The Architecture of Crypto Innovation

Some of the biggest questions in technology today opportunities for crypto center around issues of privacy and security. When you consider the very nature of crypto is decentralized networks and databases, how does information sharing that work when you don’t have a trusted third party in the middle who can verify the information being communicated? So in this talk, MIT professor (and former senior software engineer at Google) Neha Narula, who researches distributed systems and did her doctoral work on fast, scalable databases -- as well as Director of the MIT Media Lab Digital Currency Initiative -- shares what "consensus" is in the crypto and blockchain context. What is the spectrum of "decentralization" involved? What are remaining challenges in the space, such as security and scalability? And what are the tradeoffs involved, and how are different players addressing them? This talk was originally delivered at the inaugural a16z Crypto Regulatory Summit 2019, which brings together leading crypto experts and builders, other technologists, academics, industry executives, and government officials -- along with forward-thinking regulators -- to foster collaboration and the exchange of ideas around this important emerging industry.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  Hello and welcome to the A16Z Podcast. This episode is based on a fireside chat between Katie Hahn, crypto general partner, and Marc Andreessen, co-founder of the firm, during our recent inaugural crypto regulatory summit. Katie asked Mark to trace some of the possible parallels between the early days of the internet and the current state of crypto, including how the browser went mainstream and its initial usability gaps, the impact of regulators and policymakers in the early days, the beginnings of commerce, and the story of the world's first online bookstore, hint, it's not named Amazon, and the original sin, as Mark calls it, of the internet, and how crypto could revisit that original sin and help create new economic models to drive the internet and the broader economy forward.

• 0:44Speaker 01

  Mark, you had a front row seat when the internet was kind of being developed, right? At least for consumers. How did the internet go from being this kind of weird thing that academics used, and by the way, the government, to a mainstream growth engine for the economy?

• 1:01Marc Andreessen?

  Yeah, so to start with, I should say, to clarify, it wasn't actually a front row seat. It was more, it was the little kid's seat. It was the booster seat, where you still use the regular shoulder strap in the back of the internet van at the time. I was very young. So, the evolution of the internet over time has so many echoes and comparisons to what we're seeing, what I'm seeing now with cryptocurrency and blockchain. It's really quite remarkable. And so, you know, maybe the best way to think about that is kind of by actually telling the story about how the internet actually first became useful as a commercial medium.

• 1:29Speaker 01

  Yeah, that'd be great. And by the way, for those who don't know, I'm sure many know, but Mark, you co-founded Netscape and brought the first kind of internet browser to mainstream.

• 1:40Marc Andreessen?

  And then even before that, a team of us at the University of Illinois created Mosaic, which was kind of the prototype for what became Netscape. And so that was in 1993. And it actually turned out 1993 was the pivotal year. And so, so the internet has this very interesting backstory. So first of all, you know, the origins of the internet kind of go back even decades before the 90s, you know, they go back to the 50s or 60s. And so there was academic research happening over multiple decades of the form of basically how to develop a decentralized network. And right, actually decentralization was actually key to the internet early on because the original motivation for it was to have a network that could actually survive a nuclear attack. So it was sort of, it was born out of the present dangers of the Cold War. And as this revolutionary concept, you have this decentralized network with no central hub. At the time, there were data networks, but all data networks ran through centralized switches run by companies like AT&T.

• 2:22Marc Andreessen?

  And true to form, the senior executives and many of the top scientists at AT&T always maintained the internet was impossible, right? It could never exist, it could never happen, it could never work, it could never scale, it could never be secure.

• 2:31Speaker 01

  So there's one parallel right there.

• 2:33Marc Andreessen?

  Well, there's a parallel right now we could talk more about, but it did actually start to emerge, and then it started to go mainstream in academic circles, in particular in the 1980s. And it was actually at the time funded by the National Science Foundation, so it was called the NSF net, was like the precursor to the modern internet. The thing that people forget today, because everybody now today, you shop online and you do all the stuff online, buy airplane tickets, and it all seems so natural. It was actually illegal to use the internet for commercial activity or commercial transactions until 1993.

• 2:58Speaker 01

  And I want to stop you there, because I remember when you told me that, I thought, what? I've never heard that. It was illegal to use the internet for commercial transactions until 1993?

• 3:07Marc Andreessen?

  Yeah, so it's a different circumstance. So the illegality or sort of prohibition, let's say, I don't know if they could have sent you to jail, but they definitely could kick you off the network, which is kind of the same thing, at least in modern life. So it was funded by the government, it was funded by the taxpayer, and so the government got to set the policy, and the policy was, it was called the Acceptable Use Policy, or AUP, that's on Wikipedia. You can pull this up, and it's basically the Acceptable Use Policy is you can use the internet for academic, nonprofit, individual use. If you can figure out a way to connect to it, you're free to kind of play and explore and kind of do experimental things, but you specifically can't engage in commercial activity and commercial transactions. And so as an example, companies were actually connecting to the internet prior to 1993, like a lot of defense contractors were on early, and a lot of tech companies were on early, but they literally technically could not sell something to somebody.

• 3:51Marc Andreessen?

  And it was literally, it's taxpayer money. And so then there was this kind of turning point moment in around 92, 93, 94, that I was kind of sitting in the middle of at the time, which is sort of internet traffic was starting to really kind of take off, people kind of decided they liked this thing, and then there was sort of growing latent demand for actual commercial activity, but the government was kind of holding firm. And then basically what happened was a set of, I would say enlightened regulators and policymakers, you know, decided basically two things at the same time. Number one is they decided to legalize, basically revise the AUP to legalize commercial activity, but in parallel, they decided to hand off the internet backbone to the telecom companies, to, at the time, AT&T, Sprint, and MCI. And then that solved the economic problem, right? Which is we'll get companies to pay for this, so it doesn't have to be the government. But it was a very enlightened and forward-thinking move to make this change, because it wasn't

• 4:36Marc Andreessen?

  consistent with internet culture at the time, it wasn't consistent with how this thing had gotten funded, it wasn't consistent with any of the governance mechanisms. But there were a set of people in Washington who said, you know, this would be a good idea, and furthermore, it'd be a good idea for the internet to exist, it'd be a good idea for it to engage in commercial activity, and in fact, you know, in crazy upside scenarios, it might be a driver of economic growth. And for those of you who are, you know, older like me, you'll remember like 1989 to 1993 in the US was a severe recession, right? And it was like grim economic times. It's kind of gotten forgotten a little bit, because it had happened a while ago, but like it was, it was pretty dark and pretty bad. And it turned out, I think you can trace a direct line to the economic boom of 1994 to 2000, from basically legalizing commercial use of the internet.

• 5:16Speaker 01

  I just heard you call the regulators enlightened, enlightened regulators.

• 5:19Marc Andreessen?

  Yeah. My friend Al Gore, you know, gets a tremendous amount of teasing for having claimed to invent the internet. I will defend Al Gore's honor. Al Gore never said he invented the internet, he said he helped to create the internet, and that actually is true. He didn't write the code, but what he did do was he set many of the original policies for the funding of the NSF net, which led to the creation of the internet, and then he was involved in this process of basically privatization and having it become an engine for economic growth. And so he was integral, and a whole bunch of other people in Washington as well, on both sides of the aisle, but he was in the middle of that whole thing.

• 5:47Speaker 01

  And that AUP changed then in 1993?

• 5:49Marc Andreessen?

  Yep. So 1993, the policy changed, and so now it was legal to do commercial activity online. And so people started to have all these, you know, really crazy, wacky ideas, like, you know, what if you could buy things online? And there was immediate backlash, right? Response basically was, okay, it's now legal, but it will still never work, because it's not secure. Right? And so that won't possibly work. And then, and then the other thing, of course, that was immediately obvious, was it'll never work, because big companies aren't hooked up to it yet, and so, right, it's all going to be these fly-by-night operations.

• 6:14Speaker 01

  And so that's another parallel that you're seeing in the crypto space?

• 6:16Marc Andreessen?

  Yeah, so one of the things you talk about with crypto today is it's like, okay, it's like sort of where are the use cases? Yeah, I understand that there are these potential use cases, and yeah, there are specific examples of these use cases today, but like where's the amazon.com of crypto, right? Or where's the Google of crypto? And it's like, okay, well, let's think about actually where amazon.com came from. This is another kind of great analogy. So the first internet store was a bookstore. It was not Amazon. It was a cult, little, tiny, science fiction and fantasy bookstore on El Camino Real in Palo Alto called Future Fantasy Books. And it was kind of this legend, and if you like, if you liked science fiction at the time, it was like the place in the Bay Area you'd go to buy, to buy science fiction books. It was run by a guy who'd been running the store for, you know, whatever, 30 years in the same location. And it was just like a store. And so actually, another friend of mine, a guy named Brian Reed, I believe is his name,

• 7:00Marc Andreessen?

  at a research lab in Palo Alto, unconnected to this bookstore, it was a company called DEC at the time, and he was in the research department. He said, "Boy, you know, I really like this bookstore. It might be fun to be able to buy science fiction books over the internet." Right? This is, this is a new idea at the time. So he went to the owner of the store, this is in 93, he went to the owner of the store right as this policy change happened, and he said, "You know, let's put your inventory, you've got all these books, let's put your inventory online so that anybody in the world can buy books from you on the internet." And, you know, the guy's first question was like, "What's online?" Right? Because like, that's just like a weird and unusual concept. And so then Brian explained, you know, my friend explained like, "Okay, here's what this means." And then the guy's like, "Okay, that sounds great. I would love to get orders from all over the world. I don't own a computer." And so Brian said, "Not a problem. I work for a big computer company." He's like, "You know, if you don't have a computer, what do you have?" And the guy's like, "Well, I have a fax machine."

• 7:44Marc Andreessen?

  So my friend basically set up, in the research lab in Palo Alto, he set up the online web store, the site for the bookstore, and then you would order a book on the website. And then my friend set up a fax gateway, so the order would get sent to the store in a fax. And then the guy in the store would package up the book and sell it. In the first two weeks of that store going on the internet, that guy's business doubled.

• 8:05Speaker 01

  Wow.

• 8:05Marc Andreessen?

  And, you know, it's a small, absolute numbers, but a huge relative boost. And guess what? It turned out, it turned out there were like, science fiction readers all over the world, in particular, Germany and Japan, that couldn't get a lot of the English books in their domestic markets, and so they immediately started to do mail order.

• 8:17Speaker 01

  I hope he then got a computer.

• 8:19Marc Andreessen?

  He did buy a computer, unfortunately, then Amazon came along and he went under.

• 8:21Speaker 01

  Wow.

• 8:25Marc Andreessen?

  It's not a totally happy story, but however, the whole phenomenon worked and that actually, you know, that and sort of people who are experimenting like that kind of paved the way for what became e-commerce. But I hope that gives you a sense like of how, like, it was not obvious. It was not obvious that these things, it was going to make sense to have these online businesses. It was not obvious that the use cases were going to have consumer demand. It was not obvious that you were going to be able to secure it. It was not obvious that there'd ever be a business model behind it. Like, all that actually had to get invented.

• 8:49Speaker 01

  Well, you just mentioned "secure it," and I think that's an important point. So the government, with the help of the enlightened regulators, turns this thing over kind of to the private sector, right? This internet thing. And then you have SSL. And then, for those who don't know in the audience what SSL is, it's basically very strong encryption. Did the government then kind of discourage SSL?

• 9:10Marc Andreessen?

  Yes.

• 9:10Speaker 01

  So what about that?

• 9:11Marc Andreessen?

  Yeah, so then they turned around and tried to ban that.

• 9:13Speaker 01

  You did have a front row seat there.

• 9:15Marc Andreessen?

  You did have a front row seat for this, yes. So we drove encryption into kind of public use, really, for the first time in volume at Netscape. So basically what happened, so now the internet's available for commercial use, consumers are starting to log on, the whole thing's starting to work. But then the whole problem was, okay, all the connections were unsecured. So it's just plain text data, right, traveling back and forth over the network, including passwords and credit card numbers and everything else. And so, right, the immediate logical conclusion is, this is an unsafe place to do business. No consumer, you know, the newspapers at the time were just filled with stories. Consumer guided stories basically saying, whatever you do, never, ever, ever put your credit card number on the internet, right? Because it will instantly get stolen by hackers and your entire net worth will be wiped out. Now, that wasn't technically true, but that was the level of fear and anxiety. And, you know, true enough, like, it was unencrypted, so it was a bad thing. And so we did what we thought was the logical thing, which was

• 9:57Marc Andreessen?

  we built SSL, basically, as the secure protocol to be able to have, you know, a client and a server connect to each other, a web browser and a web server connect, and be able to exchange information securely without hackers being able to penetrate.

• 10:07Speaker 01

  What did the government say about that?

• 10:08Marc Andreessen?

  Well, we thought this was a really good idea, and so we brought it to market, you know, we were about ready to launch it. And then we discovered a regulatory regime that we were previously unaware of called ITAR, which some of you may know about. It's like international trade in arms and something, something, something, bad stuff. And so it's the regulatory structure, I don't know if it's still current, but at least the time it regulated Tomahawk missiles, right, and like Napalm and all kinds of fun things.

• 10:30Speaker 01

  Export controls.

• 10:31Marc Andreessen?

  Export controls. And so it basically governed what US weapons manufacturers effectively could export overseas. By the way, which, fair enough, like, I'm in favor of regulating Tomahawk missiles. It turned out encryption was on that list. Strong encryption was on that list. And so the basic deal was, at the time in the US, you were allowed to, we could build strong encryption into the browser. Strong encryption being encryption that was either hard or impossible, depending on your point of view, to break. There's a big debate about hard versus impossible, but fundamentally difficult to break. And therefore, safe for consumer use, let's say. So we were allowed to do that in the US, but we were not allowed to export it. And so that led to this very lovely compromise we came up with, which is there were two versions of Netscape Navigator. And it was the same product, but there were two different boxes. The sticker on one box said, "Strong encryption." The sticker on the other box said, "Weak encryption." Which is true. So the expert-

• 11:15Speaker 01

  What's sold better?

• 11:16Marc Andreessen?

  What's that?

• 11:17Speaker 01

  What's sold better?

• 11:18Marc Andreessen?

  Yeah, well, it's funny. Funny that the market demand would line up maybe the way that you think it would. And so, yeah, right, here's the soft drink that tastes terrible, right? Now, let's see if people want to drink that. So, we were allowed to sell the browser overseas, but it was deliberately weak. And now, we were, and by the way, by weak, I mean like it was trivial, it was, it was weak. We had to degrade the encryption to the point where basically anybody could crack it. Anybody with a computer and any level of software skills would be able to crack the encryption. So basically, useless. The good news is, all the other American companies were under the same restrictions as we were. So it's not like any of our American competitors kind of outfoxed us by launching globally. The problem was, I've heard rumors, America's not the only country in the world. And so, it turns out there are a bunch of other countries, and it turns out they have, you know, sharp software people also. And so then, we immediately started to see this kind of cascade of international competitors. It's the most obvious way to compete with American software companies.

• 12:03Marc Andreessen?

  We'll do the thing that they literally are not legally allowed to do. And so then, it became a rush of foreign competition.

• 12:08Speaker 01

  Well, speaking of parallels between crypto and the internet, one of the things that I hear a lot, I know you hear this too, is, well, crypto's so hard to use. Like mainstream consumers can't possibly figure this out. It's all janky. But I know you've talked to me about it, and I'd love to hear your take on, you know, just how easy was it to get on the internet in the early days?

• 12:28Marc Andreessen?

  Yeah, super easy. You just pressed a button. No, that's not true at all. So, I'm a consumer, I'm just a normal consumer, and let's say I'm totally comfortable with computers. And so I like, I know how to use computers, I own a computer, I'm perfectly happy with computers. And it's 1992, 1993, and I decide, I read about this internet thing, I want to go on the internet. So I go on my computer, and I look for the internet button, there's no internet button. None of the computers in 1992, 1993, the consumers had, had any internet awareness at all. They didn't have the software for access to the internet, they didn't have, they didn't, they didn't have any of that stuff. And so, so basically what you would do is, this is literally true, what you would do is, if you knew to do this, you would go to a bookstore, which is, they used to have these, you know, physical places where you'd buy books. And you would buy a book, and the book would be "How to Use the Internet." And this, there was, there was a point when it seemed like all the bookstores were just

• 13:12Marc Andreessen?

  going to be taken over by literally just internet books, because this was, this was how you got online. So you'd bring the book home, and you'd read the book, and it'd tell you all these great things about the internet, and then there would be a floppy disk, if you remember those, in the back of the book. And the floppy disk had the software that you had to load on your PC in order to get on the internet. If you're technical, if you're technical in the room, it's the TCP/IP stack, and like all the other kind of networking codes on that disk. And so, you then had to go, and it go through a 38-step process, it took you through this in the book, of sticking the floppy, and you basically had to rebuild, essentially rebuild your operating system on the fly, as an end user, to add the internet support. It was a very complicated process. By the way, you need to be very careful doing this, because if you made a mistake, you would probably break your computer forever. It would probably totally stop working, because you were mucking around in the operating system. And so you wanted to be very careful, kind of, through that part. So then you'd get all the way to the end of that process, be like, "Great, I'm ready to go on the internet."

• 13:57Marc Andreessen?

  And then it's sort of step 38, you know, the end, "Okay, congratulations, now you're ready to go buy a modem." Right? And so, then you're like, "Okay, so what's a modem?" All right? And so then it's like, "Okay, now I have to go find a computer store, and I have to pay whatever, 300 bucks to buy a modem, which is, used to be the thing that connected the computer to the network." And then you bring it out home, you had to plug the modem into the computer, and then you had to configure the computer and the modem to actually work with each other. And that was like another 38-step process that you had to do very carefully. And it was another process you didn't want to screw up, because you were mucking around in your operating system. So then you'd get all the way through that, and say, "Okay, I got a modem, ready to go on the internet." And then it would say, "Now you need to sign up for an ISP." And then that began the process of what's an ISP? Internet Service Provider, you know, that's okay. That's who your modem is going to call to go on the Internet, right? And so then that began the process of like, okay, what's an ISP? Who's an ISP?

• 14:41Marc Andreessen?

  And these were in the days, you know, this is the days where most people who were using networks at home were using AOL, right? Which was sort of a proprietary version of all this, but it was not a full ISP at the time. So you'd have to go track down, literally, a lot of these were like these little local ISPs, these little local neighborhood companies, almost like the old neighborhood cable company from, remember those days? So then you'd have to like call the ISP. Remember, you're not on the Internet, so you can't like search for the ISP on the Internet. So you gotta go in the Yellow Pages. You find your local ISP, you call their phone number and you sign up and you send them your, you know, you tell them your credit card number and they give you the code. And so basically, you know, this is like four weeks later, right?

• 15:13Speaker 01

  Yeah.

• 15:13Marc Andreessen?

  That you're going through all this. And then you cross hit dial, then you get the busy signal. That's how straightforward it was. And so, so, and there was a good.

• 15:20Speaker 01

  Makes crypto seem easy.

• 15:21Marc Andreessen?

  It does, it does. Crypto's easier, crypto is easier. Now, I would say the good news is, in the early days of Internet, you could argue like it applied a high bar to the people who got on the Internet. Like, you had to really want to use this thing. And so, you know, the people who were on the Internet early on were like incredibly enthusiastic and like just absolutely thrilled to be there because, you know, it was like joining the Marine Corps, like it was effort.

• 15:40Speaker 01

  [laughing]

• 15:41Marc Andreessen?

  And so, you know, that was the good news. The bad news is, it was obvious, I mean, this was just, this was absurd, like this was just absolutely patently absurd. And so then a whole series, you know, there were then several years of work by us and others to kind of streamline this thing to the point where you all have today and your views just come.

• 15:53Speaker 01

  Well, thank you, because I, thank you for letting me just develop products so I can now push a button.

• 15:57Marc Andreessen?

  Yes, exactly.

• 15:58Speaker 01

  But you did talk about with the Internet, now, you get on the Internet easily, but what, you've called something before, I think I've heard you call it the original sin.

• 16:07Marc Andreessen?

  Yes.

• 16:07Speaker 01

  Of the Internet? What is that?

• 16:08Marc Andreessen?

  Okay, so then you go, okay, so then you go on the Internet and then you see something that you might want to buy or pay for. Like, let's say hypothetically that you wanted to have like an online newspaper or something. And let's say hypothetically, you wanted to like read an article in that paper, you know, maybe you wanted to have an economic relationship where you were like paying for the information you're getting, because then you know there's like alignment of interest between the person writing the information and what you're reading. So you might think their logical thing to do might be, there must have been a button in the browser that was like a buy button, right? It's like you actually kind of have, you finally kind of have this in the form of like the Apple App Store with like the, you know, the one-click buy and then the.

• 16:40Speaker 01

  Amazon.

• 16:41Marc Andreessen?

  And that payments, or one-click buy on Amazon. But like, one would think like, maybe it was the most obvious thing to do would be building the browser like the ability to actually spend money, right? And so, you'll basically notice like, that didn't happen. And in a lot of ways, it's like, we don't even think it's unusual that that didn't happen because maybe that shouldn't have happened. But if you think about it, the reason I call it the original sin, and I'll explain why, but because we were unable to build payments into the browser, and I just like, flat out say, we're unable to do it at the time. I want to hear why. Yeah, well, I'll explain why. But as a consequence of being unable to do it, that is why the internet today, at least in the West and in the US, is predominantly based on advertising. Right? And so, and by the way, kind of downstream from advertising is kind of everything else that people are like anxious and worked up about online, right? Like what? All this stuff for privacy, user data collection, user data targeting, all these third-party ad networks that collect all this, harvest all this data, the data brokers.

• 17:26Marc Andreessen?

  And then the kind of the misalignment of incentives, you know, is the news site that you're reading, do they have the incentive to actually tell you the truth? Or are they actually getting their money from the advertiser? And so, they're just trying to get you all hyped up on something so that you'll click on more stories, so they can generate more ad revenue. And so, I think the original sin was, we couldn't actually build economics, which is to say money, to the core of the internet. And so, therefore, advertising became the primary business model. Did you try to build? We tried very hard to build payments into the browser. It was not possible. Why? Well, we made a huge mistake. We tried to work with the banks, and we tried to work with the credit card companies. Let me start by saying, those organizations have come a long way. And so, I'm not talking about the banks and the credit card companies today. I'm talking about them in 1993. And so, we went to, actually, it's funny, so we got embroiled early on in this kind of big fight with Microsoft. And so, we ended up aligned with Mastercard.

• 18:11Marc Andreessen?

  The basic thing was to get into the credit card switch to be able to do transactions, particularly microtransactions, like small transactions. So, we allied with Mastercard, Microsoft allied with Visa. And I bring up the Microsoft part because it's not, because you might say, okay, Mark, like, you tried to do this, but like, you, you and Netscape's a little company, and maybe you guys just screwed it up. And it's like, well, Microsoft also, Microsoft and Bill Gates also couldn't make it work with Visa. So, there was, both of us ended up frustrated by the same thing. And so, you know, then we went, so we approached the credit card companies. We sat down with our first meeting with Mastercard in 1994. You know, they said they wanted to work with us to put money in the browser. So, they, they said, well, who at Mastercard should we meet with? And they said, we've got this guy, and I've long since forgotten his name, we've got this guy Joe, who's like our technology visionary. And so, he's the guy, if anybody in an organization's going to understand this stuff, it's going to be Joe.

• 18:55Marc Andreessen?

  And so, we had Joe over to our office and sat down with Joe, and sat Joe, we're going to demo, he hadn't seen any of this, so we sat Joe down in front of a PC, hooked up to the internet with the browser, and we kind of said, okay. And I kind of just said, okay, here's the browser. I said, okay, click on, click on, you know, click on, click on this link, right? And so, you know, he does the logical thing, he takes his finger. presses it on the screen, completely ignoring the mouse that's sitting on the side of the thing. And I was like, no, no, use the mouse, right? And he's like, and I was like, okay, it's 1994, it's nine years since Apple invented the Macintosh. He has never seen a mouse.

• 19:27Speaker 01

  But he was the visionary.

• 19:28Marc Andreessen?

  He was the technology visionary at Mastercard. And I was like, oh, God, we're just completely hosed. Like, there's just no way on earth.

• 19:34Speaker 01

  And so you were hosed. You didn't get it.

• 19:36Marc Andreessen?

  We were completely hosed. We were completely hosed. And so we went through the entire thing. And he stared at us like, you know, we had three eyes. Microsoft went through the same thing with Visa at the time. Like, it was just a completely incomprehensible concept. And then the banks could not have been less interested. And so basically, it was, you know, it was, it was sort of the classic kind of single point of failure bottleneck, or at least in this case, two points of failure. Visa and Mastercard essentially had a duopoly at the time. And so they were just literally, you know, they just, if they did not want you to be in the switch, they did not want you to be able to do transactions, you just simply weren't going to do it.

• 20:00Speaker 01

  If you had had crypto then, how would that have changed the calculus?

• 20:03Marc Andreessen?

  Yeah. So this is, this is the what if. And so, you know, so the modern, what we're all here to talk about today, like modern Bitcoin blockchain, you know, sort of nuisance, effectively since 2009. So, you know, we were kind of, what, 15, 20 years too early. If we had had it, we would have been able to have a parallel transaction system, like completely parallel transaction system, that would not have been reliant on the centralized gatekeepers, right? And so hypothetically, it would not have mattered. You know, if Visa and Mastercard had wanted to cooperate, this is the time, that would have been great, we would have done it. But had they not, we still would have had a route to actually be able to build a transaction network that wasn't dependent on them. And so that, to me, is the great what if. And had we done that, the internet today, like I was just like, I don't know, 80% of the things people hate about the internet today would not be problems. And then the other kind of what if, and this is a big part of why we're excited about crypto and blockchain now, the other part of the what if is, it's like, okay, what if you could actually align economics, right, with user behavior and

• 20:48Marc Andreessen?

  with company behavior? Like, what if you could actually have, I mean, this is kind of the way the real world works. The real world works is, how do I know that something is valuable if somebody's willing to pay for it? Well, how valuable is it? Well, how much are they willing to pay for it? And then, and then, by the way, any, any individual provider of any product or service, if they're not doing a good enough job, somebody else offers it. And then the, you're right, the consumer votes with their wallet, and they go buy from somebody else. And so you have market discipline on all the players. And so the big what if is like, what business models could have existed this entire time, right? What drivers of economic, because, look, the internet generated a lot of economic growth just for the advertising model. Right, and so like what if you had had kind of the turbocharge of a real economic model based on money kind of integrated into the thing from the very beginning? What kinds of services would entrepreneurs have come up with that we haven't even thought of yet?

• 21:28Speaker 01

  Or what might they come up with?

• 21:29Marc Andreessen?

  What might they come up with now? And so that one of the reasons why we're so excited about crypto and blockchain is it's basically it's the first chance I've seen is the chance to kind of revisit the original sin. And you know, it's different now because it's like, you know, the internet cat's out of the bag and like, you know, browsers are proliferated and it's a different, it's a different game now. So it's, it's much more already fixed and established. On the other hand, it's, you know, huge scale. And so you don't need that many people experimenting with something to get critical mass on a new thing these days. And so this is a chance to kind of reexamine that original sin and kind of envision, you know, for the first time in a long time, be able to say, okay, what if we could build a different kind of system? You know, for example, a system in which, for example, advertising was not the central model. And how might that be an improvement? Either an improvement on what we have or just something completely different and, you know, better and potentially, you know, much bigger than what we have today.

• 22:12Marc Andreessen?

  And so I think, I think that, and you know, this is, and as we see every day, this is the wave of entrepreneurship that we're seeing now at our firm, which is entrepreneurs kind of thinking in these terms.

• 22:19Speaker 01

  Absolutely. Well, where are we in this crypto journey? I mean, having seen a technology like the internet go from something technical and obscure to now a technology used by billions of people, where do you think we are in the crypto journey?

• 22:31Marc Andreessen?

  So my personal view, so it's like, you know, what are the possible points? It's like, okay, it's like, the possible points are like, it's like, I don't know, 1965 or something and Vince Surf and Bob Conner coming up with the original idea for 2CP IP. It's like 1989 when it's present on university campuses, but like no normal person would use the stuff. 1992 and it's like a fringe activity, but it's not quite mainstream yet and it's still kind of too hard to use. And then kind of 1994 where it kind of goes mainstream. I mean, it feels to me it's sort of 1992, like 1992, 1993, somewhere in there. Like it clearly like there's like the usability gap still has a ways to close, although it's getting closer. You know, the use cases, they now, you know, several big use cases exist, they have to be, they have to be kind of more productized and provided to more people, but like it's starting to feel really, really close. And by close, I mean really, really close to like mainstream, the kind of knee in the

• 23:16Marc Andreessen?

  curve. By the way, it's also worth saying like, it was not, I think I made the point in kind of macro ways, but I made the larger macro, or microways, I made the larger macro point. It was not clear in 1994 the internet was going to hit the knee on the curve. Like, that was not... If you just, like, read the press coverage of the internet in 1994, it was like, "This thing is a joke, it's a toy, it's not serious."

• 23:33Speaker 01

  Okay, at least it didn't have a real-time price feed, like crypto.

• 23:36Marc Andreessen?

  That is true, yeah. We did not have a real-time price. That would have been extremely depressing.

• 23:40Speaker 01

  Yeah. Well, is that a feature or a bug?

• 23:43Marc Andreessen?

  It's both. It's a feature just in the sense of, like, it just shows the fact you can get... You know, the system is kind of alive and running, which was... The internet was kind of an open question. A lot of people didn't think the thing was even going to last. And so, it's sort of a marker of, like, let's say, existence. Like, it's a pulse. So, it's good to have a pulse. It's bad because you see this in the stock market, right, all the time, which is just... There's actually... This is a long-established theory in behavioral finance with stock prices, which is the more often you can measure something's price, the more manic-depressive the behavior comes around that price. And so, there's actually... You know, there are markets in which transactions happen... You know, venture capital's like this. Transactions happen every year or two or three. And so, there's only price kind of measures. There's only kind of core samples of price every year or two or three. And so, you know, 364 days out of the year, nobody's paying

• 24:27Marc Andreessen?

  attention to the price, nobody's talking about it. It doesn't matter. You're just working on the underlying thing. You know, once you go public, it's this radical change where now the stock... You know, if you want, you can now see a real-time stock ticker. You know, you can see the price every five seconds.

• 24:38Speaker 01

  Yeah.

• 24:38Marc Andreessen?

  And you can get all excited when it goes up, and you can get all depressed when it goes down.

• 24:41Speaker 01

  Yeah.

• 24:41Marc Andreessen?

  And so, you definitely have this psychological manic-depressive overlay that we didn't have.

• 24:45Speaker 01

  Yeah, absolutely. Well, we have time for one more question. I think I'm just under the wire. And that's a question, I think, with all the regulators here in the room and also all the crypto builders in the room. Are you... And I know you're very bullish on Silicon Valley and keeping companies here. Are you optimistic that the US can be a center for emerging financial technologies like crypto, given the regulatory environment here?

• 25:07Marc Andreessen?

  Yeah, so I think this is quite... This is very... This, again, is very analogous to the internet, because this could have gone the other way. Like, had the AUP not been updated by the government in 93, you know, we could have definitely strangled... We had two chances to strangle the internet and its crib here in the US, and we missed both of those chances, which I think is really good. And I think we got, you know, tremendous national payoff on many, many fronts, including economic growth.

• 25:31Marc Andreessen?

  Coming out the other side of that. I'm very optimistic. Like, we're in kind of the same position. We have the raw materials. Like, we're in the best position. We have the critical mass of talent. You know, we've got the entrepreneurship community. We've got venture capital. We've got the research universities where a lot of this work is happening. A lot of it's happening in the US. We've got, you know, this risk-taking culture. You know, we have a huge domestic market. We have experience building multinational technology companies that can operate all over the world. And so we've got, like, all the raw materials to have this be a major engine for economic growth. However, much like the internet and also like encryption, like, the cat is well and truly out of the bag. Like, there are crypto startups all over the world, right? This is happening today. And new kinds of fintech startups all over the world. And so it is going to be a race, I think, for sure. And I think there's a major -- by the way, I think there's a major economic argument

• 26:15Marc Andreessen?

  that we should want this to happen in the US. And that we want the US to be kind of the foundation for the global, you know, financial services industry and for the global R&D in this area. I think there's actually also a national security reason we should want that, which is, you know, it's the same reason it's good from a national security standpoint for the US internet companies to be the winning companies, because, like, we want those to be American assets. I think it's good from a national security standpoint for the global financial system to be centered in the US. And so, again, I think there's the opportunity here for enlightened policy, you know, that helps set the right guidelines and the right guardrails, but fundamentally enables American businesses to win these markets. And I think that would be better for us than most of the alternatives.

• 26:47Speaker 01

  Well, enlightened policy, enlightened regulators. And on that note, I think we're out of time, and thank you very much.

• 26:53Marc Andreessen?

  Good. Thanks, everybody.