Skip to content
PodSubs
ShowsPeopleGuestsJobsNewsEvents
More
BlogPodCred
Menu
ShowsPeopleGuestsJobsNewsEventsBlogPodCredNetworksSign in
Join nowSign in
  • Shows
  • People
  • Jobs
  • News
  • More
    • Guests
    • Events
    • Blog
    • PodCred
    • Networks
    Join nowSign in

Bitcoin Takeover Podcast

Bitcoin Takeover Podcast artwork

S15 E25: Nicholas Gregory & Tom Trevethan on MercuryLayer

Mar 24, 20241 hr 13 minAudio
ShareXLinkedInFacebookWhatsAppEmail

Thanks to the efforts of CommerceBlock CEO Nicholas Gregory and CTO Tom Trevethan, Statechains are now a reality on the MercuryLayer. Originally designed by Ruben Somsen, Statechains serve the purpose of transferring ownership of bitcoins in a private, scalable, and affordable environment. However, the original Statechain design requires the ANYPREVOUT soft fork - something Bitcoin didn't get. But thanks to the efforts of Nicholas Gregory and Thomas Trevethan, MercuryLayer works around the limitations to make Statechains happen without the need for a soft fork like ANYPREVOUT. Timestamps 00:00:00 - Intro: MercuryLayer with Nicholas Gregory & Tom Trevethan 00:00:44 - From MercuryWallet and CivKit to a generic Layer Two 00:02:54 - What counts as Layer Two: unilateral exit vs federations 00:04:22 - Tom on Lightning liquidity vs whole-UTXO statechains 00:07:19 - Sidechains, custody, Drivechains, and client-side scaling 00:10:17 - Privacy priority, blinding, and Nostr swap coordination 00:12:29 - UK legal posture and Ordinals demand 00:16:49 - Access tokens, ~$50 sessions, Lightning payments 00:19:01 - Open API, WASM libraries, testnet clients, mainnet plans 00:21:54 - Wasabi sponsor read and Tor wallet trauma 00:28:28 - Website walkthrough: MPC, protocol docs, GitHub 00:35:03 - Use cases: virtual OpenDimes, gold bars, not coffee 00:38:40 - Ark comparison: no soft fork, no server liquidity 00:41:37 - Drivechains vs statechains, ossification, Paul Sztorc cameo 00:49:38 - Lightning Latch atomic swaps explained 00:54:02 - Five-year vision and Fedimint mosquitoes 01:00:36 - What builders should ship; CTV, OP_CAT, Simplicity 01:09:26 - Where to find MercuryLayer, Telegram, docs 01:11:36 - Outro

Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.

Transcript

From the show. Tap a time to play from there.

Vlad CosteaAnd welcome to the Bitcoin Takeover Podcast. I am Vlad, and this is Season 15, Episode 25. I'm going to be talking about the MercuryLayer with the two people behind this project. I have Nicholas Gregory, who is on the show for the fourth time probably, and his CTO from CommerceBlock, Tom Trevethan. Welcome, guys.

Nicholas GregoryHey, thanks for having us.

Vlad CosteaOkay. So, Nicholas, I've had you on the show three or four times before this. And each time you spoke about, I think the first time it was about Mercury Wallet, and then you spoke about CivKit. So you had two episodes about Mercury Wallet, and you had one about CivKit. And now you're going to talk about the MercuryLayer. So what's the difference between Mercury Wallet, and then you spoke about statechains, and now MercuryLayer. What's this?

Nicholas GregoryYeah, so they're all statechains. So originally, the original idea about statechains was from Ruben Somsen, which needed ANYPREVOUT. We worked for a few years to kind of build a version of statechains without ANYPREVOUT. But at the time, we very much focused on building a kind of statechain wallet with some sort of coin swap facility. And, you know, I noticed you're sponsored by Wasabi, but our goal was to build like an alternative privacy wallet to say Wasabi and Samourai. And although we got some good volumes and good traction, it didn't really take off if we're honest.

Nicholas GregoryAnd in that period, though, there were some new things that happened in Bitcoin. Taproot came out, and Schnorr signatures, and we were able to build a more blinded version of statechains. So the original version, although the swaps were trustless, and you know, we didn't know who people were swapping to, we did see Bitcoin addresses. So, you know, as you know, you can censor those. I think you've talked about that previously with other projects you've worked on. But with the Schnorr stuff, we're totally blinded. We don't see any of the Bitcoin addresses at all.

Nicholas GregorySo we felt going forward, why not basically rewrite a lot of it, which is essentially what we did with the Schnorr stuff. And a lot of the work we did was actually deleting the code, make it a lot smaller and just make it like a generic alternative layer two. And I say alternative because there's not many true layer twos out there, but make it a true layer two that other people can build on. So we're not... We are going to release a wallet, but it's going to be, you know, kind of like an open source thing for people to play with. But we're really more focused on being a layer two. And we are working with other projects that want to build on us instead. And that's the big change. That's the big change. So even though we were always a layer two, we really focused on the wallet building a privacy solution. Now it's really just building a generic layer two like Liquid or something like that.

Vlad CosteaI guess you opened a can of worms there because some people say that Liquid is not a layer two and sidechains in general are not layer twos and only Lightning is a layer two. Well,

Nicholas GregoryI mean, look, everything's a layer two on top of Bitcoin, including an ETF and Coinbase. I guess I always think the best analogy for layer two is can you retrieve your funds without interactivity or coordination with the entity? With statechains, you can. You always have a backout transaction. So if you wake up one day and MercuryLayer is gone, you can retrieve your funds. And the same with Lightning. I think some of the others you mentioned and that's no knock on them. You can't. But yeah, that's the way I would argue a layer two is.

Vlad CosteaSo you just said that Liquid is not a layer two because you need interaction to peg out and you need permission from the federation. So by your definition, it's not a layer two.

Nicholas GregoryIt's a tool. There's nothing wrong with sidechains. We used to build sidechains before Mercury. It's just these are different tools in the box. You have spanners, hammers, screwdrivers. We're not pro or anti them. For example, on CivKit, we're probably going to be integrating either Cashu or Fedi. That's not a true layer two, but it's a better use case for CivKit than, say, integrating statechains. It's, you know, I think for Bitcoin to scale to work, we're going to need all these tools and it's not going to be a one-size-fits-all.

Nicholas GregoryAnd I think that's probably some of the challenges we have now. Lightning, which is getting bashed a lot because I think it was maybe marketed as a retail solution to us. And we build Lightning software. It's probably more B2B protocol. And it's not the final solution that's going to probably end up in someone's mobile wallet unless that's custodial.

Vlad CosteaOh, man, that's a bummer. I'm going to let Tom, maybe he's more optimistic about this. What do you think about MercuryLayer, Tom? Is it going to be better and more reliable than existing sidechains? Will it fulfill some of the promises that the Lightning Network broke in its marketing?

Tom TrevethanWell, the fundamental difference, OK, so you can compare it to Lightning. And I mean, yeah, following on from the previous kind of discussion about layer two, I mean, people get very obsessed about the meaning of this term. But lots of people are willing to define layer two as something that has this kind of unilateral exit, which is what Nick described. Basically, you can come out. If everything else fails, you always have a backout on-chain. And so in that sense, only Lightning and MercuryLayer and potentially also when, you know, if it ever gets released, Ark, have this property of this unilateral exit.

Tom TrevethanMercury is very different from Lightning and what it can do and what its limitations are. I mean, the biggest problem with Lightning is this issue of liquidity. You basically have to open channels with people. And in order to make a payment, you have to have an existing channel and has to be enough liquidity along the route. Whereas, you know, MercuryLayer statechains don't have that problem at all. You can literally just send to anyone. There's no liquidity issue. There's no doesn't need to be backed by anything.

Tom TrevethanHowever, the biggest limitation, I think, is the fact that in Mercury statechains, you can only send whole UTXOs. So this makes it difficult as a general kind of payment system where you want to pay someone an arbitrary amount, which Lightning is very good for. And that led us to our kind of initial use case of a privacy system, because obviously, it works very well for privacy if you have like fixed amounts of kind of fixed UTXOs. You can swap them with people and you have this big anonymity set that you get from that.

Tom TrevethanBut yeah, like Nick said, I think we want to explore a lot more of these cases now. And I think it really is just a case of finding where it can be very useful. And I think it can be used in combination with Lightning for the part which Lightning struggles with, which is the liquidity issue. And so some kind of combination of Lightning and statechains, I think, is definitely, yeah, there's definitely a lot of promise there.

Vlad CosteaRight. So what is the difference between a statechain and what we call a sidechain? And how is the MercuryLayer better than the average statechain?

Tom TrevethanSo the only current sidechains we have are like federated sidechains, so they're custodial. So if you want to use a sidechain, the sidechain essentially exists as a ledger to update ownership of Bitcoin, which has been sent for essentially a multisig in the case of like Liquid, same as RSK. But ultimately, that Bitcoin is under the custody of whoever has the keys to that multisig. And that's the big problem with custodial kind of federated sidechains. And I think the thing that we're most bullish about Mercury, although there's a trust assumption in Mercury, is fully non-custodial.

Tom TrevethanAnd the blinding means that it's a lot more legal to run a statechain entity server without any KYC than we think, for example, a federated sidechain. So I mean, at the moment, say Liquid and RSK are acting in a permissionless way, but yeah, we don't know how long that's going to last and how long this custodial kind of federation is going to be allowed to operate, essentially running privacy infrastructure without any kind of KYC or AML. I mean, obviously, I guess you're very aware of drivechains, and that's an alternative model for a non-custodial sidechain.

Tom TrevethanYeah, I guess it's a different thing. Obviously, that requires some kind of, you know, this soft fork, which may or may not ever happen. But yeah, I guess that's the difference really between MercuryLayer where, you know, you have... I mean, the other part of it is scalability as well. The sidechains are just other blockchains, and so fundamentally, they don't solve the scaling problem because you basically just, you know, Bitcoin doesn't scale because you have to verify everything. It's a fixed finite block size. Same with the sidechain, same with Liquid.

Tom TrevethanYou just, you know, you've just kind of moved the problem, kicked the can down the road a little bit. And if Liquid ever does become very popular, it will have exactly the same scaling problems as Bitcoin does. Whereas with statechains, you only verify the UTXO history that you have. So in that sense, it scales a lot better than the sidechains.

Vlad CosteaYeah, I guess you foreshadowed my next question, which I guess CommerceBlock is known for building privacy products. And I see MercuryLayer as something which concerns mostly privacy, but it's also about scalability. So I was going to ask you, which one do you see as the priority for the MercuryLayer?

Tom TrevethanI mean, I can... Neither and both, if that makes sense. I think the, I mean, the first Mercury Wallet was privacy focused. But you know, the world is changing. It's more difficult to have a product entirely focused on privacy, I think, as the world is changing in lots of unfortunate ways. And the, yeah, the MercuryLayer is fundamentally a lot more private than the original Mercury Wallet, because we've now completely blinded the process. So the server doesn't even know it's dealing with Bitcoin, let alone which Bitcoin. And we've incorporated features, so like atomic transfer features that are going to allow people to swap coins completely privately off chain.

Tom TrevethanAlthough we're getting out of the business of running a, like a coordinator, which we did with the previous Mercury Wallet. So we're going to, and yeah, we're working in ways that, you know, you could use like maybe Nostr as a way to coordinate swaps and allow people to kind of, you know, find each other and coordinate coin swap. I mean, one of the big improvements with MercuryLayer for privacy is that in Mercury Wallet, if you wanted to gain and do a coin swap, the server would be aware of the coins that have swapped.

Tom TrevethanAnd so you need to do lots and lots of swaps in order to kind of gain anonymity from the Mercury server. But in the new version, Mercury server is completely blind. So essentially you get, you know, 100% privacy just by doing a single coin swap. And so, yeah, we want to support this, although we, yeah, going forward, we won't kind of market it as a privacy solution or, you know, do any kind of coordinating or anything like that. I don't know if you want to add to that, Nick.

Nicholas GregoryYeah, I mean, look, we are, you can probably tell by our accents, we're British, we used to run things offshore. I mean, we now run things in the UK, we've been working pretty hard to work with the local regulation. I think that's what you have to do in this world. And, you know, the stuff we now run, it can run in the UK, which I think gives a bit more trust as well. Because, you know, one of the people questions people says, what happens if you wake up in the morning, MercuryLayer is gone or the Lightning infrastructure is gone. Yeah, there's a bit more trust because we are running this in the UK. We have got legal clarity on what we can do. A lot of our designs are based around that. And, you know, even though we're legal now, you know, laws can change, but we're hoping because we are legal now that that will stay for quite a while. I think that's quite important going forward. But you have people want to run coordinators themselves on the Nostr protocol, charge money for that. That's up to them. I mean, we're a blinded solution.

Nicholas GregoryWe don't know how people use this. And we're encouraging all sorts of use cases. I mean, I think I've been quite open about this on Twitter, probably 70% of the requests are around ordinals. Now, I'm not into JPEGs, but I'm not going to judge people who are. It's a permissionless system. It is what it is. But yeah, that's that's our position.

Vlad CosteaOh, but that's very interesting, because you have a product, basically a layer which is designed around the idea of privacy. But why would these guys who do ordinals want to have any sort of privacy? Because that...

Nicholas GregoryWell, I can explain to you why ordinals people like it, because obviously, ordinals rely on Bitcoin transaction fees to move them around. If you were to put them on a state, if you were to use our blinded signature service, they can move them around for a number of times without any on chain fee. They can do swaps. So if I've got two ordinals, and I want to swap it for one of yours, we could do like an atomic swap. And this all happens off chain with no fees. And clearly, we didn't design MercuryLayer with ordinals in mind.

Nicholas GregoryWe predate ordinals, but it seems that there is a use case there and seems ordinals is not going away. The big community they have, many of the ordinals communities have reached out to us. And as we said, we're a blinded signature service. We don't know how people use us if it's for ordinals, I have no objection, we wouldn't filter it anyway. And you know, if ordinals guys end up, you know, doing less on chain, that hopefully lowers the fees for us economic users.

Tom TrevethanYeah, I mean, you know, people criticize ordinals for basically a spam or taking up block space. But, you know, if you can provide a solution that enables us to take a lot of that transaction volume off chain, then that's good for everyone, you know. And yeah, it just seems like it's a natural kind of the fundamental property of ordinals is that it's you have individual UTXOs, which hold value. And yeah, and that's obviously, that's the limitation of MercuryLayer as well, that you can only operate on the level of individual UTXOs.

Tom TrevethanSo yeah, again, you know, I think, I don't necessarily have to be a fan of the concept to see that, yeah, if it can be used to benefit that and take transaction volume off-chain, then you know, it's good for everyone.

Nicholas GregoryYeah, I mean, we internally are focused on more research between statechains and Lightning, and we've done bits and bobs, but that's quite a hard thing to crack. But, you know, maybe the work on ordinals will give us ideas to help crack that. I mean, innovation leads to other innovation as well. And, you know, we have to fund it, we want users to be able to keep the protocol going as well. And that's, if it's funded by ordinals in the short term, while we work on research on how we can integrate statechains into Lightning or potentially into Fedimint. I mean, last week, we were at Advancing Bitcoin, we spoke to Obi, the guy from Fedimint, and they said that they've been looking at how to integrate statechains and Fedimint. So yeah, we, it doesn't stop with what we're doing now, we want to see how we can integrate into other protocols. And I think that's key. There needs to be some sort of commonality in these layer twos. And they need there needs to be some way of moving between them.

Vlad CosteaYeah, you said something interesting there that you have the Ordinals people paying the bills. And that was my question. My next question, basically, what is the economic model behind the MercuryLayer? How do you collect any kind of money? Because you said there are no transaction fees.

Nicholas GregoryYeah, so we haven't fixed it yet. But it's likely to be in the region of $50 or 50 euros, because we're in the UK, we probably should charge in euros, but likely 50 euros for a session. And that's a blinded signature session that lasts for a year. Now, every time you use like do a transition, use the server, you lose eight hours, that's part of the protocol or the way it works. But essentially, that's that's roughly the model. So those guys, instead of, you know, paying a Bitcoin transaction fee, every time that they can reduce a lot of their cost, we get the money. And that's how we're funded.

Tom TrevethanSo yeah, just to add on that, I mean, the model: in the previous Mercury Wallet, we charged like a withdrawal fee. So people could deposit any amount, and then we charged like a 0.5% withdrawal fee. And the withdrawal transaction had an additional output, which paid the fee. But we can't, we couldn't do that with a new version, because it's completely blinded. So we got no way of enforcing any fee collection on-chain. So now, yeah, we basically the setup that we've now designed and set up, you have an access token in order to be able to create a statecoin.

Tom TrevethanAnd it's through that that we can control, you know, we can control access. Now, again, it's completely privacy preserving and blinding. And obviously, yeah, whichever way we, you know, we've got an implementation that you can pay with Lightning, basically, to get one of these access tokens. So yeah, if you wanted to create a statecoin, you would go, yeah, get a Lightning invoice, pay it, get this access token, and then you'll be able to deposit into the MercuryLayer.

Vlad CosteaYeah, that's very interesting. And how can users interact or get started with the MercuryLayer? What is the easiest way to onboard?

Nicholas GregorySo we have a test environment up at the moment. And you basically we have a public API, which we I mean, everything's open source. So nothing that we have couldn't be copied and pasted by someone else. Obviously, well, there is some skill in securing a server, locking it down, doing all that infrastructure, which obviously is what we keep private. But there is a Rust-based WASM API that anyone can integrate with. And yeah, people already started playing with it. And, you know, the earlier people play with it, the easier it is to make changes for different use cases.

Nicholas GregoryBecause, as you know, things ossify once they get in production, once you get more users. So we would like any comments, complaints to be done earlier rather than later.

Tom TrevethanBut yeah, so the original Mercury Wallet, it became I mean, it's like Nick was talking about earlier, I mean, it became a bit of a closed thing. So we had this wallet, which was just a desktop app. And it became a bit of a monster, you know, we spent so much time on the UI. And a lot of the business logic got intertwined with UI. And we ended up with this kind of mix of WebAssembly and JavaScript. And it was, people came to us and said, you know, how could I integrate mercury into something else?

Tom TrevethanAnd it's, so, well, actually, it's not that easy, because of the way that the wallet ended up evolving. So one of the big things we did with MercuryLayer is we wanted a clean slate, brand new API. And to make it a lot more accessible, so that we created very easy to use, like, essentially a library with the API, which is kind of cross platform. We've done this with WebAssembly. And we've created, you know, example, like clients with JavaScript, React and Rust. And so yeah, we want it to be very easy for other people to come and use these libraries directly into other wallets, third party wallets.

Tom TrevethanSo we are going to release a desktop and mobile wallet as an example of how to do it. But yeah, as our experience with the original Mercury Wallet, I think we realized that the future is not some specific kind of monolithic wallet app that we want really to get this used. And the library is used and integrated with other projects. But I mean, if people want to try out now, I mean, you could use testnet, and you can download any one of our clients. And yeah, and obviously create statecoins, deposit testnet Bitcoin and test out the process of, you know, sending the coin from user to user and yeah, seeing how it all works.

Tom TrevethanBut yeah, we're planning to do probably turn on mainnet, I guess in a few weeks time.

Vlad CosteaYeah, I think it would be a very good idea to take a look at your web page and comment on what's there and maybe ask some questions based on your presentation. But before that, I got to present an ad for this show's sponsor, Wasabi Wallet. I guess you guys are competitors right now in some way, except that Wasabi Wallet works on...

Nicholas GregoryNo, no, no. We're not competitors of Wasabi at all. Big fans of those guys and like what they do.

Vlad CosteaYeah. So in case you're not aware and you're listening to this, Wasabi Wallet is this CoinJoin wallet, which works on the Bitcoin main chain, the base layer. And just recently, they had this very large CoinJoin, which put together more than 300 inputs and outputs. The amounts are not equal. So it's very hard to associate what comes in what goes out for remixes. So if you do a second CoinJoin in this coordinated setup, you're not going to pay any coordination fee. And much more than this, I know that the FUD about CoinJoins is that once you do CoinJoin, you're not going to be able to use your coins on regulated exchanges or whatever.

Vlad CosteaWhat Wasabi did was to integrate this buy anything button, which is a direct communication mechanism with ShopinBit, but it's routed via Tor. And you're telling them what you want to buy and where you want it to get delivered. And they make it happen. And then you pay directly with Bitcoin from your CoinJoins. No questions asked. It's super cool. They make Bitcoin fungible and easy to use in commerce. And I guess you can think of certain use cases, you can even buy cars with that. So if you do want to buy a car, you can do it with your CoinJoin Bitcoin, download the wallet from wasabiwallet.io.

Vlad CosteaAnd while I prepare the web page that I'm going to share on the screen, let me know, fellas, what do you think about Wasabi? Or if you designed Wasabi, what would you do differently? Maybe some feedback is useful.

Nicholas GregoryWell, we have a weird relationship with Wasabi in that we've worked a lot with nothingmuch. As you know, he was heavily involved in the WabiSabi protocol. The original Adam Ficsor, he reached out to us multiple times during the Mercury Wallet days, MercuryLayer days, always very complimentary. So, and, you know, and I think the CEO of Wasabi, we bumped into him quite a lot at conferences. Yeah, fans of theirs. I mean, it's a hard space. I, you know, I've used the wallet. I think it's a good product.

Nicholas GregoryObviously, you know, we work with nothingmuch. He's, you know, he's kind of told us some of the things he would like to improve. So we are a bit biased. So nothingmuch was involved a bit, he kind of reviewed a bit of Mercury Wallet and MercuryLayer. He did help us on some of the Tor issues we're having. And obviously, he's, he helped us a bit on CivKit. So we, you know, we know those guys quite well through all of this. But in terms of improving it, I don't know. I think it's a hard space. I think my experience with Mercury Wallet is I think people who don't build wallets in this space, they don't realize how hard it is to work with Tor. And that was probably a lot of our problems just dealing with Tor. So I do congratulate in the way Wasabi have kind of hidden the complexity of Tor from their users. But there were some days that Mercury Wallet was just unusable. And we did things like I mean, I know Wasabi basically ended up running their own Tor node.

Nicholas GregoryWe thought there was too much legal issues there. So we ended up providing a switch from Tor to I2P. I mean, that's probably, I would like Wasabi to maybe support other protocols other than Tor, nothing against Tor. But I think we need, we can't be so reliant on one protocol, because although Tor's improved a lot, I think any one protocol is a DDoS attack waiting to happen. But no, I think they've done a great job of like hiding the challenges of Tor very well.

Tom TrevethanYeah, it's very difficult to design this with the limitations of Tor and make a nice, usable wallet. And yeah, it's a lovely wallet, Wasabi, it's a really nice UI. Yeah, I mean, I guess yeah, it's a whole other discussion that would take a long time talking about the issues with coin swaps compared to CoinJoins, which is quite an interesting, you know, privacy discussion. But there are... Yeah, I mean, there are... I think we struggled with... I mean, potentially coin swaps can be a lot more private than CoinJoins.

Tom TrevethanBut there are big difficulties with perception. The thing is with CoinJoins, everyone understands what they are and how they work. And yeah, and coin swaps are a different kettle of fish. But yeah, maybe that's a discussion for another day.

Nicholas GregoryYeah, maybe if MercuryLayer becomes a bit more in the mindshare, maybe we could convince Max to basically integrate MercuryLayer into Wasabi and do some, they, I mean, they have experience of running, you know, CoinJoin coordinators, they could run a coin swap coordinator. I'm not putting pressure on them. This may not be necessary, but we'd happily work with them. If that's something... Maybe that's something I'll mention, but I don't know where they are mentally, where they're thinking of continuing with the product. But that might be something useful to them. Who knows? You know, we're entering a high fee environment. Clearly.

Nicholas GregoryI mean, obviously, there's a lot of focus on the Bitcoin price. But if Bitcoin price goes to what some people think it could go to, it's gonna be a high fee environment. And maybe Wasabi are gonna look at ways to get around that. I don't think Lightning is probably something that they would touch because that's, you know, they've they've been public about that. So maybe MercuryLayer and coin swaps, who knows?

Vlad CosteaYou're lucky right now, because Max Hillebrand is one of the most active listeners of the show. So he will consider it. It's pretty cool that the sponsor is also a big fan of my work. And I think this is the perfect marriage, or is it a marriage? I don't know. It's the perfect match, and also...

Nicholas GregoryDon't say marriage, because most marriages end up in divorce. Say friendship.

Vlad CosteaYeah. And also you mentioned nothingmuch. That's Yuval Kogman. I've had him in season 10, episode 7 of the show. If you want to hear him talk about Bitcoin privacy and what he did with WabiSabi, which is version 2.0 of Wasabi Wallet, then you got to check that out. It's season 10, episode 7 of the Bitcoin Takeover Podcast. Now, after all of this innuendo, let's spend some time looking at the MercuryLayer website and I'm sharing my screen right now. I hope you can both see it.

Vlad CosteaSo this is the web page of MercuryLayer. I said Mercury Wallet before, haven't I? Yeah. Sorry, old habits die hard. Bitcoin MPC Layer 2 Protocol Using Blinded Signatures. What is MPC?

Tom TrevethanMPC stands for multi-party computation. It's a general word. It gets abused a lot these days. But essentially, it means, at a fundamental level, it means that two parties can basically compute a single, perform a single computation, but with neither revealing their inputs to the computation. So in MercuryLayer, essentially, this is used for describing the fact we do a two party signature. So we generate a single signature using two key shares. Also, the process by which we update the private key shares when ownership changes from one owner to another owner, the key shares are updated in a way that the address doesn't change, but each participant's share of the key updates so that previous owner can't then, you know, can't then collude.

Vlad CosteaYes, I see that your website is pretty technical. It looks more like a scientific presentation of the concept with all the mathematics of transactions. It's not in marketing mode, to put it like this.

Tom TrevethanYeah, so, I mean, obviously, as the thing goes forward, we might update this, make it a bit more user friendly. This is, I guess, we've put in a version of the initial kind of paper for the MercuryLayer, which basically explains the protocol. And yeah, each step of the process.

Nicholas GregoryLet's just say the website was built by engineers, not by marketing people. And you can blame me for that.

Vlad CosteaNo, but it's pretty cool. Because you're honest about what this is, you're not trying to make some bullish narratives about what MercuryLayer is, you're allowing builders, developers to acquire an interest and see everything that they need to know before building on top of this layer. So I think this is a good approach. I also told this to Paul Sztorc, that since he created the company, LayerTwo Labs, the website became dumber in the sense that you have these terms about, oh, it's gonna scale, it's gonna be private, it's so nice.

Vlad CosteaBut before that, it was more interesting, because it was more academic, to put it like this. It had more references. I don't know, I guess I'm a different kind of audience.

Nicholas GregoryYeah, I mean, look, to be transparent, I built the website with a Rust templating engine for a bit of fun to learn that. But you know, as we get more, hopefully we get more users as it's more successful, we'll build cute explorers, we'll build more stuff around nice marketing, you know, around marketing, block explorers, stuff like that. And there's time for that. But this is really just to get it out there. Because a lot of people ask us, how does it work? How does it work? We just wanted it on the website, send them there. If you want to go low level, see the cryptography, it's there. Just be direct about it.

Vlad CosteaSo you said something about the API key. How do you get that? I'm on the website, you go to GitHub?

Nicholas GregoryWell, it's not live yet. So we are in the process of going live because we... But that's, there'll be an update on the website on how people can start using it.

Tom TrevethanYeah, so if you wanted to use the testnet version, if you go to our GitHub, which is github.com/commerceblock, and then mercurylayer. There's a, yeah, there's a document basically that it gives instructions on how to, how to basically install a client, how to access the testnet version and get a testnet access token.

Vlad CosteaI was gonna open the GitHub, but I have to do 2FA and input a code. And during the previous live stream that I did, I got rugged because I had Bryan Bishop on and he spoke about his Webcash system. And he showed me how it works and stuff like that. And I opened the wallet, I did everything on live stream and somebody stole the funds from my wallet. Everything was visible and I thought it was super funny. I don't want to get my GitHub account hacked right now.

Tom TrevethanI hope it was only test coins.

Vlad CosteaYeah. So by looking at this, I see that the structure of your web page is that you have an overview, you have a mention about statechains and blinding. And then you have, so there's a distinction between the MercuryLayer and the MercuryLayer protocol. Can you explain it?

Nicholas GregoryWell, it's more of an API, like that's the interaction between what you do on the, on the client side and the interaction you do with the server. So that's, that's what would be explained and cryptography used there. So it's not like a, it's not a distributed system, like a, you know, I'd say that the Lightning Network. I mean, in theory, other people could run MercuryLayer instances. You know, people always ask us, do we support it? Yeah, we'd love to, but, and we're open about it, people say that's a good thing, but no one ever comes because there's a lot of complexity in running a backend server.

Nicholas GregoryI think we're seeing that in Lightning where kind of pleb nodes are dropping off because it's not easy. It's not an easy thing to do. And this is where I have sympathy for like people who maybe criticize Liquid Federation members, you know, running a full, you know, always available 24/7 99.9% high availability Federation nodes is not easy. But yeah, we're probably going to add some stuff there later on and how people could run their own instances if there's interest.

Vlad CosteaOkay, so let's make this a bit more practical. Can you describe some use cases for the MercuryLayer? Very simple for the average user to understand.

Nicholas GregoryI mean, when I explain it to like, you know, the less technical crowd, like we go to meetups in London where they're a bit less technical: it's just virtual Opendimes. You know, I think I'm sure you've been to Bitcoin conferences where people wear like an Opendime around their neck. All MercuryLayer facilitates is virtualizing... you could call it, I mean, people now say VTXOs, which came from the Ark world, but we're just virtualizing a UTXO and able to move that off chain in terms of use cases.

Nicholas GregoryThere are many, I mean, ordinals is the obvious one because that helps visualize it. But, you know, these could be large bills. I mean, think of gold bars. What are gold bars? They're blocks of high amounts of money that can move around off chain with no settlement fee. And that's what, that's where we are. And I think as we see, you know, fees increase on the blockchain, which I think is going to happen with the price going up, these layer twos are going to get a lot more interest. I mean, it's going to be hard to move significant amounts of money between different Fedis, between different sidechains. And I think, you know, MercuryLayer facilitates that at virtually a ridiculously low cost.

Tom TrevethanYeah, the fact that there's no liquidity constraints, you can send any amount to anyone. But yeah, I guess it's visualizing this constraint that it is a fixed amount in a UTXO. So, and I think that's the challenge, you know, that we're going to face, because it's extremely useful principle. But yeah, getting getting that into other, yeah, integrating out with the services and being able to help, you know, move around these, these chunks of money. And I think that's the challenge, obviously, with, you know, limitation in terms of payments, you know, so you could you wouldn't be able to go, I think it's Yeah, I think we can be honest and say, you're not going to be able to go and pay for your coffee with a MercuryLayer.

Tom TrevethanBecause, you know, you are limited by it's a whole UTXO, you're limited by the on-chain fee. And so the amount, for example, you're having in a statecoin, it doesn't make sense for this to be a very small amount. So, yeah, it's chunks of money. I mean, other use cases, I guess, yeah, privacy is very good, because you have these fixed amounts you move off-chain. And yeah, I mean, any case where you want to make, yeah, you want to move a chunk of value off-chain, you know, instantly.

Nicholas GregoryYeah, and what's the greatest thing about it is, I mean, the server code is relatively simple, complex cryptography. But we only have two or three database tables, we store virtually no data. We essentially are a blinded signer. Think of someone who can't see signing checks, but we co-sign checks with other people. So the likelihood of us getting shut down is minimal, because all we are is a blinded signing service. I think that's why we wanted to make it an open API and not have a wallet so that we can create creativity.

Nicholas GregoryBecause I think in a high fee environment, there are going to be challenges moving Bitcoin and are going to be, people are going to search for optimizations. You know, people are moving Bitcoin all the time and transaction fees of $100, $500. You start, you're going to start seeing the creation of settlement systems and you're going to want a settlement system that's regulatory resilient. Yeah, how are you going to move large chunks of money into to Fedis, for example? Lightning, we're again, we're a fan of Lightning. We run Lightning nodes ourselves, but it is very capital inefficient.

Nicholas GregoryThat's why I think Lightning is very much suited for small payments. But you do need someone to manage liquidity there. This has no liquidity inefficiencies. I mean, it just doesn't exist. People compare MercuryLayer to Ark a lot. Now, first of all, we don't need a soft fork. A soft fork would be great. I mean, we've spoken to the CTV guys, very supportive of what they're doing. You know, it would help us for sure. It would remove some of the constraints we have. But a soft fork may not ever happen, but we don't need one.

Nicholas GregoryWe're live now without soft fork. And also, Ark has liquidity needs: to run an Ark node, you're going to have to provide liquidity for the way they kind of move those VTXOs. Here we have no liquidity. There's no money locked in our servers. We're just blindly co-signing transactions. And I think that's where people are going to say, well, this works now. It doesn't need a soft fork. It doesn't need liquidity.

Vlad CosteaYeah, you mentioned Opendimes. And this is my opportunity to promote the work of two Belgian guys who are called Bastien and Baudoin. They created the Satodime card, which is a Java card that they programmed to store Bitcoin. It's basically a very privacy friendly way to transfer Bitcoin between you and your friends. You can give this to someone.

Nicholas GregoryMe and Tom, we met the guy behind it on Thursday night, didn't we? Advancing Bitcoin, the Java card.

Tom TrevethanYeah, I don't know if that's the same.

Vlad CosteaBut is it Satodime? Because Java card is the name of the general product that you can buy.

Nicholas GregoryMay not be, but we were at a meetup, because I was at Advancing Bitcoin last week and he was showing us his Java card and we were like, wow, it's a little VM running there with a little Bitcoin signature service, which was pretty cool.

Vlad CosteaYeah, they're about 25 euros and you get, I think, a 5% discount with promo code BTCTKVR. Get them and help me buy a Supra, because that's why we're all here, right? To buy Supras. Anyway, I have to read the message that I received from Paul Sztorc and then I'm going to move on to the next question. And he said: Is Blockstream going downstream? Has RGB got you down and brown? After a long day of listening to Peter McRetard, why not feel something for a change? Feel alive, join LayerTwo Labs and make everyone feel sad and frustrated and angry.

Vlad CosteaOur next ritual is at 3 a.m. in the ancient forest. See you there. I'm Paul Sztorc and I approve this message. Or do I? Fair enough. They have a very funny way of marketing. Some people dislike it and they say they're making a joke of this whole concept of promoting Layer 2s. But I think they make some good points. Anyway, what's the difference between drivechains and statechains? I know you mentioned it briefly, but in specifics, it's that drivechains are storing everything just like Bitcoin while statechains are only about UTXOs.

Nicholas GregoryWell, I'd say maybe a dumb version is, you know, the state... You could argue that a statechain is like a client sidechain. I mean, all the states are stored on the client side. The server doesn't have any state. Whereas in a sidechain, the state is stored in another set of infrastructure, the sidechain. But yeah, they're very different protocols. I don't think you really want to compare them. They're just at the opposite ends of what they're trying to do.

Tom TrevethanI mean, a drivechain is another blockchain. Everything needs to be verified by everyone, which is the model. Obviously, a drivechain is secured by proof of work. The peg is secured by proof of work, which is still kind of decentralized, but there's some issues with how secure that is. But yeah, fundamentally, I mean, I don't know. I'm not as opposed to drivechains as a lot of people are. I think it's, I don't see it as an existential risk to Bitcoin like some people do. And you know, it's like either it works, in which case, a lot of activity would then go to the drivechain or it doesn't work. So I don't see the risk.

Tom TrevethanBut yeah, I think there's going to be an uphill battle for them trying to get that soft fork in.

Vlad CosteaI think they're gonna do it without a soft fork with BitVM. Okay, that's one way of doing it.

Tom TrevethanThere are a lot of complexities there. There's some trade-offs there. But I mean, yeah, good luck to them.

Vlad CosteaAnyway, I think drivechains find themselves in two extremes when they're being described. On one hand, people say no, they're not going to get any traction, because in the absence of the speculative token, nobody wants to use the Ethereum technology. And this is one extreme. The other extreme is when they say, oh, but if this becomes too successful, it's an existential threat for Bitcoin, because too much activity is going to happen on these sidechains. And you're gonna take away fees from the base layer. And make it less secure, I guess.

Vlad CosteaBut I guess you can also say this about statechains, right? Because you're taking away some activity from the base layer, which means that fewer fees get paid to the miners.

Nicholas GregoryYes, but I don't think anyone would use the clunkiness of a statechain unless there was a reason to it. And that would be high fees. And you could say the same with Lightning. Lightning is quite tricky to use, the whole liquidity issue. People only use it because the fees are uncooperative. Everyone wants to be on the base chain. I mean, the base chain is easier to use. In terms of sidechains, I'm not anti sidechains. My only concern is that we do have sidechains in Liquid and Rootstock.

Nicholas GregoryAnd the uptake has been not amazing. But I guess I'm very much in the anti-ossification crowd. I don't want us to ossify. I think we should allow things like drivechains, CTV to happen and let the free market decide. And, you know, Paul's a builder. He's definitely a smart guy. He should be allowed to build it and let the free market decide if there's a use case there. I mean, that's what I always try and push.

Vlad CosteaI was going to crack a joke earlier when you said that you're regulated and you work with statechains, and say you guys are statists.

Nicholas GregoryWe're legal. I wouldn't say regulated, that would be the wrong word. We're legal. We're not going to go to UK prison for running it. But they could change the law. So I don't want to shout about it too much.

Tom TrevethanYeah, so we're really designed specifically so that we aren't regulated. Yeah, so we don't need to worry about regulation.

Nicholas GregoryLaws can change, as you know. I mean, if you look, I think talking about your friends, the CoinJoin guys, I mean, they've latched on to a rule that says that, you know, as long as they're non-custodial, they're cool. But I think the Treasury in the US are trying to push changes and that could happen to us as well. So we're not going to show off or be excited about that. We've got to keep the big fight. And keep on making sure that, you know, dumb rules that prohibit innovation are dumped. Make noise about it.

Tom TrevethanBut I think, yeah, we have had some doubts or criticism about the name statechain and statecoin that we use because it kind of does imply it's like a government thing or, you know, like a CBDC or something, you know, obviously. But statecoin just, I don't know, it just seems naturally to roll off the tongue when you think about statechains. But yeah, maybe we have to rebrand.

Nicholas GregoryWhen we did Lightning stuff and called it state channels, that sounded cooler. I thought that was. But that is a hard thing to get statechains integrated with Lightning. Conceptually, it sounds easy because you go, oh, you're moving around private keys. Why can't there be private keys of Lightning channels? But then you have to think about what happens if a backup, an old state, is broadcast, if you're moving the statecoin and the Lightning channel is being used. And then, you know, you've got a lot of edge cases there that could wreck the whole thing. So we've parked that for a while, but we still think about it.

Vlad CosteaYeah, there's this guy from Liverpool. I had him on the show. His name is Matthew Croughan or something, Krogan, Kroghan. I'm not sure how you pronounce that. But he is working on NixOS and Nix has no states. And he was saying, so do you like the state? Do you want to abolish it? And everyone says, hell yeah, of course, use NixOS because there's no state in NixOS. It's a pretty tongue in cheek kind of joke.

Nicholas GregoryAbolish the state and statecoins. Yeah, we're not that Mad Max.

Vlad CosteaAnyway, I was going to remark that all of these layer twos and it doesn't matter how you define a layer two, anything that uses Bitcoin and is not on the base layer can be remotely defined a layer two, even WBTC to some extent. But I'm going to get punched in the face for that. I'm going to stop here.

Nicholas GregoryNo, I agree with you. Wrapped Bitcoin is a layer two.

Vlad CosteaSo all of these are a very nice leverage against the base layer becoming too... or having too big fees. How should I put it? Like if the fees get too expensive for the average user, then the average user is going to say, OK, I'm not going to use this anymore. I'm going to switch to this, which means that there's going to be a natural regulation mechanism to make sure that the fees are affordable for the users. And they're never going to get crazy expensive. And it's an interesting mechanism as a leverage for the users.

Nicholas GregoryYeah, I think every layer two is going to have to sacrifice a bit of trust. And that's got to play. I mean, I think sometimes people want the trust of the base layer, but they don't want, but they don't accept the layer two is not going to be as trustworthy as the base layer. And I think that that acceptance needs to be accepted. And I don't think that's understood fully. But layer twos are going to be around, they are going to exist, and they're not going to be as trustworthy as the base layer. And that's why they're layer twos, I think.

Tom TrevethanI think all these things should be part of the conversation. We talked about it earlier, this distinction that only these things are layer twos. I think if you want to call a layer two that. But I think we need to consider the custodial things, sidechains, Fedimint, and even, you know, you go to the very top, just databases, Coinbase, or, you know, people do use these to transact Bitcoin. And these are all things that people use to make payments in Bitcoin, which is what, you know, Bitcoin is all about.

Tom TrevethanAnd, yeah, unfortunately, Bitcoin doesn't scale. And all of these things have different trade-offs. And yeah, they should be part of the discussion about, you know, how can we enable people to use Bitcoin at scale, you know, which is crucial.

Vlad CosteaYeah, there is someone in the chat. His name is Volodymyr Moiseuk. Anyway, he says he's interested to hear about Lightning Latch. What's Lightning Latch?

Tom TrevethanI can talk about that. I mean, this came about from, yeah, I mean, you know, one of the use cases being if you did have a statecoin, and you wanted to swap it for a Lightning payment. So, you know, you might ask, well, what's the point of that? One of them might be privacy, you might want to purchase a coin with Lightning that you can then not trace to your Lightning payment. The other one is maybe if that coin has a value, which is different from the Bitcoin value, maybe because it's an ordinal.

Tom TrevethanAnd you may want to, yeah, make an arbitrary payment for that. And so we wanted a way to support that in a completely non-custodial way. Eventually, you know, this can be integrated in some kind of, you know, other swapping service or DEX. And so, yeah, the basic idea is that you have an atomic transfer of a Mercury statecoin and a Lightning payment. So that means that either they both complete and each party, you know, the other party receives a statecoin and the party receives a Lightning payment or neither complete, all while being non-custodial.

Tom TrevethanSo, yeah, essentially, this comes in a modification of the Mercury API and uses the concept of these hodl invoices or hodl payments on Lightning, where you hold back the pre-image of the payment hash of the Lightning invoice. And so the Lightning payment can complete, but it doesn't finalize until the pre-image of the payment hash is revealed. So that anybody receiving a Lightning payment can confirm that the payment is there, but it doesn't finalize until this pre-image is revealed. And essentially, the Mercury API will generate this pre-image and this hash.

Tom TrevethanAnd only when both parties have verified that the, you know, the statecoin transfer has been verified and a Lightning payment has been verified, then both parties unlock. And then the statecoin transfer can complete and the keys are updated and then the custody is changed. And then the Lightning payment finalizes. So, yeah, I mean, maybe that was a complex explanation, but in essence, yeah. So effectively, yeah, it's just it's a way of allowing this atomic transfer of Lightning and statecoin transfer.

Vlad CosteaI'm very happy that I have such smart listeners who are engaging with the content because I wouldn't have come up with that one. I wasn't even aware that it's possible. I wasn't sure what latching refers to.

Tom TrevethanIt was just a name we came up with. So we came up with this, we developed the protocol for doing this only like in the past couple of weeks. And yeah, we wanted to come up with a kind of a snazzy sounding name, you know. And latch is obviously something you have in a door that locks it. So, you know, Lightning Latch kind of, you know.

Vlad CosteaIt sounds like a subatomic swap or something. Submarine...

Tom TrevethanAll right. I guess it's similar conceptually to that, yeah. But essentially, yeah, you're doing a swap of a Lightning payment for a Mercury coin, and that, yeah, there's a lot of use cases for that.

Vlad CosteaWhere do you see the project in five years if it succeeds? You see it being this bridge between various layer twos of Bitcoin?

Nicholas GregoryI think more use cases are going to appear that we haven't thought about with high transaction fees. But ultimately, I think people will use this. If you're doing regular settlement, I think this is where this fits in. You know, just think of gold bars. You can't, you can move this without liquidity constraints. It's, you know, it's not suited for like specific payments like Lightning, but it doesn't have the liquidity inefficiency of Lightning. And that's where I see it fitting in. In terms of retail use, it's probably going to be novel things like ordinals.

Nicholas GregoryMaybe that's probably the limit of it being used from a retail point of view. But for things like, I mean, I'm a believer in Fedimint-type solutions. I don't know if it's going to be Fedimint, but we do look a lot at Fedimint because of the work we do at CivKit. But if you think of CivKit, they're essentially order books, which is run by kind of like a federation where people can trade with each other. They're going to have to settle with other order books. They're going to have to settle with other services. Lightning, you know, has a use case there, but I think that there's something missing. And that's where MercuryLayer would fit in.

Vlad CosteaI'm actually very curious to find out more about your takes on this Uncle Jim kind of approach, which reinvents banking with Fedimints and stuff. But before that, let me read another ad and then we can talk about that. So this is from Hodling, which is a service from Switzerland, and they provide this consultancy for large holders and businesses and anything which requires multisig. And they have this message at Hodling S.A. We're happy to help simplify the complexities of Bitcoin security, providing peace of mind for a secure, private and self-managed environment.

Vlad CosteaIn the Swiss tradition of precision and security, our services assist you in setting up, maintaining and operating a self-custodial Bitcoin infrastructure. From the selection of multisignature configurations and hardware wallets to ensuring your setup is robust and tailored to your needs. Check out their service at hodling.ch, ch being the domain for Switzerland. And now, Nicholas, tell me more about why you're so bullish on Fedimints, because to me, it seems like they sort of defeat the purpose of Bitcoin. They are custodial. You have to trust someone that they're not going to steal from you. They do benefit from scalability and privacy, but they're not trust minimized in any way.

Nicholas GregoryWell, I think we've always taken the approach we work with the Bitcoin we've got, and we don't pray for opcodes or changes that may never happen or end up in political debates. But if you look at the Bitcoin we've got right now, it's probably one of the better solutions. But there's a lot of caveats there. I don't see a world where there's a bunch of Fedimints that, you know, we all interact with and we take their token because they would get whacked, they would get shut down. But in terms of smaller quorums that basically do shared custody, maybe share management of a Lightning node, maybe optimize that Lightning node with MercuryLayer, I think that could work. But they have to stay small. I think, you know, I've been on your show before with Riard from CivKit. He says an army of mosquitoes. I mean, an army of mosquitoes is hard to take out. If we end up with 10 Fedimints, they're going to get nuked. I mean, they're going to get shut down by regulators. They're going to get shut down.

Nicholas GregoryAnd that's what we saw. If you look at what do people mean about free banking in the UK, we obviously had banks. Middle class people didn't use them because they were too expensive. A bunch of middle class people cooperated together and created building societies. Those building societies worked well, but they became very big and ended up becoming regulated like banks. I think Fedimints have to stay very small. I have no idea what the number is, but if they become big, they're going to get regulated. But if me, you and Tom and a couple of other guys have some smart software that can share a UTXO or share some Bitcoin between us, share a Lightning node, that potentially could work.

Nicholas GregoryI mean, there's a lot of caveats there, but that's the way I see Bitcoin scaling with the Bitcoin we have right now. And I'm not worried. I don't believe in this stuff that Bitcoin needs to support 8 billion people. Half those people are not going to use Bitcoin. They're going to use an ETF or Coinbase. We don't necessarily need to support 8 billion people, but we need to support enough people that stay self sovereign to keep the system honest.

Tom TrevethanYeah, I think the fact, the reality is that we're really interested in technology and want to come up with the best solutions we can, keep Bitcoin non-custodial, but the only real thing that works for payments at scale currently is this custody. And eCash is the best way of doing this, like privately, but yeah, I am worried that people aren't going to get away with this at scale with regulation and KYC requirements. I think, like Nick says, the only way this can work is with small kind of issuers, and we'll see. We'll see how this plays out, but I think it's definitely going to be part of the puzzle.

Vlad CosteaYeah, I just wanted to mention that we crossed 1,000 views on Twitter slash X, which is pretty cool because on YouTube we have about 10, which is sad. I don't know what's wrong with YouTube, but too many ads, and I guess there's something wrong with their notifying system because the subscribers don't always see that something new gets posted. I guess they changed the algorithm to prioritize some other types of content over others, but it's good. We have 1,000 people who tuned in on Twitter slash X, which I think...

Nicholas GregoryOn a Tuesday morning, I'm a bit surprised it's that high, so that's great.

Vlad CosteaYeah, it's really cool. And there's going to be more in the coming days, so thank you guys for tuning in. I guess I should ask you a last question because you said you can only stay around for one hour, and we should keep this short and efficient. How can builders create something with MercuryLayer, and ideally, what do you think should get built that's super useful right now?

Tom TrevethanWell, yeah, I mean, they can build anything that enables to transfer an amount of Bitcoin between any people, and you want to do that instantly and for zero marginal fee, but there's lots of... I mean, I think as we've said before, I think the fee pressure will start to see use cases kind of develop, or that people will be incentivized to kind of integrate this. I mean, all these scaling solutions suffered like Liquid. I guess Liquid's big issue was the fact that for most of its existence, Bitcoin fees have been quite low, and so people aren't really incentivized to use or to go through the effort of trying to integrate with it since main chain fees are so low, but I think that's going to change soon quite dramatically.

Tom TrevethanAnd then, yeah, people are going to be incentivized to use these technologies, which can give a chunk of efficiency and can save them a significant amount of money. But yeah, I think we're kind of exploring now the different ways that this can be easily integrated with the kind of existing things and exploring kind of new use cases.

Vlad CosteaI like how you try to keep it diplomatic and vague in the sense that you don't necessarily say who should use it, you say anyone who needs this. So I guess this is a message for builders.

Tom TrevethanI guess you can look at, but yeah, you can look at different classes of people. I mean, privacy, there's people that do want privacy and that's their main thing. They're not necessarily concerned about the speed of the transaction or the cost of the transaction. And yeah, the different people want different things. Some people want something that's instant and they don't mind about fees. Some people want to save money on fees. Some people want privacy. And yeah, I mean, it gives you all of those things.

Vlad CosteaYeah, that's really cool. Actually, I lied. I have another last question, but I don't want to make it too long. So you mentioned CTV and the fact that it can help MercuryLayer. And I think there are layers to this, no pun intended. So first of all, do you believe that Bitcoin needs a covenant soft fork? And which one do you prefer? Because I have had Andrew Poelstra on the show last month and he was promoting OP_CAT, saying that it's superior to OP_CTV because it opens up more use cases.

Vlad CosteaAnd then there seem to be a lot of people on Twitter saying, no, we shouldn't do CTV. We should do LNHANCE or whatever, which is CTV with something else. Do you have any opinion about this and is there anything that's useful for MercuryLayer?

Tom TrevethanSo I mean, my personal view, I mean, I don't think, I mean, there's been enough analysis of OP_CAT and OP_CTV that there's no real risk to these. And so I think people kind of arguing about it, quite depressing. And the fact that it's taken so long, I mean, in the past, a lot of these things, new opcodes, obviously SegWit was a very big change, Taproot was quite a big change, but previously to that, just activating a new opcode like OP_CLTV, it was done and there was no controversy.

Tom TrevethanPeople just did it. They thought it might be useful for some future application. So just why not just add it? It's really easy. And so, yeah, it's a bit depressing that people are kind of like fighting so hard over this. But yeah, I mean, personally, OP_CAT, I think is more general. There's a lot more use cases to OP_CAT and you can do a lot more things with it. And I think scripts should be kept as simple as possible. So the simplest like base opcodes that people can then use to make more complex scripts to implement more kind of complex functionality.

Tom TrevethanYeah, people should just be given the tools, the basic tools, but I have no objection to either OP_CAT or CTV. Yeah, I don't know, you're a fan of both, Nick?

Nicholas GregoryYeah, I mean, sometimes I wish we would focus a bit maybe on a better scripting language like Simplicity, so that, you know, every two years we don't basically have another code war, if that's the right word, or fight for another scripting change, because I think it's getting a bit tiresome. And I think it becomes very religious. And sometimes if we just had something like Simplicity added and then, you know, Paul Sztorc can go build his drivechains, we can build better versions of statechains and the Lightning guys get what they want.

Nicholas GregoryThat might be a more longer term thing, because I do think it's an energy drain. These opcode battles and they get just religious. The argument at some point does not become logical anymore. It becomes religious. I don't like the way Jeremy Rubin was treated. I think, you know, he did build some cool stuff. He was a smart. It was and is a smart person. But it became religious when, you know, he was trying to push CTV and, you know, people say there's no use case, there's no there was no POC.

Nicholas GregoryI think that's BS. We need to incentivize smart people like Jeremy to build on Bitcoin. Otherwise we are going to ossify and we're not going to get cool layer two, layer three scaling solutions.

Vlad CosteaYou know, I think Jeremy made some mistakes with the way that he promoted CTV in the sense that the average person did not understand what this is about and it was easy to frame as an attack saying, oh, no, he's trying to destroy Bitcoin because you're going to have these vaults that are going to get regulated and people are not going to be able to move their funds anymore if they join the vaults and whatever. I think this was the conversation a couple of years ago and then people started presenting more logical use cases and explaining, you know, we can open more Lightning channels.

Vlad CosteaWe can do this and do that. We can create much better trust-minimized sidechains with this.

Nicholas GregoryBut I also want to pick up here. I don't think I mean, I'm pro CTV. I mean, we've spoken a lot to those guys about how it could improve statechains. If we got CTV this year, it's going to be years before there's going to be decent software you can use. I mean, I still don't think we've really done a good job of using multisig. Yeah, and why? Because it's quite hard to get the UI right for those kind of scripts. I mean, I think you probably would admit multisig is not easy to use. I mean, a lot of people screwed that up and you can only imagine something like CTV is going to be 10x harder. So I think sometimes it is portrayed as, oh, let's get this opcode and Bitcoin suddenly scales, which is not the case.

Tom TrevethanYeah, no, it's kind of, I don't know, having worked on Bitcoin software for years now, the idea of these shared UTXO models with OP_CTV and making that into a usable UI kind of, you know, brings out like anxiety attacks. It's going to be a nightmare. But you know, these things take a lot of time to get right, you know, and it's not a magic bullet. People kind of promoting, people like promoted Lightning as this magic bullet that solves all our problems. No, I mean, obviously, CTV will be the same.

Tom TrevethanIt gives you a boost, it gives you a wedge, it enables us to get a bit further down the path of, you know, making Bitcoin scale, but there's a lot of hard work.

Vlad CosteaNo, Bitcoin is still a very young project. And that's what a lot of people don't realize. The way that we use it today may not be the way that we're going to use it in 10 years. Even if the base layer is not going to change fundamentally, the user experience is going to be very different. And it's very cool that there are all of these projects and all of these experiments. I'm very happy that MercuryLayer exists, and I'm proud of you, Nicholas, and proud of you, Tom, because you basically took a wallet that was super cool, but for some reason did not become popular, and expanded the concept to the point that now you have your own layer, your own sidechain-like being, which does not store anything except for the UTXO states, which is super cool.

Vlad CosteaI wish you best of luck with this, and I hope that there will be some developers listening to this and coming up with creative ideas about stuff that they can do. And what is the best place? So maybe the people listening on Spotify will not get access to the visual side, so they will not see that I presented the website. So I'll let you say where can they access your work and where can they find out more and maybe build on MercuryLayer?

Nicholas GregoryYeah, so we have the MercuryLayer website, which we keep very informative. We're quite active on Twitter. We try and do like one or two minute video clips every month just to explain some of the tech we do, because we realize that in long form, we kind of lose people. So we try and do these like 30-second, one-minute clips of saying, what's Lightning Latch? What's MercuryLayer? And then if you want to dig deeper and need some support, just come to Telegram. A lot of people have come to our Telegram group.

Nicholas GregoryThey DM me or they DM Tom privately to see how we can help them build or bounce some ideas. And we want that we want to create a builder's environment. And I mean, we're lucky in that one of the fans of MercuryLayer is Nostr, and, not Nostr, sorry, is fiatjaf, who invented Nostr. But we would love to have that kind of Nostr culture, builder culture on MercuryLayer. I mean, I'm not sure if that's achievable, but that's what we want. And that's yeah, we work. We kind of play with Nostr in the work we do with CivKit.

Nicholas GregoryBut yeah, we want to create that kind of builder environment, builder ecosystem on MercuryLayer. That's that's our goal here.

Tom TrevethanYeah. Also I'd add just GitHub as well: commerceblock/mercurylayer basically has all the documentation, instructions how to use the test network and yeah, any kind of, you know, people can, any issues with it or, you know, can kind of, you know, leave issues there. And yeah, Telegram is a good place to go. And just if you just want to kind of, you know, have a chat or, you know, explore some ideas.

Vlad CosteaYep. Nicholas, Tom, I'm super proud of your work. Thank you for joining my show. I wish you best of luck with the MercuryLayer. ================================================================================ Production notes: CORRECTED. Base: local Whisper medium.en (int16) transcript with automatic speaker diarization (recorded live), compared word by word against the YouTube auto-captions of the livestream (captions run about 47 s behind the feed audio); disagreement spots were checked against the audio. 49 short audio clips covering doubtful, dropped or truncated stretches were re-transcribed locally with faster-whisper medium.en (int16), including Vlad's question and Tom's "neither and both" answer at 00:11:18 (partly dropped in the base), the Wasabi Wallet introduction at 00:23:16 (dropped in the base), the Wasabi/nothingmuch passage at 00:24:56, Paul Sztorc's listener message at 00:41:20, the "statists" joke at 00:46:07, the Lightning Latch exchange at 00:53:47 and the CivKit "army of mosquitoes" passage at 00:57:53. Speaker labels were checked with a full-episode local speaker-embedding scan (TitaNet reference voices for Vlad, Nicholas and Tom) and fixed by hand (Vlad Costea, Nicholas Gregory, Tom Trevethan); turns split mid-sentence by the diarization were rejoined. Names and jargon were checked against the title, the feed show notes and the web: MercuryLayer, Mercury Wallet, CommerceBlock, CivKit, statechains / statecoins, Ruben Somsen, ANYPREVOUT, Schnorr, Taproot, blinded signatures, MPC (multi-party computation), two-party signatures / key shares, unilateral exit, Liquid, RSK / Rootstock, drivechains, Paul Sztorc and LayerTwo Labs, BitVM, Ark and VTXOs, Lightning, Lightning Latch (hodl invoices, pre-image), Cashu, Fedi / Fedimint, Obi (Fedimint), Nostr, fiatjaf, Ordinals, Wasabi Wallet, WabiSabi, nothingmuch (Yuval Kogman), Adam Ficsor, Max Hillebrand, ShopinBit, Tor, I2P, CoinJoin vs coin swaps, Samourai, Opendime, Satodime (Bastien and Baudoin, promo code BTCTKVR), Advancing Bitcoin, Bryan Bishop and Webcash, Matthew Croughan and NixOS, Hodling SA (hodling.ch), OP_CAT, OP_CTV / CTV, LNHANCE, OP_CLTV, Andrew Poelstra, Simplicity, Jeremy Rubin, WBTC. Sponsors: Wasabi Wallet, Satodime and Hodling SA, read live by Vlad, plus Paul Sztorc's LayerTwo Labs message. Light clean verbatim: ums, false starts and stutters lightly trimmed; meaning not changed; factual slips and jokes (e.g. "Peter McRetard" in Paul's message) kept as spoken; the opening music is omitted. Remaining doubtful words are marked best guess.

Open the transcript file

Hosts and Guests

Guests

  • Nicholas Gregory Guest Feed-stated
  • Tom Trevethan Guest Feed-stated

Credit the people on this episode

Be the first to rate this episode

More episodes

All episodes
  1. S17 E52: Michael Sutton on Kaspa, Covenants & Research

    Nov 19, 20262 hr 59 minAudio

    Michael Sutton has been with Kaspa for eight years: first as a researcher, then as the engineer who rewrote the entire codebase in Rust and unlocked 10 blocks per second. He is the co-author of the DAG Knight paper with...

    Credit the people on this episode

  2. S17 E51: Martti Malmi on Iris, NostrVPN and Freedom Tech

    Nov 13, 20261 hr 13 minAudio

    We recorded this at Bitcoin Amsterdam, the Bitcoin Magazine conference, with legendary cypherpunk and Satoshi contributor Martti Malmi. This is his second time on the show. Two years ago in Prague we talked about 2009,...

    Credit the people on this episode

  3. S17 E50: Daira-Emma Hopwood on Zcash at 10, Halo, ZSAs and the Fight Over Grants

    Nov 7, 20262 hr 4 minAudio

    We recorded this conversation in Mexico, at an undisclosed location, during the celebrations for Zcash's 10th anniversary. Daira-Emma Hopwood has been with the project for 11 years, since before launch, and came to it...

    Credit the people on this episode

  4. S17 E49: Kris Nuttycombe on Zcash at Ten, Hard Forks, the Dev Fund & Why Ossification Is Death

    Nov 1, 20262 hr 24 minAudio

    Kris Nuttycombe of the Zcash Open Development Lab came back to the show two years after his first visit, this time on a hotel sofa at Zcon7 in Cancun, on the last day of the conference and around Zcash's tenth birthday....

    Credit the people on this episode

  5. S17 E48: Robin Linus in the 1000 BTC Car, Driving to BitDevs Atlanta

    Oct 24, 20261 hr 20 minAudio

    This episode was recorded while driving to BitDevs in Atlanta. The car belongs to Art (at least 1/3 of it) and is the first vehicle ever bought for bitcoin in 2013: a 2011 Toyota Prius, for which the Mycelium wallet...

    Credit the people on this episode