 OK, Evan Schwartz will talk us about Interledger. So that's when you want. All right. So thank you all for coming. My name's Evan Schwartz. I'm an engineer at a company called Ripple that does payments. And I'm a co-inventor of Interledger. So I'm going to be talking about internet working for money. And I'm going to turn the lights off so that you can see these dark slides better. I think this is the right one. Hopefully the room doesn't blow up. OK, so in this presentation, I'm going to talk through why payments need internet working and a little bit of what exactly we're talking about there, how Interledger works, and then present a specific use case, which is web micropayments. So payments today are broken. This is what it looks like. This is an actual Starbucks in Turkey where they have something like eight different payment card terminals because none of them work with one another. Each of the banks has an exclusive license with one of the card networks. So you need eight different terminals in order to accept payments. This is crazy. But this is actually pretty similar to what it looks like online, where if how many people have ever tried to accept payments online? Fair number. It's a pain. You have tons of different options. This is an actual website. And this is something like, I don't know, 30 different payment methods that they accept. And again, the reason is that none of them work with one another. So the problem basically comes down to the world is never going to agree on one payment network. It's not about finding a better one. It's not about making a new one. The world just won't agree. There's always going to be lots and lots of different payment networks. So what do we do? Well, the problem here is that all of these networks are disconnected. So today payments work pretty well if you want to stay within one network. If I have PayPal and you have PayPal, it works fine. If I have Visa and you accept Visa, it works great. SEPA, same thing, Bitcoin, same thing. All of these networks work well if you're just sending within the network. But as soon as you want to send money to someone else on a different network, it completely breaks down. So this is actually surprisingly similar to the problem that the internet originally was solving, which was that you had all of these different information network technologies. And they were all disconnected. And so what the internet was about was internet working. So that's obviously where the name comes from. So internet working is being able to tie together independent networks to kind of make a bigger one. So the question, how do we fix payments? Well, the answer is we need internet working for money. Then I'm going to talk a little bit about what that looks like. So Interledger is an open protocol for connecting different types of payment systems, whether they are blockchains, banks, digital wallets, mobile money networks, et cetera. So in order to do internet working, you basically need four key ingredients. I'm going to go through each of them in turn. So the first one is routers. This is basically from how the internet works. We call them in this system connectors. So what this picture shows is you have a sender on the left and a receiver on the right. The sender is on a euro denominated ledger or payment network. And the receiver is on a US dollar denominated ledger. And so what you need is some party in the middle that we refer to as a connector that's basically going to accept money on one side and pay out on the other side. Pretty simple. Connectors also exchange currencies. So a connector can set any rate they want. I'll get into that a little bit later. But they'll take in a certain amount of euros on the left, pay out in dollars on the other side. So one issue that you might point out with this is that bilateral connections don't scale. Obviously, the number of connections you need if you're trying to tie together different networks increases very rapidly to the point where it's infeasible. So networks do scale. If you don't need full connectivity, if you don't need a direct connection between every possible network, we know that scales. So if there isn't a direct connection, you basically need multi-hop. So that's the second ingredient for this. So doing multi-hop transfers with money is quite a bit trickier than doing it with information. Well, on the internet, if a packet drops, that's kind of OK. With money, that's not ideal, because that basically means that somebody's lost money. So this is how the world works today. There's a system called Correspondent Banking, which is how most especially international transfers move and especially larger volumes. And so each of these boxes just represents a bank. And what happens when you do a wire transfer is you tell your bank to send this money. And they just basically do a blind handoff to the next bank, where they send it on and kind of hope that it gets to the other side. But sometimes it doesn't. And one of the things that we've a number of our customers is that something like 5% of wire transfers just get lost, which is kind of crazy that the system that underpins international payments, 5% of transfers just disappear. And so we heard one story where someone was applying for a job with us. And he had gone to start a company in Singapore and had wired 15,000 pounds from the UK to Singapore. And a couple weeks after he had sent the money, money didn't show up. So he called up the bank to say, excuse me, where's the money? And they said, so we don't actually know where the money is. But if you pay us 40 pounds, we'll start an investigation and find out where the money is. And so that's how payments work today, where you have this blind handoff. So in order to do secure payments and secure internet working with payments, you need a better system. So yeah, this is what happens if one of the hops fails today, money gets lost. So what you need is basically a two-phase commit. And so this was simultaneously developed both by us working on Interledger as well as the Bitcoin Lightning Network, if you're familiar with this. Same exact kind of flow. So you prepare a payment in one phase where everyone just puts money on hold, no money actually moves. And then you have a second phase where the payment is actually executed. And there's a cryptographic condition part of it, which is what triggers it. But I won't go into too much detail with that. So what this alternative protocol gives you is that this means that the sender has a guarantee that either the money will arrive at the recipient's account or they'll automatically get their money back. And so you get proof that the money got there. So that's the second component is secure multi-hop transfers. The third component is an address and packet format. So if you want to understand Interledger in one slide, it's basically this. There's an address format. That is, I would like to get this money over there. And there's an amount. And this is all encapsulated in a packet. Again, this is very similar to how the internet works, where the core building blocks of the internet are at the internet address or IP address and the packet format. So this is Interledger. So the address is a hierarchical identifier that can identify ledger, subledger, accounts, things like that. I can answer more questions about exactly what the address contains later. The fourth component, kind of racing through this to try to get through, but the fourth component are the routing protocols. So not going to go into too much detail here. This is just an example of one of the routing protocol broadcasts. Basically what it is is routing the routers or connectors broadcast their rates to one another so that they can populate local routing tables. Again, very similar to how the internet works, you can kind of think of this as BGP, but for money, which is crazy that there is such a parallel. But this is what we work on. So I've mentioned a number of times that we've taken a lot of inspiration from the internet. And as we've gone along with this project, the whole thing has turned out to be much more similar to the internet than we had ever really imagined. To the point where, at one point, one of my colleagues did a search and replace on the IPv4 RFC and kind of replaced data with money, and most of the sentences just worked, which astonished me. So taking a lot of inspiration from the history and the architecture of the internet. So familiar diagram to most, to many of you here, I'm sure. So the internet has this layered approach where, at the bottom layer, you have different networking technologies. This is your Wi-Fi Bluetooth ethernet. On top of that, you have a very simple internet protocol layer, which is just about addressing and not even routing, actually. And then on top of that, you have all of your application specific protocols built. This is the interledger architecture. Looks very similar, where at the bottom layer, you basically have all of these different ledger technologies or payment network technologies. And then on top of that, you have this very, very simple interledger layer, which is just about addressing and having this packet that kind of goes across all of these different ones. And then on top of that, you can build a lot of application specific things. Some of the ones listed there are building micropayments into HTTP or into torrents. Or the SPSP is a simple payment setup protocol. So this is interledger. And I'm going to give a quick demo of what the experience looks like. So it's kind of hard to see on a relatively small screen. But what I have here is a simple payment network that's running what we call ILP kit. And so the normal payment experience, when you go to any your PayPal or whatever kind of digital wallet, you have the normal experiences. You can type in a recipient. You can send them some money. And send goes through nothing special. So what interledger provides is a way to send across different networks. And so what I'm going to do now is send to an account on a different ledger. So this is a higher level protocol built on top of interledger. So I'm sending to an account at this other kit, completely independent. This could be denominated in a totally different currency. In this case, my ledger is in euros. This one is in dollars. So I'm going to send just a tiny little payment because otherwise my friend is going to come and actually ask me for the money. So I'm going to send it. And if you can see here, they're going to receive one cent. And I paid two cents in euros. So that kind of covers those fees. So that's a quick demo of interledger. And now I just want to give one kind of one taste of what this type of open payment protocol can do. So today, the fact that payments are so broken, I would argue is breaking the web as well. This is a great quote from CEO of Apple that when an online service is free, you're not the customer, you're the product. Everybody here, obviously, very familiar with all of these services that exist on just sucking up loads and loads of data. One of the reasons why it's so difficult for any kind of service to not be a data service or an advertising company is because it's very difficult to pay online. People are probably familiar with the very annoying Wikipedia banner ads that come down and say, hey, please, please give us money. Wikipedia is one of the most used services online, and yet they still have such a hard time raising money. And in part, it's because the payment experience is so difficult. As a result of this, there's very few companies that can actually convince us to take out our credit cards and fill out some details. So what we end up with is a small number of services that can act as a gatekeeper, skim tons of money off the top, and pay out very poorly to the actual content creators. And some of this has to do with unwillingness to pay, and a lot of it has to do with just it's very annoying and insecure to put your credit card information into just some website. And so if you had this open payment protocol, which is what we work on, what it would enable is direct payments from you to the content creator or an individual website. So a lot of peer-to-peer models are suffering from the lack of an easy way to pay. And some now are kind of building in special tokens, especially in the blockchain space. But the problem you're running into is the same one where everyone needs to be on the same network. So in order to support this kind of decentralized web, one of the things that you need is a payment protocol that's neutral, that's not tied to a specific network. So in conclusion, this is Interledger. This provides internet working for money, which increases connectivity and competition. It can lower the cost of actually doing payments to the point where it can enable totally new use cases. This is one of the great things that we saw with the internet. And some key facts about the project. This is an open protocol that anybody can use. We're in the process of starting a nonprofit foundation to own all of the IP. The code is you can find online on GitHub. Most of it is Apache 2 licensed. We're working on it as part of a W3C community group, which has 201, I think it's actually 230 members now, ranging from banks to central banks to payment companies to blockchain startups, et cetera. And so yeah, come help us connect the world's ledgers. I may have time for one or two questions, but I'm not sure. So thank you. Sorry for speaking so fast as well. So the question is, is there security built into the routing protocols? This is something that we're working on a lot. One of the things that you saw with the internet is that the routing protocols have evolved many, many times since its inception. So we're kind of starting off with something. But if you're interested in routing and these types of issues, please come get involved in the project. There's tons of interesting issues like this because you have a lot of the same issues that come up with the internet. But even more so and in a kind of tweaked way because you're dealing with money. So the answer is yes, this is something we're working on and thinking about a lot. But please come talk to me if you're interested in hearing more about how it works. Good question. So the question was, is it part of the W3C web payments group? And one of my colleagues is actually co-chairing that effort. This is not directly part of that. That effort is a W3C effort to standardize a payment API in the browser. So this is kind of a new payment method that we are hoping will be supported in that. But it's a different effort. And I, one minute? OK, that's time. Thank you very much. I will be outside. Come talk to me if you'd like a sticker.