 Oh, my God. I'm so excited. I was like, hey, you guys better come. If you RSVP, man, you better be here. And sure enough, you did. Wow, this is just so awesome. I'm so excited. You know, I just, I started this meetup because I really thought we had a lot here in Boston and Cambridge. You know, I mean, New York, San Francisco, no, it's happening here in Boston. So I wanted to bring together some of these researchers. And today, you know, wow, I just, I don't even know what to say. Thank you so much G20 Ventures. I'm not even going to talk too long because I want to let our main speaker get up here. But thank you all of you for coming. Thank you MIT Bitcoin Club for the help that you've given us. And just thank you, all of you have stepped in and helped with the meetup and everything. So tonight we have a really special presenter and you all know who he is. He has a long history in this space. And I've heard some people call him the grandfather of blockchain. Well, that's something when you're only, well, 31 years old. But that tells you how young this space is. Anyhow, Charles Hoskinson has gone on to do some really cool things and he started input output Hong Kong. And they are the firm behind the Cardano blockchain. So without further ado, ladies and gentlemen, Charles Hoskinson. Hi everybody. I shaved. Everybody noticed that. You all mind if I take a selfie? This is going on Twitter. All right, thank you so much. Appeal to vanity. So first, thank you so much, Amy for putting this on. And thank you, Mike and G20 Ventures for your hospitality. Thank you everybody for showing up. Amy asked me to do this event for a while. And I said, okay, you know, we'll have five, ten people. She said, oh, no, it'll be like 40. And they're like, okay, that's a big group. And now what are we like, 200, 300? 400. Wow. Okay, so let's play the distance game. How far did everybody travel? Anybody here from outside of Massachusetts? Wow. All right. So what about you? Burlington, Vermont. Wow. Can anybody beat that? That's Florida, yeah. Singapore. Wow. They win. I don't know what the prize is, but give them something. Okay, so the topic of tonight is third generation blockchain technology. And that's kind of funny because many people have just recently or have only been in the space for a little while. We're already in the third generation, apparently. So that's going to be fun. So I've been in the space for a long time. I've been in the space first as kind of a lurker and a watcher since Bitcoin was a dollar. And back then it was a very different environment. We would never fill a room like this. I mean, you could be like, come, free money, pizza, everything you want, and we'd get like six people. Okay? Now this has been pretty common. And it's amazing to see the growth of passion and excitement. But back in the first generation, back in the day, and this was like 2009 to 2012, 2013, basically we were just trying to solve a very simple problem, which was could you create a form of decentralized money that actually has value and nobody controls it, doesn't have a central party behind it? Today it seems like obvious, right? Yeah, we have that. We have Bitcoin. But back then it was actually a very controversial and interesting question. You'd mind something, but who would buy it? Where would you trade it? Where would you sell it? I remember back when Bitcoin was a dollar, I'd be like, I have all this Bitcoin and my friends would be like, yeah, but what can you do with it? I said, but you don't understand, one day I can do a lot with it. And they were like, ah, you're crazy. Okay, go away. So it was a heck of a lot of fun. But here's what happened. It got valuable. So it went from a dollar in a very short period of time up to about $250 during the Cyprus crisis. And what that meant was that people started taking it seriously. And when they started taking it seriously, what's the first thing that happens? People start commenting, yes, but they want to do something to it. They want to build a business on it. And they start discovering that Bitcoin wasn't good enough. So there was an invitation for augmentation. So we started seeing lots of things like altcoins. We started seeing overlay protocols like color coins and master coin materialize. So right around that time period, I said, you know, what the heck do I have to lose? I'll just go into the Bitcoin space and do something professionally, have some fun. The problem was I didn't know anybody. I knew of Bitcoin. I had a forum handle of Bitcoin Talk and Bitcoin Reddit. I mean, that's basically the extent of my exposure. So I said, here's what I'll do. I'm an academic. I'll teach a class. So I'll create a free online course and release it for Bitcoin. Yay. I'm a Peter Sellers fan, so I named it Bitcoin or How I Learned to Stop Wearing and Love Crypto after Dr. Strangelove, right? And so I put on a Udemy, put on a Recreative Commons license. I released it and I got, today I think I like 70,000 students or something like that. I got 5,000 emails and I replied to every single one of them. And what I gained from that is I got to meet everybody. I met Anna Blavine. I met Andrea Santinopoulos. I met a lot of venture capitalists. It was just an amazing experience. And from that I was able to leverage a career and build some companies. But the thing I'm known for is the project of frustration. That's Ethereum. And that came a little bit later. So what happened was that there was a convergence of forces. So I had worked on a project called BitShares and I'd worked with many other people in the space and we were all kind of in a rock and a hard place where every time we built a protocol we'd want to add something to it. But after you deploy a cryptocurrency it's really hard to change it. That's kind of the point. So how do you upgrade after you've deployed? You know, it would be so nice if you had a programming language or a scripting language that let you do that. And right around the same time Vitalik Buterin was working on Color Coins Master Coin and he was just in development hell. He was having all of this trouble trying to upgrade Color Coins and Master Coins to do reasonable things, okay? So he said, wouldn't it be nice if it was like the web when JavaScript came to the web browser? And we just got a programming language and now we have Gmail and Facebook and all these other SaaS products, right? So wouldn't it be so cool if we had a programming language for a blockchain? Now this wasn't a completely new idea. It has its roots back in the 1998 and 1990s from Nixzabo. It's called the smart contract. And actually, Sergio Lerner, who's another pioneer in the Bitcoin space who's underappreciated, was one of the first people to talk about Turing Complete scripting contracts and now he's one of the co-founders of the Rootstock project down in Buenos Aires. So Vitalik wrote a white paper and I got dragged into it and boy, we had a lot of fun along with a lot of other people and we built Ethereum. And that's the second generation. It's I want my transactions to be smart and I want my transactions to do meaningful things, okay? So it's not good enough to have money, it's not good enough to have a decentralized database, a blockchain, whatever the heck that means, but you also want the things that are happening to be eventful, stateful, rich, and programmable. So for example, imagine if you have a business logic where you have a board of directors and you have normally three of five people are required to sign for a transaction. But on December, it goes down to two of three because two of those directors go on vacation. That's your business's custom logic. I can't predict that. But with a smart contract, you can write that kind of a relationship financially speaking and it can be as arbitrarily complex or arbitrarily simple as you want. The point was that you're in charge of that transaction. So this was a crazy idea. We took enormous criticism for it. We had no thought that it was ever going to succeed. We released it, now it's worth $100 billion and everything's gone mad, right? It's paper. So here's what's happened, is that the world is kind of caught up and they said, decentralized money, that's a really good idea. Okay, check the box. Smart contracts, we like them. There's a lot of cool things you can do and now we're having questions of how do we want to do them? They want to be turned complete, turned incomplete. Do we want to write special languages for them? Do we want to write them in normal programming languages? Are we going to run them on blockchain, off blockchain? But the concept is okay, check. Do they work for billions of people? No, CryptoKitties killed us. Be killed by a cat. Oh well, at least we have a few more lives. So anyway, they don't work at scale. The other problem is we have thousands of them, right? If you go to CoinMarketCap, you'll see all these cryptos all the way down. A few I built and guess what? They don't talk to each other, a lot of them are blind, deaf, and dumb. So that's not a good thing. Finally, how do you decide how to pay for things and to govern things with cryptocurrencies? So let's say that you are a genius cryptographer and you're like, yeah man, I'm going to solve this quantum computer crisis. I have XMSS, it's the best signature scheme ever. And he's also a genius cryptographer. He's like, nah man, lattices are like so much better. We're going to go with bliss or dilithium or something like that. Okay, these are two proposals and let's say they're both great guys. How do we decide which one to go with? Who gets to decide that? Do we just go to Bitcoin Talk? Do we go to Reddit? Whoever has more Twitter followers? What is the empirical, meaningful, systematic way upon which we are going to govern these open source, not centralized, not curated protocols without an appeal to a cult of personality or something along those lines? That's a challenging question. We're seeing the fractures of governance. Anybody here hear the block size debate? Yeah? Anybody here have any Bitcoin cash? Anybody have any Ethereum classic? Oh, there we go. I'm just going home now. Okay, so anyway. So anyway, basically we have governance crisis and it's a very challenging one and it's one that creates a barrier to consumer and enterprise adoption because you as an executive can't go to your boss and say, hey, CEO, I think we should use this blockchain for this thing. It's like, yeah, but we have no idea how it's going to change, who's going to change it, when it's going to change and what it's going to do, and we have no say in the matter. It's like, yeah, that's a good idea. I should build a billion dollar product on that. I mean, think about that. It's not a really good and meaningful conversation. You know, the other thing is how do you pay for stuff? So we have the ICO. Anybody here have an ICO? Yeah, right, SEC is not in here, I think. Right, okay, good. So what is the ICO? Well, you know, we inadvertently created a Rube Goldberg machine with Ethereum for venture capital, which really makes Mike's life difficult, right? People are like, wait, why do I have to give you equity? I can just issue a token and raise 10 million or something like that. But that's temporary. Don't worry, you guys will be okay. But basically you just issue a token like an ERC-20 token. You go do a crowd sale, whatever that means and you raise a bunch of money and now you have a bunch of money. You can go spend it and it's a donation, so no strings attached. But no, levity aside, the ICOs are great and they're becoming regulated. There's lots of ideas from Smart Valor to Blockhouse and others, Project Brooklyn from Consensus about how to do these things, how to govern the funds that are collected, the checks and balances and controls that ought to be put together. So that's wonderful, but it's an injection of capital. And like an IPO, these are injections of capital that you can't do consistently and repeatedly, right? So how do you pay for things long-term? And how do you pay for things without having a big pool of money laying around that somebody's going to control? Okay, so we have that problem again. So we have a sustainability problem as well. So we have a scalability problem. We have billions of users that are coming in and we don't know how to handle them and crypto-kitties is killing us. We have an interoperability problem in the ecosystem where we have thousands of cryptocurrencies and they're all just not talking to each other well. And we have a sustainability problem where basically you can't figure out a pay for things without these damn ICOs and also it's hard to figure out where to go. So whenever you have problems that are indicative of your size, scope and scale, whenever you have problems that occur as a consequence of success, that's a good symptom that you have a new generation coming. Just like in the old days when we went from first generation to second generation, we were saying, gosh, everybody's trying to get smarter transactions. Everybody's trying to do more, but every time we try to do it, it's painful. Now we're in the same situation where I want to do all this cool stuff, but I want to do it on a global scale. I want to do it with traditional systems and other people's systems. And I also want to build a community with having to hold a billion dollars of their money and trying to play venture capitalists. That's a big challenge. And that is basically an invitation for the third generation of cryptocurrencies. It's an invitation for new classes of technologies. So congratulations. You've lived through two generations already. Isn't that cool? You all should have white beers and be like Vint Cerf and be like, ah, back in my day, we screwed up so bad. Right? Okay, so what is IOHK? What is Cardano? And what are the other third generations? So after Ethereum, I kind of had the luxury of being able to start a kind of company that I wanted to start. So I'm kind of an academic. I studied mathematics. I like analytic number theory. And so I said, you know, there's a deficit in our space with respect to rigor. There's both a peer review and academic deficit and there is an engineering deficit. So on the peer review side, we have a lot of people writing papers. And those papers are not really rigorously checked by people who should be reading them. And these papers are playing with not simple things. They're playing with cryptography, with programming language theory. They're playing with game theory. They're playing with all kinds of things that people spend a lifetime studying and rigorously reviewing and having to go to great schools like Harvard and MIT in Boston University and CU Boulder to... All right, that's for last. Go Ralphie. But anyway, they have to go to these great schools to learn how to do these things and then they have to go submit their papers to a conference for the computer science side or a journal for other sciences. And guess what happens? You get your paper rejected. Or if it gets accepted, it's like conditionally accepted and they yell at you. And then you have to go and show the paper off and then they yell at you. And it's like a very masochistic process. There must be a special kind of crazy that goes into these fields. But the end result is that you actually have some degree of assurance that the paper may be right, that the claims may be reasonable, that the things we're trying to do may be accomplishable. And that they've been checked by people who've spent their lives thinking about this stuff. So I said to myself, it'd be so cool if I started a company that did that. But that alone is kind of not a business model. It's kind of a lost leader. Spend money, but there's like not money coming from the other side. So I said, you know, I do need to do something. So why don't we build cryptocurrencies for a living? That would be a heck of a lot of fun. So I said, okay, but if we're going to build them, we're going to build them like a mathematician would build them. So we're going to build them in Haskell So what the bloody hell is Haskell? Or what is functional programming? Well basically, it's just a different way of writing code that brings it closer to math. And it's a different way of writing code that allows you to use all kinds of new tools to test and check what you're doing to make sure it corresponds with reality. It's not new. It's been around for a long time. The problem is it's a little harder to write Haskell code. It's a problem. It's a little harder to test it. A little harder to find developers. So it's not quite as popular as JavaScript. But the other thing that Haskell can do that's really magical is Haskell also is really easy to work with formal verification tools. So what the heck is formal verification? Why should we care? Okay, so you have a paper. Okay, I don't have another problem. Actually, I do have the stand. Okay, and so you have your code. Code lives here. So your paper lives in like the Billiken land. Billiken is like a Japanese spirit of the god of the way things ought to be, not the way that things are. Okay, so this is an ideal. It has ideal functionality, all these cool properties and features, and this lives in the academic's mind. Now let's assume the paper is right. Then at some point an engineer comes along and says, I'm going to implement that paper. Cool, it goes over here and writes a bunch of code. And there's this gap. It's called a semantic gap. And basically there's going to be a difference between what the academic did and what the engineer did. There's a thousand assumptions. There's a thousand ideal functionalities that had to be turned into real functionalities for the sake of practicality. For example, you can't instantaneously transport a blockchain to everybody at the beginning of the launch of the network. It doesn't happen. But you can assume that to make a proof work. So there's this gap right here. And if you're not careful, the thing you end up building here isn't the thing that you wrote here. So all the promises, security guarantees, and things that you'd care about and formal methods do is they say, we're going to do something about this gap. We're going to write a specification that allows us to close it up a little bit. And it's going to extract all the ambiguity and be machine understandable. But it's not a full implementation. Then what you can do is use what's called a proof of bisimulation here to here. And actually show that every output here is the same as that specification. So in other words, it has no bugs. As long as the paper is okay, you're like, well, but Charles, that sounds great. Why doesn't all software do that? It is horrendously time consuming and expensive. And the French are involved with it. Who would want to be with these guys? I'm serious. They screwed us, man. The most popular formal programming methods language is called COQ. I kid you not. So this is what we deal with. So out of principle, we use Isabelle, which is slightly less popular, but we didn't want to say COQ every day. And at levity aside, it takes a lot of time. For example, let's think of things you'd like to be verified. How many people here have operating systems? Well, you do. You have phones and computers, right? So the kernel of an operating system would be really nice to know that that's properly built. Because you kind of rely on that every single day for every single thing you do on the digital world. Somebody did that. It's called the SEL-4 micro-kernel project, and it took over five years, millions of dollars, 180,000 lines of Isabelle code to produce 7,000 lines of generated C that had some nice properties. That is not easy. So I said, you know, I want to build an engineering company because I'm kind of a crazy entrepreneur. Not only do I want to do the paper stuff right and go through peer review, but I also want to do this, but I want to do it quickly. It's kind of a crazy idea. So that was the dream of IowHK. Get the engineering right, do functional first and formal methods, and get the academia, right? Started in 2015. There was just two people, my co-founder and I. And now we're 100. We operate in 10 countries. We've got nine figures of revenue. We have a lot of projects we work on, the largest of which is Cardano. And we're building Cardano with this type of approach. So what is Cardano and why is it a third-generation cryptocurrency, and why do we think it's fun and interesting? Okay. So as I mentioned before, there are three characteristics, scalability, interoperability, and sustainability that justify the third generation. Okay. So in terms of scalability, our belief is the best way to achieve that is as you gain users, you gain resources. Pretty simple. Good ideas should be simple. So do we can think of any other protocol that has this property? Anybody have any ideas? Yes, sir. Maybe Yoda. Maybe. BitTorrent. Well, that's the one I was going for. It's more classical, less controversial. Okay. So BitTorrent. What is BitTorrent about? So first off, it has over 250 million concurrent active users. That's a lot. And it carries about a third of the internet's traffic. The only way you could have a protocol like that run consistently for more than 15 years is if you have a property that when someone enters the protocol, the protocol gains resources, it gains capabilities so that you actually get faster or stay at the same performance despite the fact that you're gaining hundreds of thousands, millions, billions of users. That's a basic characteristic of scalability. Now BitTorrent is quite simplistic. It's just moving data around. So cryptocurrencies are a bit more complicated and nuanced because you have more than just moving raw data around. You have transaction processing capability. And this is where things like Hashgraph and IOTA are trying to make innovations. You have things like data storage. Okay. So where do you put all the information? Who stores the blockchain? Is it sharded or not? Are you trusting somebody to hold it? And there's the actual data itself that you have to port and move around. Think about the consequences of having a system that runs at a million transactions per second. How much bandwidth do you have for a node to process that? It's not cheap. It's not small and it's not coming from grandma's wifi. So our current models that we have for replicated networks for generation two, generation one cryptocurrencies exist in a world that everybody has a full copy of the blockchain, that everybody has a reliable, stable internet connection and can keep up with the network. These are assumptions that will become untrue because we become victims of our own success. So to achieve real scalability you need a protocol that gains transaction processing power. You need a protocol that gains network resources. You need a protocol that gives you more data storage as you gain more users. So that's the first property that we try to resolve with Cardano. Now the problem is that this is a horrendously complex task. A lot of people can claim that they've solved it or they know exactly how to do it or what it is. But it requires lots of protocols. It requires you to design network protocols. It requires you to shard large data sets. It requires you to implement new authenticated data structures. It requires you to put in new cryptography. This is a challenge and no one solution can actually do all these things together. And also let's say you think you have it. How do you know you have it? You need peer review. You need some bedrock to sit on. So in 2015 when we first started working on this project the first question we asked is what is bedrock for a cryptocurrency? It turns out not until 2015 was there even an answer for what is bedrock? We didn't know what a ledger is. We had an idea of a blockchain, like an append only link list. There's been some ideas of directing and cyclographs, these types of things like Spectre for example. And yet we didn't actually know what properties they should have or how to write a security proof. So we had to do, our chief scientist had to write a paper on it. It's called GKL 15 where we defined this is a secure ledger. This is the standard upon which we ought to be judged. Great. So the first question you ask is Bitcoin is a secure ledger. And actually in the paper they prove that Nakamoto Consensus forms a secure ledger. So congratulations. Satoshi did something really magical. If it's Craig Kudos because guess what? It actually turned out to have all the security properties you'd want to have without providing any security proofs that did it years before anybody bothered to clean it up. That almost never happens. So Bitcoin is pretty magical and there's a reason why it's so loved. So proof of work solved that problem. Great. The next question is what other types of protocols that are much more scalable have these properties? So we'd like to use proof of stake. There's a lot of good conversation around it. That's floating about. But does it have an equivalent level of security as proof of work? This is a horrendously controversial statement amongst the Bitcoin people. They say of course not. But we're academics. So we said let's use this foundation we've generated that we've proven that proof of work is secure with. And then let's go ahead and see if proof of stake has the same properties. And surprisingly the answer is yes. Turns out that given certain assumptions proof of stake can have an equivalent level of security as proof of work. That to me said we're in business. Now the question is how do we make it practical? So what protocols do we have to marry it with so that we gain that scalability property? And what protocols do we have to marry it with so that it becomes a sustainable system? And so in 2016 we wrote up our proof of stake protocol called Ouroboros. We had to go through six revisions. We first submitted it to Oakland, got rejected. That's a tough conference to get into. We dusted ourselves off. We got into that conference in Santa Barbara. We had to go hang out with Addie Shamir and Whitlock Diffie and the rest of the old guys and some of the new guys and get yelled at which is our favorite thing to do in the whole wide world. And then after that we said you know we're not done we have like six more papers to publish because this is hard stuff. One of which we presented this morning and we wrote another paper called Ouroboros Proust that's making our protocol more practical and we have more papers on the way. One of them is making that heartbeat that consensus algorithm of our system a bit more resilient, a bit more robust, a bit faster, and eventually it will gain resources as people join the network. Now that alone is not enough. We need a network protocol. We've chosen RENA by the way. Kudos to Boston. Part of RENA was developed here at Boston University. Any BU grads? Okay. So the guy named John Day helped create the internet along with a lot of other people and John is the guy who created Recursive Internet Work Architecture. Wrote a lovely book about it and he was one of many who looked at it but looked at networking as a distributed computing problem. It's complicated and involved so I won't go into it tonight but it's a new type of protocol it's a little different from normal network protocols and the power and advantage of it is you gain much more granular control over how you build these large scale distributed networks so you can model them so that you don't have to download everything and if we just did those two things alone we'd have built a pretty good cryptocurrency and it would be worth some money in trading but remember what I was saying is you have to be interoperable and also sustainable plus it would be nice if we had good smart contracts and stuff like that, right? This is just generation one stuff we're just trying to make generation one faster. Okay. So how are we doing smart contracts? Now here's the problem. We screwed up with Ethereum. Sorry. The virtual machine wasn't really well designed and it was not well designed because we didn't put enough time into it enough people into it we didn't have enough resources to really carefully think about what are we doing? The languages that we wrote, like Solidity anybody here a Solidity developer? Okay. Is that your favorite thing in the whole world? You just like wake up every day and be like man that Solidity is like Zen. Right, no. No one has ever said that about Solidity hanging out, painting to Bob Ross like loving life, everything's great and you made a Solidity developer like in a basement with whiskey doing cold shots saying why won't this work? Okay, so that's an indication we kind of screwed up on language design so Ethereum has a bit of problems Ethereum itself acknowledges it and they're fixing them they're creating new languages like Viper and Bamboo they're cleaning up the virtual machine and these types of things they're even thinking about pivoting to WebAssembly so evolution is on the way but the first thing we do because we're academics is we say well what should the machine look like so let's study what we did so we worked with a team called Runtime Verification they're based at the University of Illinois Urbana-Champaign, huge campus, little city wonderful place Gregorio Roshu and his team wrote the first set of formal semantics using something called the K Framework where they actually modeled the operational semantics of the Ethereum virtual machine Huzzah the hell does that mean it means that we very carefully read the documentation of the papers and translated it into a formal language that's machine executable in other words we actually made the computer understand what we wanted to do as opposed to what we actually did in code okay what does that mean it means that now you can actually rigorously look at the system and say what did we do well what didn't we do well and what needs to be completely thrown away oh god what were you thinking it was ugly of systems design so Gregorio's team took the KEVM which they published last year they updated it, upgraded it and created a new register based assembly virtual machine called Yella named after a Romanian fairy you can you can guess where Gregorio is from so anyway great machine we have the semantics for that and we think that this is a much better foundation because it looks a lot like LLVM developers are a bit more familiar with that it's easier to build compilers for that it's much better studied than the current foundations that we have but the problem is we also need to be able to write smart contracts in the languages that you want to write them in not the language that Charles Hoskinson wants to write them in so we said wouldn't it be so cool if we had some sort of framework where all we have to do is just define a language and then suddenly the framework itself build a compiler for you and then you can just write your smart contracts and you don't have to do anything if I ever update the virtual machine I just do it once all the compilers are going to build but Charles if only well it turns out that we may be able to do this and we think we can there's a thing called semantics based compilation and there's a thing called the K framework and what you do is you write the semantics of a language and K takes care of all of that translate your language into another language so we have K semantics for Java for C for JavaScript for Solidity for Plutus which is our own in-house language and Viper and we'll just keep writing them we'll just keep making the framework better the long and the short is that if we pull this off you'll just write your code in the language you want click a button compiles runs on our framework you don't have to worry about it or build your own compiler or these types of things that's pretty cool and it's actually a big innovation in computer science and we're pretty excited about where we can take it we're a very large team led by runtime verification that's working on that we think that's fun okay so we're pretty comfortable with the smart contract side of things we're pretty comfortable with how this stuff is evolving we love the fact that we're introducing lots of great formal methods we even have conversations about like proof carrying code and actually we've even specified the first formalized ERC-20 token spec this type of stuff that would be a cool product but what we've just done is built a better Ethereum and I didn't sign up for that I signed up to build a third generation cryptocurrency so briefly there's two last things I'll mention about Cardano and then we'll take some questions because I know you guys are probably dying for them first is interoperability so what is interoperability really all about what are we really trying to solve there they're actually it's amazing how there's two things right there's actually two things to worry about with interoperability one is the legacy guys the chases of the world and the wealth froggos of the world the banks exchanges traditional regulated financial actors two are the cryptocurrencies of the world okay so the problem here is a problem of attribution and metadata and standards and it's a problem of consensus within a community and we just have to accept what they want to talk to them because they kind of own their side of the system it's a negotiation but this is a problem resolved by committee and it's a problem resolved through things like interledger and what Ripple's doing and other such things okay and that'll come when it comes I don't have any control over it the problem with the cryptocurrency world is actually a cryptographic problem it's actually very easy to understand cryptographic really hard problem to solve so the problem here is if you're Bitcoin and there's litecoin over there if somebody sends you a litecoin transaction wants to put some litecoin on your Bitcoin blockchain you want to honor that you have two questions you need to answer first are the litecoins real do they exist do they come from somewhere second have they been double spent or not these are two foundational questions the proof of existence and the non-existence of a double spend pretty simple right so how do we do that well we trust the consensus of the system we have a full copy of the blockchain so wait a minute if we have a thousand cryptocurrencies how can we have a full copy of every chain no you can't so wouldn't it be super cool if you construct a proof a very tiny proof in the kilobytes the megabytes for each cryptocurrency that can act basically as a bridge between you and the other currencies so that when you get those transactions you can verify that they exist and they haven't been double spent we wrote a paper on this it's called non-interactive proof of work and it's awesome we love it we just recently pushed it out and we're translating it to proof of stake and our hope is to pull this into as many proof of work cryptocurrencies as well the carrot outside of interoperability is the fact that you also get super efficient liteweight clients because the very same thing that allows you to validate the transactions are legitimate from external blockchains allows you to validate the transactions are legitimate within your own blockchain without possessing the whole copy that's pretty cool science and it's going to take some time and effort but that's what we're doing there now for the sustainability side comes down to voting, voting, voting now we had an election recently if you guys didn't know did your favorite candidate win? mine didn't although I worked for Ron Paul so I never will win at anything guys, right? okay so the problem with that election was less about the candidates running it was more about the voting system itself and the fact that we were given poor choices by the time we had the right to vote so it's not just being able to vote that matters it's what you get to vote on when you get to vote your incentives to vote how you vote who gets to be on the ballot these types of things that are much much more meaningful and it's been studied by cryptographers and political scientists and other academics for a long time but come up with all these cool names like Condorcet, voting systems and Borda, voting systems and so forth and the theory is quite interesting so we have to take a position and we think the best of the worst is liquid democracy it actually has a lot of really cool interesting properties we get to do things like additive homomorphic encryption God help us but at the end of the day what this is all about is being able to just delegate your vote to other people for each situation whether it be funding for a ballot or it be voting on a change to the underlying protocol and that's the matter of setting thresholds and assigning incentives closely so like all good academics we did literature review so we started with Dash BitShares and other systems that had been released in the past that had some notion of voting or treasury and we extracted from them the good, the bad and the ugly similar to how we extracted from the Ethereum virtual machine the good, the bad and the ugly and we wrote a paper put it on our website then we put a whole team of academics together at Lancaster University wonderful place out in the UK led by Bin Cheng and we're now actually writing papers on how to build a voting system and how to build a treasury that we can abstract away from a blockchain and then put in as a reference model for any blockchain that wants to adopt that's similar to our interoperability work so our hope is that eventually this can do what Tezos is trying to accomplish which is saying that you can mutate or change the chain you can fund things, these types of things but it can do it in a way where it has some good cryptographic guarantees and some fine game theoretic properties that your voting will result in good outcomes as prior elections have shown getting an election system right is super hard and in itself it's worth a generation but in my view it's the only way that we actually succeed or do anything interesting in the long term because at the end of the day these systems need management they're getting too big the money is getting too big the risk of their failure is getting too big if we can't find a way to govern ourselves the governments are going to just govern us for us and then we've lost so that's what Cardano is in a nutshell we have some ideas about scalability that starts with the Ouroboros agenda the conceptual idea is just increase resources as you get people we have some ideas about interoperability and that's just all about saying let's do some side-chain stuff and some magic there and for the legacy stuff let's join the right committees and just adopt the standards and then we have some ideas about sustainability and that starts from let's build a good voting system and let's go ahead and build a great treasury and a great way to change the cryptocurrency all of our work is open source all of our research we try to make peer-reviewed and we try to collaborate with as many people as possible I which has three research centers one at University of Edinburgh one at Tokyo Tech so that does mean I get to go to Japan a lot and one at University of Athens we're also setting some up here in the United States and we'll be announcing that soon hopefully and we work with lots of different academics from researchers and open source developers and it's been a heck of a lot of fun learning how to build this type of a project now I mentioned there are some other competitors and you know, well, mentioned them you have guys like IOTA and AOS and they're in the scalability bucket they're really trying to push that we can do millions of transactions per second and these things can handle the load there are people who are in the governance bucket for like the dashes and the tazosis of the world who are really emphasizing that element of the system and then you have people like Ripple and AEON who are living a lot in the interoperability side of the spectrum so that's okay that's their vision and they're all executing and they have money and people and we love reading their papers and we love seeing where they're going we try to be a little bit like the Capybara try to be like the Zen master of the animal kingdom we try to be in the center where we do a little bit of each and provide that to people so thank you so much for coming I hope this was a little educating and I'd love to get your questions okay, yeah, for questions we're gonna line up here in front of the mic Hi, Ty Danko massive Cardano fan and thank you for elevating the level of transparency and all of this what I don't get is when you make Cardano, I'm gonna call it backwards compatible or being able to accept Solidity and Java and all these other languages and your essence is being more robust and more rigorous than the other languages how do you keep the problems of for example the parody wallet exploit from migrating into your system including you that is a great question you're not the first person to ask me that question Phil Wadler and I had a long conversation about it, he said Charles if your backwards compatible with Solidity you basically accept all the sins of Solidity at the same time that's a good point, Phil so the same way that poor countries protect people they have walled happy rich people land that's segregated and everybody's safe and then they have the hinterlands world so you do the same thing in protocol design so you have Cardano SL and you have Cardano CL so SL is where we do our accounting and then it's connected via sidechain to CL where we do the computation once you've done this abstraction there's no reason you can't run multiple control layers one of which is basically a clone of the Ethereum virtual machine which is 100% backwards compatible with the VM by code and the other one is Yella so as a coder you have to make a decision of what you value do you value safety security or do you value ease of development and your old tools that you're used to the network will provide resources to both but the security guarantees are only with one and it's encapsulated yeah thanks so yes sir John Bottoms I'm the inventor of the browser 1987 congratulations I'm sorry that JavaScript got put in it I feel pretty bad about that I have a few things to say on that going forward letters of credit paperless letters of credit involve escrow which fall into the Fed what do you see as the way forward paperless letters of credit involving escrow so we could talk about lending in general do you want to go down that road mostly transactions where there's an agreement by a financial institution controlled by the Fed okay so anything that's Fedland is not going to run on a permissionless ledger this is not going to happen it's going to go permission ledger at some point so you have solutions like hyper ledger enterprise Ethereum there will be an enterprise Cardano will announce that at some point and that's where you say okay there's an institution issue between trusted known actors who don't trust each other fully so Wells Fargo and Chase and others they all know each other they're public they're regulated institutions they won't show each other their customer ledgers without some arm twisting but they do have a degree of trust that these are legitimate entities or that they have recourse if that entity lies so the permission blockchain problem has always been a coordination problem I have a collection of actors that must do business with each other due to the dynamics of the market whether that be telecommunications with spectrum you're moving from one telco network to another and you want to have roaming or that be medical records where you move from one hospital to another hospital or that be financial transactions where you move from one bank ledger to another bank ledger so what ends up usually happening is that they designate a protocol and a standard error to be the central authority we trust to help reconcile everything so we say swift or fix protocol or whatever that may be and then the lender of Lazarus or it's usually the central bank so if they really screw it up the central bank is there to come in and kind of resolve these types of things so we have this nice hierarchy of necessity and hierarchy of centralization that's been constructed to coordinate institutions that don't want to work together but have to work together because of nature of the customer where a blockchain is nice is a blockchain gives you the ability to say you can remove these centralized consortia of necessity and now you can form a common state between your organizations where it allows you to agree on facts and circumstances that allows you to agree on time stamps it allows you to agree on a litany of little things that are just difficult and usually require god mode some central actor to do that now in the existence of this you start having to bake the question do we actually still need central banks it's an interesting question and that's probably going to be the next wave of monetary policy discussion that occurs saying how much decentralization could be introduced into the central banks the Federal Reserve itself is a decentralized entity it's not just one single building sitting in Washington DC with one guy at the top who's like really smart or gal it's a committee and a grouping of banks and there's some degree of decentralization there so they've already admitted that there needs to be federated control so similarly the same discussion that can happen in inter institutional transactions can also happen with central banking itself it's not going to happen in America to begin with it likely will happen in small central banks for example in Barbados and Jamaica and all these other little places where they actually have their own central bank because they print their money and they do trilateral settlements so you can't go from the Bayesian dollar to the Jamaican dollar you go from Bayesian dollar, US dollar, US dollar Jamaican dollar and they don't like doing that it'd be nice to do bilateral swaps so you can introduce coordination there if it starts working you can pull it into the stack a bit and then gradually can open things up now what are the other reasons you do this because you gain things like proof of solvency you gain fraud proofs you can do real time continuous auditing you know it's much easier to reason about how customers and money is going to flow through that system on the dark side it gives you much easier ways of enforcing capital controls which is why the Bank of China is incredibly interested in this type of stuff so it's an interesting question and it's something that's right now not academic it's actually happening it's happening within our 3CEV it's happening within the enterprise groups all these people are talking to financial institutions Barclays has its own research desk they're publishing papers on this stuff the back is writing things like smart contract templates and so forth digital asset holdings has a whole line of research that they're doing for this type of stuff they even create a programming language called DAML and they have a whole enterprise blockchain solution they're thinking about so I suspect over the coming years this will organically work its way in now the question is how does that system talk to my system and Vitalik's system and the IOTA system that requires some discussion and that requires some standards and that requires some careful thinking because these guys if money is in the system live within a regulated world meaning they have metadata they have attribution so why did you do the transaction where did you do the transaction at where did you get the money from who the hell are you are you on a sanctions list there is no de minimis clause for money laundering so they have a whole bunch of regulated needs here which none of these transactions natively have installed into them and these guys aren't going to willingly put that stuff in and I don't think they should I think we have every right to resist that encroachment okay but if you're going to move your money from that system into JPMorgan Chase's ledger you can't be like yeah I know you have to have a relationship with the regulator but for me can you just be principled if you want to do business with that guy you have to play by their rules so there needs to be some discussion about what that looks like and what rights do you lose when you go from one enclave to the next enclave and that's the next big conversation that's going to be happening over the next few years and I don't have an answer for you I mean we're going to write papers we do participate lobbying policy I sit on several committees one in Barbados where we're doing a sandbox for ICOs so it's like I do have a little bit of say but these discussions are above my pay grade and I'm just going to deal with them yes sir hey my name is Shiv Patel I'm a software engineer at AWS we use your services thank you I'm just wondering why is IOHK committed to working on Ethereum Classic what do you think is the market fit given Cardano's existence yeah that's a good question you have kids when you have kids you'll know this one I have two kids which one do you love more okay so Ethereum is a project I worked on I love the social contract in my view of Ethereum was code as law I wanted a better Bitcoin that's why I worked on it Vitalik wanted a world computer these are very different things when I'm building with Cardano looks a lot more like a world computer than it does a commodity with principles and utility so those visions were okay until they weren't what happened was that an event occurred where one group of people were like oh obviously we're over here and another group of people were like hang on guys we're here what are you doing over there come on over and there's irreconcilable differences so I happened to be in the minority and I put my money where my mouth was I hired a Scala team we built a client and we'll just keep working on it as a donation and keep it coming along but you know there's room in the world for digital commodities by the way we think Bitcoin is going to go away just because Ethereum is here and IOTA is here and these other guys are here it's going to be here forever and it's going to slowly lumber its way around as digital gold so look at Ethereum as digital silver it's a commodity in its own right but it has more utility silver can be used in medical applications and semiconductors and things like that so similarly it has a smart contract language and we'll try to innovate we'll innovate on the proof of work side and make it a self-sustaining system and it's a great experiment in decentralized governance I don't have control over ETC there's several groups of people from Barrie to the Commonwealth to Ethdev and these people are equally powerful if not more so than our say in the matter yet we have to coordinate to upgrade that in itself is worth our participation because we're learning how to coordinate with a decentralized system instead of being like what Ethereum does you have a god mode and whatever comes broadcast down we just accept it I'm not Steve Jobs I don't come out with say one more thing here's the iPhone we're all in this together guys we're building these things from the bottom up and if we really are serious about these being decentralized systems then we have to get serious about updating and upgrading systems in a federation where people actively don't like you or don't like your ideas on the other side of the federation without turning into anything if ETC gives us that knowledge okay you know that's great for me and I'm also very proud of the software we wrote we wrote the Mantis client the growth and team spent a whole year writing 10,000 line Scala code it's super concise it's fast it's beautiful it's just insanely great still not Jobs and you're just going to love it and by the way we'll have it run in Ethereum mode at some point so it can both download the ETC blockchain so guess what we're Ethereum developers too maybe we can join the EEA maybe they'll send us invitation so I understand it's controversial we have segregated teams the Mantis team doesn't talk to or work with the Cardano team they write in different programming languages and there's not a lot of common DNA there second Cardano is a proof of stake system Ethereum Classic will remain as a proof of work system it looks like and so we can innovate in different directions and run the experiment of which one seems to be better for the market and we place Ethereum Classic as a commodity and we place Cardano as a third generation financial system that will be the stack for the 3 billion people who don't have one said that a lot yes sir hey I'm Sid Ramesh I recently launched a weekly newsletter on Cardano I saw that how much aid did you get I said give this guy some aid I don't think we got any because the wallet was still in maintenance in Betrex damn it we'll check it out we launched our first issue and Charles actually retweeted it thanks to that but check us out subscribeweekincardano.com so my question is for a lot of people here at least in North America Cardano came as a surprise a lot of people didn't hear about it so you guys were focusing a lot of your efforts in Japan sort of the Southeast Asian side could you talk a little bit more about the history of Cardano and why you guys decided to do your initial token sale in Japan that's a very good question in North America I was shocked to learn this when I was traveling like there's other countries and stuff it's really amazing no but levity aside Asia is a great place it's the future of finance you have 4 billion customers and guess what they all have bad financial systems that are old in Japan's case it's like 30 years out of date it's still a cash economy like the first time I went to Japan I got a credit card and it was $300 for a train ticket it's like you gotta pay cash you're the teller for the train tickets how do you not accept credit cards this is insane but this is Japan, this is where they live so to me it automatically made sense that if you wanted to have a meaningful sustained impact on the evolution of finance and money that it was very important to innovate in places where there was an actual need to change system number one, number two they're willing to listen to you the problem with America in Western Europe is the regulators do not listen to us the regulators just don't give a shit about innovators or how we go about our business they just decide how do we keep the people who are in charge in charge that's why no one goes to jail when crisis has happened and that's why we have the system we have so I don't want to play in a rigged system I would much rather go to the Caribbean I'd much rather go to South America to South Asia I'd much rather go to places where there is a need there are customers, there is value in some cases more so than the U.S. markets and go build something there second, at the time the project began in 2015 cryptocurrency penetration in Asia was actually very light the largest country by far was China Korea had almost not heard at all and Japan hadn't really done very much they were like oh that Mt. Gox thing didn't the CEO of Bitcoin get arrested and it's only recently that we've seen a humongous speculative surge in the Asian markets so we actually timed the markets correctly in that respect so it was a strategic bet it was a cultural bet and it's something that did pay off but Asia's not the only place we are going to enter Africa this year we're planning to go into Ethiopia we're planning to go into Ghana and to Kenya we're already in Barbados in fact as I speak there's a Haskell class being taught at the University of West Indies and Barbados is structurally very similar to Ghana and Kenya and Ethiopia if you look at any index like the amount of people who are unbanked the average median income education these types of things it's a good sandbox for us to test our expansion plans into Africa so we're going to enter those markets and we're already in Argentina we've been there since 2016 we have an office with ATICS and it's right in the Bitcoin center there we talked to Diego and Circeo they talk about monetary collapse pretty regularly so it's great to be in a jurisdiction where when you talk about monetary collapse they get it in fact as a side story one of my favorite guys I talk to a lot every time I go to Argentina he took me to his house one time he says Charles let me show you something really cool he said okay so this is where I buried the silver so when the money collapsed the police they came and they stole the silver so let me show you something else he's right behind this bush and he's like this is where I buried the real silver and that's the decoy silver over there you learn these things so I love being in countries like that you know the people are just so much more real they're more genuine they have nothing and if you succeed you build something and frankly that's in my view the only way we're ever going to get a good banking system financial system in the western world you can't go and ask for permission please chase in Goldman Sachs change the rigged system that makes you rich they're not going to do that and no matter who you elect they're not going to do that but you build a system with 3 billion people that's worth more than the system we have and the only way to do business with that system is to change your system that I think that I think is going to change the world and so that's why we focus on that area it means that unfortunately you guys you speculators didn't make quite as much money so sorry but you know it's time for other people to play yes sir hi my name is Joe Miano I'm a data scientist at CVS Health my question is around the scalability so we think about the popularity of coins like iota ryblox things with non-blockchain technology I wanted to ask how does Cardano achieve increased scalability and what are the theoretical I guess limitations limitations and maximums in terms of transactions per second right so we actually don't know what the peak TPS rate is and even if you have a high TPS rate the question is can you propagate all the message through a gossip protocol it's a really interesting question so we're studying the network side using a science called DeltaQ we have a firm out at Bristol led by a bunch of network theory academics and basically they're called PNSOL and we have some videos explaining how that works as the network topology changes and as we gain users proactively as opposed to retroactively through empirical analysis that's one of the first steps because you need to know where can you go with what you have in terms of Ouroboros itself the reason why we designed it the way we did is that we have this idea of an epic so how Ouroboros works is that you have two inputs that hold an election and they elect a bunch of slot leaders to run consensus for a few days five days in the case of the current implementation so you need a sorts of randomness and then you need some sort of election set of people who are eligible to become slot leaders and that's called cryptographic sortition not a new idea it's called Follow Satoshi by Ito Bentov Sylvia Makali also has a similar system and so forth and the sorts of randomness you can get from a hash from multi-party computation all these things is very scalable now the first thing to do is to say that if you've done this election in just the right way that that epic is secure and these slots in the epic are honest as long as that committee is majority honest which is a similar assumption to Bitcoin majority of miners being honest the next thing to do is to say can you take that epic and instead of having one of them run another one in parallel and another one, and another one, another one up to n whatever the needs of the network happen to be for its historical throughput it's kind of like a feedback mechanism like mining has with difficulty right we can go up and down based upon some evidence of the endogenous within the network and that's the next generation of our protocol after we finish Orvore's prowess up and it finishes peer review is to explore how do we run epics in parallel and then once we do that then we get n performance it's n times the amount of the constant for whatever the TPS limit for that is that's your raw theoretical throughput and as you gain users it's easier to gain committee members so it seems to me like this will scale fairly well the problem is still moving that data and storing that data we're fairly along with the moving the data part and using rena we think we have some good solutions as for chopping up the blockchain and putting it into its own little buckets and no one has to have a petabyte or exabyte of data still very hard problem the good news is there's some very well funded ventures that are trying to solve that and we're reading their papers right now and we're still kind of in the due diligence and literature review phase of the project but that will become a necessity in 2018, 2019, 2020 we become more successful and we gain more users we imagine that these chains could become exabyte scale or yoda byte scale at some point if they become very useful so there's just no way that anybody's never going to be able to possess all of them so you need to be able to chop up history now the good news is the same interoperability research we're doing that allows you to have a compressed representation of a chain but still verify things are correct in addition to interoperability can be repurposed for lightweight clients so it is our belief that over an arc of time that research will bear fruit for the sharding effort of the data side as well so that's kind of the best short answer I can give you now if you want to know more go to our website we have a library with tons of papers and take a look at some of the papers read through them send some emails our way and also I can put you in contact with Neil and Peter who work on the more data sciencey network sciencey side of our organization we're actively testing about how do we simulate and test these networks as if they had a million users and a billion users and so forth this is a difficult systems problem and it's one that we're working our way through yes sir Matt a data scientist I had a question more about the practical side you've mentioned not only proof of stake but also being able to have a bit more control of the nature of the language that smart contracts are written in as someone who's programming solidity thank you so much for that a bit more on the practical side do you think this could sort of look potentially lower the barrier for creating mobile decentralized apps well that's a really good question so how do we run these things on cell phones and what does that experience look like problem there is that that's not a fair game either Apple and Google have a little bit of a say what goes on their phones right so we can't maybe they like crypto but maybe they won't like crypto silk road so you have to think carefully about that okay so first for language design itself because this is an area that I'm a little passionate about I love it a lot the first question you have to ask is what problem are you trying to solve with any programming language no one language is universal language for everybody you're just dealing with a troll and they say oh my language can do everything you're just dealing with that so for financial contracts there's actually a very meaningful thread of discussion that's occurring about whether those actually need to have gas and be turned complete can we build a domain specific language that has a collection of combinations and that is complete and sufficient for all financial relationships and there's some great research that's been done along this line for example there's a project called Project Actus and Actus has this idea of 30 primitive contracts that are composable that can allow you to basically rebuild any bank or rebuild any insurance company and so forth so you can build a domain specific language wrapped around that there's even better research done at Microsoft Research by Simon Payton-Jones it was called Adventures in Financial Engineering this was back in 2000 and what Simon did he was one of the creators of Haskell a brilliant guy, very charismatic but what Simon did is he said let me reimagine the bank as a functional language and let's just see how far we can go with the minimal amount of functionality possible Chris Clack has done the same with smart contract templates as I mentioned at Barclays and then also Blockstream is doing this research as well it's called Simplicity where they're actually thinking carefully about how do we get something that's better than Bitcoin script for the developer but doesn't allow you to shoot yourself from the foot, chop your leg off okay, we actually have a project at Kent University under Simon Thompson and as postdoc Pablo we're creating a language called Marlowe where we're taking a wrap up of many of these things we've learned and trying to create that turing incomplete DSL for developers now what does that mean? It means briefly you're like a paint by number style contract so you're kind of like Bob Ross put a tree here, put a cloud here so similarly I want to do this many multi-sig and I want to do escrow here to have an oracle here and whatever and you just kind of plug those things in and then you just know the contract is going to work as you've written it you as the developer don't have to think about that once that becomes a standard it's a lot easier to have a meaningful discussion about how do we get these things to run on machines like the EVM or Yella or that be on cell phones and these types of things so that's a discussion as a conversation we have limited control over what can go on to Android and what can go into iOS you do what you can now the other side of it is do you have ideal languages that are really good for robust applications for programming smart contracts we have one called Plutus we're developing with Phil Wadler and Daryl McAdams it's the next generation man we're really pushing that hard and we think we built a really nice language so that's the best non-answer that I can give you to your particular question and I have to kind of punt it a little bit because the reality is that it's not in my control to create a good development experience for mobile clients I can build my own client and I can make it good I can write in F sharp and use Samaran and have like a beautiful cell phone app great and I can go beg to get it on an app store but at the end of the day that is a walled garden and that's the problem with walled gardens is that you don't get certainty with projects you have to have negotiations banking is a walled garden I have to negotiate with them and cell phones are a walled garden they have to be negotiated with as well now if they're too draconian we'll do what we did with money we'll build our own hardware and that might come and we might have to so yes ma'am and then we're going to have to call it a night well thank you I'm honored to have the final question and I want to thank you again well what's your name? my name is Yusef nice to meet you I want to thank you for your contribution to the future incredible as we all know blockchains are technology but also economic systems and social consensus networks you intimately have experience in a contentious hard fork and after years of civil war about the scaling debate bitcoin and bitcoin cash diverged and although on face that was a technological issue in reality it was a power struggle to be future oligarchs what lessons have you learned in the nine years that this has been going on about what blockchain 3.0 means in terms of economy as well as social consensus and ideology that's a great question thank you so much you saved the best for last you ever watch The Simpsons? building these protocols is almost like this one scene from The Simpsons so Mr. Burns gets angry at Springfield and he's like I'm going to turn the power off for the city so he and Smithers they go through this elaborate series of doors and hidden passageways and retina scanners and so forth to get to the heart of the power plant so he can turn off the power so just as he arrives in the heart of the power plant there's this rickety old wooden door that leads to the outside and there's like a dog inside the room and it kicks the dog and shifts the door and that's unfortunately where we're at right now in the design of cryptocurrencies the sticker price is very expensive you have to learn all of these extremely bizarre complicated terms and concepts like public keys and private keys and you have to go ahead and download this strange client that's peer to peer and run it on your computer and then you have to be really careful with the way you use it and back it up and oh and by the way a hacker is always trying to steal it from you and if something happens that's on you so you pay a huge consumer unfriendly price to use the system and you have to go to the lab tunnels and at the end of the rainbow you have that backdoor of oh by the way there's like six people who basically run this thing and they run it in the shadows like a colstered emperor because they write the code and you trust them to make the code correct because they run the mining pools or they own a large chunk of the currency or they run the exchanges and they decide who gets listed and who doesn't get listed or they run the media systems where there is power and money there will be attempts to centralize and to control that power and money no one is resistant to this ever in any capacity that's the lesson of Satoshi that's why Satoshi had to die and disappear you can't build a cult of personality around someone because no matter how good that person is how smart that person is they will let you down they will betray you or they will get killed or corrupted or coerced or forced act the way it breaks your heart there is no greater example in my mind than Ripple I like Ripple I like David I like the guys there I think they're good guys and they do good engineering and for a long time I a lot of respect they get kicked all the time and people piss on them and for a long time I always defended them I said you know Arthur at the end of the day and Chris these guys are good guys they're okay and the entire day I really did I felt sick to my stomach because it wasn't like we brought them as an advisor we hired them to do some legal work we put them on the board of directors to decide the philosophy of our company and who we are as people because it's necessary for the next step in their evolution so they can do their IPO and all become billionaires and ride off into the sunset with Mark Zuckerberg great but that's the nature of humanity and who we are so the first lesson I've learned is no one is above corruption boom everyone will become corrupted at some point so how do we establish that things will work right process over people if I make a science claim don't trust me trust the peer review process it's been around for a long time it's given us the planes trains and automobiles we have and the physics we have it's pretty good it's not perfect but it's pretty good especially in the computer science world that's the first point second when people write code ask who watches them who's reading the code we have an external auditor it's called fp complete we'll start publishing those reports February most likely who they say here's the good and bad in the ugly of what Charles has done it's not all swimmingly perfect but you know you accept these things so make sure that there are checks and balances that are in place behind how the code is written and the claims that are made that'll get you so far your system needs to also have a way of governing itself within the system you need a constitution you need rights the reason why I still have the freedom of speech and the reason why I still have the freedom of religion is because some very smart people a long time ago wrote that down and they put it in the document it's really hard to change and boy there's a lot of people who love to change it if they could change it so protocols are all about how do we enshrine those principles that we started with before everybody got rich before everybody got powerful into the protocol so that when they get rich and powerful they get tempted to change things for their personal benefits they can't this is why I was so against the Dow fork with Ethereum because I felt like okay well now for the rest of time every single time a bad event happens you have to make a decision and somebody is expected to make a decision one way or the other parity has like what 200 million locked somebody has to make that decision maybe they won't but admitting that a decision is there to be made by a person is admitting you've actually given a CEO to your system and that just felt wrong to me the bitcoin core developers resisted that when mount gox happened they personally some of them had millions of dollars in mount gox and they chose not to reimburse themselves so I think those are some of the basic lessons scientific claims need to be peer reviewed not by random normal weirdos over the internet but by scientists if you're making scientific claims because they spent decades of their lives at great institutions like CU Boulder and Boston University to learn how to do this stuff and code claims should be reviewed by credible third parties and bake a constitution into your protocol itself and give a voting system to people who have it alright well this has been a lot of fun thank you so much thank you Amy thank you Charles Hoskinson thank you Charles all the way to Cambridge to see us we appreciate it so very much thank you all of you for coming tonight please make sure to tweet Cambridge blockchain Cambridge let people know