch. And we said, and the engineering building, the site of the crime, was scheduled to be torn down shortly. So we said, you know, maybe we ought to recreate this. And one of our friends, by this time friends are full professors, instead of instead of kids said, I'll bring cameras. So we congregated outside this building. The next morning, same window is still open. Vint is still climbing on my shoulders, not me. And so we have artworks, you know, somewhere that somebody can. 

I have the video.

Joe Catapano  12:28  
Anyway, yeah. And so now right. So now we kind of fast forward now. Since, you know, to present day and kind of the, the just expansion of the network and what we're dealing with now, and now Reema you'll take over with a couple of questions about that.

Reema Moussa  12:49  
So some decades pass, and we're in the early sort of 2000s, and the internet is scaling. Here we are now in 2022. How has the internet sort of evolved from your lens? Since its initial inception? And what are some of the challenges and opportunities with the internet that you sort of foresaw? And how has your lens on that evolved since then?

Vint Cerf  13:26  
So it's a question that comes up, as you might imagine, from time to time, you know, what, what could we say to do? Did we know what we were doing and how far it goes? And the answer is very straightforward. Everything's proceeding exactly on schedule. The more serious answer is that there is a certain surprising amount that we could indeed see. Because the ARPANET, which was the sort of seedling that then led to the internet, and developed by ARPA, was in a environment where all of the participants were at research projects that ARPA was supporting. And in essence, were living in the future. And by living in the future, I mean that in 1968, for example, at Doug Engelbart's laboratory at SRI in Menlo Park, he had invented the mouse. He had interactive graphics and structured text, and was able to give demos and people were using that kind of system on a daily basis. It would be just a few more years before it was commercially feasible, and it would be embodied in Macs that you could buy from Apple and so forth. But, but there was no question what the vision was. And the same was true of time sharing systems, interactive systems. Artificial intelligence was a main stream of the work that was going on, even though it was sort of ignored in the general thing. So we had a fairly decent picture of the future now that a lot of things that we didn't see. If somebody had explained Facebook to me, I would have said, Yeah. And, you know, and all the social media and disinformation with simply: why would you do that? I'll just stop there.

So, several things. First of all, the way you ask that question reminds me of, please explain the universe in 25 words or less, give three examples. So, so I'll do the best I can. First of all, the ARPANET project that Steve was so centrally involved in for host to host protocols, tackled some of the key hard problems. The actual packet switch network was developed by Bolt, Beranek, and Newman, but the host to host protocols had to deal with considerable diversity, different brands of computers, different memory sizes, and different character encodings, different operating systems. And so trying to solve the interoperability among such a diverse set of machines was one of the core problems that Steve and the network working group had to tackle. When the ARPANET project, I'm sorry, when the Internet Project gets started, Bob Kahn comes to Stanford in the spring of 73, and sits down and looks at me and he says, we have a problem. And I look at him and I say, what do you mean we? And he said, Well, you know, the ARPANET worked really well. Electronic mail got invented in 1971, networked electronic mail, and he had already started thinking about how do we use this technology in command and control. So he had already started working on a mobile packet radio system that would eventually be installed in the San Francisco Bay area, and a packet satellite system over the Atlantic to link the East East Coast of the US with the west coast of Europe, emulating the situation where you have mobile vehicles with computers in them in a command and control environment, or ships at sea that need long distance communication ship to ship and ship to shore. So he lays out this multi network problem, and it's the internet problem that needs to be solved. We took everything we learned from the ARPANET project, including all the host host protocols, and essentially translated that into a multi network environment. And what was that, for me? A particular high point of all of this is that I followed Steve to ARPA in 1976. In fact, that's what brought me to this town. So I've been here for 46 years now. And the in 1977, and we had gotten enough work done on the TCP IP protocols to test it. So we did a three network test with a mobile vehicle running around in the San Francisco Bay Area radio connected and passing the traffic all the way to Europe, back again over the packet satellite network and down from from SRI International's USC Information Sciences Institute. So it's only about 400 miles between San Francisco and Los Angeles. But the packets went 100,000 miles because they had gone through two internals, Geosynchronous Satellite hops back and forth. And it worked. And I remember leaping around my office saying "It works! It works!" You know, as if it couldn't possibly work, in any view, realize that when software works, it's a miracle. So this was a big deal for me to be able to show that. So to come back to a core part of your question, the basic architecture is still as it was, in the sense of multiple networks, each with their own boundary running their own internal operations. With a well defined interface between those autonomous systems. The network still functions that way. It still uses the basic protocols that were invented way back almost 50 years ago. It's

Unknown Speaker  19:14  
[Cell phone ringing]

Steve Crocker  19:15  
That's either you or me, it's you. 

Vint Cerf  19:18  
I turned mine off. 

Unknown Speaker  19:21  
[Audience laughter. Cell phone stopes ringing]

Vint Cerf  19:24  
So so it amazes me, when you think about it, that this thing has managed to scale by seven orders of magnitude on most dimensions, the number of hosts, the number of networks, the speeds, we were running 50 kilobits per second in the backbone on the ARPANET, 128 kilobits a second on the satellite net, and between 100 and 400 kilobits a second on the mobile radio network. Today at Google, we run an optical fiber network thats minimum speed is 400 gigabits per second per optical channel, and we're shooting for 800. So it is astonishing that a system of this kind could expand to that degree. So on the purely technical side, between the Internet Engineering Task Force and the World Wide Web Consortium, the invention of the World Wide Web, the system has scaled dramatically well. The thing which hasn't scaled very well is our ability to use it in a safe and secure way. In the original formulation that Steve was describing, we were all friends, we knew each other, we were engineers, and all we cared about was getting it to work. Fast forward to 2022, there are people who like would like it not to work, there are people who would like to spread misinformation and disinformation. And so one of the biggest lessons that I've taken away over these several decades is that this platform, which has been more or less neutral, is an amplifier of everything, including falsehood, and you know, hate speech and all the other bad things that you can think of. So now we have sociological problems that we need to deal with. And those are not going to be solved with engineering. Those are going to be solved by having common agreements, possibly some regulation, and certainly cooperation across international boundaries to hold people accountable for what they do on the network, and to provide agency to people and institutions and organizations to protect themselves online. So despite 50 years of history, we still have a lot of work to do.

Reema Moussa  21:37  
So despite being a worldwide tool, the internet still isn't accessible to everyone, especially those in rural communities, developing countries. So what do you make of the lack of sort of widespread accessibility to the internet? And how do you think we can address that?

Steve Crocker  22:04  
All right. Vint has spent a lot more time on this problem of ubiquity, and digital divide, and so forth. And I have but a couple of comments. It's it's 100%, predictable that when you have a new technology, not everybody's going to get it all at once. And so there's going to be some period out there. Now, the question is, how long should that be? And what do you do for when you're in the middle of that transition in some sense, which is, arguably where we are. Some of that just time will take care of some of it, you have to put energy into, I got a big surprise, I guess it was during the Clinton Administration, I was talking to a guy in the White House. And having sort of a similar discussion, so that was, you know, a few decades ago. And he pointed out that in the US, only 95% of the population had access to a telephone, which just, you know, took my breath away. And he pointed out that there are pockets, some rural pockets, and some urban pockets where just people didn't have access to the telephone. So you know, we're in that ugly period where it gets harder and harder and harder to squeeze out and get to the very end of that process. And how you gauge it and what you do about it along the way, is a continuing challenge. [To Vint] You can fill in much more useful stuff than I can.

Vint Cerf  23:35  
First of all, just for fun, I'm chairman of the Marconi Society and one of the things that we've been doing for the last couple of years is looking at this whole digital divide digital inclusion problem. So if you go to marconisociety.org, you'll see some of the things that we've been doing among them is helping NTIA spend its $42 billion dollars. Now, they haven't sent any of it to us, unfortunately, but we are trying to help with the mapping problems so that you can figure out where do we Internet where we don't have it? Or where do we have it, but it isn't adequate? So right now, the global statistics look like about 60% of the world has access to the internet, the other 40% Doesn't, I'm not so sure about those data, because there's 7 billion mobile phones out there, a significant fraction of them are smartphones that have access to the internet. So I am not so sure about the 60/40 split. In any case, an awful lot of people are going to be encountering the internet for the first time with a mobile phone. And the awkward thing about that is that mobile phones are mostly internet enabled through apps. And so there are millions of those. And so the impression that a lot of people get on the internet is through an app on the mobile phone, which is a very modest slice of things that you can actually do. So I have some concern about that. The good news, I think, is that we're seeing an extraordinary increase in alternative ways of getting access to the system. So two things that have surprised me the first one, of course, is the massive expansion of low Earth orbiting satellites, those sky Starlink effort that SpaceX is putting in plus two others, Kuiper and OneWeb. And so they're well on their way to making it impossible to avoid access to the internet, it'll cover you know, the entire surface of the planet. Now, whether it's affordable, whether the speeds are suitable, and all those things still remain to be seen. The other thing that has surprised me is an incredible amount of investment in optical fiber networks undersea fiber linking continents together at super high speeds. At Google, we've even gotten to the point where we're even building our own cable as opposed to having to join with somebody else to deal with the cost. So that's rapidly evolving to the point where there are islands in the Pacific and in the Atlantic that are getting optical fiber connectivity, I would have lost money on that bet. So we are seeing some serious effort for physical infrastructure. There are also real barriers when it comes to people knowing how to use it, how to build and operate pieces of it, how to cope with some of the hazards that I mentioned earlier. And so those those deficiencies are also barriers to uptake and use of the internet. So there's still plenty of work to be done to make this a useful and safer environment for people to adopt. I do the one story to tell though about mobiles. The original design of the mobile handheld phone was done starting in 1973, a guy named Marty Cooper at Motorola has this idea. He sees car phones, but that takes up you know, all the space in the trunk for the radio gear and everything else. He wants to have a handheld machine. So he starts designing this in 1973, same year that Bob Kahn and I start working on the internet. And we just go along in parallel. We don't know anything about each other. And then a mutual friend of ours, Danny Cohen, in 1983, when the internet gets turned on for the first time, on a production basis, and the handheld mobile service from Motorola gets turned on the same time. So Danny calls me up and he says, "come and have lunch I have something to show you." I said "what is it?" He says "come and have lunch I'll show it to you." So I show up. And he has his thing sitting on the table. It's about you know, foot tall, and it's got a whip antenna. And I look at it and what's that? He says "it's a phone." And I sad "where are the wires?" He says "there aren't any." I said "how's it work?" So I asked him a bunch of questions and eventually he says, "you know, I don't know the answer that why don't you call the guy that built it?" So I call Marty Cooper on this handheld, we call it a Motorola brick, because it's a big honking, heavy thing. So the first question I asked Marty was, "hey, Marty, how long does the battery last on this thing?" And he says "about 20 minutes, but you can't hold the phone up longer than that anyway, so it's okay." So what is interesting is that these two technologies went in parallel in not interacting with each other at all until 2007. That's only 15 years ago. And of course, the smartphone shows up in the form of the iPhone from Apple. And what I want to emphasize here is how much has changed in just 15 years. And so just imagine what can happen in the next 15, or the next 15 that we don't know about that we might not even be prepared for. I just find that the most astonishing thing that in this short decade and a half, we become so dependent on those smartphones, that when they don't work, there are all kinds of cascade potential failures, like I couldn't get logged in. So I couldn't see the message that so I missed the business opportunity and my company collapses and it just goes on and on. So I am worried that we are becoming overly dependent on that one piece of technology. And I would really like to see serious effort to make all the other screens and devices that are internet enabled be backup capability and substitutes in the case that we don't get a mobile signal or the battery is dead or something else has happened.

Steve Crocker  29:27  
Yeah, I want to I want to add a counterpoint to what you're asking about, how do you reach the rest of the population that Vint has taken us through a tour de force of the 15 years of technology and thats changed in that same 15 years has been another change as well, which is we all got 15 years older. 

Vint Cerf  29:47  
[In the background] You told me you wouldn't bring that up.

Steve Crocker  29:50  
And it's not just that we got older, but there's a new crop of people who didn't exist 15 years ago who are now ages zero through 15. One of the spectacular changes is that you approach in the societies that we live in a child and I'm not talking about a school aged child, I'm talking about a two year old. And they go up to a television and they try to swipe it and are very confused that nothing happens. So the the literacy in a sense, has also changed. And so part of what's happening is that the population is changing along with the technology.

Reema Moussa  30:34  
So, in addition to a lack of access, there's a number of challenges with the evolution of the Internet. And you already started mentioning a few misinformation, political polarization, cybersecurity issues, cyber warfare, algorithmic bias, etc. Is there one particular issue that gives you pause or concern, in particular?

Steve Crocker  31:09  
So many to choose from. I think I'm just going to pass on this because in addition to living, and having had some small part of developing the technology, I'm a citizen and an inhabitant just like everybody here, and subject to all the same forces. [To Vint] I'll let you try to say something more coherent than I would likely say.

Vint Cerf  31:38  
Look there are a bunch of challenges, as you say. Some of them have to do with safety and security. Some of them have to do with the rapid propagation of misinformation and disinformation. I have a small theory about some of this. Warnings are probably an important survival factor, our response to warnings. If you fail to notice it's a bear over there, you probably disappeared from the gene pool. The people that stayed were the ones that said "It's a bear, run." So warnings often propagate more quickly, even if they're false, that because they fall into the warning category. So I think part of the experience we're seeing with rapid propagation of misinformation, there's a lot of it is a warning about the secret thing that somebody is trying to do that will be harmful to you. The only solution that I see to that is not technical. There's this wetware up here and there's this process called critical thinking. And it's the thing that I believe we should be teaching our kids and adopting ourselves, which is to ask questions like, "where did this information come from?" And so provenance may turn out to be a really important property that we would like information to have, that we can have access to. The second question, of course, is, is there any corroborating evidence for assertions that are being made? Why is somebody handing me this information, are they trying to convince me of something? So critical thinking, I think is very important. The problem with it is it takes work. And some people may not be willing to do the work, somebody else has already figured it out, I'll do whatever they tell me to do. That's not a good solution, if you want people to be thinking more critically. So between that and the other things that I mentioned before, which is agency and accountability, which I believe we should start deliberately building into our legal and entreaty structures as a respect for agency and a respect for an insistence on accountability. I spent half a day at the State Department on the cybercrime and cybersecurity which are distinct topics. And a good deal of that conversation had to do with how we find ways to cooperate with each other in order to cope with the fact that a harmful act in one jurisdiction can harm a party in another one. And one of the most interesting principles that came out of that discussion was that if something is, is criminal in one country, it's going to be hard to get cooperation unless it's also criminal and the other one, and so this notion of dual criminality might turn out to be a core component. There's an awful lot of work to be done to make this a safer and more secure system. So guess what? The job isn't over yet.

Joe Catapano  34:39  
For Vint, this this plays into this next question, too. So I'm glad you brought this up. And I was thinking about it as as Reema was you know, asking her question about challenges and things and you know, as us you know fellows at Foundry well what brought you to the foundry or what brought you to the company? Well we have an interest in tech policy. And it's like when you guys were developing this thing was tech policy like, even a thing like, would that ever happen? It's just interesting how it works out. But so in recent years, there's been talk about the internet undergoing a, I've heard decoupling with certain countries, like you were talking about for operating in one sphere and others operating in a completely different one. There seems to be kind of a lack of harmonization, in some ways. So can you comment on whether you agree with that sentiment? And if you if you do, what issues do you think are coming up and if there are, not solutions, but pathways to solutions?

Vint Cerf  35:48  
First, the first thing I would observe is that countries have the ability to turn the internet off, and some of them do. In the course of elections, sometimes they just shut off the underlying communications capability. Without that you can't carry the packets. So that happened during the Arab Spring in Egypt, for example, when you see this in other parts of the world as well. In fact, there are some organizations that track this sort of internet shutdowns. There's nothing that will stop a country from doing that. They have control over those assets within their political boundaries. There has been, however, an increasing mantra called data sovereignty. And this is a mirror of the Westphalian notion of sovereignty and in geo political sense. I don't, I'm not a happy camper about that mantra, because I don't believe that it's necessary to draw geopolitical boundaries around where information is in order to protect it. And in order to share it. I think the what makes the internet useful is the connectivity that is created among all the machines that are part of the system, all the data centers, all the clouds. And so if you care about protecting the data, if that's if that's the reason that you're asserting data sovereignty, you should encrypt the data and offer keys only to people who should have access to it, as opposed to building hard barriers. And the Chinese, of course, have done a very good job of that, and the Russians are trying to emulate it with some lack of success, I would say. But that's not helpful. But they can do that. There's nothing we can do to stop that. So I don't worry so much about that, you want to cut yourself off from the internet fine, go ahead. However, the thing I do worry about is people or entities taking action that interferes with other people's ability to use the net. So as an example, there's a routing system called the Border Gateway Protocol, that the autonomous networks of the internet used to tell each other how they're interconnected. It's possible under the current design to issue false information about what the connectivity is, we have ways of fixing that. We have protocols for that. But they haven't been broadly implemented. We need to do things and provide incentives for operators to take up those mechanisms in order to make the system safer and more secure. We need for people to have access to two factor authentication that's robust. And by that I mean not just on your mobile phone, because mobile phones can be defeated in a two factor authentication mode, maybe having little Titan chips or YubiKey chips or other things that establish cryptographic communications is really helpful as a second factor. We need to make that easy for people to use. And one thing I've noticed about security is that there's always some, you know, irreducible inconvenience associated with security. And that is used as an excuse to not follow the rules. And so everything we can do as technologists, to make it really easy, maybe almost impossible to avoid following the rules would be a good thing. At Google, one of the things that we think of is you've heard the term zero trust, we assume all networks are compromised, including our own. And so we build in mechanisms that above the layer of the of the network in order to establish bona fie days, there's still plenty of work to be done in that space in order to create an environment that's much safer than it is today. Steve?

Steve Crocker  39:32  
Thank you. I want to I want to pick up on two points that you touched on. One is we talked about a country could isolate itself to take itself off the internet. There's the complementary aspect, which you which you mentioned, which is pressure applied sometimes to take a different country off the net. And one of the more visible recent experiences was Ukraine sent a formal note to ICANN, which we've both been involved in asking to take Russia off the internet and effectively remove it from the domain name system. I had to smile a little bit, because about a decade and a half earlier, I had been in Russia and had very direct interaction with a official there who expressed concern, what would happen if the US took us off of the internet took us out of the domain name system was the particular thing. And I went to great pains to explain what all the controls were and that it was not only against policy, but there were a lot of checks and balance checks, balanced type inhibitions against that, that would, would stop that. And I was also thinking in the back of my mind, I didn't want to say it, that it would be a very, very bad move because it would instantaneously put the US in a very negative position, it would be viewed negatively by everybody in the world. And the reaction worldwide would be a huge backlash. And worse yet, it would be relatively quick for system administrators and programmers and hackers all over the world to develop workarounds, boy scouts would come out of the woodwork basically and develop all that. And so what I wanted to say to this guy, and this guy was, was a very nice guy. And he was he was quite smart. He was the equivalent of deputy head of the Federal Communications Commission counterpart. That's not what they call it there. And he was leading a delegation Down to Rio de Janeiro for Internet Governance Forum. And the, what I wanted to tell him was, if he could cause the US to do something so stupid, that the result would be so beneficial to the to Russia, that he would instantly be made a hero of the Russian Federation, and he would do everything possible to cause that to happen. But the humorous side, one of the lessons that has been learned pretty well, not only for the internet, but think about the phone system, you can place a phone call to Cuba, you can place a phone call to Russia, there's been no disruption, no matter what, throughout the Cold War throughout all of these. So the the accumulated wisdom is that there are things that you can use when you're in a confrontational situation, and other things that you best leave alone, because the results would be far worse, and we're all better off, leaving things in place. So that's related to this issue of cutting people off. And then you mentioned Westphalian issues. So that's 1642, I guess 48, excuse me, decision that countries, countries had a certain degree of sovereignty and they shouldn't interfere with each other. The internet is not the only thing that brings that model into a challenge position that provides stress, you have other commonalities, you know, we all share the same air, we share the same oceans, and so forth. So the the models that have served very well, from a certain point of view, historically, over the last several 100 years, are showing some signs of stress. And it's not clear how to evolve. But it is clear that there's going to have to be some degree of evolution. And the choices are either going to be incremental and relatively smooth, or not incremental and not so smooth. We'll just have to see how it plays out. And the Internet is a piece of but not the sole participant in that evolution that is making the world smaller and flatter, if you will. And then of course, we have a few issues with just respect to the to our planet versus our people, different topics.

Joe Catapano  44:33  
So I have one more and then I think we got some from the other fellows. So playing off that another thing that I've heard discussed is the idea that we have moved from a values driven internet to an interest driven one. As two gentlemen who were there at the beginning, do you agree with that sentiment? And do you think those things are mutually exclusive? 

Vint Cerf  45:01  
Can you ask the question one more time?

Joe Catapano  45:02  
Yes, of course. So another thing that I've heard discussed is the idea that we've moved from a values driven internet to an interest driven one. Do you agree with that sentiment? And and or do you think these things are mutually exclusive?

Vint Cerf  45:21  
Well, I don't you know, to be really honest with you, I think interests have always driven an awful lot of everything people do. Of course, there are a wide range of different kinds of interests. Some of those are driven by values. So I think it's, it may be a false dichotomy to answer the question quite that way. If you want an interpretation of the values side of things, I think that there was a certain idealism associated with the early days of the Internet, getting computers to interconnect with each other and making things happen 3000 miles away, it was nothing short of miraculous and fun. And, and we were all you know, looking at this is, is amazing. Can we actually get this to work? You know, and now that it's working, what can we do with it? I think what happens is that as it becomes commercialized, which begins in the 1980s, we see it Cisco Systems in 1984, Juniper comes later, Protean. Start to monetize these the equipment and software associated with internet and then the world wide web shows up. First, in its very open variety in 1991, 93 the Mosaic browser comes out of NCSA. It gets Jim Clark's attention, who did Silicon Graphics, and he sees the world wide web and he says, there's a business there, grabs the guys from NCSA and drags him out to the West Coast and starts Netscape Communications in 1994. In which year, I had gone back to MCI to start building the internet's for them, and they wanted to build an MCI Mall. So I bought $7 million worth of licenses from Netscape Communications to build this mall, I should have bought $7 million worth of stock, but I didn't. That was the company's money, it wasn't mine. So the thing is that that this stuff starts to get seriously commercialized. Internet service shows up in commercial form in 1989. And so, you know, we're talking 30 plus years ago. So a lot of today's investment is driven by commercial interests. And and to the to the extent that that was the distinction you were trying to make. I agree with that. I think that the heavy evolution of the system now is being driven by business opportunity more than anything else. You still see a lot of open source software, though, which is there, the spirit of sharing is not dead.

Steve Crocker  47:53  
Yeah, that's that, I think, is a crucial point, that it's easy to take phrases like interests versus values, set up a dichotomy, and then, you know, try to have that argument. These things basically coexist in in ever changing mixtures. And if one insists on seeing it only one way, then you're, you're off by quite a bit in terms of having a useful model. So it's more a question of degree. And the degrees are shifting. Commercial interests obviously play a huge role, but so does idealism, and just the sheer fun of being able to explore and create. And so, you know, I think one has to look at these things in a more delicate and nuanced way than just a simple dichotomy of within and particularly trying to choose one side or the other.

Reema Moussa  49:03  
So to finish with our moderated questions, and then we'll get into one or two audience questions before our time is up, what opportunities on the horizon keep you excited about the future of the internet? And what sort of advice in that vein do you have for future generations interacting with with this technology?

Steve Crocker  49:35  
One of the things that is very different, and just speak very personally from my point of view, you know, back from the 1960s and 70s, versus here, so you know, 50 year span more or less, it's that the world that we could, that I could touch was principally within the US and a little little bit of foreign travel. Nowadays, you live in a completely connected to the rest of the world without any trouble at all, you know, subject to the digital divide issue, but just in terms of geography, and with the geography comes cultures and language and tremendous amount of access to history and to other effects. And so I watch my children and I watch their children. And they, you know, the the diversity in every dimension, not only just diversity of races, and religions and so forth, but every sort of thing that they're exposed to, and that they take as a given, and that they enjoy and that they make use of, is really quite phenomenal. That's quite enlivening. That's, that's a very big positive thing.

Vint Cerf  51:03  
So I started taking some notes, the first observation I'd make is that essentially everything that you see in the Internet is a result of software. I mean, that's what animates the whole thing. So if you're looking at opportunities, software is still a huge opportunity. If you can figure out how to program something And we're certainly starting to see tools that allow people to effectively program without programming. I mean these sorts of, you know, audio, speech interactions that cause things to happen, there's lots of software being executed behind Google's answers to your voiced questions. And there are even some exploration of the ability to describe what it is you want to have happen and have the system try to prepare some software to to achieve that objective. Open source is another area where the sharing allows people to learn how things work, you know, in the early days of routers, the router was was built by getting a computer and a graduate student, and you wrapped the graduate student around the computer. And eventually, we ran out of graduate students to do that with so Cisco and Protean and others started selling those things. But the early days, you got to know what was going on inside because it was still a research project, the software was available, then it became commercial and all the software was hidden. Now more of it is available. So people have the chance to learn more about how to program all kinds of things that are have computers in them. So I think that's an important element. One thing I will observe, our experience with the pandemic, has taught us a number of interesting lessons. One of them is that not everybody can work at home for a variety of different reasons, by need the job that requires proximity, or you don't have a place that's, you know, quiet, or you fought with the kids over the Wi Fi because there wasn't enough capacity. But we have learned that flexibility is is possible. Companies have discovered that people actually can work and will work at home. So that's an important lesson. The other thing, very important lesson is that the supply chains are fragile. And we discovered very quickly how broken they could be. We also discovered that the internet as good as it is about erasing distance doesn't do anything about time zones. And so when you're trying to do things online with a whole bunch of people, some of them are up at 6am and others at 3am and somebody else at 10pm. And we haven't figured out how to deal with that problem yet. We're working on it. So my general sense is that there is infinite opportunity lying ahead. Just to add one more thing on the AI and machine learning which gets overhyped. It's also fair to observe that machine learning and neural networks have done some pretty damn amazing things DeepMind just announced that it had modelled 200 million, the folding of 200 million different proteins, which are essentially all the ones that we even know about. So that I guess is going to turn it into opportunity to understand you know, medical opportunities or understand all kinds of other complex biochemical opportunities, interactions and the like. So there are lots of things where machine learning has actually done something good speech, understanding speech recognition, speech generation. Interactive systems, are all feasible because of those technologies. But at the same time, we need to remember that they are potentially very brittle. Depending on what their training is and on what problem or situation the machine learning network has been exposed to. It could make very are bad decisions. And we aren't too good yet at noting ahead of time, what kinds of mistakes those systems are going to make. I keep wondering about self driving cars, we have a big investment in that in the Waymo part of the Alphabet Company, and they keep trying to figure out how does the how do I know that the car knows that I'm there. If I can catch somebody's eye if there's a person behind the wheel, and that sort of gives me some sense that they at least they know I'm there. But I don't know how to do that with a car that doesn't have a driver. And so that's an example of one of the conundrums that this sort of thing presents.

Reema Moussa  55:39  
All right, so we have just a couple of minutes left. And so I think we can get into maybe one or two of the questions that was submitted. So one question we had is that many of the Foundry Fellows are looking at the emergence of decentralized protocols, Blockchain, smart contracts, etc. The building blocks of the decentralized web as an exciting place for innovation, as one of the as some of the creators of the original version of the internet, do you see any promising developments in the decentralized web?

Vint Cerf  56:23  
Sure. Well, look, first of all, the Internet was designed to be decentralized. And it still is, I mean, it's just the fact that you see concentrations of implementation of the network doesn't mean that it's dissent that it is centralized. It just means that some of its concentrated. So that's the first point. The second point is that we've actually discovered that some centralization is very useful. An example of this is, is the open flow networking software defined networking system, which has a central component to it. It does computations very efficiently and uses the resources better than some of the decentralized algorithms do. So centralization is not necessarily evil or anything like that. When people talk about decentralization, and they point to blockchain, and they talk about, you know, the anonymous blockchains. And how wonderful that is, my first reaction is well, first of all, anonymous blockchain means that I don't know who I'm depending on for the operation of the system, then that, frankly, makes me nervous. So if I'm going to do blockchain at all, I'd like to know who's involved. So that's called permission blockchains. Second, not all blockchains have to have crazy computational requirements, like Bitcoin does, there are others that don't have that. So that's helpful. I'm not sure that this mantra of pushing for decentralization, is necessarily going to solve all problems. Think for a minute, suppose you say, well, here's what we're gonna do. Everybody's got all these laptops, and they all have huge amounts of storage. So why don't we agreed you to store each other's information? We'll just fully distributed everything. Let's even pretend that it's technically possible to do that the data is encrypted so you know, you don't your your privacy isn't at risk, then the question will be Joe, who's got some of your information on his laptop, decides to replace the laptop? And the question is, how important is it to Joe, to make sure that all your data got replicated on the next laptop? And the answer might be not very, which is why sometimes you want to turn to organizations who see it as their job to make sure that all of your data is is replicated and retained and not lost and backed up and everything else. So I'm not persuaded these days that central is decentralization isn't necessarily something to be sought. For it's its own value. I think it's useful for resilience, but maybe not so much for some other reasons.

Steve Crocker  59:10  
So the first the first thought about blockchain is a very clever technology, but it's been hyped way beyond appropriate level, in my view. Almost to the point of a joke, Blockchain is the answer. What's your problem? Kind of thing. You know, most of what I've said today, has fallen into the broad category of avoiding the extremes and looking at things from a balanced point of view. And, equally, you know, looking to the future, the problems and the opportunities are all going to fall into the, into the middle ground of how do you look more deeply into the way things are working? Where are the opportunities to improve things? And it is not going to be a simple thing of replace everything with this or, you know, this answer will, you know, this piece of technology will solve all your problems, you have to really look and see which things matter in terms of the problem that you're trying to solve. So the kind of a general, mild rant, if you will. And then you can pick whatever problem you like, and I can, I can expand down that path, almost, you know, without exception, you know, pick, pick your problem, you don't even have to name them. And they will all have this quality, that the extremes are probably not the right answer, and that the hard problems, the challenges, and the opportunities are, are trying to dissect where the what the interplay of the various forces are, and then new and that is worthwhile work for us. And for next generation and generations after that.

Vint Cerf  1:01:17  
A couple of things never resist an open microphone. A couple of things. First of all, you mentioned smart contracts. And setting aside buzzwords for just a second, it's a very interesting idea. Because imagine that you got a contract with someone for some service. And it's possible for the software that that represents the agreement, the commitments that are in that smart contract, it's possible for that software to monitor the state of the service that's being delivered. And so if it can tell how things are going, whether it was a contract to develop a piece of software, or to provide a regular service of some kind, if it can tell how things are going, then it can automate, you can automate the process of alarming in the event that some term or condition of the contract has not been satisfied. And so the idea of having some kind of a piece of software in the background that's monitoring things for you, at the right level of granularity is actually kind of a really interesting thought. And so I don't want to lose some of these ideas in the course of the overhype that shows up. When when blockchain first showed up, my first reaction was blockchain is the new Brylcreem. And not everyone will recognize that reference, because it's a 20th century reference to some stuff that you were supposed to put in your hair, that you've improved your romance life. Of course, I don't have any hair left. So Brylcreem isn't helping me any. But blockchain is not that. It's also not a stupid technology, though, there there. We use it at Google for a number of applications. But it has to be sized to the application to be appropriate at scale, at the right scale. So I get excited about what you can do with software. But here's the problem. In the 80 plus years that humans have been programming computers, we've never figured out how to avoid making mistakes. And those mistakes we often call bugs. And if you talk to programmers, you often will find that they all have a little dent here in their foreheads from the millions of times they've gone "God how can I make such a stupid mistake?" So one of the things that I would love to see the next generation challenge to do is to build better programming environments that will tell us when we're about to do something dumb, like reference a variable whenever we got set to make a decision, or, you know, going through a loop too many times or not enough times. Those are or reading in a piece of data into a buffer that's too small. So those buffer overflows and you execute software that you shouldn't be executing. We really need better programming tools. And so that's the challenge to the research community.

Reema Moussa  1:04:09  
All right. So with that, I think we'll close out our Q&A session. A huge thank you to Drs. Vint Cerf and Steve Crocker. Love a round of applause.

Unknown Speaker  1:04:23  
[Applause]

Reema Moussa  1:04:29  
And anyone on our livestream or listening to this podcast later, thanks for tuning in to the Tech Policy Grind. With that I'll hand it over to Joe to close us out.

Joe Catapano  1:04:43  
Thank you very much Reema. Sincere thanks to both of you Dr. Cerf, Dr. Crocker, for being with us. This has really been a pleasure. Just a few more thank yous. Thank you to Reema, for putting making all this happen. Thank you to Tim Lordan and the Internet Education Foundation because this this whole thing wouldn't have happened without them either. As well as all my colleagues, the fourth class Foundry Fellows for their support. I just want to thank you again for joining and and we hope to see you again soon.

Reema Moussa  1:05:18  
Thanks for listening to this episode of The Tech Policy Grind Podcast. Be sure to check out the foundry on LinkedIn and Twitter. And if you enjoyed this episode, leave us a review and give us a five star rating. It really helps out the show. If you're interested in supporting the show, reach out to us at foundrypodcasts@ILPfoundry.us. You can find our email in the show notes as well. You can see the full show notes and download the episode transcript for every episode on our website ILP foundry.us/podcast. The Tech Policy Grind podcast comes out every Thursday. See you next time. The Tech Policy Grind Podcast was created by the fellows at the Internet Law and Policy Foundry. It's produced and edited by me Reema Moussa, with support from the incredible Foundry Fellows.

Transcribed by https://otter.ai