Epicenter - Learn about Crypto, Blockchain, Ethereum, Bitcoin and Distributed Technologies - Evan Schwartz & Stefan Thomas: Building the Internet of Payments with Interledger

Episode Date: May 16, 2016

One of the foundational problems in payment networks is that they are mostly uninteroperable. This problem exists at all levels, from consumer payment solutions like PayPal, to national and multi-national banks. This complexity is brought on by the proprietary nature of payment networks, and moving value from one to another requires a negotiation between parties on which common payment network to use in a transaction. We saw similar problems in the early days of the Internet, assembled around protocols which allow for data to be routed and move between networks in a standardized way. We’re joined by Stefan Thomas and Evan Schwartz, co-creators of Interledger. This neutral protocol would bring the same level of interoperability we know take for granted around the flow of data, to payments, thus allowing money to move freely across networks. A market maker, who holds accounts in both networks, would receive funds in escrow from a sender, and move funds to an escrow account with the receiver, getting paid by the sender when he shows the proof the funds were delivered to the receiver. Topics covered in this episode: What is Interledger and what problem is it trying to solve Interledger’s architecture How connectors and routing works, and how we may compare it to the way data flows on the Internet Cryptographic Escrow and its role in Interledger Requirements for payment solutions to become Interledger compatible Interledger’s community group at the W3C How Interledger applies to micropayments Ripple’s role in Interledger Episode links: Interledger Interledger Whitepaper Interledger + WebTorrent - Demo by Evan Schwartz Interledger Architecture Interledger Github Interledter Slideshare Ripple Lab Episode 92 with Stefan Thomas This episode is hosted by Brian Fabian Crain and Sébastien Couture. Show notes and listening options: epicenter.tv/131

This is Epicenter Bitcoin episode 131 with guests, Stefan Thomas and Evan Schwartz. This episode of Epicenter Bitcoin is brought you by high.combe. Protect yourself against hackers and safeguard your identity online with a first class DPN. Go to hide.m.me slash epicenter and sign up for a free account today. Hi, welcome to Epicenter Bitcoin. The show which talks about the technologies, projects, and people driving decentralization and the global cryptocurrency revolution. My name is Sebastian Kucho. And my name is Brian Fabian Crane.

We're heard today with Stefan Thomas and Evan Schwartz. Stefan, many listeners will probably know because he's been on the show before. He's the CEO of Ripple and Evan is a software developer at Ripple Labs. And they are also the co-creators of the Interledger Protocol. Interleisure is very exciting because they're trying to tackle one of the really, really hard problems, which is interoperability when you have 100 new projects coming up with their own blockchain and people trying to integrate that into all. kinds of other systems it just becomes a bigger and bigger problems so we're glad that

they're trying to tackle that and we're excited to speak about that project today so thanks so much for joining us guys thank you so yeah even actually actually i we've been we've been sort of harassing you guys to come on for i don't know how long six months perhaps and no no i'm excited that it's finally working out and and i think i mean you also gave a talk at to meet up in Berlin about this. Yeah. And you also gave, we should also add, I think there was maybe a year before,

or maybe even longer than that, a workshop on Codeus ones. So can you talk briefly just sort of where does Interlegia come from and how does that relate to Codeus? Yeah, so when we're working on the Codeus projects, one of the project, one of the things that really came out of that was like starting to think beyond a blockchain,

and thinking beyond having one cryptocurrency that everything's sort of tied to. We didn't want everyone to have to use XRP or Bitcoin or just one specific cryptocurrency to pay for Codius contracts. And we also wanted the contracts to then be able to trigger payments without having to integrate with 10 different types of blockchains. And so it was really during that time that we started to think about interoperability and like how do you make it so that the user can pay with one thing and the Codius host can accept another thing

and have it still work. And then also have sort of a standard protocol to kick off payments and to control where money is, no matter what kind of ledger you're talking about could be, you know, blockchain, a centralized ledger, et cetera. And so that's where the sort of key, sort of core of interoperability came from for us. Yeah, and I'll just add that a lot of, like today, if you want to make payments, most payment systems in the world work fine as long as you want to send a payment to someone else who's on that same payment system. So if I'm in Europe and to send a payment using SEPA, works well. If I have Bitcoin and you have Bitcoin, works well. If I have PayPal and you have PayPal, probably works well. The problem comes in, if I have one of those and you have a different one, and we have no match, and maybe there's no way for, we don't want to get another account. We'd like, when we talk about this vision of the Internet of value, what we see is an experience where it's much more like how the Internet actually works, where I have whatever money,

or whatever, I'm on whatever network I want, and I can send money to you. So that's a lot of where the interoperability goal comes from. But wasn't this originally kind of also the idea of ripple that you can sort of then connect into all kinds of different things and, you know, people can issue assets on ripple and then everything kind of moves smoothly as well? Heck, it was even the idea of Bitcoin. I mean, if you think about it, that was one of the promises of Bitcoin is that you can use Bitcoin to send money anywhere.

around the world without having used banks and that and Bitcoin provides a layer of interoperability. Although, although Ripple I guess was, did make the step that you say, okay, you can issue like US dollar on Ripple or something like that. Yeah, this obviously, you know, really speaks to me the way that you're putting it because that has been really my journey in terms of the goals always been the same. It's been to make, you know, the movement of money as frictionless as information. But that was the reason I got into Bitcoin and what I noticed within Bitcoin, and what I noticed within Bitcoin, was that, well, you needed support for different currencies. So I joined Ripple.

And then I noticed with Ripple that, yes, now we have different currencies, but you still have sort of this issue that you all have to be connected to one ledger. It doesn't address how to move the money from that ledger to all the other systems out there. There's never going to be one ledger that all data is stored on and all transactions happen on. And the reason is not even scalability. I think we could solve scalability with sharding and with other cool concepts. The problem is that there's a sort of scalability of functionality as sort of diversity in terms of the use cases where you can't have one system that really perfectly serves each use case. It's the same system that's doing micropayments going to be good for moving billions of dollars.

It's the same system that's really decentralized good for someone who wants like the lowest latency possible, probably not. So it's probably good to have some layer on top that allows a little bit of diversity in terms of the ledgers and still makes it possible. to transact without thinking about the difference of ledger and thinking about I have to go from this network to this network. And what we've also seen is that because there's so much that goes into the design of a specific ledger and there's so many decisions that are made and often there's some, there's economic interests in people using your ledger, what you end up having is someone's, let's say I come up with a ledger that's going to be the ledger for interoperability, it has all the best features, etc.

So I go to you and I say, here, come get on my ledger. But you already have the ledger that you're using. And you say, no, you should get on my ledger, not the other way around. And then you're at an impasse and you end up with a very fractured system where everyone is saying, my ledger is the best one for interoperability. But you don't end up with interoperability because everyone stays in their own camps. And so what we saw the need for was something that could bridge that, that allows everybody, to use whatever the ledger they want and it doesn't matter and we can still get that

payment experience we're looking for. So what's interesting here is that you use this term ledger. Is a ledger an accurate description for all these different systems that hold value? Because I've always kind of heard ledger in the context of blockchains, right? We speak about of distributed ledgers. But does it make sense to speak of ledgers? When you speak about PayPal and all these other systems, we speak about, that you want to be interoperable here?

So in general, the term ledger just refers back in the day, really back in the day, and in some places still, a ledger is just a book with accounts and balances. And it records changes to those. And ultimately what you care about is like, I have this account, how much of the thing that I'm tracking, does that one own?

And you can see everything from, we use that as a kind of general term term because every one of these different cryptocurrencies, banks, PayPal, etc., are, they all operate ledgers. Maybe it's possible that they don't think of them, that as being their primary function. They may think of themselves as a payment network, for example. But underlying all of that, you have some listing of accounts and balances. And that's what we talk about when we say ledger.

And one should also add, right, that interledger itself is actually not a ledger, right? Correct. Yeah, there's no ledger built in. It's purely a protocol for interoperability, or it's a protocol suite for interoperability between different ledgers. I think also interesting to point out, and perhaps when people initially think of these things, they think of a network. And interleger is not a network even. It's a protocol. Clearly, at the base layer of what a protocol is, it allows people to organize around the same. way to communicate. Yeah, so there's actually a pretty close analog between interledger and the internet. That's also partly why we chose the name. So for example, on interledger, there is a way to address different ledgers and address accounts in different ledgers, just like the internet protocol IP has IP addresses to address hosts

on different networks. There's also other similarities. So for example, on an internet protocol suite, you have TCP and you have UDP, which are what's called transport protocols. And so they take care of things like retries and retransmissions, and so there's an equivalent intelligence as well. So it's just a parallel in terms of internet is not a network, it is just something that ties different networks together in the minimal way possible.

That's a really key point because when, so a lot of different cryptocurrencies have been called at different times protocols for money. And a key thing to understand about that is that all of these cryptocurrencies, you know, that Bitcoin, ripple, et cetera, have protocols that they use to communicate with one another. But the really key point is that in order for me to use the Bitcoin protocol usefully, I have to be connected to the one true Bitcoin network. There's not really any purpose for you and I to use the Bitcoin protocol with one another if we're not both talking about the exact same network,

because then basically what you're talking about is a fork or an alt-coin or a different chain and that's not that's not bitcoin that's your other thing and so while they do have protocols in order to use them usefully with both bitcoin and ripple you have to be connected to the network of specific validators etc whereas with interledger it's really just a way for different networks to peer with one another so there's no there is no one interledger in the same way that there's no there isn't really one internet it's just it's a network of telecommunications networks and what we talk about it as one thing, but it's really an abstract concept that's just made up of this network of networks. Yeah, and what's interesting in the literature that you've written in

the white paper is that you reference these RFCs from the late 80s that describe the internet protocol and describe, yeah, IP. And it's really fascinating to me that, I mean, these ideas have been around for so long. And I mean, we can get into this later, but it seems to me that Interledger itself, the fact that it's happening now, and of course, like you guys are in the Bitcoin world and ecosystem and what have you, but I mean, Interledger could have existed a long time ago, like even before Bitcoin. There's nothing that I don't think anyway, that Bitcoin brought anything new to the table that made Interledger inherently possible. So it's sort of fascinating that it's happening now, like 25 years after the web was created and whatever, how long

after the internet was created. And it took so long to get here when all these principles have been around for quite some time. That's a really good observation. I gave some guest lectures recently for some students. And one thing that I told them is that I think the biggest contribution that Bitcoin has really made is that it's brought a lot of really smart people into the space of finance where, you know, a lot of them might have not even thought of themselves as financial engineers or anything like that or something that that would be something they could ever have an impact on. And think Bitcoin gave everybody sort of this realization that, no, you can have a huge impact by working in this space and by working on these technologies. And so I think

that's a really powerful contribution. And I think that's why these types of efforts are happening now. It also got a lot of financial networks to say, oh, hey, we should probably look at this new technology that's coming out and what you know what ever everyone nowadays is like what are we doing about blockchains and that that's also a really big thing to get existing existing financial players to come along and say yeah we actually we like that vision that you're pitching how can we be part of it the doors are all open right now yeah i think what bitcoin may have brought also says you know this change in mindset and this idea that you can innovate without permission in in this financial space which is highly regulated and sort of, you know, cordoned off.

So let's talk about interledger, the architecture. Can you give us a description of what are the different layers of the architecture of the protocol? So we use a model with currently four layers. And the four layers are the ledger layer, the interledger layer, the transport layer, and the application layer from bottom to top. So the bottom layer, the ledger layer, is basically where you find the ledger protocols. This is where you find the Bitcoin protocol, the ripple protocol, you know, ISO 20-022, which is what a lot of banks use. So this is where you find sort of the nuts and bolts of making

money movements, making transfers on individual ledgers. The next layer is the interledger layer. And on the interledger layer, what you find is basically the header, the information that is necessary to associate a transfer with what's supposed to happen on another ledger afterwards. So if you see an interledger transfer on one ledger, a connector can look at that header and can look at that interledger layer data and knows what to do with that transfer going forward. Now, with that, you have sort of this graph of nodes of ledgers and connectors. And the next thing you need to do is you need to actually send something across that without without dropping the money or things falling apart halfway.

And so for that, we have a transport layer. And so just like TCP takes care of retries, we have a transport protocol called a UTP, universal transport protocol, which basically takes care of retries and takes care of putting money in escrow such that it can't just get lost on the way. And then finally, you have an application layer. And so on the TCP IP side, on the internet side, you would find the HTTP there.

you'd find SMTPs. There's lots of different application layer protocols. That's probably the layer where there's the most diversity. And so we think with Intellegic could be very similar where there are sort of a bread and butter standard protocol, which we haven't designed yet, which is the equivalent to HTTP. But then there's lots of other ones there for specific use cases. So for example, within Ripple, we have one called Payment Services Protocol

that is proprietary, that is completely just used within Ripple. That's an application layer protocol that builds on the open ILP process. calls. Let's take a short break to talk about high.mee. You know you need a VPN provider to protect yourself against those nasty hackers trying to steal your private information. With high.db, it couldn't be easier. You just install their application on all your devices, iOS or Android, log in, and you have

a cushiony, cozy tunnel in which your data can move freely and unencumbered all the while protecting you from those mealed hackers. Now that's boom. To sign up for the free plan, go to High Top Me slash Epicenter. The best thing is when you use that URL, if you ever go premium, later, you're going to get 35% off. And premium comes with unlimited bandwidth, using up to five devices at the same time. You can use all their servers worldwide. You can pay with Bitcoin.

And best of all, it comes with a feeling of peace and satisfaction like having tea with the Dalai Lama. We would like to thank HaidotMe for their support of Epicenter Bitcoin. So talking about the first one, the ledger. layer as sort of the fundamental, I guess, would it be correct to say the ledger layer is really kind of a set of requirements that different ledgers have to fulfill in order to be part of interleger? There's a couple different ways you can look at it. So the ledger layer is the, as Stefan was saying, the protocols that individual ledgers use.

So the Bitcoin protocol, for example. And so we put out this architecture document and in there, and we put out this architecture document and in there, one of the things it talks about is different levels of functionality that an individual ledger might use in order to support interledger best. So the absolute minimum is you have some protocol for expressing transfers from one account to another. That's like the really the absolute minimum and having a way to either have a unique identifier or a memo, some kind of memo attached to that transfer is on is like on the lower lower end of functionality. So you have basic transfers from one account to another with some data attached to be able to say,

this is that transfer we talked about. So would an example of this be, let's say a separate payment where you have an IBAN and, you know, you send an amount and then you can also send some texts with the payment? Yeah, exactly. Exactly. Today we're kind of in that phase where, like, when you talk about SEPA as a ledger within to ledger, like the first thing that comes to my mind is as fast as SEPA is and as efficient as SEPA is, is it really that kind of, you know, ledger that could participate in an interledger transaction?

And something that's really put that in context for me recently is like looking at some old videos when people were first experimenting with Internet protocols, and they were also using a lot of not-so-great networks. You know, you can kind of think of the phone receiver that you're putting onto the thing, it makes a noise and then the phone receiver picks it out, right? That sort of old school stuff. And so that's kind of the phase that we're in right now. Yeah, you could make an interledger payment over SEPA, but we also imagine that interledger is going to put a lot of pressure on ledgers

to get a lot faster, a lot more efficient, and the true interledger vision won't be realized until ledgers have, sort of, yeah, ledgers have caught up with it, essentially. Yeah, sorry, I can speak a little bit more to what some of those requirements would look like. So, Stefan mentioned things like speed and, like, latency and scalability. So that's one piece of it that a lot of ledgers right now are not very scalable. And so that would sort of limit the throughput, obviously. Another part of it is when Stefan was talking about the different transport protocols.

So the way the universal transport protocol works, specifically what it's trying to address is we have the way interledger works is you have ledgers and you have connectors sitting in between them. And a connector is some party that has an account on both ledgers. And what you want to do is I want to send my money, to a connector that shares a ledger with me, and I want them to pass it on. The immediate problem you run into when you're talking about that is, what happens if they steal it or lose it?

And so that was the first thing that we had to address with interledger, and that's what the universal mode, in the white paper, it's called Universal Mode, now we're calling it the universal transport protocol. So what that is all about is I don't actually want to just send the money to, let's say, Stefan is the connector. And he may be trustworthy, maybe not, but there's some other people behind him as well. And so I really only, I want one of two things. I either want a guarantee that either I will get a signed receipt from my final recipient saying,

yep, I got the money, or I want my money back. And so that's the guarantee that the universal transport protocol has, is that either you have your money, either you keep your money or you get a signed receipt. And so the way that works briefly is basically with escrow, where instead of sending the money directly to Stefan, because he might run away with it or steal it, what I'm going to do is put it in escrow on my ledger. So with Bitcoin, you'd use hash time lock. With other things, you could use a dedicated sort of escrow thing. But the key point is that the ledger itself is providing the escrow. So I put my money in escrow, and it only comes out if Stefan can deliver me that cryptographically signed receipt from the recipient.

And so we set up this chain of escrows, and so everybody's putting money in escrow all the way down the chain. Then the recipient sees, oh, there's money in escrow for me, just pending my signing this receipt. They sign the receipt, and that trickles back all the way along the chain. Okay. So I think in one of your talks, you describe it as the escrow goes from, so if you have payments going from left to right, right, the escrow comes back right to left. Can you perhaps give some more detail as to how the connectors are sending money back and forth in the escrow, like in a transaction?

And by the way, so for people who are listening who want to get a better sort of visual understanding what this looks like, you can go to slideshow.net slash interleisure and there's some slides there. and I'll put the link in the show notes as well that sort of illustrate what this looks like. Yeah, I would say that, you know, you can dive as deep as you want.

If you really want to get the defense and the really explanation of this protocol, I would refer you to the white paper, which is on Intelligent.org, but I'll try to give sort of a very high-level description based on what Evans already said. So if I'm a connector, and kind of keeping with that same example,

everyone wants to send money, I'm the first connector, I don't just want to put money into escrow on some other ledger until I know that Evans actually committed to this payment. So I'm waiting for him to put money in escrow first. And that's why the escrow is kind of going from left to right. And then the execution, well, the recipient is the one who generates that receipt. So clearly the receipt has to come back from right to left. And so that's why you kind of see that like left to right preparation, right to left execution.

Right. So Stefan is the connector doesn't trust me as. the sender. So he's not going to do anything I ask until he sees money. But the thing he does trust is there's some ledger that shared between us. Let's say that's the Bitcoin ledger. And so he trusts the Bitcoin ledger. I trust the Bitcoin ledger. If he sees that there's a transfer setup with money on the Bitcoin ledger for him, and that could be, yeah, we can get into exactly how that would work, but there's money sitting there that's committed to this specific transaction. Then he's like, okay, I don't trust Evan, but I do trust Bitcoin.

And so I'm going to take that, and I'm going to go and set up a corresponding transfer on a later ledger. And what I'm interested in as the connector is that the two conditions are the same. Because again, I trust the two ledgers on either side of me. And so if the two conditions are exactly the same, I know that whenever the right-hand side transfer is going to be fulfilled, I can take that fulfillment, that signature, and pass it on to the left-hand side ledger. and they'll be valid there as well.

So in many ways, it's very similar to sort of the hash lock concept in Lightning, if you're familiar with that. And one should probably add that this is actually a really brilliant cryptographic, neat little trick, right? Because it allows you to do these things in a trustless way, even though those ledgers may be very different. And it's sort of, you know, simple, works universal. And it's not, it's not very intuitive, right?

It doesn't make sense. I think if one just sort of explains why this will work. So I think to kind of really understand it, one needs to like sort of really walk through it. Yeah, you just mentioned the trustless point. And one way to look at this is if Bitcoin and all of these cryptocurrencies have brought away so that you can make an individual ledger trustless, what interledger does is make it so that you can trust,

you don't have to trust a connector that's something. party that's relaying money between different ledgers. Right. Okay. So as long as you're, yeah, because you're within that payment network, if that payment network provides an escrow, then you trust the payment network, not the connector. Okay. So coming back on this idea of connectors, you know, you've compared them to relay nodes

on the internet protocol. What would they be analogous to? Would they be analogous to an ISP or like an organization like Level 3? How can we compare our connectors to existing Internet infrastructure, you know, participants? So there's one role in the Internet Protocol that's a very, very close analog. And that role used to be called the Gateway. Today we most usually call it the router. And it's basically a system which is on two networks.

It's usually a lower-level network, a high-level network, or on the tier-level network. or on the tier one level, it's sort of connecting two ISPs networks together. And it basically knows how to take a packet that's coming in on one network and figure out which of the other networks that it's connected to, to forward it on. And so the gateways and the routers are responsible for the routing. And one thing that's really powerful about that is that it means that the sender and the receiver are not responsible for the routing.

It means that the ledgers are, the networks themselves are not responsible for the routing. And it keeps them all very simple. And so the complexity within the Internet protocols is very much sort of concentrated in the routers. And even there, like, it's concentrated in the largest routers, the routers that are actually used at the Tier 1 providers. And so your home router can also be pretty simple. And what that does is it means that the only people who have to upgrade the protocol regularly for it to work well are people who are going to have companies. They're going to be very invested in it. They're going to be competing with each other to be.

to be on the latest version. And so it tends to be a thing that gets upgraded very easily. And so you can have your laptop, you can have your phone, stay pretty much the same and still benefit from these improvements. So you would compare it to sort of the Tier 1 Internet routing, like companies that do writing, like, I don't know, like companies that do, like level three, for example. I don't really remember all the names of the companies.

So they're not the only ones to do routing, right? Like they do routing on a large scale, but then you have a home router and you do routing on a small scale. And you could make sort of a mesh router and do routing a different way. That's really all up to you. You can create your own routing protocol if you want. But I think most of the complexity is concentrated among those larger companies. Oh, I see.

Okay. So you could have, for instance, some lower level connectors that maybe do like connecting SEPA to PayPal and then you may have some smaller connectors that are connecting some obscure cryptocurrency to some other obscure cryptocurrency at some. And so you have different levels there. And depending on how big the network is that you're trying to send payments to,

you may stay on that lower level network where there's a lot of traffic and a lot of volume. But if you want to get into something like some little payment service provider in China, for example, then you may have to go through some smaller connector router and find a route to it. Exactly.

Yeah, so the connectors are like routers, and just like there's various degrees of routers and various complexities, same thing we expect to see with the connectors. So what kind of companies are going to be running this? Like will we see companies like Google in the end running those big, the big more complex routing things or ISPs, or will this be more of a financial service thing? Banks will do that, or is it more like in Bitcoin, way,

sort of independent entities that maybe be anonymous in some location? Like, what's that going to look like? It's kind of funny because it's kind of like the first person who discovers gold in California trying to imagine, like, who's going to be rich off of this? Well, whoever is there first, you know, like whoever can capitalize on that opportunity, the quickest. So it could be banks because they already have some of the regulatory and compliance of the background to do this, but they're kind of more slow moving. So maybe it's more like payments companies like Venmo, Square, Google, Apple, could be those guys.

It could be Bitcoin startups. It could be that Bitcoin exchanges are just the only ones that adopt a protocol like this. And so they just grow to be, you know, very large network providers. Whoever it is, we think it's a huge opportunity. And so it's just a matter of like who's going to be there first. And speaking of exchanges, you know, if at some point interledger becomes as big as the internet. And like everybody's using interledger, all payment platforms and payment service providers have integrated all the conditions to be fully supported. And we just have this network of payments between different types of ledgers.

What does that mean for, you know, things like exchanges that, you know, things like exchanges that, traditionally have served that role, or even the stock market, like if I can just send money to myself from one payment network to another, then doesn't that whole infrastructure of, you know, third-party exchanges sort of crumble? So I think that exchanges are essentially connectors, right? They, are connected to more than one ledger, so it could be a Bitcoin ledger and SEPA, and they provide liquidity between those two ledgers and their customers provide liquidity between those two ledgers. And so one of the primary models that we see that connectors could work is by being exchanges, by having people trading and then using some of that liquidity to facilitate payments.

So what would be different then in a world where those exchanges are talking to their customers through interledger rather than what we have now? So if you just want to compare exchanges today with how connectors work in interledger is, primarily I think the difference would be the degree of automation. And so if I want to use an exchange today, I actually need to go on one network and like manually type in a bunch of things in order to send a payment to them. And they may maybe have custom instructions to say, you know, include this transaction ID or send to this Bitcoin address or something like that. have custom instructions to be able to correlate an incoming transfer from me to them. But I need to go and actually type that in myself, and then they will go and they'll have some process for doing that on the other ledger.

What Interledger does is standardizes that way of doing it so that you can automate more of this, so that you could even see it as like a standardized protocol between exchanges so that I can talk to each of my exchanges on different networks in the same way, which just makes it a lot easier to route a lot more payments through them. And so one way you could see it is that exchanges would need to be very competitive because they could, having a standardized protocol in some ways makes them more interchangeable. But the way I would look at it is that you also make it a lot more useful because if I could, you know, if I can pay with Bitcoin really for anything and not just to people who accept Bitcoin, then Bitcoin is purely competing on the strength of it

itself as a ledger, as opposed to needing to go and get all the network effects of getting lots of people to accept it and use it for everything. And so you could have exchanges and ledgers be both more competitive, but also be used for a lot more because if you can automate things, it makes it a lot easier to route payments through that. I want to stay on that point really quick, it's an extremely important point. You think of the internet as this extremely frictionless system that you don't really have to worry about it. I don't worry which path my messages take because it pretty much works.

And one of the reasons for that isn't necessarily that the protocols are so amazing or so special, but because of the competitive environment that the internet creates. If I'm a network that drops a bunch of packets all the time, people are not going to want appear with me. and it'll automatically find the routes that are the best. And so compared to the financial network, like today, if I want to get money from Bitcoin to Lightcoin and someone wants to provide that service, they have to actually have a brand.

I have to kind of trust them a little bit. And I have to figure out that they exist. I have to find them. So you have to do marketing, branding, all these things. And so there are companies out there already that are connectors, there's shapeshift connecting different cryptocurrencies. There's companies that are connecting, let's say, PayPal to M-Pesa, and you can use them.

So they are out there, but they have to have all this overhead of being branded and having customer service and all that kind of stuff. Whereas really, it should be automated and it should be something that is just competitive. So if I can offer a better rate, I should be able to just put that in the system and have traffic go through me immediately. Right. So before you mentioned that this was a big opportunity to become such a connector. I mean, if there's no brand involved, really, isn't this just kind of like a race to bottom and will be done at cost? Or why do you think there's such an opportunity here? So there are in the internet, again, speaking from that model, if you are the best connector, you will grow. And once you have economies of scale, you will be permanently cheaper than your competitor.

So, I mean, it's not the kind of opportunity like being a correspondent bank today where you literally can, you know, have a cost basis of one or two basis points and charge people 20. I don't think that'll be like that because there's too much competition for that. But you can definitely capture a margin of the economies of scale. And so if you get to be very large, you can make a lot of money with that. The other thing I really want to highlight, though, is if you compare this to the development of the Internet and looked at just the scope, the scale of how many, for example, payments are being sent today, we very much expect that the volume of payments sent will explode. In the similar way to how much information was being sent, exploded on the internet, because if you can dramatically reduce the cost of sending an individual

message to the point where you don't really think about it, and then if you have that interoperability so that you also don't have to think about, like if I want to send you money today, we have this whole negotiation about, you know, do you have PayPal? Do you have PayPal? No, you don't have PayPal. Do you accept Bitcoin? No, you don't accept Bitcoin. Then, no. And like, we have this whole manual negotiation, which just means we don't use it a lot. I think the statistic is like people in the U.S. spend 2.1 payments per day. I have no idea how many packets I'm sending across the internet today, because I don't care. It doesn't, you know, I'm thinking on a different scale. I'm thinking about my cell phone's data plan, which is like huge numbers of packets.

And so we fully expect that to happen with payments as well, because if you can automate all of this and then stop thinking about it where, you know, I'm just going to build payments into this other thing that I'm doing where you never would have thought of doing that before because you have to do signups, you have to do all of this kind of overhead. It wouldn't be worth it to build payments into all of these things. But if you get that interoperability right, then you can build payments into a lot more stuff. fully expect the volume to just explode with something like this. Yeah, absolutely.

I totally see that. So you mentioned economies of scale regarding connectors. Is there some kind of centralization risk here that you'd say like, well, that gives undue power to those entities if they become massive? Yeah. So, I mean, we obviously get that question a lot. And so having been in the quote unquote, the centralization industry for almost six years now, It's given me a lot of time to think about that.

And thinking back at what decentralization really means is it means that you have the choice, you have the ability to go with whatever provider gives you the best option. It's almost like more about competitiveness than it is about what the actual physical topology is. So think about something like Bitcoin. Bitcoin isn't centralized just because there is one network. Bitcoin is a decentralized network underneath. And so the reason for that is because anyone can become a minor. And so for something like Interledger, if anyone can become a connector, then even if the connector is very large, it doesn't mean that they have this power over you.

It's not like Facebook where if I leave all my contacts are gone, right? If I switch to a different connector, I just get the new rate and that's it. And so if anyone out there can offer a better rate, they can become the new big connector. that really puts a check on even a monopolist. In competition, it really doesn't matter how many players you have, only how hard it would be for someone else to enter the market and undercut them. It's actually a good thing if you have some very large players that are able to take advantage of economies of scale,

just as long as they don't become able to keep other people out of the market. Yeah, I know that makes sense. Are there some other attack vectors? Like, could they abuse, connectors, abuse, abuse their position in some other way and potentially, you know, steal money from people? I mean, so the protocol kind of keeps you safe from the connector pretty much. Like, as you mentioned, there's escrow. A connector can't really steal anything. There's certainly the ability for people to deos each other. One of the things that's actually kind of interesting about Interledger is that

it's both a way to send payments, but we also can use Interledger to prevent DOS attacks against interledger, right? You can actually make certain requests that are expensive to process cost something. And so that's true for any kind of distributed system that you can use incentives to solve a lot of these type of the tech vectors. So, I mean, I can't predict the future, but certainly if I play it out in my head, I can't think of any way that a very large connector would be able to just outright disable the competition because it is just a protocol because I can set up a path however I want. So as soon as someone misbehaves to a certain extent, it'll be worth it for people to make sure they route around them. I want to just highlight that point that you made, because I think it's a really cool one, that when you talk about denial of surface attacks today,

there's a lot of different ways that people address denial of surface attacks. But fundamentally where the problem arguably comes from is that when you make a request to a server, you're asking it to do some work for you, but you're not giving it anything in exchange. And so the servers have to try really hard to figure out who's doing it legitimately and who's not. But if you could just send these little tiny payments, and this gets back to what I was saying about, I think the volume of payments will explode. It's like if you just made some very, very tiny payment to every server that you talked to in the world, it would be a completely negligible cost to you.

But that would completely cut out denial of service attacks. not only in, so we can talk about that in a general sense. So if I'm operating some service, an API service, I could use Interledger along with that to cut down on denial of service. What Stefan was saying is that we can also use that very feature within Interledger to cut down on denial of service attacks against Interledger itself, because after all, the connectors and things are just internet services that are similar to other ones.

And just to add to that and sort of what's going on for a long time. time, but it's something that's very exciting for us, is a lot of distributed systems have these incentive problems. And so a lot of the people who come to us looking for solutions are people working on mesh networks, on, you know, distributed file systems and that kind of stuff, where they're trying to figure out how to, you know, pay for the utility people add to the system in a decentralized way. And I think Intelleger is a great solution for those types of protocols. And so I'm excited about what interleger can do for the decentralized future, where, you know, I've almost gotten a little bit, I know, disillusioned with this idea of like, okay, we're going to have this, like, decentralized ledger, really huge ledger, but there's still going to be some people who kind of control that. They're kind of setting the rules. They're kind of defining the protocol. But intelligence really lets you get away from that. And so it allows you to build decentralized systems that are, I guess, more decentralized or decentralized in a more real sense, right? Because there's no, not even a group of developers. that kind of controls the whole thing.

Yeah, it's decentralized. I wouldn't call it decentralized, but I would call it distributed and having more for you to choose, I guess, whereas with something like Bitcoin, you're sort of stuck with one system. You mentioned, you know,

economic incentives. You know, coming to sort of, you know, this is, This is very interesting. And theoretically, it's very cool. And, and, and, and, and, and, if, uh, if, you know, it's somewhere down the line, we expect companies to start using interledger and making their ledgers cryptographically, um, uh, so, so inter, integrated cryptographic escrow into those services.

Uh, what is the incentive for a company like PayPal where, you know, their, their interest is to have as many users as possible and to be using their service? If you're creating really easy avenues for interoperability and perhaps using other services, that might be better, what is the incentive there for payment companies to come around this protocol? Yeah, so that's actually kind of an interesting question that I was struggling with as well. And I kind of looked at the history of the Internet for some inspiration. And when you think about the history of especially the web, where you kind of had this transition from the days of the online service providers, the online services networks where you had AOL, you had

CompuServe, Genie, which were all sort of closed networks, and they were competing on the number of services that were on each network, just like PayPal's competing on the number of users and merchants that are on their ledger. And so what happened was that some of the smaller guys, not necessarily the number one, number two, but maybe the number three, four, and five, saw the interoperability that the web provided as a way to expand the services that they could offer. We expand the content that you could get through their network. And so that gives them a way to compete with some of the larger ones on the reach. And eventually it gets to a point where this sort of open protocol has enough content on it,

that if you're not on it, it's a pretty major omission. And so if you're one of the largest guys, you start getting that request from your customers saying, like, why aren't you on that? You know, like, this is the biggest set of content out there. Why can't I access that through AOL? crazy. And so AOL, the way that they responded to that was they eventually added that support. And they thought about it as, well, we're going to connect to that, but we're still going to have all the extra AOL services and people are going to value those extra AOL services, and so they are

going to stay on AOL. But what happened long term was the internet just became like one global market. And so the most successful companies were the ones that stopped worrying about the size of their network and stop worrying about the added value that they could provide and rather just try to be as efficient as possible and try to provide the best customer access to the internet itself. That's really interesting. I love these internet analogies with interludger because I think they really tie into how networks develop. So tell us, where is the project at right now? What have you accomplished? Maybe talk about some of the efforts going on with the W3C to try to bring this to some sort of standard level?

Sure. So we came out with the white paper last October. So that's available on interledger.org. That's kind of the technical defense of it. We've been working on putting out more documentation, things like that. We have the, there's an open source reference implementation that's under heavy development that you can find that on GitHub.gitub.com slash interledger. And then a lot of the kind of core development is going on.

The W3C is one of the Web Standards organizations, and we have a community group there that's focused on Interledger, and so that's where a lot of the discussions and technical development is happening. So if you're interested in tracking the progress, you can find us on, we have on Interlegia.org, it says a bunch of different ways. You can find us. There's a mailing list, IRC, things like, that but the main avenue is to join the community group that's where a lot of the

development is happening and what what kind of remains to be done like what do you see interleger at the year from now two years from now like how long do you think this adoption is going to take and how does it have to happen so I would say that the first thing the most low-level thing that we had to achieve was the creative specification for how ledgers express and validate the The reason why that's the most fundamental thing is because all the ledgers have to validate the escrow in the same way. Going back to that example where I'm a connector, if the ledger on the left of me, let's say Bitcoin, the ledger on the right of me, validate different types of cryptographic conditions, of escrow conditions, then I can't facilitate a payment between them.

And so it's really key that those cryptographic conditions are standardized. Now, obviously, there are cryptographic standards already. So, for example, Bitcoin already implements shot to 56. So one thing that we did was just sort of adopt some of the standards that are already most widely used so that as many existing ledgers as possible, including Bitcoin, most other cryptocurrencies, and Ripple, of course, all already support interledger as is without any changes. The next thing, though, was to sort of plan for the future. And we wanted, for example, to have distributed systems, decentralized systems, be able to sign things that then trigger interledger payments.

And so for that, we needed the ability to express signatures with multiple signers, as standard multisig. And so we kind of came up with respect for that. So that's the first part. The second part was to work on routing. And so the crypto conditions being pretty much final. I think that we're still looking for input, of course. but we have a working spec for that. The second part being routing.

So we need to figure out how do the routing tables at the connectors actually get populated. And so that's the thing that we're most actively working on right now. We just added it to a reference implementation, the way that we think it should work initially. Obviously, there's going to be a lot of innovation over decades, I think, on this. But we have a working sort of model right now. And so you can take the reference implementation today and make payments that use these cryptos, conditions and use that routing. And so the next part for us is kind of to make a lot more easy interfaces and kind of

clients and other things that make it easy for people to use the system, as well as make it easier to run sort of scalable ledgers, scalable connectors and so on. We might do a lot of optimization, kind of trying to make it faster, trying to reduce the latency of payments. But really the two key pieces where the cryptoconditions and the routing, we're just getting to a point now where both of those are starting to stabilize the lot. Yeah, and so the point that it's at is that if people are interested in trying it out, you can go and look at the code, you can download it. There's a demo that kind of shows off a lot of this functionality. That's all working and there. There are some of the internal details that are still being worked on, but as Stefan was explaining before, one of the goals with this project is to make it so that the interface to using interledger can stay the same, and the internals of it can keep getting better and better over time.

So we're trying to make it as easy as possible for developers who, I kind of think of as there's two classes of developers that might be interested in this. One may be really interested in the technical details and understanding the routing protocols and everything about how this works internally. And so for that kind of person I would say to get involved in the community group, to read the white paper, look at some of the routing docs that are going to be coming out shortly, and really dive into that and definitely looking for people who are interested in that. I would say there's another group of people who don't care how all of this works internally.

They basically just want to use payments in their other thing that they're building. And so for that audience, we're trying to have really good documentation and very simple interfaces to use this so that you don't have to understand all of the details of how this works, similar to how I don't understand all of the details about how the internet works, but I can still use it. And even if the internet protocols or the routing protocols and whatnot are being constantly upgraded, I still just have basically the same interface to use the internet. So just to boil that all down for someone who's watching, kind of thinking about,

okay, how do I get involved in this? The best thing right now is to go to Intelligento.org and go to the W3C community group and join that and sort of start participating on the mailing list and the IRC channel. And then we are very aware that our website is severely lacking right now, so we are working on a better one. That kind of brings all the different documentation that we have and all different specs and everything into one place. Right now that stuff is kind of all spread all over the place. So expect a better website in the next couple of weeks, next couple of months. Today's magic word is Bells, B-E-L-L-S.

Head over to lessstock bitcoin.com to sign in, enter the magic word, and claim your part, of the listener award. So just one last question on this thing here. Like, what is the adoption? Is there any adoption happening in terms of like businesses or banks or maybe payments companies or like working with you guys on integrating that? And do you have plans on sort of actively trying to get this adopted? Yeah.

So right now, what we're mostly interested is. in is like people can help us figure out the technology right so individual developers individual contributors who either have some background with designing protocols or simply just have the free time to dedicate to a project like this which is already you know a lot of people aren't that lucky you know so I think the right now what we have is like a number of contributors some coming more from a community background so for example there's that as somebody who who's been working on mesh networks and sort of looking at

Intelligent for that there's someone who works for for well-known a payment startup and he's kind of providing some input on the protocol. He's worked on, I think TLS before. So he has some background in that space on cryptography and so on. There's a scribe, the guys behind Big ChainDB who've been very, very involved in kind of helping us every step of the way. They have one of the most scalable ledgers. And so they have the interest in interoperability because, you know, it's going to make their ledger look really good compared to all the other ones. You know, whoever has the best ledgers has the most interest in interoperability, I think.

And so, yeah, there's been some early contributors. We're definitely eager to grow that now that we've had a little bit more to show. It's always a little bit hard to, you know, advertise too much when you're still just writing the initial specs and still just writing the initial stuff down. But, yeah, we're at the point where we're hoping that more people will get interested. So briefly before we talk about Ripple and wrap up, so Evan, you brought up the use case of micropayments. You mentioned it a few times. like why is that such a good fit with interledger and like why is that going to be so important? Yeah. So right now there's a lot of different people are talking about micro payments.

And one of the, I think the drawbacks with a lot of the, a lot of the non-interledger solutions for micro payments is that in the same way with regular payments, if you tie it to, for example, one currency or one ledger, you have the limitation of you need to get everybody to use your one ledger. And we've already talked about the reasons why we think it's more powerful if you have interoperability between these and it's more interchangeable. And so when talking about micropayment use cases, if you had a system that had all of the benefits that people talk about of micropayments, but made it so that I can accept dollars and you can pay with Bitcoins or whatever combination of those things, that's much more powerful than having to go and sign up for a different thing than what you already have. So that's one of the reasons I think it's a really good use case for it, as well as because interledger, because there's no network and there's no scalability problems to interledger, because it's not a thing in and of itself.

It's just a protocol between different networks. You could use interledger for an unlimited number of micropayments, which is one of the big, big issues to address with micropayments. And obviously, you need the ledgers to support that, but if you have, say, dedicated, micropayment ledgers that are really, really scalable and can handle that kind of volume, then it becomes a non-issue to send lots and lots of micropayments through. Yeah, and I actually add one more thing to that, which is a lot of people talk about including micro payments in other standards, and so that's everything from talking about HTTP micropayments to a project that I've been working on recently as trying to include micropayments in torrents.

And what I would say is that if your micropayment solution is tied to one specific currency or one specific ledger, you're going to run into a problem because you run into that same issue of you come and say, oh, I have the best micropayment solution and you have to use PayPal. And then someone else is like, no, no, no,

I have the best micropayment solution but you have to use Bitcoin. And everybody will end up with lots of different systems like that. Again, you have no interoperability between them and you're kind of sunk because the way that standards work is that in order to build something like payments into other standards, whether it's BitTorrent or HDP or other things, it has to be truly neutral where there's no party that gains from using that. And so you can't tie it

to a specific thing. And so I think it's much more powerful if you're interested in micropayments to think about what you could do with something like interledger where you can express payments in a generic way that can use any underlying ledger to carry them out. So that's why we talk about micro payments. So you guys are doing all this work on interledger, you know, as part of your job at Ripple. Why is Ripple investing so much in this? Why is this such an important project for Ripple? Or Ripple lapse, I should.

Yeah, so Ripple, we are now 125 people, about 4,000. or five of those are working on Interledger and just to put things in perspective. And the reason we do that is because Intellegers used in Ripple's own products. It's kind of like how, you know, a company that's using a database internally might open source that database, the way that we've open source Ripple D. You know, it's kind of something that is very foundational. And we've built a lot of, you know, value add and proprietary stuff on top of that. But with interoperability, it's kind of pointless to keep that in-house.

You know, like what's the point of having HTTP if you're going to make it closed-source? So for us, it's just a way of open-sourcing that protocol. And also, it helps us a little bit in terms of, you know, why should banks upgrade? Why should banks upgrade their infrastructure? And a lot of, especially the largest banks, are pretty comfortable right now because the system sucks for their customers, but their customers have no other alternative. And so the way that I described earlier, you know, Intelliger makes it so that if someone has a better path that even goes around some of the big hubs of liquidity and is a little bit cheaper, Intelligent will be able to take advantage of those paths.

So it basically gives the smaller players a way to make their systems more interoperable and therefore make themselves more efficient as an alternative to some of the big hubs that are out there. And that, of course, makes the big hubs one to upgrade, and that's where Ripple provides products. And so it also creates a little bit more demand for our solutions. I'll also add that in general, Ripple's vision has always been talking about this Internet of Value concept where it's as easy to move money as information and connecting lots of different currencies. And previously the setup has been focused on the Ripple consensus ledger, but we've discussed endlessly now the problems with trying to attract everyone to one ledger. And so the vision of Ripple from the beginning was to create this Internet of value.

And we think this is a really good way of doing this, as well as there's also lots of interesting use cases outside of the ones that Ripple as a company are interested in that are more focused on banks using this. So we want to get lots of other people who have other use cases that they're interested in to use this as well. I want to quickly mention something since you brought up the Ripple Consensus ledger, and I'm sure there's a lot of people out there that sort of have the question, like what becomes of that, what's the future of Ripple Consensus. Sledge what's the future of XRP. So we think that that Ripple consensus ledger is going to continue to be the root ledger for XRP. And we believe that XRP will be an attractive currency to use as a bridge currency. So whenever you're going through a multi-currency transaction, a currency exchange, because it's very inefficient to create markets between all possible pairs that tend to be bridge

currencies that emerge. And so Ripple, you know, we'll try our best to to position XRP as a bridge currency, not the only one, of course, but as one of the better ones. And so from our perspective, it's always been something that, you know, people know about Bitcoin so they use Bitcoin. And so it's been frustrating for us having, you know, let's say, you know, higher transaction processing or lower latency and so on and not having it be used because people weren't aware of it or weren't using it. And so we're hoping that it actually creates a bigger market for XRP because, you know,

if it's the cheapest, if it's the fastest, which we believe it is, it'll have an advantage over other systems that are slower. So you mean because once you have interleger widely adopted, if I want to send US dollars to your account, I'll have to hop through something and then you think because Ripple is going to be easy to ride through anything, then Ripple's ledger and XRP might become one of those routing stations where the actual currency exchange takes place? Yeah, so I said earlier that it was very attractive to, you know, try to scale up your ledger and try to become a very efficient hub for liquidity. And so we certainly want to play that game. We certainly want to compete for that revenue. So whether we'll be successful, I don't know.

there's nothing just to be totally clear, put all the conspiracy theories to rest, like there's nothing in Intellegger that would particularly advantage XRP. But we are thinking that it is the best ledger out there right now. It's the best decentralized ledger. And so as that, it should do really well in the system. Okay, cool. Well, thanks so much, guys. Thanks for coming on.

And it will be very exciting, I think, to see how interledger progresses and how it's going to be adopted, how companies are going to use it. and what the reception will be. Thanks for having us. And yeah, of course, we will have links to the white paper, to some of the presentations they've done, to the working group and all of those in the show notes, so listeners who want to dive deeper or get involved,

you'll know where to go. So we're part of the LTV network. There's lots of shows you can find on let's off Bitcoin.com and we put out new episodes every Monday. You can subscribe it on on your favorite podcast app or watch the videos on YouTube.com slash episode of Bitcoin. And well, thanks so much,

and we look forward to seeing you next week.