{ "episodeId": "SLP651", "speakers": { "stephan": { "name": "Stephan Livera", "role": "host", "tag": "STEPHAN" }, "eric_joschi": { "name": "Eric & Joschi", "role": "guest", "tag": "ERIC" }, "guest_2": { "name": "Guest 2", "role": "guest", "tag": "GUEST" } }, "segments": [ { "speaker": "stephan", "time": "00:12", "start": 11.69, "text": "Hi everyone, and welcome back to Stephan Livera podcast, brought to you by Bold. For American listeners, you can buy Bitcoin over at getbold.io. Now, we're doing an update on Fedimint because there's been some updates on this. So rejoining me on the show, Eric Sirion, he is the, basically, you can call him the The founder of Fedimint. Now, long-time listeners might know we did a, an episode a long time ago, called Mini Mint, kind of a related idea. And also joining us is Joscha, Fedimint core contributor. So guys, maybe just take a minute and, introduce yourselves just for listeners who don't know you and what you're working on. Hey," }, { "speaker": "eric_joschi", "time": "00:50", "start": 49.97, "text": "yeah," }, { "speaker": "stephan", "time": "00:51", "start": 50.57, "text": "thanks for" }, { "speaker": "eric_joschi", "time": "00:51", "start": 51.09, "text": "having me on again, and, yeah, great to see you again. I'm Eric. I have a background in computer science, distributed systems, and cryptography, and in two thousand twenty-one, I started the Fedimint project, the Fedimint open source project. And like previous to that, I had been like just, talking to fellow cyberpunks about this idea for a few years, and yeah, ever since, I've been developing it. with a great team, so one of, my team members is joining me today, Joscha. do you want to introduce yourself?" }, { "speaker": "guest_2", "time": "01:31", "start": 91.49, "text": "yeah, my, GitHub handle is JoshiSun if you're looking for contributions. I joined the project, two years ago. My original background is abstract mathematics. so I have, done a lot of that, I'm in a way our in-house mathematician, which comes in quite handy when using some, Yes, specialized custom mathematics to make the system work well and efficient. for the last two years, I've been mostly focused on, the consensus logic. the first six to eight months was focused on that with the transition to a different," }, { "speaker": "guest_2", "time": "02:16", "start": 136.34, "text": "to a different consensus algorithm called LFBFT, which significantly lowered our latency, especially in challenging networking environments. So we get more reliable low latency in practice. so the, the, the performance of the system stays low even with, stays, Good. We have low latency even with a high number of guardians or with some of the guardians, offline. Yeah." }, { "speaker": "stephan", "time": "02:44", "start": 163.81, "text": "I mean, we should explain some of the, the concepts before we sort of get really into the-- Yeah, yeah. So let's, so let's just do a quick overview just for people who don't really know what eCash is, a-and then we can sort of take it, you know, one step down at a time. So just, you know, quick overview, what is eCash? How is it being used in a Bitcoin context?" }, { "speaker": "eric_joschi", "time": "03:04", "start": 183.87, "text": "So at a very high level, eCash, is an IOU, that you get from, an entity, we, we call it the mint. and it represents, in our case, Bitcoin. Like you give, you give Bitcoin to a mint, it gives eCash in return, and later on you can exchange the eCash for Bitcoin again. that alone doesn't sound that interesting, but the protocol, how it runs ensures that the mint cannot tell, when you redeem your eCash for Bitcoin again who deposited or who created the eCash in the first place? So it gives you very, very good privacy, and that's what, made me interested in eCash in the first place because, Bitcoin's privacy is a challenge And there are many approaches to solving it, and eCash is a very promising one in my opinion. if we want to go into a little bit more detail, we can explain blind signing, but we might not have the time for that today. Yeah. Okay. Well, I think, I think that's-- Yeah." }, { "speaker": "stephan", "time": "04:02", "start": 242.0, "text": "That's, yeah, I think it's good just as an overview for people, and also if you could explain, people might have heard of Cashu versus Fedimint, so can you just quickly explain for people what's the difference there? Yeah, so" }, { "speaker": "eric_joschi", "time": "04:16", "start": 256.47, "text": "the, both systems are eCash systems. Both systems, have Bitcoin backing and issue these, IUs that, people can use to send money to get their Bitcoin back. but the, the main difference is how are these Bitcoin being held? And there, like with Fedimint, we have a federation. It's a distributed system where, multiple people hold the keys to the Bitcoin. It's a multisig. So for example, you can have a three out of four federation or five out of seven, and that means as long as, like the first number, like for example, the five out of seven people are online, the system will keep Keep working. Like two of them can be just offline and it keeps checking along. So that gives you a lot, lot more resiliency against attacks against like just operator error. And also As long as less than five of them conspire against the users, they can't drag pull the, the mint. Whi-while with, system like Cashu, the money is in a centralized place with one operator, and they could just take the money and run away" }, { "speaker": "stephan", "time": "05:26", "start": 326.1, "text": "with it at any time. I see. So in simple terms, Cashu is like a single custodian for your eCash, and you, the end user might have, I don't know, some app on your phone, and you might be tran-trading them around, but you're trusting a single custodian. In the Fedimint case, it's more like there are multiple guardians, and they sort of effectively have a- Some kind of threshold multisig that is underlying, that's holding the actual Bitcoin underlying your eCash tokens." }, { "speaker": "guest_2", "time": "05:55", "start": 355.2, "text": "Yeah, exactly. Yeah." }, { "speaker": "stephan", "time": "05:57", "start": 357.32, "text": "Okay. And then the other one, just to help disambiguate or, you know, stop the confusion, Fedi versus Fedimint, the protocol. Could you just explain, because I know Eric, you're probably involved on both sides there, so maybe you just explain for people what's the difference there?" }, { "speaker": "eric_joschi", "time": "06:13", "start": 372.72, "text": "So Fedimint started as an open source project and, yeah, is being supported by a lot of the open source grant bodies, OpenSats, Spiral, HIF in the past and that's great for building a protocol, but at some point we noticed if we really want to, make the protocol successful, we also need, Like some entity that, can iterate on a great user experience to, and to figure out where's the product market fit and how can we actually apply this protocol to the real world, and that's where Fedimint came into the picture as a company that builds on top of Fedimint and other freedom technologies. And, I co-founded it with, Obinavosu and Justin Moon in, I think, twenty twenty-two. It's been going for three years now. And, yeah, Fedimint is deploying Fedimint in the real world. that's, yeah, great help to the open source project because, we get testing, we get, user feedback, and in return, like Fedimint gets, Like a very reliable custody and transaction protocol on top of Bitcoin." }, { "speaker": "stephan", "time": "07:29", "start": 448.55, "text": "I see. And so, I guess in simple terms, Fedimint is like the, is a, is an example of a commercial, you know, company, for-profit company that is using Fedimint in the background, but it has, you know, an app, and it has, you know, it's building communities and so on. Whereas, as I'm understanding, Fedimint, the protocol is more like, no, this is like the protocol level of how, you know, it actually works works under the hood, or at least that's, let's say, the main component of it." }, { "speaker": "eric_joschi", "time": "07:57", "start": 476.85, "text": "And, also there's, multiple, companies integrating with, Fedimint actually, there's also Bitzacco, which is, Like a community savings, company in Kenya. And, yeah, there's other projects like, Harbor Wallet or, Wipe are also building wallets on top of Fedimint. So Fedimint is, only one of the, like- upstream consumers of our protocol." }, { "speaker": "stephan", "time": "08:25", "start": 504.57, "text": "Got it. And I guess, yeah, just for-- because it might be confusing for people because you're involved on both sides of that, whereas, Joschi, are you just at the protocol side or are you also involved in Fedimint? Yeah, I'm just working on" }, { "speaker": "guest_2", "time": "08:35", "start": 515.0, "text": "Fedimint. I'm Spiral funded. For the time, and yeah, completely focused on that. Excellent." }, { "speaker": "stephan", "time": "08:44", "start": 523.86, "text": "Okay, and so, yeah, I think you were touching on this earlier, Eric, but I guess it would be good just if you guys could give people an overview, how is Fedimint being used so far? Like, what are the big applications, at least so far, a few years in? What, what is it being used for and where is it being used?" }, { "speaker": "eric_joschi", "time": "09:00", "start": 540.25, "text": "Yeah, so, from personal experience, I can say I totally switched from running my own Lightning node to just using one of the public federations as my daily driver and It works for, it has worked for any amount so far that, I wanted to, to pay, and it is very fast. It's way faster than any payment I ever made with my own Lightning node because I just don't have the time to manage that. And so for me, it was a replacement for running my own Lightning node, which is a lot of work. for other people, like especially in the global south, and we are seeing that with Fedimint, it's even more of a necessity, Because they might not even have the ability to own new xOs because the average income is so low that, like for me, the custody isn't a problem. For me, I can use, like hardware signing devices, Or, yeah, hold Bitcoin on chain, but not everyone is in such fortun-fortunate position, and the only other way of holding Bitcoin would be in centralized exchanges basically. And Fedimint gives a way to take back custody into the community, where you actually trust the people that are holding your Bitcoin for you, and then you can build a circular economy, you can, transact and, grow your wealth that way." }, { "speaker": "stephan", "time": "10:23", "start": 623.37, "text": "Yeah, Joscha, anything to add there in terms of how you see it being used?" }, { "speaker": "guest_2", "time": "10:27", "start": 627.16, "text": "Yeah, I had a, I had a similar experience, living in Africa right now, where you have a lot more opportunities to pay with it, especially in South Africa. I used it as a daily driver there myself, to pay for groceries, for example. And at some restaurants they take it. And, yeah, I think here, what you see used here everywhere is Blink. people rarely use on-chain, rarely use self-custody. Now with the fee level, it would be manageable for many, but as soon as you go to a few, dollars per transaction, it is out of the question here for many for the amounts that they use. So, they are completely dependent on- On, custody in a separate country even, not even on, on, on a separate continent, no. And, yeah, we're trying, we're trying to mitigate that. So we are seeing interest here, strong interest from like, Advanced Bitcoin communities, like Bitcoin EKASI for example, they'd wanna take local custody of The funds for their own community, yeah. They have a few people there that are using on chain, but most of them will, will not get there, yeah. So that's what we are building for, for them, the federated custody is a, is a good solution." }, { "speaker": "guest_2", "time": "12:02", "start": 721.7, "text": "yeah, as soon as, as soon as the amounts to the u-- become significant to the user, you would wanna go with a Federated custody solution." }, { "speaker": "stephan", "time": "12:11", "start": 731.1, "text": "I see. Okay, yeah, and so let's talk a little bit about, I guess some of the big updates. I know you have some updates to share. There was a recent, new version. let me just check the number. Oh, seven. Okay, gotcha. Yeah, yeah. So, yeah, if you could tell us a little bit about zero point seven going full stack, it looks like you've got this, new Guardian UI, recurring payment support, and Iroh. So, yeah, I guess we'll go into those. Maybe just give us a high-level overview and, and we can sort of go into specifics." }, { "speaker": "eric_joschi", "time": "12:41", "start": 761.29, "text": "Yeah, I think the most important there is, the Iroh support, and it's worth explaining because pe- most people probably haven't heard of that, yeah, the term. Yeah, exactly." }, { "speaker": "stephan", "time": "12:52", "start": 771.76, "text": "I hadn't heard of this either. So what is Iroh?" }, { "speaker": "eric_joschi", "time": "12:55", "start": 775.14, "text": "so at a very high level, Iroh allows you to run applications, behind, like a firewall or a home router, and have them still be accessible from the outside, like what you'd use Tor hidden services for, but without the performance penalty. So you can have, low latency, high bandwidth connections to servers you run at home on the Raspberry Pi, for example, or StarNine, like that What we are shipping a package for. Maybe, Joscha wants to go into the technical detail a little bit more because he has been driving the integration and, yeah. Yeah. So," }, { "speaker": "guest_2", "time": "13:34", "start": 814.15, "text": "It is generally quite difficult for a server that you are running at home and not in the cloud to accept incoming connections, you know, like a client making requests, even if you, even if you run a Bitcoin Core node at home, as long as you don't have, significant dev upskills to open a port in your firewall, your Bitcoin Core node will be limited in the same way and will only need to connect to other nodes first, but can't accept incoming connections right away. there is," }, { "speaker": "guest_2", "time": "14:10", "start": 849.7, "text": "and there has been for, for decades ways to get around this, but so far there was To, to, but to make it work reliably, it's, it's, it's quite a challenge and outside of the scope of a project even like Fedimint was several, capable engineers So there has not been so far like production ready solution for this and, but now there's one on the horizon. It's implemented in Rust, the same language Fedimint is implemented in, and they are approaching a 1.0. And, yeah, we've been integrating this over the last, six months. And have been, are at the end of like a first manual testing campaign, allowing us to set up a Fedimint in a, in a call where everyone runs the local guardian on their laptop. Before everyone would have to have like a, like a cloud virtual machine to run the Fedimint on, now we just run them on our laptops when we set it up in the call. It makes the whole process much simpler. it was by far the biggest pain point of setting up a Fedimint. and required the most amount of, I would say, specialized expertise setting up the networking. Also, you don't need domains anymore. before you had to configure a domain and you were like reliant on it as well. So, we had this hiccup there about a year ago with a public federation that lost the domain. it actually continued operating, It's normal that some latency issues, but it, it worked fine. That's the upside of being federated, if you lose one domain, it still works. But of course, that's not a, it's, it's not a sustainable practice, like having people to register domains, it's a huge amount of work, it's specialized knowledge, and you also have, the risk of losing access to your domain or to your clients, losing access to your domain and thereby to your clients. So Yeah, we can avoid all that with a library that is called Iroh and, yeah, be able to run, run on local machines such as laptops or start nines. making this a true one-click deployment." }, { "speaker": "stephan", "time": "16:33", "start": 993.14, "text": "I see. Yeah. So this reminds me of how, you know, when people were setting up their Lightning nodes and they had, they needed a way to, quote-unquote, phone home to their node and use their, like, to use their, phone app as like a remote control. So they were using things like Tor. I think after a while, people using like WireGuard. Nowadays, there are a lot of people using, Tailscale, which is like another way to sort of-- But I understand that So can you explain a little bit about Iroh? Why, why and how does it work in a different way to the other ideas?" }, { "speaker": "eric_joschi", "time": "17:08", "start": 1028.32, "text": "Yeah, so the, the special thing about, Iroh is compared to a lot of the other solutions that, once the connection is established, it's a direct connection between the two devices that are communicating. Like a lot of other solutions, like for example also, not the web services, as far as I'm aware. Always have a relay in the middle, like there's, some server that, acts as like the public, message broker that, like communicates with the two parties and just relays all the messages. While with Iroh, you have something like that at the beginning, like you have an introduction point of sorts, but that only helps the two parties to establish a direct connection. And so once that happens, then you don't have a man in the middle anymore, like you, you don't incur the cost of that because Like otherwise you have the central server that, has to, pay for all the traffic that goes through it, and that's not scalable. I think from what you" }, { "speaker": "guest_2", "time": "18:04", "start": 1083.81, "text": "mentioned, it's the most similar to Tailscale, they were, I was heavily inspired by Tailscale. the problem though with Tailscale, it's, they, they don't, they don't provide you with a library on the right level of abstraction. So like integrating this in a custom protocol, is, is- is, is not possible. Yeah, you could build Tailscale with Iroh quite easily, but not the other way around. Iroh is on a, on a lower, lower level, it's a, it's a library approach, it's more customizable while, I think Tailscale provides you with a ready-made daemon that you're just starting to connect your machine. yeah, and that's, that's not quite the right" }, { "speaker": "stephan", "time": "18:49", "start": 1129.02, "text": "fit for us," }, { "speaker": "guest_2", "time": "18:50", "start": 1130.02, "text": "yeah." }, { "speaker": "stephan", "time": "18:51", "start": 1130.84, "text": "Back to the show in a moment. This show brought to you by CoinKite dot com, the creators of the best Bitcoin hardware security devices such as the Coldcard Mark IV and the new Coldcard Q. Now, we use Bitcoin hardware security devices to keep our keys offline, our private keys offline. Now, the way these work is you can do that setup, write down your twelve or twenty-four words on those, the seed word cards, and keep that secure. Now, you can use this device to interact with the Bitcoin network work using software such as Sparrow Wallet, Electrum, or Vector Desktop or Nunchuk as a few examples. Now, you have a range of security features that you can use with these devices such as passphrase, you can use seed x or my favorite is multi-signature. Now, if you're starting in a basic way, just start with the device and the USB-C cable, plug it directly to the computer and use it that way, and then later improve your setup. But I believe these devices are great at helping secure your coins. Especially as you start to migrate up into multisig security. But don't be disheartened or don't be, scared away. They are accessible, and I think you actually do learn about Bitcoin in the process. So to get yours, go to coinkite dot com, use code livera to get a discount on your cold card. This episode brought to you by Galloy. They are building banking software for the Bitcoin age. So if you are with a bank, a fintech, or a startup looking to offer some kind of Bitcoin product, whether that is a Bitcoin collateralized loan Deposit accounts or payments, Galloy can help you. Their latest product is called Lana. It is a loans management platform, and you can use this to come to market quickly and offer a loans or Bitcoin collateralized lending product for your customers. Now, Galloy have a lot of experience in the space. They started with Blink Wallet in twenty twenty, and they've since grown this to become a community favorite over time, and so they have a lot of experience making things work in a secure, reliable, and scalable way. So if you need assistance coming to market Market quickly with a Bitcoin banking product such as lending or deposits or payments. Talk to the team at Galloy, you can email them, the email is biz at galloy dot io, or go to the website galloy dot io. And now back to the show. And so, as I'm understanding, and, you know, the my updating, my, update me, 'cause my understanding may be wrong here, but the idea with Fedimint is there are, you know, these guardians, that's the concept, and they, they are the The, let's say, the multisig holders, and then, you know, with that there might be a lightning gateway that is how it, you know, does the lightning payments aspect of it. And so what we're talking about here with the Iro part is you might be an end user of that Fedimint, like just an everyday person with your phone, and so Iro is how you can, let's say, connect with your phone or your end user wallet or app to the Fedimint, or is-- and it's also, I guess, as I'm understanding The way that the guardians talk to each other as well, right? Because you're talking about that guardian setup process, right? Exactly," }, { "speaker": "guest_2", "time": "22:02", "start": 1322.18, "text": "exactly. So the peer-to-peer, interactivity goes over Iroh, and the client-to-federation interactivity also goes o-over Iroh. Yeah. So the, every guardian has, has a public key, and that is all you need to connect to it, and those public keys are obviously static. So you can even move one guardian from one machine to a different machine and all your tra-network traffic will automatically be redirected. Yeah." }, { "speaker": "stephan", "time": "22:37", "start": 1356.88, "text": "Wow, it sounds kind of like a magic. I mean, how does it actually, how does it know where to find, you know, how does it know where to find? Well," }, { "speaker": "guest_2", "time": "22:43", "start": 1363.33, "text": "they have different, different discovery mechanisms. So one of them, is the, mainline DHT used for BitTorrent. So if you, yeah, if you spin up an Iroh node it will publish, this information, how to connect to it, to the DHT and further like DNS servers, et cetera, et cetera, to speed up the process and all those, the, all these discovery mechanisms, they run in parallel and whatever is the fastest will be used, yeah." }, { "speaker": "stephan", "time": "23:15", "start": 1394.52, "text": "Interesting. Okay, so that sounds similar to like even what the, the Keat g-gu-hole punch guys are doing as well. They're doing something sort of similar with the whole bunching" }, { "speaker": "guest_2", "time": "23:23", "start": 1403.23, "text": "is one aspect to it, yeah. one of the techniques used to establish a connection is hole punching. I see." }, { "speaker": "stephan", "time": "23:31", "start": 1410.88, "text": "Okay, and so as I'm understanding then, it's like Iroh is, you've, you've built it into this Fedimint software so that it's kind of already inbuilt and it does it, you know, automatically in the background." }, { "speaker": "guest_2", "time": "23:44", "start": 1424.36, "text": "So in, in, in practice, when you're deploying an, a Fedimint Guardian based on Iroh, the only thing that you have to specify is the, the directory to store your data in and the Bitcoin backend you wanna run against. everything else will be completely taken care of, like the, the whole concept of networking is completely automated a-and, run in the background. You as a, as a user don't have don't even have to have a concept, or as an operator, don't even have to have a concept of how it works." }, { "speaker": "stephan", "time": "24:18", "start": 1458.04, "text": "I see. So I guess what you're saying then is it's, it's like, it's a big deal in terms of making it easy to set up a Fedimint, right? Because I guess historically, like you were saying, the, the, those, let's say it was five guardians or whatever, those five community leaders or whoever, they needed a certain level of technical competence, technical ability, and as you said, a public IP and A guardian in this Fedimint, whereas now it's more like actually get a Startnine and sort of install the app, and that's gonna be the setup relatively easy. Yeah," }, { "speaker": "guest_2", "time": "24:54", "start": 1494.34, "text": "the Startnine, the Startnine is smooth, smooth as the last two points because you already have a Bitcoin Core node running, the environment is always the same, so we can also mat- automatically point a Fedimint D on a Startnine to the local Bitcoin Core node, and the data directory to store will also be handled by Startnine, so it's The, if, if you have a start nine, you have a Bitcoin Core node running, you, it's a one-click install of Fedimint D, and, after then, you, you just have to, you need to exchange, connection code, so every, every, guardian will get a, a string, a short text to exchange with the other guardians, and that's all the machines need to connect to each other, run the distributed key generation process. once, and then the system will start up automatically and is ready to use, yeah. Okay." }, { "speaker": "eric_joschi", "time": "25:49", "start": 1548.59, "text": "One, one important, caveat though that, we should be giving here, we are talking about Starlight already, but Iroh itself is still waiting the one dot O release, and they still want to make one breaking change, like one last, protocol cleanup on their side. before they commit to it long term. And so we right now, don't, yeah, I want to encourage people too much to run this in production yet. so we already have, Like, some subline images built, but it's only for MutinyNet, because we know that eventually the, the Iro compatibility will be broken, but When do we expect the 1.0? I think it was later this year." }, { "speaker": "guest_2", "time": "26:39", "start": 1598.86, "text": "yes. So, the last update from, the Iroh folks was they expect to do a 1.0 in September, and we try to release, a Fedimint D As soon after that as possible, so we are doing all the work in advance. We have, we just had the 0.7 release, alongside that, we released, an we released a start nine package, and we will further do like testing of this, et cetera, in the, over the last three weeks, we maybe set up fifty to a hundred federations over Iroh, and it's been quite promising. So we will do more manual testing over the next month, so such that when Iroh dot seven dot o release, for us, it's, just a version bump and another release of our own software, and then, you know, integration into Fedimint and the other clients, et cetera. which they get it for free, they just then need to bump to the oh eight or oh nine version of, of, Fedimint D and our, of, of the Fedimint client on, and our integrators won't be even, they won't be able to tell if the connection that their client makes goes over Iroh or, the current Web Socket protocol, which is a nice, nice thing about having a client like ours, a client library that we provide, we can release A feature like that without our integrators having to do anything besides bumping the version and doing the manual testing that they do on every release. yeah, so we hope once Iroha does a, once Iroha does a 1.0, we hope in, in ten to release this quite quickly and smoothly. Yeah, and then we will also be releasing a StarNine MainNet package." }, { "speaker": "guest_2", "time": "28:32", "start": 1712.39, "text": "yeah." }, { "speaker": "stephan", "time": "28:34", "start": 1713.67, "text": "Okay, great. And so, that was gonna be one of my next questions, which is, kind of client agnosticism, but I guess you kind of answered that because it's like Fedimint D is the underlying- Damon and clients, it's, it's, you know, it's kind of agnostic to what client actually connects to that, right? So" }, { "speaker": "guest_2", "time": "28:52", "start": 1731.8, "text": "we, I think there's a technical di-difference to Cashu, maybe they do these days, but when they started out, they just-- the reference implementation was only, the mint itself, not the client. We always also provided, A, a client library. so the client that we provide already has interfaces for pay Lightning invoice, send eCash, receive e, eCash, create Lightning invoice, so, so quite a very high level. So the only thing left for our integrators is," }, { "speaker": "guest_2", "time": "29:29", "start": 1769.2, "text": "like scanning the QR code, turning it into text, and then calling the method on our client to, for example, pay the invoice and all the, the graphic design around it. So our integrators are a relatively thin layer Over our quite complex Rust client that we ship. So once you have done this work, once you have integrated our, our Rust library, what goes inside of that Rust library you, you can't see and you don't really care either. Yeah." }, { "speaker": "eric_joschi", "time": "30:01", "start": 1800.51, "text": "And there's, a really good reason for that, like Fedimint never really had a protocol definition as, eCash does, but It also started out as, way more of a research project, like, there were so many, like problems to solve and so many things to figure out and we needed that flexibility of being able to iterate on the protocol level, and by providing both the client and the server, we can ensure that, like all our clients are compatible with all the server versions from half a year back and half a year in the future, like we have some strong, stability guarantees there that allow, like, a current Fedimint client to communicate with a pretty old Fedimint D, federation or Fedimint D version running in the federation. And like by keeping both the client and the server in the same hands basically, we can actually pull it off. Like it would be way harder if we, had like a protocol and multiple client implementations, then it's much easier to get stuck with, less than optimal" }, { "speaker": "stephan", "time": "31:06", "start": 1866.03, "text": "Right, right. You might, you might have run into some kind of cross compatibility issue with like, I don't know, some older, whatever, o-others thing. Gotcha. okay. And then, also in the release notes I saw LNURL and possibly Bolt twelve in the future. So can you guys explain what's, what's going on there?" }, { "speaker": "eric_joschi", "time": "31:25", "start": 1884.54, "text": "Yeah, that's, the, the thing I've been working on for the last, quarter or so. it was a much requested feature, like Fedimint should be able, or Fedimint users should be able to generate LN URLs, which then also allows them to generate LN addresses And receive, recurring payments without regenerating invoices all the time, in a non-interactive way for the user. And it was quite a challenge to Like make it work with Fedimint protocol, like what we implemented is, I think, the, the best, you can do, like it's not fully trustless, you will have like one trusted entity, but it's pretty much the same, like for any other LNURL server, like you trust the server that gives you the, the final, lightning invoice And most people use it in a custodial way anyway. So, but then once the payment is made, the funds are in the federated trust model, the user will be able to claim them whenever they come online again. And from the UX perspective, it's a huge unlock. You see that in Africa quite a bit when there are shop owners accepting Bitcoin, they aren't fiddling around with their phone. They have a static QR code printed next, to their shop, and they just tell the, the customer, \"Hey, can you send me, the two hundred Ken Links here to that Lightning address instead of me generating the QR code and like maybe handling, some food on the side. Yeah. Much more clunky experience." }, { "speaker": "stephan", "time": "32:54", "start": 1974.5, "text": "And I think that, I mean, you guys might know more than me, but from what I've heard in China with like WePay and Alipay and things like this, people just use QR codes in the shops as well. It's like a similar thing where they don't necessarily have to ring it all up the same way, they just sort of scan and pay, right?" }, { "speaker": "eric_joschi", "time": "33:10", "start": 1990.05, "text": "Yeah, exactly. And pretty much the same, with, Bolt 12, like it's also about recurring payments, and that's something I want to, build next, like using the same infrastructure basically, just Bolt 12 is a bit more of a sophisticated protocol, it's not just spinning up a web server and it, gives you some lightning invoices, but rather you have to deal with, like lightning onion messages, and it's a bit more complicated to integrate, but, yeah, it's definitely the more pure approach from a Bitcoin pers-perspective, so I want to also see it supported." }, { "speaker": "stephan", "time": "33:49", "start": 2029.04, "text": "Yeah, interesting, and I think that is really cool from, implications perspective because it might mean contact apps or contact lists inside our Lightning apps become much more feasible, much easier to do, because now there's just a Lightning address or in the future a Bolt 12, code for all your contacts. And so then when you're regularly paying back and forth, it's just a lot easier because as probably any, you know, I'm sure you guys see this as well. Well, we do sometimes when we're doing payments, we sort of do this, I call it the dance. We do a, we do a dance of like, \"Oh, hey, do you want me to pay you on chain or lightning? \" \"Oh, okay, now invoice me this amount, \" and then, \"Oh, wait, no, maybe the other guy is in another time zone and he was asleep when you sent the invoice through and it's expired, \" or, you know, \"Or it's the wrong amount, \" or \"Maybe they sent an invoice, sorry, an amount and of course, things like BOLT twelve and LN address help with that, right?" }, { "speaker": "eric_joschi", "time": "34:47", "start": 2086.8, "text": "Yeah, and, BIP three five three, which, kind of combines them all into, one human readable name standard" }, { "speaker": "stephan", "time": "34:55", "start": 2095.29, "text": "Yeah, so hopefully we can, have less of this kind of, invoice dance back and forward or whatever. but actually, I'm also curious as well, how does that work from an eCash context, right? Because we're talking about eCash and Lightning. Is it just that you're invoicing for a certain amount of satoshis and you have to pay whatever amount of eCash tokens is equivalent to that, or how, how do you-- Is there a translation there or how, how do you, how does that part work?" }, { "speaker": "guest_2", "time": "35:22", "start": 2122.3, "text": "Yeah, the translation Creation, for us is, is handled by an external service called the gateway. So, we have, in Fedimint we have two types of smart contracts for outgoing lightning payments and incoming lightning payments. That is a smart contract that the user and the gateway service," }, { "speaker": "guest_2", "time": "35:45", "start": 2144.63, "text": "they go and like pay into this contract together. depending on sorts of an outgoing payment, the user funds the contract, for an incoming payment, the gateway funds the contract. and this is in a, in a trustless way, in a sense that both parties just trust the federation to correctly adjudicate this contract, but they don't trust each other." }, { "speaker": "guest_2", "time": "36:08", "start": 2167.92, "text": "this It gives us an additional benefit that the, that the federation operators don't need to run their own gateway, and the same gateway can be shared by multiple federations. So the, the cost of operating a gateway, the technical knowledge, the time, the channels can be amortized between multiple federations that, again, each of them have multiple custodians. so there's a, a, another," }, { "speaker": "guest_2", "time": "36:39", "start": 2199.21, "text": "another, a level of interaction there. Yeah, yeah, interesting." }, { "speaker": "stephan", "time": "36:43", "start": 2202.61, "text": "And so as an example Let's say we're, we're all users on this federation, and, you know, you issue a, a lightning invoice out of this federation using, you know, this LNURL, then The invoice obviously is denominated in sats, right? And so then, whoever's paying that, the outcome of that then it becomes like, let's say my balance as a user on that Fedimint. That's kind of The, the net result of it, right? And then, let's say I have an eCash balance, let's say I'm a user on this Fedimint, I have an eCash balance, and I want to pay out, you know, whatever, ten thousand sats or whatever. That again is handled by this, I guess, the combination of the gateway and the guardian sort of saying, \"Oh, Stephan's balance is, you know, enough to pay this ten thousand sat invoice, take away his eCash or take away that eCash balance and make the lightning payment, and we're all good.\"" }, { "speaker": "guest_2", "time": "37:43", "start": 2263.0, "text": "Yeah. So on an outgoing payment, the smart contract we have is just an HTLC, so a hash time lock contract similar to Lightning, where, it is funded by the user with eCash, and then the user, pings the gateway and says, \"Hey, I have funded this smart contract, I wanna have this invoice paid. I have locked this, this eCash to the public-- sorry, to the, to the pre-image of the hash that is in the invoice. So if you pay this invoice for me, you get the pre-image, you can get the eCash I locked in this smart contract.\" Yeah, that is, that is the flaw." }, { "speaker": "stephan", "time": "38:27", "start": 2306.65, "text": "okay. And so then the, and what you're referring to there is the lightning gateway, because that lightning gateway guy, he has to hold a bit of a balance both in lightning and eCash, right? Because he has to obviously place some trust in the federation and the guardians to not rug him, obviously, that's part of the model. But obviously, there's also a profit model for him too, right? He's making money, maybe he's charging a percent on the transactions, right? That's, is that broadly- How it's working there?" }, { "speaker": "guest_2", "time": "38:53", "start": 2332.89, "text": "Yeah, that's, that's how it's, how it would be working out. And, he has an a-additional advantage to charge there because, as you said, he's trusting the federation, so he would think that the, the set of operators is, is, limited. It's not like a, lightning routing node where only the largest are profitable, but like a smaller gateway, that is just connected to like one local federation, could also be quite profitable because it would be able to charge maybe point one, point two percent on a payment, yeah." }, { "speaker": "eric_joschi", "time": "39:30", "start": 2370.22, "text": "Yeah. And the additional benefit, for serving a federation versus just being a large lightning routing node is that For, for locking like a certain amount of, Bitcoin and eCash, which is equivalent to like one Lightning channel otherwise, you get access to a way bigger user base. Like if one federation, for example, has a thousand users, then you amortize that locked capital over all these users. And if you're even luckier, then the users' net transactions balance out. Like there will be some users that spend more, then there will be some users that receive more, and then you- You might not even have to go on chain to rebalance your channels or eCash, quite that often." }, { "speaker": "stephan", "time": "40:12", "start": 2411.54, "text": "Gotcha. I guess it just kinda depends on the flows in and out and that kind of thing. and then With Fedimint, is it that every Fedimint can only have one Lightning gateway, like I-- or can you have multiple? Or how does that work? You can" }, { "speaker": "guest_2", "time": "40:24", "start": 2424.43, "text": "have, you can have multiple and, the client will, just choose randomly, whichever is online so they serve as fallback. If one of the gateways goes down, the clients will immediately use the-- or will automatically use the other ones to route. Yeah, I see." }, { "speaker": "stephan", "time": "40:42", "start": 2442.36, "text": "And when you say client there, are you referring to like the end user on his phone, or are we talking about the guardians there? Or the guardians of the serv-- guardians of the server in this context, right?" }, { "speaker": "guest_2", "time": "40:53", "start": 2452.84, "text": "Yeah, the guardians are the server in this contract, context, but, the way Fedimint works, this choice will be made by the client. Yeah. We push a lot of complexity in the clients, the servers are relatively dumb, since they are also completely automated, choices like that are made by the client. Yeah." }, { "speaker": "stephan", "time": "41:13", "start": 2473.04, "text": "I see. So choices that the guardians would make would be more like which Lightning gateways we want to support or what?" }, { "speaker": "guest_2", "time": "41:22", "start": 2481.58, "text": "So, one of the few things that you can do in the, admin guardian UI is configure a list of gateways. So, to make it even easier to set up and also for small deployments, we are running our own gateways. so there is a, small message there that you can Reach out to us on Discord and we will, get you started with one of our own gateways. And then, you know, once you get bigger or for some reason you wanna be running your own gateway, you can, add this in the UI. gateways do have com- do have domains, so you just add the, add the domain there. They will, will be running side by side for a while if there are any issues with your own gateway, if it goes down, for example, users, clients will- Automatically fall back, to, to the other gateway in that, in their case, ours. And, yeah, once you, once you grow bigger, you could add a third gateway, and maybe then at this point you would be dropping, you would be dropping on, the Fedimint gateway that we are running." }, { "speaker": "eric_joschi", "time": "42:36", "start": 2555.7, "text": "you made a really important point there that, basically for running a Fedimint guardian, a Fedimint guardian, the only thing you have to do is set it up initially, maybe configure it, and then you just keep it plugged in, give it internet, give it power, and that's all you really have to do. Like all the complexities of managing, for example, the lightning node, you can offload that to someone specialized in that. so you kinda, remove the, like you, you split the, the responsibility for the money from the responsibility on the technical side, and it's much easier, I think, that way because finding people that are both trustworthy and technically competent and willing to, like, operate such infrastructure for their community, it's much harder, ask than if you only have to be trustworthy and willing" }, { "speaker": "stephan", "time": "43:28", "start": 2608.18, "text": "Interesting. Yeah, 'cause that's actually what's gonna be kind of related to my next question, which is, how professional do you have to be to be involved in all this? And I guess you're touching on some of this, right? The, let's say, the end user just, just basically needs to be someone who can use an app on a phone. The guardian, as you're saying, could be someone who can just run a start nine, as an example, and plug into the internet, keep it online, some of those things. and then you're The highest level of technical competence and service provision, because they need to, obviously, they need a dom-domain name, they need to be online, they have liveness requirement, they have to trust, the federation as well with an eCash balance, so I guess a-as, and I guess maybe that's a bit of an open comment, I'm curious to see what you guys think. How big do you see this getting? Do you see this being like, there could be kind of like the garage federation, you know, kind of just a small, and then as you professionalize and Further up, how, how big do you see it getting?" }, { "speaker": "guest_2", "time": "44:33", "start": 2672.88, "text": "Well, I think in a way it's kind of the," }, { "speaker": "guest_2", "time": "44:39", "start": 2678.69, "text": "it's, it's, it's kind of the, the second wave of Bitcoin adoption because you need a, you need a certain density of Bitcoin use for it to actually make sense. something like, like a, like you would need a, like a circular economy there where it would actually make sense. Well, like Maybe, a few people already have start nines with Bitcoin Core nodes on them, and they know enough people that actually require their servers." }, { "speaker": "guest_2", "time": "45:09", "start": 2709.35, "text": "Yeah, so maybe a community would start out on, on Blink, for example, when it's only a handful of people, but once, more business is driven, over Bitcoin, then they would go out of their way and like Buy some start nines and deploy a Fedimint, yeah. So it, it has this initial setup co-cost. so I think it needs a, A certain amount of users to be worthwhile, but then it is quite easy to provide a service that is, comparable to something that is professionally managed, even though you're, you're barely doing anything that, then the quality of your service largely depends on your lightning operator or on the, on the lightning gateway operator. Yeah," }, { "speaker": "eric_joschi", "time": "45:58", "start": 2758.45, "text": "and, I think in the end what we'll be seeing is, pretty much a power law distribution of federation sizes. There will be some really big public ones, most likely, and the limit there will probably be the, just the technological limits in Fedimint, because Fedimint isn't built to be, like, the one, A big federation to rule them all. So it's really meant for there to be many federations, and that's, like a core difference to Liquid, for example, which was built as like this really super robust, reliable, Like protocol, but that can really only be run by this one set of people, or like if you'd want to run a second, liquid network, you'd need to, invest enormous resources to set it up with Fedimint, it's more targeted at like the medium, small size community, and I hope that we'll see many federations in the range like ten thousand users, maybe hundred thousand Like that's where it's still, we still have some connection to the people actually running it. It's like a small town basically." }, { "speaker": "stephan", "time": "47:03", "start": 2823.2, "text": "Do you have any ideas on-- I mean, I don't know if you could say, but do you have any ideas on how many people are using this today or what is, what is the size of, let's say, the biggest Fedimint today in operation?" }, { "speaker": "eric_joschi", "time": "47:14", "start": 2834.43, "text": "So it's really hard to tell, how many users there are because of the great privacy properties. What we can see is the number of transactions. And like the bigger federations, when you look at, actual transaction numbers, like there's some federations that seem to have, like some test transactions going on, so it's a little bit skewed there, but, it's on the order of magnitude like a hundred transactions a day maybe," }, { "speaker": "guest_2", "time": "47:40", "start": 2860.34, "text": "Yeah. Personally, I gotta say the, the federations that I use are all private, so I don't use the public ones at all. I only use like primarily, well, two private federations that I don't wanna mention who's running them here, but it's like exactly that, it's like people that I know, that are running them. And, yeah, I don't, I don't use public ones at all. For me, that is kind of, not the point if I don't know who's running them." }, { "speaker": "stephan", "time": "48:11", "start": 2890.83, "text": "Yeah. Interesting. So I guess there's different types of Fedimint and federations, as you said, one, the probably a common or at least ideal, i, you know, ideal case might be, okay, a community people who live near each other, so they know who the guardians are. But I guess what you're saying is it, it can also be, you know, people around the world who still know each other, but they wanna get in on a federation together. For, I guess, I, I guess this would also be interesting for listeners, like, why would you use that? Is it for the privacy reason? Is it for the convenience reason? Like, what are the main, drivers for you there?" }, { "speaker": "eric_joschi", "time": "48:46", "start": 2926.34, "text": "I think it's a lot about the convenience and at the same time trust minimization, because I can have convenience with wallet, wallet of Satoshi, but then I trust like some random dudes I don't know, in exactly one jurisdiction that might become hostile to what they're doing, while the federation, like the like a month ago, I think we tried out running, like one federation setup running on our laptops, and we were on two different continents, three different countries, and we still had, reliable, yeah, service and It just worked. Like, yeah, you were on a mobile" }, { "speaker": "guest_2", "time": "49:24", "start": 2963.58, "text": "connection in a hotel, if I remember correctly." }, { "speaker": "eric_joschi", "time": "49:27", "start": 2966.65, "text": "Yeah, yeah, exactly. And someone was on Starlink, it was between the US and Africa. Yeah, it's, it's amazing that you can have a distributed system now that is so georedundant that even if something happens to, like, one country, it becomes unstable for some reason, the operator has to flee and can't continue running that, node anymore, they can just settle somewhere else, migrate there, and in the meantime, everything keeps running, you wouldn't even notice. Right. That's the beauty of federation." }, { "speaker": "stephan", "time": "49:57", "start": 2996.51, "text": "That guardian in that particular jurisdiction to a guardian in some other country and do it that way, I see." }, { "speaker": "guest_2", "time": "50:02", "start": 3002.48, "text": "Yeah, even just Practically, if you have a StarNine at home and you happen to move, it isn't a problem. You just unplug it, move to, move it to your new home, plug in your internet connection there, it will reconnect, the users won't even be able to tell." }, { "speaker": "stephan", "time": "50:19", "start": 3019.33, "text": "The lead sponsor of this show is Bold, the best place to buy, sell, and save Bitcoin. For listeners in the US, Bold lets you secure your financial future with complete peace of mind by integrating a low fee Bitcoin only brokerage with next gen multisig vaults. With Bold, you can smash buy Bitcoin or set a DCA plan for only 0.99% fees and seamlessly deposit the Bitcoin direct to your Bold Vault. The Bold Vault is a 2 of 3 collaborative multisig where you hold 2 keys and Bold holds 1, as a Standard backup protecting against loss or theft. You can use Trezor, Ledger or Coldcard hardware wallets to spin up a Bold Vault in just a few minutes, and the Bold Vault is the only collaborative custody vault available with zero monthly fees. They're also offering zero fees on your first ten thousand dollars of Bitcoin buys and twenty five dollars of free Bitcoin when you buy a hundred dollars of Bitcoin or more. Try Bold today and upgrade your stacking experience over at getbold dot io. And now back to the show. I see. And so let's talk a little bit about the, I guess, comparative analysis, right? Because this is bringing up the question. So I guess for users and even for builders, they might be interested to understand, okay, what are, what are the different alternatives, right? So on one level, okay, go use a custodial, you know, think like Blink or Wallet of Satoshi. On another level, you could just use, I guess, like a Phoenix or a Zeus or something like this, where it's self-custodial, but you're paying yourself. on another Liquid, right? So like Aqua or, you know, some of these things that are kind of liquid-based wallets, let's say, that use kind of swaps in the background. and then of course there's Cashu as well, it's like the Cashu style stuff as well. So I guess as you sort of look out at the different- You know, comparative, possibilities. how, how, where do you see Fedimint competing against those?" }, { "speaker": "eric_joschi", "time": "52:11", "start": 3131.49, "text": "Yeah, I think, first, we can talk about self-custodial lightning because in, in my opinion, that is the, like gold standard still. It's just, if you actually run it yourself, it's a lot of work for managing, and if you use something like, Phoenix or like other LSP based implementations, then you still trust your LSP to provide you with service. Like they could stop providing you with service, service, the only guarantee you have is that you will get your Bitcoin back And, you're still paying for all the on-chain fees when managing your channel when you have to grow it or shrink it, and that's also a bit of a UX limitation because you suddenly have to, if, if I wanted to onboard my mom I had to explain to her, okay, I can send you money for like a hundred bucks, and then eventually the app will tell you, \"Hey, to receive more money, you will have to pay one fifty,\" and it's pretty random at that point. And, yeah, that's, UX that, you don't have with, custodial systems like Catcher, like Fedimint, like Liquid. Where in the end, the, someone else is managing the liquidity for you." }, { "speaker": "guest_2", "time": "53:26", "start": 3205.97, "text": "one important driver I think of the use case is also, on-chain fee. these days it's negligible, but when we released the system in like early 2023, I think, no, 2024, we, we at the first release, on-chain fees were around like twenty dollars And they stayed there for like, I think the next five, six months, so we could already see it, s- with, people that were doing onboarding in the global south that they were running into issues because they had onboarded people onto, non-Custodial solutions and suddenly it was forty bucks to close a Lightning channel. so that was completely unworkable, they had to fall back to custodial solutions, and," }, { "speaker": "guest_2", "time": "54:20", "start": 3259.98, "text": "you would arrive at a situation where, you simply need to custody a signif-- to you significant amount of, Bitcoin with someone else because it would be uneconomical to move it on chain. So, Probably you would need to custody up to like a hundred times the on-chain costs to, for, for, for it to be still viable, because you know, you need to withdraw and then, deposit it again, and the two percent losses Quite significant, even in terms of the current financial system. So at a, ten dollar fee for a transaction or twenty dollar fee for a transaction that is a thousand dollars, two thousand dollars that you wanna custody, yeah? And for me, that comes to the point that I'm like most to, to, to the border of me being comfortable to hold this with like a centralized custodian. So maybe like a thousand dollars where it's like, okay, if I lose it, it's obviously not great and my day is ruined, but I can stomach it. Everything beyond that, you would, want to, you, you, you need a, like a, you, you wanna have a more secure custodian that you can still withdraw with Lightning in. And there, a Fedimint is probably your only option because you can withdraw, s- withdraw into a multisig, over Lightning, yeah. And for, someone in the global south, this threshold is much, Quicker reached and a thousand dollars, may very well be like a year's worth of savings." }, { "speaker": "stephan", "time": "56:00", "start": 3360.31, "text": "I see. So I guess what I'm thinking then And like not attacking you guys at all, but just a, as a question that people might be thinking, okay, let's say you're in the global south, what's stopping that person just using Blink or Wallet of Satoshi versus like a Fedimint style solution?" }, { "speaker": "guest_2", "time": "56:17", "start": 3377.47, "text": "Yeah, well, with Blink, you're trusting your custody to, someone on a different continent even, so you have no recourse whatsoever on the custody there. And," }, { "speaker": "eric_joschi", "time": "56:33", "start": 3392.66, "text": "S-same for wallet of Satoshi, it's just, I think Australia and, yeah, actually, it's a Salvador-based company." }, { "speaker": "stephan", "time": "56:39", "start": 3398.7, "text": "Yeah, the wallet of Satoshi founder as well. and, but, yeah, go on." }, { "speaker": "eric_joschi", "time": "56:44", "start": 3404.04, "text": "Yeah, I'm not even saying that, the people running that have any bad, bad intentions, that's not at all the point, but also the jurisdictions they're operating in, they might just become, not as friendly to what they are doing. And we, we saw that when, for example, Phoenix World had to withdraw from the US, then suddenly a lot of people had to get their money out, were left without a way of paying in Bitcoin. And in the US, it's not a big deal, because realistically, in the Western world, we don't really need Bitcoin. It's more of a like nerdy thing or like a libertarian ideological thing to, to use. But in some places, like, circular economies in Africa, just like your financial infrastructure being wrecked, pulled From under you, could actually be catastrophic for people." }, { "speaker": "guest_2", "time": "57:32", "start": 3452.31, "text": "Yeah, that could throw Bitcoin adoption back by a decade or two easily. Imagine you're onboarding an entire community on like, what we're seeing right now where like entire communities are on boarded onto Blink, for them that is like syno-synonymous in a way. and if they, would even just suffer like a A severe technical outage or, like ju- judicial issues, for them, Blink is synonymous with Bitcoin, so they, they won't touch it for the next decade or so." }, { "speaker": "stephan", "time": "58:11", "start": 3490.97, "text": "And I mean, to some extent, in El Salvador with Chivo wallet, I think that, you know, people, I think a lot of people didn't like the Chivo wallet, it wasn't very reliable, and in, in some sense, there was a bit of brand damage there, but, you know, so be it. I think as I'm reading you and trying to understand and driving towards, where is the, let's say, the unique places or where are the places that it really would make more sense? It, to me, it sounds like It makes more sense in a, let's say, a high fee on-chain environment and potentially if the government cracks down or if it looks like, you know, and now, I guess, nowadays in the Trump pro crypto era, it seems relatively permissive, but I mean, just, you know, one or two years ago, it was pretty bad, right? Like Biden and Warren and all these people, like it sounded like there, it seemed like there was a lot of things going against Bitcoin and Lightning companies, and that was obviously, as you mentioned, causing Phoenix and Async to, to exit the US. Wallet of Satoshi left the US. Now Phoenix have reentered the US. I think Wallet of Satoshi might still be out, but as an example, Wallet of Satoshi, right, based in Australia, maybe if the Australian government came out with more KYC laws or something like this that- stopped that use, then maybe there'd be more of a, a use or a need for, more of a demand for a Fedimint-style solution, You know, possibly, but even then, I'm, I'm curious as well, how would you compare that with, let's say, a liquid-based kind of solution where, you know, it's, now they're distributing it across the, what is it, fifteen functionaries or so? I think that number might change, but I think last I heard, it's, I think it's eleven or fifteen multisig or something like this distributed around the world. How would a Fedimint solution compare with, let's say, a Liquid style solution." }, { "speaker": "eric_joschi", "time": "01:00:02", "start": 3602.73, "text": "I'd say the main difference there, and similar to Blink actually, I want to add something there, is that you can easily run a Fedimint yourself. Like with Liquid, there's this one federation, it's run very reliably, and it's great. You, I actually don't have a problem with using it, but It's one federation, and if at some point something happened to that federation, that would be like a systemic, risk to Bitcoin if it became widely adopted. While with, Fedimint, if one federation goes down, that's Like very bad for some people, but it's not a systemic risk to Bitcoin because, in like the future we envision, there will be many federations and, so You, you don't get these single points of failure, with Liquid, the Aqua wallet's really impressive, I like what they are doing there and like by bridging into Lightning, I think via, what are they called, Bolt's exchange, yeah, and that's very similar to what we are doing in a way. Like we have our lightning gateway, they have Bolt's exchange, the federation is holding the money. privacy-wise, Fedimint is a little bit better, I think, because eCash has a slightly stronger privacy properties than confidential, assets and confidential transactions, but for end users, that's probably not like the main selling point. So I would actually argue that They're very similar, the main difference is can I run it myself? Like if I want to, get the trust closer to my community, then can I do that? I, the only, viable way that I see right now for that is Fedimint really, and where it's not totally centralized. Like you can also run your own CashuMin, but then you trust one person in your community At" }, { "speaker": "guest_2", "time": "01:01:48", "start": 3708.02, "text": "least you have it in the community. I think Shivo is an example of, you know, people often criticize them for like doing their own thing, the technical issues they had, et cetera, et cetera. but I think it's an example of the This strong preference to, to, to take local custody if you do anything significant. So they were like, they wanted to onboard an entire country. And our answer in the Bitcoin ecosystem would have been, well, you can, you can onboard them to Wallet of Satoshi or you can onboard them to Liquid. In both cases, the, the custody of your funds will be completely outside of your country. That isn't, that isn't, Probably not even responsible as, as a country's leader. Like, ask yourself if you, want to onboard your entire community with money that is significant to them, how comfortable would be, would you be with like onboarding them in a solution that you have like zero, zero control over? And so far," }, { "speaker": "guest_2", "time": "01:02:55", "start": 3775.33, "text": "We, we didn't have any answer for them whatsoever. There was no, like, no custody protocol in any way, shape or form that they could run themselves. Like, I, I think an interesting point that is like often completely overlooked in eCash and Fedimint is that they are like custody protocols, where you can have different clients and you have a standardized way to interact, with a server, so you could be, so you could be running it yourself. You don't even need eCash for that, you could just- Do like regular accounts, et cetera." }, { "speaker": "guest_2", "time": "01:03:29", "start": 3809.15, "text": "but this was so far not the case, everyone was doing their own thing. You had like one provider connected to one app, And no, no way there to like reuse software. So if you wanted to take custody, you had to go the way the Jivo Wallet went, where you do your own wallet and your own server implementation, so I completely write your own software. There was nothing off the shelf to pick." }, { "speaker": "eric_joschi", "time": "01:03:56", "start": 3836.81, "text": "I, I think there were some things like that in Lightning space with, like BlueWallet, I think for a while had a service or like something that you could run and then, connect different frontends to it, but, yeah, it wasn't widely adopted at all." }, { "speaker": "stephan", "time": "01:04:11", "start": 3851.28, "text": "Yeah, I think it was, was it LND Hub, something like that? It was like a wrapper for LND, something? I can't remember the detail now, and I think they've stopped that now anyway. Yeah. Okay. So, interesting. But you make the point there that, let's say, if a nation state were to try to create a Bitcoin wallet for their citizens, could they use a Fedimint-style solution for that? maybe. But I guess, and I mean, even with, even in the EU, they're talking about like CBDCs and saying, \"Oh, no, we're still gonna give you privacy and stuff like that.\" I mean, theoretically, they could try to do some-- if they really, I mean, I think" }, { "speaker": "stephan", "time": "01:04:50", "start": 3890.32, "text": "care about people's, the individual right to privacy or anything like this, but hypothetically they could do something like an eCash, you know, style system if they did care about this." }, { "speaker": "eric_joschi", "time": "01:04:59", "start": 3899.59, "text": "Yeah, one eCash protocol that, is working together with the European Union, it's called GNU Tala, and what they built is, like a- Kind of limited version of eCash where only the, sender has privacy, but not the recipient, so they can still enforce taxation. And they, attended one of the eCash hack days in Berlin, we had some interesting discussions, they're using similar cryptography to ours and Yeah, but, I, I don't think any, CBDC would, ever consider using like a truly private, eCash system." }, { "speaker": "stephan", "time": "01:05:32", "start": 3932.35, "text": "Yeah, for sure. and I guess just zooming out more broadly, obviously we, we can have our ideology, whether it's kind of libertarian, sci-fi punk ideologies It seems to me like bottom line, there's still a lot of people coming in through Bitcoin ETFs and Bitcoin treasury companies like Strategy, Sumo Scientific, MetaPlanet, et cetera. So, do you guys have any comment on that or is it just more like, well, look, people are just-- some people are just gonna use, let's say, KYC custodial things. Obviously, that's not the cypherpunk-approved way, and, you know, the self-custodial, peer-to-peer, no KYC aspect of it is just totally the opposite side of any comments, any reactions on that?" }, { "speaker": "eric_joschi", "time": "01:06:16", "start": 3976.43, "text": "I, I think we have to distinguish two different use cases that people see in Bitcoin. Like there's the one use case which is seeing Bitcoin as an investment And that's mostly how people perceive it, in the West, and because we have good banking, like, I'm overbanked, I'd say. Like, I, I never had problems getting bank accounts. while in other parts of the world, people are actually unbanked, they don't have access to international financial systems, and they might perceive Bitcoin more of a means of payment, rather than an investment, maybe a savings vehicle, but not, like some get-rich-quick scheme or like, yeah, like a commodity. Like, probably if you're investing in the ETF, then you kind of see Bitcoin as a commodity, like gold, you want to hold it because it appreciates in value. and you don't, you're not in it to use it for day-to-day transactions. And for that use case, it might just be the right way for you to use it because you might not have the technical expertise to securely hold your own Bitcoin, and you don't have the ideological convictions, that will make you learn that. So in a way, I, I can't really complain about it. I hope that, more people will learn to cust-- to custody their own Bitcoin because otherwise these ETFs could become a systemic risk to Bitcoin But, at the same time, I see the incentives at play and it is what it is. yeah. Yeah." }, { "speaker": "stephan", "time": "01:07:44", "start": 4064.0, "text": "Joschi?" }, { "speaker": "guest_2", "time": "01:07:45", "start": 4065.25, "text": "I can't really add to that. Eric put it well." }, { "speaker": "stephan", "time": "01:07:49", "start": 4069.47, "text": "Yeah, I see. Okay, well, let's summarize. I guess there were a couple other smaller things. I think you mentioned, single binary, so just maybe you quickly update people, what's the difference there with that update?" }, { "speaker": "eric_joschi", "time": "01:08:00", "start": 4080.32, "text": "Yeah, so for a while, when you were running, Fedimint, you had to, like also spin up, like admin UI and like it was a whole process of setting up a Fedimint, and we're now at a point again with the Iroh integration, like once it stabilizes and once it," }, { "speaker": "eric_joschi", "time": "01:08:18", "start": 4098.61, "text": "And just take a single binary, like a EXE file basically if you're on Windows, and you run it, and you have set up, you of anyone guardian. And we actually did that in one of the setups, because someone didn't want to use Docker, for deploying that and It worked, it was beautiful, like you don't have to have any DevOps knowledge. That's certainly a bias I had in the beginning, when starting it, was still a student at university with a strong background in, like, IT and networking. So I was like, \"Ah, anyone can run something that needs a domain, a reverse proxy, like a certificate, mechanism like that that gets certificates from, yeah, some authority, and it's no problem, people can do that.\" Well, guess what? No, that's not how it works. And now that I've become way busier, I also see like even though I can do it, I don't want to do it. And so having this single binary that you can just run, it spins up Fedimint, you get your admin UI that is now integrated it's way more convenient. Yeah. So my" }, { "speaker": "guest_2", "time": "01:09:24", "start": 4164.66, "text": "lack of expertise in that field and very different focus was, was the asset there. So I had that experience where, you know, I wanted to set up a Fedimint, and for the first time in my life, I was asked to configure a domain, and even though I had a- Conceptual understanding of it, I've never done it in practice, and then there's so much stuff you need to, you need to figure out. You need, you know, you need to figure out the website, to, to, to register a domain, and you, those questions start, and you're like, \"Well, now I'm dependent on this. Like, what are the guarantees they are giving me, et cetera?\" Then you need to con-connect this, you need to configure the IP, you need to con-- like, you need to configure the DNS record. Also" }, { "speaker": "guest_2", "time": "01:10:09", "start": 4209.01, "text": "You can be a software engineer and never done this before, depending on what area you work on. so me going through that process, I had the realization that, \"Oh, well, this, this needs to become, this needs to become significantly simpler, you know?\" such that someone that has like a" }, { "speaker": "guest_2", "time": "01:10:30", "start": 4230.89, "text": "well, on, on start nine, no software engineering skills whatsoever, just being able to graphically click through it. But if, if you wanna run it like inside Docker, or, or on your laptop as a binary, just like The most common denominator of like software understanding, like, what is a binary, I can run it, you know, I can run a program, is sufficient to do it now, you know?" }, { "speaker": "stephan", "time": "01:11:00", "start": 4260.85, "text": "Okay, yeah, so as I'm reading you guys, it's like, there's this new update, it has some new features like, as you mentioned, this Iroh, although, as you said, there's the important caveat that that's, you know, gonna be a new update coming, so watch out for that, update in the future. But the point being, it's now going to be easier to set up a guardian and set up a Fedimint, and so as we spoke about, it sounds like, of course, there's, there's various The ones that seem the most compelling, at least as we've spoken about it, it seems like if Bitcoin were to have a high fee environment again, which seems likely again, and, and/or- Well, it has to have it, you know?" }, { "speaker": "guest_2", "time": "01:11:42", "start": 4302.45, "text": "Otherwise we've got a big problem." }, { "speaker": "stephan", "time": "01:11:45", "start": 4305.78, "text": "Yeah, so even there, I, I don't know, people sort of go back and forth. I'm, I'm not even sure. I used to think that as well. I, I'm not sure. Anyway, I think that's another whole, rabbit hole. But I, it sounds to me like in terms of thirty mint, the case that makes the most sense is, yeah, high fee environment and or if government were to crack down, maybe a future administration or other countries, other jurisdictions, maybe, the state is cracking down on it harder," }, { "speaker": "stephan", "time": "01:12:13", "start": 4333.9, "text": "then privacy and a convenience win to using some kind of Fedimint solution. So that's how I'm seeing it. any closing thoughts, to any last message you wanna leave listeners with?" }, { "speaker": "eric_joschi", "time": "01:12:27", "start": 4347.19, "text": "Yes, so the first time, I joined you on your podcast, was all about like, gathering developers around me to build a thing, 'cause it was, Audacious goal, and, yeah, now I, I feel we're at the point that, we built it far enough that we can go out there and deploy it and So I think you'll be hearing more Fedimint and, yeah, just decentralized Bitcoin custody. That's, my message. don't, centralize it in a few places. That's a systemic risk to Bitcoin. Let's not do that. Bitcoin's too important to that, for that." }, { "speaker": "stephan", "time": "01:13:10", "start": 4390.84, "text": "Yeah, Yoshua, anything to add?" }, { "speaker": "guest_2", "time": "01:13:12", "start": 4392.22, "text": "Again, nothing to add to that." }, { "speaker": "stephan", "time": "01:13:14", "start": 4394.72, "text": "Okay, fantastic. Well, I'll put all the links in the show notes. The main website is fedimint dot org, but yeah, all the links to find Eric and Yoshua will be online. Also, Eric and Yoshua, thanks for joining me today. Thanks for having us." } ] }