0:00 Good morning. Good morning. 0:04 All right, that's the Drivechains fan club over there. Let's go. Let's go. 0:12 So, very excited to have this conversation. 0:15 We're joined today by 0:17 two of the most OG Bitcoiners, in my opinion, two of the the wizards of Bitcoin. 0:22 And, you know, y'all are very lucky to have an 80 IQ person moderating this discussion. 0:28 I think it's going to make it very valuable for everyone here. So, 0:32 to get started. 0:34 First off, maybe introduce yourselves very, very quickly. 0:38 Yeah, I'm Peter Todd. I seem to be mainly known for going to a ton of conferences. 0:44 Okay, how long have you been in Bitcoin? A little bit more information, a little bit more. I read the white paper in, like, 2009. 0:49 You know, started doing some stuff in 2014 and just kept on doing stuff. Done a bit of stuff in Bitcoin Core. Cool. 0:55 Paul Sztorc. I think, you know, I've been like a Bitcoin researcher for, like, more than 10 years. A long time ago, 1:03 I wrote this essay called 1:05 Nothing is cheaper than proof-of-work. And that was a sequel to a related essay 1:10 I wrote first that Adam Back at one point linked to on BitcoinTalk.org. So, I guess that was sort of, like, my big break or whatever. 1:18 Cool. 1:19 So, we have, I think, 30 minutes. 1:22 And I wrote 20 questions I wanted to ask. So, we are going to jump around a little bit. We're going to move quickly. 1:28 Some of these are going to have to be a sideline. 1:32 So, 1:33 to start with, let's just level set very quickly. Sidechains. Can someone briefly... Peter, 1:38 can you briefly explain what a sidechain is? And can we level set, do we think sidechains are a valuable thing? Well, very briefly, 1:47 sidechain is an altcoin, but tied to Bitcoin. So, you're trading Bitcoin. Okay, so all sidechains. 1:55 That's where the term came from. 1:56 All right. 1:57 You know, it was meant to be like a something to the side of Bitcoin, 2:00 denominated in Bitcoin. And there's a whole bunch of different varieties of sidechains. 2:03 But, you know, the key concept was, hey, can we do another coin without doing another coin? Yeah. And do you think they're valuable? 2:10 Sometimes. Okay, so we do think there are sometimes that they are valuable. 2:13 I mean, Liquid has a small market of people who actually use it. Not very big markets, but for some circumstances, it could be useful. 2:20 Okay. Would you have a different answer to that question of what a sidechain is? I think a sidechain is like a 2:26 universal altcoin like emulator or something. So, whatever Zcash is doing, you send your Bitcoin over there and 2:34 then you can do the Zcash things and then you can bring it back to L1. So, I would emphasize 2:40 about the sidechain that it 2:43 I would say that it's like the key to really understanding it is actually competition. 2:48 You have, instead of only having Bitcoin core, you have developers compete over who gets the coins. 2:54 I think that's really the essence of it, actually. Okay, so Drivechains is really an implementation of how to do a sidechain. 3:00 I remember when I was in Bitcoin. 3:04 Framework. A framework. A framework. Okay. Thank you. 3:08 Again, 80 IQ, 80 IQ. All right. 90. 3:12 So, when I was a young lad in Bitcoin, 3:16 sidechains were always, not always, maybe depending on what part of the community you're from, but was like a holy grail of Bitcoin. 3:23 It's like, okay, it's the 3:26 technology to end all the shit coins. All right, though we didn't really call it that at the time. 3:33 Now, it's like we can't really exactly get agreement on whether sidechains were ever a thing that was desirable in the first place. 3:39 So, it's strange to see the perception, I guess, within the community change about, like, do we even want sidechains? 3:46 Adam recently described the Drivechains implementation as 3:52 DMMS, 3:54 dynamic membership. 3:56 Yeah, or multi-sig. Multi-sig, I think. Yeah, dynamic membership, multi-signature. Yeah, yeah, yeah, yeah, yeah. 4:03 It has the word dynamic and signature in it. Yeah. 4:05 So, do you all think that's a correct description of what Drivechains is? I think it's a cool, cool concept. 4:13 Well, I'll disagree with you on the cool concept, but it matches, you know, what Blockstream originally proposed in their 4:20 Merge Mine sidechain paper. 4:23 Do you agree? 4:26 That is a DMMS, but I don't know. I mean, the important thing about Drivechain is, you know, 4:32 it has potential to give us planetary scale, like, 4:36 immediately, and Zcash-level privacy, and it doesn't affect the L1 4:42 features of Bitcoin for the people who aren't using it, and you have security levels comparable to Lightning. 4:46 Maybe we should explain for a quick second, like, what 4:49 Blockstream meant by DMMS, which is the idea that 4:53 you can have a signature scheme. The question is, who signs it? Well, in Blockstream's proposal, 4:58 it's miners sign it, which, in the context of something like Bitcoin, means miners can decide to go and take 5:03 coins and move them somewhere else, and it's purely based on what miners vote to do. 5:07 It means those miners can change dynamically with whoever are the miners, rather than having set identities. 5:13 Yes, the dynamic refers to the fact that anyone with hash power can start mining. Right. That's the part 5:19 I think is cool. But I, so, as a framework, though, people in the audience, they can think of Drivechains, 5:24 it's this sidechain that uses some sort of framework around a 5:30 multi-signature setup, where the miners are the determinant of... 5:33 So, I think from a tech point of view, DMMS makes a lot of sense, 5:36 but I think for, like, a general audience, it really does come down to saying, hey, 5:41 in normal Bitcoin, miners can't go move money, because they can't fake a digital signature. In 5:45 Drivechains, as well as mergemine sidechains, the key console would say, hey, why don't we let miners decide where money goes? 5:52 Okay, so, 5:56 maybe we come back to that question, but, so, I want to talk about 6:01 some of the risks around, specifically, 6:05 Drivechains. So, 6:07 first off, everyone up here has been cancelled before. It's part of the reason I respect both of y'all a lot, alright? 6:13 One of your cancellable opinions that you have is that Bitcoin has a security budget problem, and that we need to 6:21 re-evaluate the 21 million hard cap as a potential solution to Bitcoin's security problem. Emphasis on potential solution. 6:27 There's a bunch of different ones. Potential solution. Okay, 6:30 you know, when I look at Drivechains, 6:33 part of the reason I think Paul's excited about Drivechains is he sees it as a potential solution to the security budget problem. 6:39 You know, why, 6:41 you know, why do you think that a proposal of increasing the hard cap of Bitcoin is less risky than a Drivechain's 6:49 implementation is? Oh, I wouldn't necessarily say it's less risky. 6:51 I think you're more likely to have other ways of fixing that problem. 6:54 But, you know, certainly someone needed to go and talk about how that, you know, how that kind of thing works. 6:59 So in, you know, 10-20 years into the future, there's plenty of time to go think about the different trade-offs. Sure. 7:04 But, you know, I also got to stress, like, I don't believe Drivechains actually does solve that problem. 7:10 You know, the problem with Drivechains is there's no block size limit. 7:13 So there's no reason to expect transaction fee demand to be anything but minimal. Much the same way is that, you know, 7:18 in the block size wars, those exact arguments that people gave against IDA, well, we'll just go increase the block size, transaction fees will flow in. 7:25 It's like, well, unless you actually have a limit, you know, it's a supply and demand situation. Supply is infinite. 7:33 Whatever the demand is, price is still going to wind up being zero. Right. 7:36 So that was my next question is, like, do you think it'll actually generate more fees? Now, Paul, you have a different perspective on this. 7:42 Obviously, I mean, because it won't literally be zero, of course. 7:45 It can be any small amount. And then as long as you're increasing the quantity of sales, so to speak, you sell more transactions, even if it's one cent per transaction. 7:53 Of course, if each new transaction will be increasing the amount, it's only if it's literally zero that it would have no effect, which obviously won't happen. 7:59 I mean, nearly literally zero times a finite number of transactions is still basically zero. 8:06 But this isn't a math proof we're talking about. This is economics. 8:09 So, you know, Visa charges only a small amount, but they generate, you know, hundreds of millions of dollars per day. 8:15 Visa charges, like, multiple percentages for a lot of people. I mean, it's not a small amount. They've got a monopoly. 8:20 So I started off by asking, are sidechains a valuable thing? OK, so we're talking about fees. 8:26 We're talking about, OK, is there something that people want to actually do that they're going to pay fees for? 8:30 I guess that's the question, though. 8:31 No, I don't think that's the question, though, because the problem is not, is there something people want to do that they'll pay fees for? 8:36 It's, what is the mechanism by which there's a reason for them to pay anything but a trivial amount? 8:42 You know, this is the same problem we have with infinite block sizes. 8:45 The fact is, fee revenue gets driven down to zero because there's nothing stopping liners from underbidding each other. 8:51 So I don't think that's actually an argument for Drivechains. 8:54 So you think that if a Drivechain got popular, it could result in less fees being paid on the layer one? 9:01 Oh, absolutely, yes, because Drivechains in that circumstance, you know, 9:04 if the Drivechains are scaling, can represent people moving away to a cheaper system and less fee revenue coming in, 9:10 versus something like Lightning, where there are trade-offs between different layers, 9:14 and there's a very clear thing of, well, if you need to do an on-chain transaction, there is that much space, and you better bid for it. 9:19 Now, I'm not going to argue that Lightning's perfect in that respect. 9:22 I mean, this is why I advocate for built-in suspenders, but I don't think Drivechains can make that argument. 9:28 Well, it's very easy to make the argument, though, because the miners themselves are the ones who set, like, the minimum fee rate. 9:33 So if they find that their fortunes have decreased by going this route, they could just disable the Drivechain. 9:38 But you're arguing for a cartel. 9:40 I mean, you're arguing for a cartel. 9:42 If it was truly a free market, someone would come in and underbid. 9:45 No, absolutely not. 9:45 I'm saying that if they don't want the block size effect to reduce the fee rate, 9:51 they can either cap the size of the sidechain blocks or cap the min fee rate. 9:56 It's the exact same thing. 9:57 But again, when you say they can, you're really talking about a cartel here who can make a decision. 10:01 In a free market system, that's not what would happen. 10:03 Absolutely not. 10:03 You could just have some formula do it. 10:05 And if it really maximizes miners' revenue, then they will do it, of course. 10:09 Obviously. 10:09 You advocate that people can go create new Drivechains. 10:11 Surely, if there's some formula, people will create another Drivechain that underbids this. 10:15 I mean, there is no mechanism to set that limit. 10:18 The mechanism is that the miners will maximize their fee revenue. 10:21 Again, it's a cartel. 10:22 This is a good— 10:23 No, but it's not a cartel. 10:24 You can do it with a formula. 10:25 A good point of contention here. 10:27 Very easy formula. 10:28 I can underbid your formula. 10:30 One question I have is Lightning Network. 10:31 Is Lightning Network net adding to the on-chain demand on Layer 1? 10:37 Or is it cannibalizing more fees than— 10:41 I think it can definitely go both ways. 10:44 But the thing with the Lightning Network is with current Comp Sci, there is no other alternative. 10:49 Nobody's figured out a way to send single Satoshi payments over a blockchain. 10:55 And it's not going to happen. 10:56 And that's a strong need. 10:59 Now, maybe we'll figure out something truly novel in Comp Sci. 11:01 Like Drivechains. 11:02 Well, I don't think Drivechains is novel. 11:05 But, you know, I mean, that's just an inevitability with Lightning. 11:08 Now, certainly in a lot of circumstances, Lightning can mean that the fees you do pay on Lightning channels 11:14 can then flow back to Lightning nodes, which can then go pay much higher fees to bid to open and close channels, 11:20 as well as adjust capacity with splicing. 11:23 And in that circumstance, yeah, it can definitely flow more fees into miners. 11:27 But, you know, I would not make the argument that that's a guarantee. 11:31 Okay, so I want to keep moving just because there's other topics I want to hit on. 11:33 But this is, I think, an important one to highlight is, will it actually drive more security budget? 11:40 Okay. 11:41 So, quick question. 11:43 I think I know Paul's answer to this question about how many soft forks will exist in the future 11:49 and maybe what Drivechains can do there. 11:51 How many soft forks do you think, over the next hundred years, Bitcoin's going to have? 11:55 Well, I mean, it absolutely has to have at least one, because Bitcoin ceases to exist in, you know... 12:01 That's a hard fork, though. 12:02 Well, that particular example is kind of a bit of both. 12:05 It's a weird example. 12:06 But, you know, if you're talking about hard forks, I mean, soft forks is basically the same discussion, right? 12:12 It's like an even more extreme version of fork. 12:14 Now, I think it would not be surprising, you know, backing away from that extreme, 12:20 that we get Bitcoin to a state where it's essentially perfect. 12:24 In much the same way as that TCPIP. 12:27 Other than IPv6, which comes from the 90s and is gradually getting adopted, 12:31 it really hasn't fundamentally changed much for a very long time. 12:35 Because we figured out something that was essentially perfect. 12:38 It might not have been absolutely perfect. 12:39 It was more than close enough. 12:40 And what did we do? 12:41 Well, we built layers on top of it. 12:42 In much the same way as that we're taking Bitcoin, we figured out how to go make lightning work. 12:47 And we built that as a layer. 12:48 And now you have this enormous amount of innovation happening above that layer. 12:52 But we needed a soft fork for it. 12:53 We needed one. 12:55 Technically, the way we implemented it was like three. 12:58 But we needed, you know, one set of changes and that was that. 13:03 Now, in the future, with things like ARK as an example, 13:05 it won't surprise me if the community as a whole says, 13:07 hey, here's this new thing we need. 13:10 Everyone's going to use it. 13:11 Let's do that one fork and that's that. 13:13 But this is a rare event. 13:15 Whereas most of the innovation that happens is things that happen on layers above. 13:18 I mean, RGB is a wonderful example. 13:20 They're implementing all this semi-off-chain, like layer two-ish stuff 13:27 that can trade assets in very efficient ways. 13:31 But they don't actually need a fork to do that. 13:33 My own Open Timestamps project is a much more boring example of the same thing. 13:37 Open Timestamps, not only does it not need forks to exist, 13:39 like it's basically impossible to stop. 13:42 That's the kind of innovation that excites me. 13:45 So to summarize, you think over the next 100 years, 13:48 we're talking about a small number of soft forks. 13:51 It could easily be less than a dozen. 13:53 Easily. 13:54 Okay. 13:54 So, Paul, one reason why I like Drivechains is because I feel like 14:00 there's a lot of changes in Bitcoin's future, 14:02 like a lot of things that need to happen. 14:04 And it creates an environment where we can do some of those changes in a low-risk way. 14:08 Do you think that that's a correct description of the value prop of Drivechains? 14:13 And yeah, how do you see that playing out? 14:16 Well, yeah, certainly the developers are going to continue to invent things. 14:19 And so the question is, what are they supposed to actually do with their invention? 14:24 And one route is to try to get the invention, persuade people to add it to Bitcoin Core L1. 14:31 But that takes a very long time. 14:32 And I think it's been getting more and more difficult as the years go by. 14:36 And there's also more and more people inventing more and more things. 14:38 So there's more specialists in different directions. 14:41 And so what are these people supposed to do with their invention that they make 14:45 and that they really like and that maybe they've convinced other people is useful? 14:48 ARK is a perfect example of something that we would have already had, 14:51 probably in 2019 or even earlier. 14:54 Because if you look at the timeline of when Jeremy Rubin was inventing CTV, 15:00 that was a long time ago. 15:01 And he tried to get it through the Bitcoin Core process, 15:04 but it just found it to be too time-consuming. 15:07 Well, he failed to convince people it was useful enough. 15:09 I mean, it wasn't important. 15:09 But the irony is now people have flipped on that. 15:13 And then now they think he was right and it is important. 15:17 Not yet. 15:18 ARK is interesting. 15:19 It's not there yet. 15:21 So this kind of moves a little bit towards the activation discussion. 15:24 But something I wanted to ask you about, Peter, is BIP-125. 15:29 All right. 15:29 You have to refresh me. 15:30 Which one was that? 15:32 Full replaced by fee. 15:34 Well, no. 15:34 Optin replaced by fee. 15:37 Optin is 125? 15:38 OK. 15:39 Yeah, yeah. 15:40 So the thing is, though, BIP-125, it's got nothing to do with consensus. 15:45 I mean, it's not related to this discussion directly other than saying, 15:49 yeah, you can go build things on top of Bitcoin. 15:51 Well, I guess where I see parallels is that there's conflicting visions of Bitcoin. 15:57 And I remember when everyone who was maybe wanting to make Bitcoin good for payments, 16:03 they're like, we want zero conf because it's a good user experience. 16:06 See, this is not a great example of this. 16:09 Because this is a matter of conflicting visions of Bitcoin in general. 16:12 This is more, there's a set of vision for Bitcoin that's just insecure and doesn't really work. 16:17 And then there's a set of vision of Bitcoin, which basically all devs are saying, 16:20 well, obviously you do it that way. 16:21 In fact, I mean, you know, could ask the audience, go raise your hands, 16:25 like how many people would wait a confirmation before accepting payment from a stranger? 16:29 I mean, people have always said, we'll go wait. 16:31 Yeah, I think replaced by fee has won the day. 16:34 So I'm not questioning that. 16:35 I'm saying like the full replaced by fee is not standard in Bitcoin. 16:42 Well, like, I think right now, roughly 35% of hash power. 16:47 When was it the first, when was it first proposed? 16:52 Well, I mean, a decade ago, the first time actual code was written was a bit under a decade ago. 16:59 So talk about just the process of like activate that. 17:01 Well, I don't let her complain. 17:03 I don't think that's a good example. 17:05 I have written a software, check-lock-time-verify. 17:07 That's a much better example because it's an actual software. 17:11 And the process for that was, well, all right, I go have a somewhat good idea. 17:17 Isn't that clever? 17:17 Check-lock-time-verify. 17:19 We start realizing payment channels need it. 17:22 Simply that basically bidirectional payment channels. 17:25 And very quickly, people figured out, well, you know, this looks really useful. 17:28 And we're going to go need it for payment channels. 17:31 We're going to need like one other mock code, 17:33 check-sequence-verify for payment channels too. 17:36 And from there, frankly, getting in was a fairly short process. 17:39 Because once you've done that kind of groundwork, 17:43 as things roll over to, hey, this is really needed, 17:47 we were probably able to get that in in like roughly a year and a half, 17:49 which was sooner than actual good implementations. 17:52 So what caused the success? 17:57 Did you go talk directly to miners to get miner support for it? 18:01 I didn't need to do like a ton of groundwork on that, 18:04 because it was such an obviously good idea. 18:07 It was something, I mean, bidirectional payment channels, 18:10 that's basically what's under the hood in Lightning. 18:14 It's obviously really important. 18:16 Everyone, with few exceptions, agreed this was a useful feature. 18:20 So the nuance of what exactly should check-lock-time-verify do 18:24 was really just a matter of discussing among some tech people 18:28 certain nuances of exactly how the opcode should do things. 18:31 We hit on the design very quickly. 18:33 Where I'd contrast that to, say, check-template-verify 18:36 is people just haven't been convinced that it's actually that useful. 18:40 Now, something like ARC could change people's minds on that. 18:43 But ARC still is not an idea that there's wide consensus that actually works. 18:48 It's an idea that not that many people actually understand. 18:51 Whereas payment channels, people figured out very quickly, 18:53 hey, this is obviously a good idea. 18:56 We should definitely do this. 18:57 And it got in quickly. 18:59 During that process, did you have critics? 19:06 Frankly, not really, because it was a very hard idea to argue against. 19:10 Taproot is probably your better example for arguing against. 19:13 Most of that was really nuance of exactly how detail should work. 19:17 Again, Taproot is an idea. 19:18 Once people got their heads around it, it seemed very obvious. 19:21 Oh, yeah, this is how Bitcoin scripts should have always worked. 19:24 It took a while, I think, for people to realize that. 19:27 But the idea spread very widely, because it's just such a good one. 19:33 Speak about your experience with this. 19:35 How was the activation discussion? 19:37 Check-lock-time-verify, that was back in the pre... 19:42 You had two eras. 19:45 I think the first seven years of Bitcoin, we had done 15, 16-something soft forks. 19:50 We had a ton of stuff to fix. 19:51 And then we had SegWit, which was very dramatic. 19:55 And then SegWit took 20 months from when it was 19:59 proposed to when it was finally coded and activated. 20:01 And then Taproot took 46 months. 20:04 So that's really what it is. 20:05 It's just a slowing over time of... 20:08 And it's really the two eras. 20:10 SegWit was so contentious that for many years, 20:13 people refused to even discuss soft fork activation, 20:15 which they still kind of do today, where they say, 20:18 we don't want to touch anything controversial. 20:20 Well, I should be clear. 20:21 SegWit is another key part of Lightning. 20:23 And again, I think SegWit falls in this example of, 20:26 this was so obviously needed. 20:28 And it also falls in the example of... 20:29 Oh, but it was highly contentious. 20:31 Well, it was contentious because it was a block-size debate. 20:33 It wasn't contentious because tech people thought it was a bad idea. 20:36 With, you know, a few exceptions of crazy people. 20:37 Why do you think this idea is bad? 20:40 Well, I think Drivechains is bad because it's basically a 20:43 let's-go-trust-miners system. 20:44 Yeah, but you're mixing up L1 and L2, though. 20:47 But again, it's... 20:48 It's only the L2 people who opt in. 20:50 You are saying that we should have large amounts of coins 20:53 tied in such a way that miners can just go steal it. 20:55 That fundamentally changes how mining works right now. 20:57 Yeah, but if miners steal from the Lightning network, it's the same problem. 21:01 But they're not able to do that on the scale of Drivechains. 21:04 Also, they're not able to do that, for instance, under a court order. 21:06 You know, one of the things that I think a lot of people get really... 21:09 The court order thing is silly. 21:10 ...get really worried about with Drivechains is 21:12 you're now creating Bitcoin in a situation where 21:14 there are targets that can actually do things. 21:16 I mean, I personally, for instance, am being sued by Craig Wright 21:19 to try to go reassign Bitcoin. 21:21 And it's a very real lawsuit. 21:22 You know, it's quite a silly thing to worry about this argument from the lawsuit. 21:27 You could even flip it around and say, 21:28 this is what will decentralize mining 21:30 because no one will be able to find them or something, if you want. 21:32 But the whole thing is kind of silly. 21:33 Anyone could be sued at any time for any reason. 21:36 Yeah, but it's a much bigger threat if the lawsuit's valid. 21:39 It's easy to... 21:40 See, I can easily defend my lawsuit because I can't actually do that. 21:44 The Lightning network, at different scales, that is part of the problem, 21:47 is that on the Lightning network, 21:49 miners can, like, mosquito attack individual channels. 21:53 They need a lot of coordination with a lot of people. 21:56 No, it's the same 51% cartel. It's exactly the same. 21:58 All they have to do is censor this justice transaction. 22:00 But the truth is that they won't do either 22:02 because miners want Bitcoin to be a success. 22:04 So, of course, they're not really going to steal from any... 22:07 In the Drivechain, though, all of the fees... 22:09 This is a key difference. 22:11 In Drivechain, all of the fees go to the miners. 22:14 So this is going to be, like, eventually... 22:15 Well, again, I don't believe fees will be significant. 22:18 Each new fee is... 22:20 It's like planetary scale merge mine transactions 22:23 would be hundreds of millions of dollars per day. 22:25 It's already on... 22:25 Ethereum is already... 22:27 Bitcoin only does, like, $500,000 a day, 22:29 and Ethereum is already doing millions of dollars a day. 22:32 To be clear, Ethereum does have a block size limit 22:34 because Ethereum is breaking at its current scale. 22:37 I mean, they can't increase the scale of Ethereum right now. 22:39 Yeah, but there is... 22:40 Each individual sidechain will have a block size limit. 22:42 Why do you say that? 22:43 I mean, I can out-compete your sidechain with a block size limit 22:48 The real question is, why would the miners mine it at all 22:51 if it generates $0 in revenue? 22:53 The whole point is to collect the revenue. 22:55 I mean, again, this, I think, gets back to your fundamental difference 22:58 about what centralization pressure is. 23:01 Overhead matters. 23:02 If I'm a miner that's mining... 23:04 Like, for example, say I'm Foundry USA, 23:06 and I'm mining, like, 30% of the Bitcoin hash power. 23:09 I can afford the overhead to do things, 23:12 like go pay guys to go get Drivechains 23:16 that bring in, you know, $1,000 a day revenue. 23:19 And that could be, like, a huge number of them. 23:22 But it's a very different kind of thing 23:23 when you're talking about smaller-scale miners 23:25 where now I've got to go get all of these possible Drivechains 23:29 that, for me personally, might go bring in, like, $1 a day of revenue. 23:32 And that really screws over the small miner. 23:33 But if it brings in $1 a day of revenue, it doesn't matter. 23:35 I talked to a miner yesterday... 23:36 It does matter, collectively. 23:38 I talked to a miner yesterday who's in the audience. 23:41 They have 15 megawatts of hash rate. 23:44 They said that if they could get a 1% or 5% boost to revenue 23:50 by adopting Drivechains, they would do it in a second. 23:53 But the problem is... 23:53 It won't be 1%. 23:54 It will be 10x, 100x, or even 1,000x. 23:57 I mean, hang on. 23:58 Let's be clear, though. 23:59 There's a misconception there, 24:01 which is that Bitcoin difficulty adjusts. 24:04 By adopting Drivechains, 24:05 they're not going to get, you know, a 5% increase in revenue. 24:10 They will only get that if they are the centralized miner 24:13 who can do it in a way that other people can't. 24:16 That's, you know, that's one of the key things. 24:17 What it actually does is increase the overhead to get into mining. 24:20 Why does it increase the overhead? 24:22 Because if you want to go run a full node 24:24 that actually properly processes this stuff 24:26 and allows withdrawals to happen... 24:27 The miners don't literally need to run anything. 24:29 Yeah, don't I get a fit if I'm just... 24:31 They just need to obtain the knowledge of, you know... 24:34 They can run a node, obtain... 24:36 It gives you absolute certainty. 24:37 I mean, again, I think this has been touched... 24:40 It doesn't really matter anyway 24:41 because the software cannot possibly be so expensive to run. 24:44 The regular full node users will have to be running the software. 24:48 Wait, hold on. Let's just be super clear. 24:50 Miners don't need to run a sidechain to profit from increased fees. 24:54 That's correct. 24:55 Sidechains that work properly need miners to go and run full nodes on them. 24:59 Or the withdrawal process doesn't work properly. 25:01 No, they just need to obtain knowledge from the hash from somewhere. 25:05 If you wanted to be absolutely certain that there's not a single atom of titanium in some bread that you're going to eat or something, 25:16 you'd have to go into a laboratory and make it from scratch. 25:19 But you know that if you go down to the supermarket, you have a different type of knowledge. 25:23 Well, let's back up. Let's explain to the audience what on earth you're talking about. 25:27 So, like I said with Drivechains, the model is that miners can go vote to decide where coins go. 25:34 Now, if you don't run a node, obviously you cannot validate any of these votes. 25:39 You have no idea where coins are supposed to go. You can't validate that hash for that vote. 25:43 That's not true. You just have indirect knowledge. 25:47 Well, again, like I say, you can't validate it. You have to go and trust that someone else does it. 25:53 It's like Bertrand Russell's teapot argument or whatever. Is there a teapot orbiting Jupiter or something? 25:59 The only way to be absolutely certain would be to go and check, but we also know that it's not. 26:05 I mean, if the teapot orbiting Jupiter was very important to a coin... 26:10 But miners don't need to run any nodes either. That's what's very curious about it. 26:13 I asked you what the problem with this idea was. You just said that you mixed up L1 and L2, 26:17 and you said that I don't like miners having control over the L2 withdrawals, 26:21 even though they already have control over the Lightning Network justice transactions. 26:24 So it's exactly the same thing. 26:26 Now we're in this weird odyssey on miner costs. 26:29 merge mining the sidechain is essentially free. 26:33 They don't even need their own node. They can just coordinate with someone else who has the node. 26:38 When you say that miners have to cooperate for justice transactions, that's just categorically false. 26:43 51% hash rate can censor the justice transaction. 26:48 If they do that, they can drain the Lightning channel for free all the time. 26:51 Hang on. Let's be clear. 26:53 That's not going to happen. 26:54 To profit from this, they have to coordinate with the Lightning channels doing this. 26:59 Yeah, but they also must coordinate in the case of Drivechains. 27:02 It's actually much more difficult for them to coordinate in the Drivechain case 27:05 because they have to pre-establish a withdrawal to themselves that is going to take effect months, 27:11 three to six months later. 27:13 They don't necessarily know if the pie chart will be the same. 27:15 So quickly respond, and then we're going to move on because we have a few minutes. 27:18 We're going to go over on our time. 27:20 I'm sorry for the people who said don't go over on your time. 27:22 We're going to go a little bit over. 27:23 I mean, look, this is silly discussion. 27:25 Go read the blog post I just posted for this detail. 27:28 Okay, so I want to move just before we talk activation on exactly how something gets activated, 27:36 one thing that you're very clear on, Paul, is the way that Drivechains will be implemented, 27:43 would be implemented, it's unwindable. 27:45 All right, can you explain what that means and if we agree that that's a valid view? 27:50 Yes, of course. 27:51 It takes this op-nop5 that's unused. 27:52 It's the exact same thing that happened with the check-lock-time verify, 27:55 and it reuses it to count to 13,000, which is all that it does on L1. 28:01 And then if there's some kind of problem with that, all we have to do is the exact same soft fork 28:04 that turned it on, you can just turn it off. 28:06 You just censor op-nop5 from ever appearing again in the blockchain. 28:10 So there's literally zero risk to this idea. 28:12 Do you share that view that that would be effective in terms of unwinding Drivechains? 28:16 I don't think you'd be able to pull that off in practice. 28:19 And one of the really ugly things about this, of course, is that if this idea does go catch on, 28:25 you wind up having funds tied up, which causes a whole lot of political ugliness. 28:29 Yeah, but again, this only affects the people who opt in. 28:32 This is crucial to the whole point, the distinction between L1 and L2 is the entire point. 28:37 You know what I mean? 28:38 You may not care about a certain thing like the EVM or something, but other people care about it. 28:43 And you may not care about large blocks. 28:45 This isn't about what other people like. 28:47 It's about their ability to choose their own thing. 28:49 So some people want, you know, Roger Ver wanted the block size to increase on his node. 28:54 So if you can't keep L2 and L1 organized, then you will never understand this idea. 28:59 To be clear, you're not talking about L2 and L1 here. 29:01 That's not how these terms get used. 29:03 You're talking about L1 changes. 29:05 All of this is L1. 29:06 The L1 change is just the OPTNOT5 counting to 13,000. 29:09 That's all it is. 29:10 Okay, so I want to talk activation for a quick second. 29:13 So, you know, I'm a simple man. 29:16 Why is it so difficult to unwind? 29:19 He's got 40 seconds. 29:20 Let him talk about his activation. 29:21 All right, all right. 29:22 It's very easy to unwind. 29:23 Let him talk about his activation. 29:25 Thank you. 29:27 Thank you, Pops. 29:29 So, activation. 29:33 As far as I'm aware, all soft forks get activated by miners at the end of the day. 29:38 That's a complex question. 29:40 I know it's a complex and controversial question, but, I mean, am I wrong? 29:47 In certain circumstances, yes. 29:49 So I think a great circumstance was how SegWit got activated. 29:53 But that's still a miner activation. 29:54 Hang on a second. 29:56 So what happened with SegWit is the community threatened the miners with an unobservable threat. 30:00 Namely, we are running UASF nodes that activate SegWit. 30:04 Thus, if at least one miner does anything SegWit related, suddenly a big chunk of the network will go switch off. 30:10 And they will be mining on a useless change. 30:13 And this is why it's complex, because that was an unobservable thing. 30:16 So the majority of miners decided in that circumstance, arguably, hey, they're making this threat. 30:22 We will just go and do SegWit. 30:24 I guess what I'm a little bit confused about is, like, okay, what you just described, this user-activated soft fork, 30:30 where there was basically a message to the miners that if you don't do this, we're going to fork you off the network. 30:35 We're going to reject your blocks. 30:37 How is that different than Paul and Drivechain supporters going to miners and saying, run this software. 30:44 And if you don't run it, at some point in time in the future, our community is going to reject your blocks. 30:53 Well, I mean, to be clear, I'm not making a statement that it's different at all. 30:56 I'm just saying, like, that is the nature of this type of activation. 31:00 There's other types of activation which looks more clearly like miners activating, like my own check-locked-on-verify, 31:06 which you could argue was a miner vote, say, hey, on this threshold, now check-locked-on-verify is activating. 31:11 I guess my point is that, like, even a user-activated soft fork, you are petitioning the miners to activate something, 31:20 or you're going to do something. 31:22 So it's a communication to the miners, a signal to the miners of, hey, we're demanding this. 31:26 I mean, it's a tricky signal because it's a signal which, maybe I'll talk about it from a more pragmatic point of view. 31:32 If we have a UASF that doesn't have overwhelming support, which is what would happen with Drivechains, 31:38 if this goes badly, suddenly Bitcoin forks, and very bad stuff can happen where, you know, 31:44 one exchange is simply not on the same Bitcoin as another exchange, yet they still have the same ticker price. 31:49 And I think we can see how a disaster this would be. 31:52 This particular soft fork should be miner-activated because it's so miner-centric anyway, 31:58 and also because it does not affect, like, there's certain things like requiring the Coinbase to have the block height or something. 32:04 There's certain things, this one is more opt-in, much more opt-in than usual, 32:08 which is to say that people who aren't using it can ignore it completely. 32:11 To be clear, like, the Coinbase thing is fixing a serious bug. 32:16 The miner-centric one should be activated by miners because if they're not really into this idea, 32:21 and if they don't really understand this idea, that they would make a huge amount of money by, you know, 32:26 like, having this enormous extensibility upgrade to Bitcoin. 32:30 If they don't really get it, then they don't get it. 32:33 So, you know, I guess my final question here is, like, okay, if I'm Paul, 32:40 I'm going directly to the miners and I'm making the case for why they need to adopt Drivechains, 32:45 and I'm selling them on the benefits of it, and, you know, frankly, I don't really care what anyone else's views are 32:51 because it's the miners at the end of the day I have to convince here. 32:55 If the community had a problem with that, how does the community stop it? 33:01 Well, I mean, I think the main thing that happens here is, what you're telling miners is, 33:05 this is going to be a shit show. 33:07 And if you go through this and you add all this uncertainty to Bitcoin, 33:12 ultimately you're probably going to either just reduce the price in general, 33:16 or you will go have a user-activated soft fork to undo Drivechains. 33:22 And that is a huge mess. 33:24 It is a much bigger mess than... 33:26 So why do you say that it would be a shit show? 33:28 Because it's just if the miners activate it, then everyone is on the same chain. 33:31 Hang on a second. 33:32 It is a much bigger mess than just doing nothing. 33:36 I mean, if you're a miner, you think, well, you know, I get my money in Bitcoin. 33:40 If I go through uncertainty in a shit show, reduce the value of Bitcoin by, like, 10%, 33:45 was that really worth it for the hypothetical Drivechain that a bunch of tech experts think is a bad idea? 33:51 You know, I think that's the main pushback. 33:54 Now, obviously, if, say, a 51% majority of miners said, you know what, screw this, we're going to do this, 34:00 they can go make a real mess, and they can go force other people to react. 34:05 But, you know, this has never happened in Bitcoin's history, 34:08 and we'd much rather it not play out that way. 34:11 I mean, because it's a soft fork, couldn't they have already... 34:13 I mean, it hasn't been... 34:15 It's theoretically possible that they already activated BIP300. 34:18 Yeah, but, I mean, if people don't know about it, it's not very relevant. 34:21 And people have put things like off-and-off 5 and other stuff in blocks, 34:26 so it's unlikely to be the case. 34:28 Okay, so I'm going to conclude here, because we're already over, and the please wrap up is going. 34:34 But I think, you know, I've found that in Bitcoin, it's now impossible to get everyone to agree on anything. 34:41 It's just impossible. 34:42 It always was. 34:43 Yeah, it's even harder now. 34:45 And so I feel like, you know, why Drivechains, I think, is important 34:50 is because whether Drivechains goes forward or doesn't go forward, 34:53 it helps define the boundaries of how Bitcoin advances in general. 34:57 Okay, you have a problem with Drivechains, like, this is how you resist Drivechains. 35:02 You want Drivechains, this is how you advocate for Drivechains. 35:05 And right now, that's so unclear. 35:08 It's unclear how to do it. 35:10 I think that makes it very uninteresting, because from a tech point of view, 35:15 you're creating something that has all of this software complexity and so on, 35:19 when it's so much easier to just build other types of systems where you don't have to ask for permission. 35:24 You know, to build interesting stuff on Lightning, you don't need to ask for permission. 35:28 You just go do it, and you write code, and you release it. 35:30 You know, open timestamps, I've never asked anyone. 35:32 Well, you had asked for SegWit, though. 35:34 Well, once we created the building blocks, but, you know, we've created a very thorough set of building blocks, 35:38 so you can build an enormous amount of interesting stuff. 35:40 You know, we took the Bitcoin protocol and fixed it. 35:42 The problem with Lightning, though, of course, is that even the proponents don't expect 8 billion people to be able to use it. 35:47 Whereas with this, they could, and so that's a pretty big difference. 35:50 The tech people don't agree with you on that. 35:52 No, you could go to drivechain.info/misconceptions, and you could search. 35:57 I have a part where I have hyperlinked a bunch of things. 35:59 So if you search this, this, this, this, this, and this, I have a link to a bunch of people, 36:03 the top experts in Lightning. 36:05 Don't take it from me. 36:06 They have all these different interviews and posts to our blog. 36:10 Let's be clear. 36:11 You're talking about experts saying Lightning doesn't scale. 36:13 That's not that Drivechains work. 36:15 So I'm going to get in a lot of trouble already. 36:17 We're way over. 36:18 But can we get a round of applause for these two wizards of Bitcoin? 36:23 Very important conversation. 36:27 Paul, where can they follow more about Drivechains? 36:29 I would just follow my Twitter on Truthcoin on Twitter. 36:34 And, Peter, you're going to write the definitive takedown of Drivechains. 36:38 Is that coming? 36:39 I mean, I just posted it. 36:40 Go to petertodd.org. 36:41 Okay. 36:42 Let's go. 36:43 All right. 36:45 Welcome back to the Bitcoin Amsterdam Live Desk, sponsored by Bitcoin Magazine. 36:53 I'm joined by my esteemed panelist, Nifty, the founder of Base58, 36:57 and Iago, a core contributor to Sovereign. 36:59 We just heard a great panel about BIP300, obviously with Paul Sztorc, Peter Todd, and David Bailey. 37:05 So I want to go right to you, Iago, with the first question. 37:08 Does Bitcoin have a security budget problem to you? 37:11 Yes or no? 37:12 Yeah, clearly it does. 37:13 We're seeing no growth in the value of transaction fees collected by Bitcoin miners over time. 37:22 And we're not seeing a substantial increase in the number of transactions. 37:26 And what Paul and Peter both want to do is try and find a way to increase that. 37:33 I don't think that Drivechains is the way it used to be. 37:37 For many years, I was a big supporter of Drivechains. 37:39 But at this point, I think it's obvious that it's a really obsolete technology. 37:43 What we need are roll-ups. 37:45 That is the ability to add more transactions into every block using Bitcoin transactions. 37:50 And it helps solve Peter Todd's concern with Drivechains, 37:54 that you would have infinite transactions and no scarcity, 37:58 because the scarcity of Bitcoin would remain. 38:00 20 seconds to you, Lisa. 38:01 Do you think that we have a security budget problem with Bitcoin as well? 38:04 No, I think I'm a security budget whatever. 38:09 I don't think there's a security budget problem. 38:12 Okay. 38:13 Yeah, no worries. 38:14 So I know we were talking about Lightning adding fees to L1 and helping out miners. 38:18 Or is it cannibalizing miner revenues? 38:20 Or is it a little bit of both? 38:22 I'll go to you, Lisa. 38:23 I think it's a little bit of both, honestly. 38:25 I think that having Lightning there makes transactions that wouldn't be possible to happen, happen. 38:30 So I think that's great. 38:31 I do think there's a little bit of tradeoff between now we have transactions that are happening on Lightning 38:36 that maybe could have happened on the base layer. 38:39 Yeah. 38:40 15 seconds to you, Iago. 38:41 What are your thoughts on Lightning? 38:42 Cannibalizing miner revenue or adding to it? 38:44 I think its overall impact is fairly marginal, 38:47 because the level of adoption that we've seen with Lightning is quite marginal as well. 38:51 And so what its theoretical impact would be, different from its practical impact. 38:56 Yeah. 38:57 Lisa, last question to you. 38:59 Miners control the justice transactions. 39:01 And I know that's what Paul was bringing up. 39:03 And you were kind of shaking your head over here. 39:05 What are your thoughts on him saying that they control the justice transactions? 39:07 I think that's completely untrue. 39:09 I think it's true that miners decide what transactions go into blocks. 39:13 But I think that being able to choose between justice transactions or not, 39:18 that's completely, I think, not how it actually works. 39:22 All right. 39:23 Thank you so much. 39:24 We're going to bring it back to the main stage or the Genesis stage here 39:26 to talk about the Nostr protocol. 39:28 Catch us after that. 39:30 Thank you, Miami, for the last three years in this amazing city. 39:35 The whole world shut down. 39:37 But Miami welcomed us with open arms. 39:40 We want to show Bitcoin to the whole world. 39:45 We are taking the conference on the road to set the stage for Bitcoin in a new city. 39:54 Nashville. 39:57 Bitcoin 2024 is coming to Nashville in Tennessee, 40:01 a city that is known as a music and freedom city. 40:05 Bitcoin 2024 in Nashville from July 25th to 27th.