0:00 Thank you. 0:30 Thank you. 1:00 Thank you. 1:33 Hello, 1:36 Here we all are, and we have a special guest this week. 1:42 think I will just say a couple sentences and then we'll just 1:48 go for it, 1:52 I've known about Syscoin for a very long time. 1:56 since in 2014, designed my prediction market thing, Truthcoin, peer -to -peer oracle, 2:01 which is supposed to a Bitcoin prediction market, but thanks to the slow arrival of 2:06 chance. It's still yet to materialize even 2:09 though the work is there. 2:11 could go to BitcoinHybeline.com if you were curious about that particular topic. 2:16 But the point is, I wrote this Truthcoin paper in January 2014 or something. 2:22 And there were only a few people who were smart enough to immediately pick up on how 2:28 great it was, 2:29 since two of them ended up joining Ethereum or even as very early Ethereum people and 2:34 they took the project to Ethereum basically where it became sort of auger in Gnosis 2:40 and even it became the answer to people who would ask Vitalik, what is Ethereum 2:46 for? And he would say, oh, he said nothing 2:48 for a while. 2:49 then after my paper, he said, oh, prediction markets. 2:53 So it kind of created an awful lot indirectly. 2:57 And it created the modern market maker. 2:59 But the point of this story is that one of the four was Bitcoin, two the four ended up 3:03 being Ethereum. were a bunch of, one was, there's actually five, suppose, 3:08 actually quite a few, one was this guy, Zach, who did his own interesting project, I 3:13 think I think called the MOVO or something. 3:15 But one of them was this Syscoin guy. 3:20 And I don't remember too many of the details. 3:24 was Sebastian, yeah. 3:27 And 3:27 he was really into it. 3:29 then I was like, kind of like, well, that's neat. 3:32 I told him to reach out to you that time. 3:35 Because I really liked the proof of work versus proof of stake piece you did. 3:39 I actually referred a lot of people to that, just because it wastes a lot of time 3:43 explaining and just like, just go to Paul's blog here and just read that. 3:47 And if 3:47 you have a rebuttal, let's talk about it after. 3:51 Nice, nice. yeah, so I had this opinion of Syscoin in 2014 that was 3:57 like, oh, these guys must be on top of something. 4:00 But it has the same problem with all altcoins, which is since it's gone from being a 4:05 really big problem to a 4:06 much smaller one over time. 4:09 But the big problem is always just like live by the fork, die by the fork. 4:13 We want to get everyone in the world on one coin, and so it's not helpful to have 4:17 people keep launching their own coin. 4:21 But now I guess it doesn't make much of a difference because Bitcoin is moving very 4:24 slowly on the whole sidechains project, 4:26 which is part of the whole point of this space. 4:30 But I think there's never been a better time to be an altcoin than now, I 4:36 think, actually, over time people have gotten... 4:40 I mean, you can look at the Bitcoin word dropout and the crypto word rise up. 4:45 Almost no 4:46 one like on a television show or something uses the word Bitcoin anymore they just say 4:50 crypto and people love USDT people love Ethereum has more transaction 4:56 fees than Bitcoin and Tron USDT has more more actual 5:02 payments than the other so 5:05 But anyway, love to, I thought maybe I researched this 5:11 before doing the space, but then also I didn't have time. 5:15 But also I thought it's probably better if I just don't research it and we just, we 5:19 just have a fun conversation about, because probably most of the people who 5:24 attend 5:25 this weekly space have absolutely no idea at all what Syscoin is in the slightest, so 5:31 maybe we should just go from there. 5:37 Yeah, for sure. 5:38 Bradley, want give a sort of a broad thousand -foot view, I guess? 5:43 Yeah, sure. Sorry about 5:44 earlier. I had to restart X. 5:46 I couldn't hear anything. 5:47 But hey, everybody. 5:49 I'm Bradley with Syscoin Foundation. 5:51 Jack's here too. 5:52 He's the president of Syscoin Foundation. 5:54 Yeah, I can offer a pretty quick thousand foot view. 5:59 Syscoin is merge mine with Bitcoin. 6:03 We're chain ID 6:04 16. We've been merged mine with Bitcoin, I think, since 2016. 6:09 Mainnet since 2014. 6:10 We've been around the block, so to speak. 6:13 We've been in this space for a while and still innovating. 6:17 our approach and what we have today with 6:22 Syscoin is 6:24 a modular full -stack blockchain that has its 6:29 main base as a UTXO chain. 6:33 Right now we have something like about 20 % of Bitcoin's hashrate supporting us. 6:38 We also have an EVM running alongside it, so it's sort of a dual -chain LayerOne, 6:43 have a functioning LayerTwo in the form of an optimistic 6:49 roll -up, what's currently an optimistic roll -up called Rolex that is mainnet. 6:56 And our view basically is we want to keep that provenance of value that Bitcoin's 7:02 proof -of -work provides 7:05 We've always been proponents of proof of work. 7:09 And we feel that the best way to tap into that is through merge mining. 7:16 we don't throw proof of stake out with the bathwater, so to speak. 7:21 We feel at the bottom of the stack is 7:22 have 7:30 some 7:33 additional security information. 7:37 by people like Luke Dash Jr. 7:40 regard to merge mining. 7:42 we have some additional security in place. 7:47 in order to 51 % attack Syscoin, basically you would not only 7:53 need control at least that percentage of the miners supporting Syscoin, but you would 7:59 also need to have a super 8:01 majority of masternodes that are serving multi -core chain locks that provide 8:07 finality on top. 8:09 the cool thing about that is we can achieve finality in a way where if for 8:15 some reason it can't be achieved, like it will resolve back down to 8:20 Nakamoto consensus 8:21 and it's not a breaking event. 8:23 but at any rate, yeah, we're, to sum it up, we're a modular full stack 8:28 ecosystem 8:31 with Syscoin as player. 8:33 Before you go a little bit further, Paul calls more deeper, I think 8:39 for his sake, 8:41 you qualify because as soon as you start to say validators, alarms 8:47 probably go off in his head. 8:49 I think you validate the fact that we don't relegate back down to proof 8:54 of stake. And that's a really interesting statement that you can have validators not go 8:58 down to proof of stake. 8:59 It's 9:00 It's because we choose not to enforce this finality. 9:05 We choose to have it as an option so that under the optimistic 9:11 scenario finality is achieved. 9:14 Under the pessimistic scenario, you fall back to longest chain rule. 9:18 how we do that is there 9:20 are some nuances in chain locking, and we can show you some nice article around that in 9:25 a marketing piece we did recently on the method to the madness. 9:29 If you Google that, we can show you that it will fall back to longest chain rule 9:34 without the risk of falling back to a purely proof of stake model. 9:39 and specifically 9:39 you're probably thinking about if you have proof of stake and proof of work together 9:44 you have the option of those super majority validators to grind the difficulty down to 9:50 control the chain and that specific problem is what is the trade -off you get when you 9:55 when you introduce some validators you have difficulty grinding and 9:59 that is something we've solved for as well. 10:03 So, just thing. 10:04 you know, Robin Linus, who's inventor of BitVM, he vaguely likes, I don't know to what 10:10 extent yours is similar, but he vaguely likes the, there a lot of people like the idea 10:16 of proof of work base and then other 10:19 stuff on top to improve the finality and they push back against 10:26 unrefined merge mining and so there has been a lot of 10:32 commentary about that and a lot of people like different things so for example 10:37 who's 10:38 Who's that guy, Sonny Ray or whatever? 10:41 A lot of people have liked the whole, and then they say that that's, I actually kind of 10:46 find it a little weird and sort of annoying, but it's not a big deal. 10:52 But then they say, we made proof of stake work by putting it on top of proof of work. 10:56 I was like, no, not really. 10:57 mean, the proof of work is the thing that works. 10:59 then there's like a refinement. 11:00 But I think that that's smart. 11:01 mean, merge mining sort of just similar like that, where it's like the L1 coins that 11:05 were mined Let's prove it work. 11:07 used to like change. 11:11 can we make it make sense for anyone in the audience? 11:14 which is that? 11:15 I mean, it's cool that this merge mining thing has been around for a while, and on top 11:21 of that, you bolt something to make it to have to give the... 11:26 this... mean, I guess the probably the smart thing to ask would be is like, what 11:29 problem is this coin supposed to solve? 11:32 And I don't know if the answer 11:34 is add more flexibility plus EVM, number one. 11:38 then number two is like increase the reliability of merge mined 11:44 blocks on BTC. 11:49 the reason why we introduce finality, It's 11:53 It's not we're trying to create fast blocks and all that dumb stuff. 11:56 It's for a specific reason. 11:58 that's when you get into, as you know, you guys are probably modular, modular 12:02 blockchain enthusiasts, separate the execution off from the base layer, moving the CPU 12:07 and memory off. 12:08 That's how we scale. 12:10 are actually pretty 12:12 big proponents of the genesis of Bitcoin as well. 12:18 So we actually enforce, we enforce merge mining, which means you can't solo mine 12:23 Sys, you have to merge mine, meaning you're mining Bitcoin to mine Sys, there's no 12:29 other way. The drivechain does that as 12:31 well, just say bye. 12:33 it's basically an extension that way, like you can't go in by itself, so it's not 12:38 parasitic on the tokenomics side. 12:41 But the reason why we introduced Finality was specifically under the modular blockchain 12:46 design the data availability component, because at that point 12:50 you've introduced the data availability system where you can start to after finality 12:55 plus some time you could remove that data because it's not hashed on the chain. 12:59 But that data availability is critical to scale up the modular stack. 13:03 know the the recursive layers if you will if it's BIP300 then the chains 13:09 within the chains. 13:10 once you have the data availability constraints, you have to have the LayerOne serving 13:14 that as sort of your court system and the finality is needed there. 13:18 Otherwise, 13:20 there's asymptotic cost that happened on the data itself and we try to remove that. 13:24 That's specifically what EIP4844 of Ethereum or ProtoDank sharding of Ethereum was 13:29 trying 13:29 to solve with their DA solution except they fell back into sharding and 13:35 because they're using proof of stake anyways it made sense to distribute the trust 13:39 across all the other nodes. 13:42 some censorship concerns there you know with sharding you know as 13:48 an artifact 13:49 of the architecture itself, know, if 50 % of the nodes are unavailable, 13:54 then you have data unavailability, essentially. 13:59 we managed to come up with an approach to DA that sort of keeps the Bitcoin 14:05 assumption of one honest node is all that's needed. 14:09 we have that, we call it proof of data availability, and it's functioning on our 14:13 LayerOne. One thing that's worth pointing out about our ecosystem 14:19 is that Sys is the native coin, not BTC. 14:23 right? 14:25 feel philosophically 14:27 that Bitcoin's economics really are not ideal for utility. 14:33 And I think the market itself sort of expresses that whenever you look at projects sort 14:38 of like, like RSK, for example, it hasn't really been utilized 14:44 much. And of course, there's a contingent of 14:46 is 14:54 a lot simpler 14:59 than Ethereum. Bitcoiners that basically are in the closet with their 15:05 Ethereum 15:05 wallet, right? why aren't they using RSK? 15:10 would think you would think it would have a lot more usage and I think what it really 15:14 boils down to is people don't want to spend their Bitcoin on 15:20 gas. 15:23 Maybe, I don't know, 15:25 a little all over the place here. 15:26 So the problem 15:28 that Drivechain is supposed to solve is that we're not supposed to need a new coin. 15:33 If we don't, we don't need one, you can always bolt one on. 15:39 so SysCoin is a different coin and you have 15:44 a coin not only because it's L2 and you need some kind of battery to store value and 15:50 move it around, 15:52 but you also just need something to... 15:56 you also say that you will change the economics of the coin, like the supply schedule 16:00 and the... what do you mean by that specifically? 16:03 I 16:05 Yeah, so our SIS is based on EIP 1559, 16:11 so it's essentially So the supply will shrink even if the fees are high. 16:16 Correct, and there's a burning component there and there's no maximum cap. 16:21 It's really, really 16:23 for utility. there's also the validators we 16:29 added to 1559 by the rent paid to the miners and validators 16:34 is still paid, and the validator rent will not go under some value, so it doesn't 16:40 deflate to zero. 16:41 The rent paid 16:42 to the masternodes will deflate only up to a certain point, so there's that static 16:46 cost. if the revenue on the EVM chain is 16:52 higher than all the fees plus the static burn deflated 16:58 maximally, then it does become deflationary during 17:01 that event but the fee market also would be susceptible to high costs at that point 17:06 people are moving off of the base layer for their transactionality because it'll be too 17:12 costly. The other cool effect of going this route and being merge mined 17:17 is that we can sort of perpetually 17:20 miners support Syscoin 17:24 with Sys. Even in the 17:30 future, havings continue on and the block rewards 17:35 diminish towards nothingness, maybe even 17:39 potentially fees aren't sufficient at some point in the future. 17:44 They're getting something. 17:45 They're getting CIS. 17:46 that CIS is incentivizing them to potentially, I mean, at the level of the architecture 17:51 itself, it's there, The idea that Bitcoin miners would have 17:57 something incentivizing them to 17:59 continue working Bitcoin and to keep the network decentralized. 18:08 mean, I think actually that decision of ETH to change it around, ultrasound money is 18:13 kind of smart. 18:14 it's part of their plan. 18:16 Ethereum's plan to destroy 18:18 BTC and replace it and take it over. 18:20 It's just quietly, without really making too big of a deal about it, change the 18:26 economics of ETH so that the supply tends to shrink, whereas BTC is going to to grow 18:31 from 19 million to 21 million. 18:33 But I think they're very aggressive on that shrinking as well. 18:36 It's really, really aggressive 18:37 to a point where even right now we had events where supply was actually decreasing. 18:43 the hardness of coins is also affected now because of proof of stake. 18:47 That's the two problems we see. 18:49 Yeah, ETH should never have done full proof of stake one, but maybe they know, maybe 18:55 they have, you know, information I don't have 18:57 or whatever, the... 19:01 Well, I think they're just trying to do something different, but I think let's get 19:04 into... I don't remember the last time I saw Syscoin 19:09 on CoinMarketCap. 19:13 I don't remember what the ranking is, 19:16 it must be... 19:18 Like 320 or something like that. 19:21 320. So that's kind of low. 19:26 And yet you have a shrinking supply, you have all these features, 19:32 and do you have some kind of story 19:35 for why you think that is or do you think it doesn't matter? 19:39 I know. There's definitely reasons there. 19:42 It's taken eight years to develop our infrastructure. 19:45 We're not thinking of ourselves as let's deflate to zero and 19:50 holders of Sys will benefit. 19:52 It's more when 19:54 AGI within 18 months is coming and there's billions of agents onboarding into crypto 20:00 because they're not using fiat There needs to be a place that will help scale up which 20:05 doesn't affect the asymptotic costs of the validators of full nodes and 20:11 That rent that they pay in SysGas needs to be economically abundant and available and 20:16 not super costly when people are trying to onboard into it. 20:22 we've taken that time to figure out what the tokenomics and the security mechanism 20:27 should be and how we all should relate to each other. 20:37 We have a of different ways to do this. 20:41 So there was no activity on the Sys chain because that's by design. 20:45 At the same time though, Jag, we have a pretty substantial community 20:52 and it's continued to grow. We're very active. We're constantly developing. Now we have Rolex 20:59 that's got, you know, a fairly rapidly growing ecosystem. I mean, we've had to kind of bootstrap 21:05 it, but we've got a dedicated BD team. We're doing a lot, okay? We're doing a lot, and we're 21:11 accomplishing a lot compared to even projects that have, like, a big war chest that are higher 21:18 market cap. So it is interesting, you know, the fact that we are at where we are in the market, 21:26 and, you know, there's probably some other reasons for that, and I don't know that it has 21:33 anything as much to do with our project itself. It might be a symptom of the market. I don't know. 21:40 We never did this crazy VC background where the market makers and VCs would pump us to distribute 21:48 the bags. We're very organic. Our community is mostly Bitcoin OGs, some of the Eth OGs, and 21:56 the people that really believe in change rather than trying to look at, like, if you just visit 22:00 Discord, the price talk is always about how will Syscoin help the world rather than, you know, 22:08 when can I sell and when's voluntary? Well, that's the smartest thing that Ricardo Spogne, 22:14 Fluffy Pony, leader, cultural leader of Monero for a time. The smartest thing he ever did was 22:21 they banned all discussion of the price, which I think was key, and instead focused on their 22:26 problem, which was private cryptocurrency transactions. Yeah, man. I remember back in 22:34 the day when, I think it might have been him, but I remember watching a Monero guy 22:39 speaking, and he basically told people, I don't, you know, I don't give a shit about 22:45 the price of Monero. Don't talk to me about that. We're here to develop it. 22:48 And yeah, that led to a whole, you know, hellstorm of comments and drama back then. 22:54 Okay, I'm sorry. My phone did something weird. Yes, yeah, that's very important. And of course, 23:00 see, Monero is not very high either. It's, let's see, I think it's like in the 20s or so, 23:08 but it is beating many coins that did do the VC route. VC route, you know, it does help, 23:14 probably, I would assume the coin, you know, the price of Monero. I don't know if it's 23:21 probably, I would assume the coin, but I think it's huge long run costs, 23:26 but I guess we would say, I mean, like Solana is like the biggest VC coin in my head 23:32 that I would say, and it's number seven now. So, I mean, you know, it's like, we can't base 23:38 everything off of the rankings because it's very circular, you know, circular reasoning, 23:43 you know, you're low until you're, but yeah, I just think it's, it's worth paying attention, 23:49 especially because it's all open source. So everyone can just copy any idea. 23:58 And that means that even if you have a great idea or even the best actual shipped code, 24:04 it doesn't necessarily mean anything, which is why I just kind of hang out in 24:10 BTC world, even though it's, you know, it's not going super great in the BTC world, 24:15 but it is number one. Right. 24:22 Okay. Yes. Yeah, that's true about open source. I mean, that's certainly true, 24:25 but there is another component as well. And that is the network that's the networks that's running 24:33 the blockchain. You know, you can't replicate, you can't just replicate that. I mean, maybe, 24:37 maybe arguably merge mining is kind of an exception to that rule, but, you know, you can't 24:43 fork CIS coins code and basically have our community and our network of masternodes. So 24:50 there's that. Yeah, I know I would say it's, yeah, it's a little, there's a bit of a little 24:58 bit of marketing brainwash that's happening, but there's been cycles that we've been through that 25:01 other coins are still going to experience. You know, we've had four or five bear bull markets, 25:09 whereas many of the projects, maybe one or two max, and they'll experience the same drawdowns 25:15 that we've experienced. And, you know, in the end it'll be the people that stuck around, 25:20 that kept developing, that put their heads to the ground, that thought about things the right way. 25:24 Think about value first. You know, if you have VCs backing you, they're going to say, when, 25:29 when profits, when bottom line, when revenue, when, how can you pump this? And you're reporting 25:34 to them versus your organic communities asking when can we change the world? It's a totally 25:40 different dynamic and allows you to freely express yourself and develop and think about 25:46 things in the right way. You know, we feel. Okay. So now should we talk about the real 25:53 meat of the substance, which is the L2 or for example, whatever, whatever it is that makes you 25:59 say that, um, you will, did you say something on Twitter earlier this week that was like, 26:08 whatever Cisco inverse BIP300 or something. 26:14 Right. Don't you have some kind of, maybe you're talking about the optimistic L2, 26:19 the optimistic rollup or just the ability to move coins. 26:24 It might've been Zhao, but he might've mentioned something that, uh, I didn't catch the tweet, 26:29 but I know he said something that there was some, you know, we should look to each other 26:35 or talk to each other about we're solving the same sort of problems and potentially, 26:40 you know, there's opportunities to collaborate there. 26:43 Yeah. There was a tweet that said that we outshined BIP300 and the idea was okay, 26:47 well, come on and tell us how. Um, I feel that we, you know, we can look at Bitcoin, 26:55 from two different angles. There's BTC, the economics of BTC, and then there's the proof 26:59 of work network that is supporting it. Um, the perspective of our project is that, 27:07 that proof of work is extensible, of course, through merge mining. And so what we're doing 27:11 with Bitcoin is we are providing utility essentially that is supported by Bitcoin's 27:19 extensible proof of work. And we are saying that essentially we are, uh, we are providing 27:27 the security of Bitcoin, um, for that, um, for that utility use case. Um, and I don't think, 27:37 um, uh, I'm, I'm not sure how BIP300, um, with, with that being said, uh, can, um, like is, 27:50 is useful or something maybe that Bitcoin should necessarily implement if merge mining is already 27:56 in place. And there's a means to scale Bitcoin's proof of work through that. Now, if you wanted to, 28:01 I think, um, extend BTC itself onto other chains than it would be, um, uh, 28:10 is a type of merge mining that, uh, and the merge mining field is a little crowded or whatever. 28:19 Um, because we have many ways of not only doing merge mining, including vanilla merge mining that 28:24 is invented by Satoshi a long time ago, and that many coins have been merged mind with, 28:30 which I assume would describe you from 2014 to 2016, based on what you said, uh, earlier. 28:36 And then there are, there's in fact a BIP301, which is different form of merge mining that was 28:41 sort of invented by me. And then, uh, BIP301 has been emulated in many ways by the Super Testnet 28:49 or even, um, um, who else, uh, Ruben Somsen, people invented different ways of doing it today. 29:00 There are, there's like, um, I think that the, uh, the sunny Ray proof of work, proof of stake, 29:07 like it's not in merge mining, but there's that, like that type of technique is possible where you 29:12 have merge mining and then you do other stuff. You have people stake the merge-mined coins. 29:18 And then there's like some second round or something like that. So that's 301 and 301 is 29:26 sort of optional. I mean, drivechain is considered like something that uses 300 and 301 and how the 29:33 sidechain comes to consensus. That's kind of like, uh, however they want to do that. That's 29:38 up to them. You know, as far as I'm concerned, it's a different piece of software and they can 29:42 do whatever they want. BIP300 is about sending BTC to and from a different blockchain network. 29:50 So it's deposits and withdrawals. And that works based on paying off the miners in all sorts of 29:59 ways and making it very difficult for them to steal and making it so that if they do steal, 30:05 it will be economically devastating for them. Uh, and so that certain chains where the chain 30:13 deposit, the total amount of coins deposited in the chain is less than the turnover in transaction 30:20 fees for certain chains, it will be economically secure. And that is a BIP300 idea. So together, 30:29 those ideas are drivechain. I think the 301, you know, it was invented in like January, 2017, 30:35 and we have different ways of emulating it. I think it's, I think 301 is a pretty good idea. 30:41 I don't think it's like, it's nowhere near as important, 30:45 especially since we can already do it on Bitcoin today with just in a very inefficient way. 30:54 And, uh, we can probably do, honestly, I haven't, I haven't really checked, 30:58 but since it's just counting the 13,000, we can probably do BIP300 today also, 31:03 but that's not really the point. The point is to, uh, is it 1306 blocks is the, is the number 31:10 13,150, which is like three months worth. So like the huge overkill on our test network, 31:18 it's only 131 now for fast testing. That's fine. 31:24 That's PV proof base. So the state gets put on, um, the sidechain state gets fully, um, 31:31 come to consensus. And then that gets put on the Bitcoin chain, like, uh, like an anchor type of 31:36 thing. And then people, is that how you do it? We don't, I don't, I make no, I, I kind of flip it 31:44 around completely where I say nothing on L2 will ever be enforceable on L1. And so L1 can ignore 31:51 L2 completely. And then it's just the deposits and withdrawals. There's the withdrawal system 31:56 is very, very, very slow and very transparent. And with merge mining, the L1 miners collect all 32:02 the transaction fees, which I imagine will be enormous in total because there's no limit on 32:07 quantity. So it's like the whole world's transaction fees, like just Visa's transaction 32:12 fees today. It's just like hundreds of millions of dollars a day, which is much, much more than 32:17 what is collected on BTC now, which is only hundreds of thousands. 32:21 The miners need to mine the, uh, the sidechain blocks. 32:26 They don't necessarily, they, if they want additional certainty, they always can, 32:30 but there's no actual requirement. And in fact, many different shortcuts are built in 32:38 to make it very easy for them to learn the sidechain state, even if they do absolutely nothing 32:44 at all. So that's blind merge mining is some people, different people who are running side 32:52 chain nodes and they know the state perfectly. They just bid on what the next block will be with 32:58 L1 coins. And they say, put this, put this little code, put this little, uh, Merkle root or whatever, 33:06 put this little Merkle root in L1 and I'll pay you these L1 coins. The miner, just the L1 miner 33:12 just does that to advance the state. And, uh, when that block is found over there, then the person 33:19 who paid L1 coins, someone who ran a sidechain node, they earn the trans pay the transaction 33:24 visa themselves. So I don't know. I explained it. If the Merkle root is uploaded, uh, you come to 33:31 a consensus on the Merkle root, the layer two sidechain has the witness data to create the Merkle 33:38 root or is it that everyone's own UTXO positions, they can do an SPV proof against that Merkle root 33:45 to exit. Uh, well again, the finding the next block is in my view, completely separate in every 33:53 way. So finding the next block is the merge mining part and the deposits and withdrawals. 34:00 Uh, those, those are two completely separate and inseparable things in my view, but not in 34:05 many people's view. Many people don't have that view, but I do. And they are totally separate. 34:10 From state to state on layer two happens on its own, but at some point you're putting the 34:15 Merkle root back onto layer one so users can exit, right? Uh, well, it's not the Merkle root that 34:24 allows people to exit, but it is the hash of the withdrawal transaction ID. And when the state 34:29 advances in the merge mining dimension, also there was a hash of the hash of the sidechains block 34:36 header. So sidechain has, I see we have a hand, but let me try to continue to answer this question 34:42 for a second, which is that the, okay, what happens is you're a sidechain node and I'm a 34:48 layer one miner and you are a sidechain node. You, you build the next block paying yourself 34:55 the transaction fees. And let's say that's four Bitcoin worth of transaction fees. 35:00 You've built that now, but you, but you have, no one has found the block per se. You've just 35:04 built it. You know what it should look like. You know what, that's what you, you aspire this to be 35:10 the next block. You go on L1 and then you pay the miners 3.99 BTC and L1 coins. You say 3.99 BTC L1 35:21 You say 3.99 BTC L1 coins to 35:28 to put this code in this special spot in L1 block. And then that is what is the sidechain 35:34 node software interprets. They interpret that as being what counts as meeting the, you know, 35:42 what would be the difficulty requirement and the sidechain L2. Great. Yeah. Yeah. It's putting the 35:48 state like as in anchoring and that lets people exit, right? Later on, they can use that. 35:54 Well, that doesn't necessarily, I mean, indirectly it does because the sidechain 35:57 state has to advance at some point so that you, someone can brought on L2 say, I want to withdraw 36:05 and have the sidechain become aware of that fact and then bundle all the withdrawals up. 36:10 Yeah. So how does that withdraw work? Like I'm a user of L2 and I move up back to, 36:15 back to base. Well, the first all like cashed and sort of collected on L2. So the withdrawals 36:22 only happen one at a time, no matter how many people are trying to withdraw. So there could 36:27 be many like 12,000 people trying to withdraw, but the way for the first withdrawal to finish, 36:31 and then the sidechain, which is aware of L1, it spends its UTXO. It says here's selects the 36:41 UTXO that has all the BTC in it, you know, the UTXO of all the sidechains coins. So the sidechain, 36:50 maybe it has like 50,000 coins or something. It selects that UTXO, and then it pays out to the 36:57 list of the list of 12,000 people who wanted to withdraw, who waited patiently. And the sidechain 37:04 L2 is, yeah, I haven't explained how it happens yet, but I'm just saying this again, the sidechain 37:08 calculates what it, what aspires to be the case. So it knows what it's sort of aiming for. 37:15 And it constructs, what would the transaction ID be of that? If it would, you know, what would 37:23 the transaction ID idea that be if it were included on L1? And then it does a tiny detail 37:31 where it zeroes out part because it's like a SIG hash, no input. But, but other than that, 37:36 it just broadcasts this hash everywhere in L2. It's in every L2 header. It's in the, 37:42 you know, ribbon of the GUI. It's in, so, so it's a, so this hash is everywhere. And then 37:50 the miners pick it up since the miners are human beings and not nodes. The miners can, 37:56 the miners just need to learn from, learn it from anyone who could make it available. 38:00 Okay. So there's a commitment of the hash and then someone needs to reveal the data behind it 38:06 so that you can disclose the state that that's the state that's being changed. 38:11 Well, there's not necessarily the state. It's only the, it's only the one transaction, 38:18 but you could call it the state if you wish. You could call it like a state changes every three 38:23 months, maybe. Anyway, we have someone who wanted to ask a question. They have a hand at least, 38:28 crumbs. Hello. Hello. Hey guys. Thanks for letting me up. And yeah, Paul, I'm really 38:37 impressed with sort of all the work you've been doing on, on Drivechain since like, I guess, 38:41 2014 or around that time. So I was wondering given that the Bitcoin protocol has changed a lot, 38:50 right? SegWit has been adopted, taproot has been adopted, replaced by fee, you know, other changes. 38:56 Why, why was Drivechain never adopted on a, on a protocol level? And can you talk a little bit 39:01 about your efforts to, to kind of get a consensus around this protocol change and what, what has 39:09 been the biggest barrier to adoption? Well, I think it went through different time 39:16 phases and you can also chop the problem down different ways. So for example, 39:20 to this day, there is no complete, we never made, even though we made test software, 39:27 we never made a fully complete pull request into Bitcoin Core that would have done all the 39:34 Drivechain stuff. So we never did that. And so in that way, it wouldn't have been possible, 39:41 but that's not really an answer to your question at all, because what could have happened could 39:45 have been that I published this blog post in November, 2015. 39:50 And then everyone reads it and then everyone's like, Oh, we should drop what we're doing 39:53 and work on this because this is clearly the way to go. So even though that is the 39:58 kind of a literal answer to your question, it's not like really a good, 40:03 it doesn't really answer the thrust of the question. And I give kind of a rambly answer 40:09 about this sometimes, but I'll try to like those three sort of phases. And in phase one, 40:17 everyone was distracted by the scaling war, which was resolved with these conferences, 40:24 scaling Bitcoin. And the solution that everyone wanted to do first was SegWit. And I was very 40:31 supportive of this. I also thought that was a good idea to do that first. And it seemed like 40:37 that would be done soon. And then there was a second phase when people were even more frustrated 40:44 and the scaling war actually erupted, which was like a miners versus developers thing. So anything 40:48 mining based was kind of stigmatized maybe, or just not people weren't as interested in it. 40:56 And then now there is a, I would say the third sort of phase is that Bitcoin has rejected the 41:06 original idea that it would do everything. And it is now much more of a 41:14 it wants to just stay. People have rejected the idea that anything, it's kind of a paradox because 41:24 altcoins went through a phase when they used to suck completely. And everyone, you know, 41:29 like in 2011, they were all terrible. And everyone, the most shameful thing you could 41:34 do is even discuss them and Cisco and people, I'm sure they know the attitude that I'm talking about. 41:41 And then over time, some of the altcoins have gotten like kind of good at things, 41:45 but Bitcoin has kind of rejected this. I think also there's a whole trajectory of the scaling 41:51 war made it so that anyone who deviated from like the, there was like two polarized edges, 41:59 which was like large blocks on L1 was the large blocker group. And then there's like lightning 42:04 network. We'll fix everything in the small blocker group. And this idea is kind of like in 42:10 between because it's not large blocks on L1. And in fact, we have, we have people who come to these 42:16 spaces. They're very funny and they're great. Like Hector is one with, you know, we love Hector, 42:22 of course, but there are people, there are a few people who come here and they love large blocks 42:28 and they don't want even the large block sidechain. They get annoyed and even offended at 42:33 being asked. So it's not, it's not that, but it's not, sorry. 42:38 They can invest in Solana then. That's a monolithic like end game. 42:42 Well, they're in like a Bitcoin SV or whatever. I mean, Solana is an interesting case because 42:48 what I think about, I would always make fun of Solana all the time. And I use it as a joke, 42:53 you know, and I use it many times. So I was at Baltic County Badger. I was like, Hey, 42:56 if people want to put their money on Solana or whatever, why can't they, they sell their Bitcoin 43:02 and buy Solana? They can, should be, why can't they use Solana the sidechain? But I actually 43:08 kind of admire how just unapologetic it is, you know, like they have found their niche of just, 43:15 it's kind of like Tron also with USDT, where it's just like, we are what we are. And that's 43:22 much more threatening than hypocrisy, I think. Anyway, that's my rambly answer. It's not very, 43:28 it's not a very good answer. A lot of it, I think also is weird social nonsense. Like, 43:33 I should, I think I should write a big blog post because I get this question all the time. Like, 43:37 people always ask this, this idea seems like it's pretty good or worth trying. And it's just like, 43:43 why? But I don't, there's no technical answer. The chief technical critic of this idea, you know, 43:55 is Peter Todd. And he didn't, it's extremely obvious that he didn't, he hadn't visited the 44:00 website ever. Because even after we paid him to write this article, he didn't know that the 44:06 software exists. So what was his argument to, you know, not like it? Because I read that little 44:12 spat. And, you know, I thought that was really funny, but it seems like he just took the money 44:20 and didn't really provide like a real explanation or, or what did you make of that? 44:25 Well, you know, last week I actually read through the article because I had nothing else to do. But 44:28 I think I should make a, I should make a video like, because we're supposed to discuss it 44:34 on a podcast, but we haven't scheduled it yet. And I don't think, I don't think that he's trying 44:42 to weasel out of it per se, but I also don't think that he's in any rush. I mean, I think 44:46 subconsciously he's trying to evade. He lost his Bitcoin, right? 44:51 No, that's Luke Dashjr. Thinking of the wrong ideologue. 44:58 But Peter Todd, I think was thinking more like small blocks because of, I guess, just keep it, 45:05 keep the base protocol as efficient as possible for full nodes, right? And reorgs and all that 45:10 stuff just to reduce. Yes. And he was right about that, but he went too far and he, Peter Todd's, 45:16 if I had to summarize it, Mr. Crumbs and also Peter, so Peter Todd has hated merge mining for 45:21 a very long time, even back to 2014. And they did a let's talk Bitcoin episode about sidechains, 45:27 which ended up transforming into like a commercial for his own thing, tree chains that never 45:32 materialized and that it was complete vaporware and no one could figure out what had happened to 45:37 it. And I even asked people at the scaling for Bitcoin program committee, when we were trying to 45:41 figure out what were like, what, whatever happened to that idea. And even in very senior Bitcoin, 45:46 people were like, I have no idea at scaling Bitcoin three. Well, it's a long story, but 45:51 the Peter's core mistake is that he confuses nodes and miners. And he thinks that it's good for nodes 46:00 to be cheap and he's right. But then he also thinks it's, it's good for mining to be cheap. 46:06 And that's not possible. That's, that's a rejection of proof of work actually. 46:10 So, so he has this belief that miners, that miners must do something. And then that matters 46:20 and neither are true. It's not true that miners must run a sidechain node and it wouldn't matter 46:26 if that was required. And he thinks that basically, if you feel that, like if the 46:35 miners invent like these shortcuts, like I was saying, like if they have a shortcut to get the 46:40 state that doesn't involve them running a node, like we call this a, what would we call this? 46:46 Like a node knockoff or like a cheap node or something like that. If they do that and it works 46:52 a hundred percent of the time, then to me, that's just some normal thing that we do. Like, 47:02 I guess, but to him it's like a logically impossible or something. So I should make 47:07 it a little video with like, I mean, you could give a metaphor where like a node is something 47:12 that removes a problem because a node increases your confidence and trust. So a node makes it 47:19 so that you don't need to trust someone. They say, oh, I sent you the money. And you say, well, 47:26 how do I know you sent me the money? And then when you run a node, then you're like, okay, 47:31 now I know that I got the money. And you can look at your own confirmations. So that's like 47:35 what a node does. So we could instead say like a node, instead of deleting paranoia, 47:42 it like deletes heat or something. Like it's like an air conditioner. 47:48 And so we at the Bitcoin Core universe, we like build air conditioners and we want them to be 47:56 cheap in addition to working. So that's why how decentralized the air conditioner is, 48:02 is how cheap it is. We want everyone to be able to have air conditioners because we want all 48:06 the trust in the world. We want all the distrust to be destroyed. So we want the world to be very 48:11 cold indoors. And so is this a metaphor starting to make any sense? And then Peter saying like, 48:19 miners, if they have some, any other way of being cold, then that's illegitimate or something. I 48:25 don't know if that makes any sense at all. But this is Peter's problem is he knows that it's 48:29 bad if for the blockchain costs to go up, but he, to him, an L2 cost is an L1 cost. So he, 48:36 so that's why he just can't grasp the idea. But either you're a monolithic engineer or you're 48:42 modular, your thinking is, it feels like it's in one of the camps. And if he's monolithic, 48:47 he thinks it has to happen all in one layer just for security concerns. And the miner should 48:52 upfront that cost. And the Nakamoto coefficient of the full nodes should enforce that the security 48:59 of the miners to not decrease that coefficient if that work is happening off chain somewhere. 49:04 But he doesn't see that you could offload a lot of that cost without introducing the trade-offs, 49:09 like what you're doing with some of the, what we're doing with the rollups and the similar concept. 49:17 Yeah, that's exactly it. That was very well put. Like he doesn't see, 49:21 I almost want you to repeat it so that I can write it down because it was so well put. It was like, 49:26 there's no, he doesn't really, he doesn't actually believe that you can move the compute, 49:34 like to him, every mining cost is an L1 node cost, which is, if you go down that road, 49:42 and that's why I asked him, at the TabCom debate, I asked him, is every single ASIC 49:49 that someone buys, is that a block size increase? And he sort of dodged the question because 49:56 you can see it, when you put it as starkly as that, it doesn't make any sense. 50:00 Paul, would you say that the Bitcoin devs, by and large, I guess, and maybe even the Bitcoin community 50:09 have come to terms with the non-viability of lightning networks for fulfilling, 50:16 I guess, Bitcoin's earlier vision of electronic cash at global scale? 50:21 I think that, well, let me tell you, because the question is very emotionally charged, 50:32 in addition to being politically sensitive, which is why it's a great question, and I hope to give 50:38 an answer that's half as good as the question, but it's like, I'm not exactly sure how to phrase it 50:43 in a way, but one way of phrasing it is that I had someone come up to me at TabCom, 50:51 very senior person, I don't want to say who, but the person was talking to me, and they were kind 50:56 of annoyed, and they said, last year, they were talking about, Matt Corallo gave a talk, and the 51:02 talk was very good, I remember seeing the talk, last year at TabCom, Matt Corallo gave a talk 51:06 called, Lightning is Broken, and he and I remembered the talk, and we were like, oh yeah, this is not, 51:12 the person I'm talking about is a very senior person, and they're not particularly, 51:20 they worked for a lightning company at one point, they're like pro-lightning, 51:25 but they are kind of annoyed, because the talk was, Lightning is Broken, and it was a very 51:30 convincing talk by Matt Corallo, but then what he said was, and he's like, but nothing really changed, 51:36 was the gist of it, and so that was last TabCom, that was one TabCom ago, that was a whole year ago, 51:43 so now we're at the next TabCom, and yeah, there was more, each year, there is more of like, 51:49 the stuff that the very, very expert people complain about, expert people complain about a 51:54 lot of things, and then they'd say, we'll solve it with factories or something, and over time, 52:00 and over time, the complaining has started to filter down from like, a very few technically 52:09 elite people, and it's in the process of slowly filtering down to the less experienced 52:17 engineers and commentators and media people, and then down to the end users, 52:26 so that is something that is happening. Another thing that is happening is that the people are 52:30 doubling down on denial, and they are referring to things that aren't the Lightning Network at all, 52:35 like for example, if what the Lightning Network is, is this network of nodes connected by 52:41 bi-directional channels and HTLCs, now the bi-directional channel, that is very old, 52:48 that long predates the Lightning Network, Satoshi invented that also, that's why there's n-sequence, 52:55 it didn't really work in earlier versions of Bitcoin, but the idea of it was old, and even 53:00 people such as the hated Mike Hearn, everyone hates him now, he was working on it in like 2014, 53:10 so the bi-directional duplex channels between two people, that's not the Lightning Network, 53:16 and what now has emerged is a network based on credit, which has the disadvantage of requiring 53:25 you to care a lot about the reputation of who your node, your channel counterparty is, 53:32 so because, well the weird way it always did though, because the best case scenario was, 53:38 in many cases, you open a channel with someone, your funds are locked for two weeks, like if they 53:45 just go AWOL, so that's a cost, having your money locked for two weeks is a very expensive 53:52 cost if you really think about it, and then paying down one transaction fee on top of that, 53:59 and so it kind of always mattered who you opened the channel with, but now we just see it just 54:04 matters more and more and more and more and more, and as a result you, and so this is a rambly 54:12 answer, but what I was trying to get at is people are, you know, they'll say, 54:18 I just sent a one Satoshi payment through the Lightning Network, 54:22 but not really, you just advanced the Satoshi, you know, for free to a third party who could 54:30 just keep it, there's no HTLC enforcing that it goes to the end person, and that's because 54:37 the stakes are so small, and because you already must trust the nodes for so much, 54:41 so it's a complicated answer, but I think that there's a mix between people who are 54:47 coming to terms with it, and people who are doubling down on lies, and also the definition 54:54 which has changed, which is not necessarily the worst thing, but the definition is mutating. 55:01 If the definition continues to mutate though, eventually it will just be the case that, 55:04 you know, Venmo is a Lightning lot. Exactly. Yeah, that makes sense. I mean, 55:12 it makes sense, you know, the momentum of the noise has to die down gradually, and then of 55:19 course you're going to have people in denial who have really put themselves out on a limb, 55:25 you know, for the Lightning Network's idea. So, I mean, what we kind of saw with Lightning 55:30 Networks was that the idea was cool, but there were some issues with it. I think, you know, 55:37 JAG understands pretty well, probably better than I do, the flood and loot vulnerability, 55:42 and then you have the simple fact that even in the Lightning Network's white paper, 55:46 it clearly states that there's a certain amount of L1 bandwidth that's needed to really scale, 55:55 and a real scalability solution is one that works at the level of the architecture, 56:01 which I guess kind of Lightning Networks does, but in practice where the rubber meets the road, 56:08 it seems like Bitcoin doesn't have the bandwidth to really make it work for electronic cash at 56:15 global scale, and people have just kind of not mentioned that that's directly in the Lightning 56:24 Network's white paper. So, for me, there's two problems with LN. I think we covered that in some 56:30 of our research, but I think there was a notable developer as well that just quit LN Lightning 56:36 Labs or something like that, but the security researchers came up with the flood and loot 56:43 attack. That's something that's just inherently unsolvable. That goes back to the Lightning 56:47 Network white paper, which clearly says if you want eight billion people to do one transaction 56:52 a year, you need so much bandwidth, and that's where the new age big blockers came from because 56:58 they wanted to scale through LN, but in order for you to stop those attacks, if all eight billion 57:05 people are using this, and this is discounting IoT, right, when AI and IoT comes on, but when people 57:11 are using this thing to do one transaction to move your value off the LN back to the base chain, 57:16 so you don't have IOUs anymore, you have actual Bitcoin, you don't have the bandwidth, and it just 57:21 goes to the fact that Bitcoin is more of a court system. It's not meant to be that transaction 57:26 processor. There's two problems. Obviously, the flood and loot attack is unsolvable, so they start to 57:32 in the peering and onion filtering, they start to try to detect that, which is inherently unsolvable 57:37 because you don't have consensus, and the second is if you try to do other tokens, like you start to 57:43 think about USDT and under hyper Bitcoinization, it may not matter, but in today's world, there are 57:48 other tokens. If you start to try to do this with other tokens, even if they're colored coins, 57:53 you end up hitting the American call option problem, which again is unsolvable, and so you 57:58 can't, you'll have this arbitrage opportunity where people will be able to steal, it'll be easier to 58:05 create theft than it costs, and both flood and loot and the call option problem are easier to 58:12 take money than to spend on taking that money, so that's something that you just can't solve for 58:18 unless you have the consensus around it. And again, BIP300 and sidechains and putting the trust back 58:26 into the miners and 51% security kind of gets around a lot of that, and you think of Bitcoin 58:33 as more of an anchor settlement court system rather than some mechanism in which it's a 58:40 substrate of a transaction processor where people are doing Visa MasterCard type things. 58:48 Jag, there's some interesting potential in client-side proving though, like 58:53 ZK client-side proving for Bitcoin, is there not? I've heard you talk about that. 58:59 You end up at the data availability problem and Bitcoin doesn't have a solution for that, so 59:04 the Lightning Networks is all a DA play because you need to prove that the HTLCs, you can close 59:10 off and stop a spam attack and take your Bitcoin back out of the channel, and if there's not enough 59:15 bandwidth to hear your court case in enough time, the judge will simply put the hammer down 59:20 and you lost your money. And even in a ZK-based system, you could avoid some of the problems, 59:25 like maybe the call option problem could be mitigated to a point, but you still have the 59:30 problem of hearing the court case on the base blockchain, and if you don't have that availability, 59:36 you run out of time to put your dispute up. So unless the chain is ZK itself, 59:42 verifying it doesn't really make too much sense. 59:46 Yeah, what about the idea of a ZK, like a succinct proof being stored directly on Bitcoin for DA? Is there any viability in that? 59:58 No, the DA itself is not ZK-able. You have KZG proofs, which can prove sort of through cryptography that you're holding a subset of data. 1:00:09 But again, you're not under the Bitcoin philosophy, where every node is trusting itself only. 1:00:17 But the ZK is computational based. 1:00:19 So if you want to solve a computation execution somewhere externally, like RISC-0 with continuations, and you distribute that ZK proof amongst a thousand computers, and you're able to do sub-second block verifications of ZK proofs, for example, you still have the problem of, okay, if people are using this with Bitcoin or value, 1:00:39 how do they enter and exit in a safe way? 1:00:41 And that requires, especially once you're moving the computation off, some mechanism where the witness data or the data of who owns what starts to be stored somewhere. 1:00:52 And the more you think about it, the more it feels like the base layer job is just to be a court system for the DA itself and not be worried about the transactionality, like not where people are spending, receiving money for everyday transactions. 1:01:12 And that's just against the Solana monolithic philosophy. 1:01:16 And as soon as you get to modularity, you start to think about how can we just move the executions off? 1:01:22 And having what Starknet wants to do is put that opcode on the blockchain, but you still end up, you don't solve for the DA problem. 1:01:30 And the costs will still be pretty high because you're going to be using scripts to store that data and put a ZK opcode for the verification, but you're still storing a lot of data that never gets removed. 1:01:43 And it'll be better than what it is today. 1:01:45 It still won't scale to what we're talking about. 1:01:48 Yeah, I guess it wouldn't get to the point where you could, like just I guess an example use case, participate in a prediction market that has a small turnaround time, pay an incredibly low fee, wager a very small amount, like 10 seconds before the prediction expires, which by the way, you can do that on Rolex. 1:02:14 Well, if you take the Lightning Network volume that you're trying to anticipate, and you throw that into a DA layer, it also won't scale, right? 1:02:24 You're talking about potentially millions of transactions between two people even within a year. 1:02:30 And if you have to throw that on the base layer, it won't scale. 1:02:34 So if you show a commitment to that, and that's what Drivechains is doing as well, the base chain never has to learn. 1:02:42 There's no asymptotic cost or asynchronous assumption between the layer two and the base. 1:02:47 If you tie in any way the base and the layer two or layer three or whatever, then you have this asynchronous assumption not only of bandwidth, but also of some security constraint the miner has to worry about. 1:02:59 And that is where you start to get broken down. 1:03:02 You either have some mad, mutually assured destruction there, or you end up having some sort of takeover or political centralization issue where someone doesn't believe in the philosophy and gives up and tries to attack that chain. 1:03:21 These are really good answers. 1:03:24 I love them. 1:03:25 I just want to comment on something interesting about this whole data availability. 1:03:35 It sounds like it's just like one... 1:03:37 It's presented sometimes as though there's this big list of things. 1:03:40 Oh, this, this, this, this, this, oh, but then, oh, number 14, we couldn't check that box, data availability. 1:03:48 But really, this is the whole, it's the whole game, to some extent, because the L1 has a limited amount of space. 1:03:55 And the whole idea is how do we make the transaction happen, but not consume this space, and it's only make, you know, whatever, one megabytes worth of data available every 10 minutes. 1:04:10 So it's actually quite, quite at the heart of the matter. 1:04:14 And for some reason, I'm not really sure why, but in Bitcoin, the technical community or whatever, has evolved over time to really prioritize what you might call the unilateral exit, which is to say, we're using L1, and we want to use the Lightning Network. 1:04:38 The Lightning Network has to always give me something that I could broadcast on L1, I individually, could broadcast on L1 in order to get my money back from the L2. 1:04:54 After all, what if all the people I'm participating in the L2 with just leave, they go AWOL, absent without leave, they just disappear, then I have to have something to come to L1. 1:05:06 Now, that is the that has always been the philosophy. And that strikes through to like ARK, and other things also do this. And federated model might do this, I know, state chains do something like this. 1:05:22 But that is never that doesn't actually work. Because if you have eight, if 8 billion people, 8 billion people won't fit on L1, and then they go to L2, you have 8 billion people, well, then they can't, all of them can't go from L2 to L1. 1:05:39 So there's kind of a little bit of a paradox there, about this data availability, to push all the data away for everyone, but then give everyone something where they can come back to L1, everyone gets the unilateral exit. 1:05:53 That's actually a paradox. And there was something else that I really wanted to say that seemed important, very important about oh, data availability. Yes. 1:06:02 It's a paradox to this is, it's actually related, which is giving everyone something they can broadcast on L1. It's the same thing as solving the data availability problem, because it's just the data that you would have to put on L1 is split up among everyone who owns the L2, who's using the L2. 1:06:24 And so I just wanted to try to comment that what drivechain does is it says, well, we have this second blockchain where everyone over there has this new, common data structure that they all share, and that they all need in order to run a node. 1:06:42 And that they, that's kind of how I solve the data. That's how I dodge the data availability problem and also solve it. 1:06:50 And I don't know if this is making any sense as a rambling answer, but I think the only way of actually solving it is to dodge it because of what I was saying before about you giving everyone unilateral exit just recreates the problem on L1 and it will never work. 1:07:04 And this is exactly why what you were talking about before with the lightning white paper, saying that, and this actually dies into many other things about the lightning network, such as the creators emphasizing the HTLCs only work for high value payments and for low value payments, they don't, and just the whole thing is just a credit network anyway, which it is. 1:07:30 And that the quantity of bytes required to enforce the HTLC or the PTLC or whatever the thing is, if that quantity of bytes is already basically almost as much as in the L1, a small, small, small L1 transaction. 1:07:46 And as a result, the fee rate, you know, the fees on both networks and it being the same, the three, the dust to their shoulders are being the same anyway. 1:07:57 And under the flood and loot, you just flood every channel regardless of the amount that's in there and force them to hit the main chain. 1:08:06 But I guess in that scenario, the flood usually will target the low value channels and not the high value. 1:08:14 So if you're using, maybe that's why they mean to, it's a high value payment channel system and not low value penny transactions or low value. 1:08:23 Flood and loot is a completely different, there's many unsolvable problems with lightning. 1:08:27 I think of which flood and loot is only one, I think that's related to the channel risk. 1:08:32 But the HTLC is that if you actually were to pull the, if you were to say, let's say it's an ABC, A is paying C and I'm B and I have a channel open to A and C. 1:08:47 If the fee rate is currently, let's say the fee rate is currently $10 for a transaction, that's 220 bytes, that's the L1 fee rate. 1:08:56 So $10 for one transaction. 1:09:02 If I have a channel with A and a channel open with B, my channel with A is going to increase by one HTLC and the one on B is going to increase. 1:09:13 The AB channel is going to, the size of the transaction that would be broadcast, that is going to grow to include an HTLC. 1:09:24 And then potentially it depends, you know, I could, the BC one could also grow. 1:09:29 And now to, just to do, just to grow that I need the hash pre-image 32 bytes. 1:09:35 I need to have another output. 1:09:38 I need to have another, that's eight bytes just to count how many coins. 1:09:41 And then to spend these later, they're going to become, each of this is going to become an input where I have to provide the pre-image or just 32 bytes or I have to provide something. 1:09:50 And I have to provide another signature. 1:09:53 The signature has to cover this thing, 64 bytes. 1:09:59 So just to spend the HTLC is going to cost, like, it depends completely on the scenario, but it's going to cost like 100, 200 bytes. 1:10:10 Even in best case, like point time lock. 1:10:13 In any reasonable case, the H, just using the HTLC, the amount of ink it would require to write to the transaction to L1. 1:10:23 In the worst case scenario of everyone going AWOL, and I have to broadcast on L1. 1:10:30 That's going to cost me hundreds of bytes. 1:10:33 If it's already 220 bytes, it's already costing $10 worth of transaction fees. 1:10:38 Is this making sense? 1:10:41 In that scenario, I can never route a payment that's less than $10. 1:10:44 It's not even worth it to go to court. 1:10:47 It's like your court analogy. 1:10:49 It's like the court charges, the court is charging $10 to hear an HTLC case. 1:10:55 If I route $7 and then they just leave, I'm not going to court to collect $7 if the court is going to charge $10. 1:11:05 So that's just one of the many problems is the sensitivity to L1 fees. 1:11:10 The onboarding is the biggest problem with the Lightning Network is that just to join the network, people need two of two multi-sig output. 1:11:17 They need their own two of two multi-sig output. 1:11:21 Even in the most absurdly generous case, this is not unrealistic and will never happen and has terrible UX. 1:11:27 They would need to have joined some kind of group, some kind of coalition, provided the public key, the XPUB, 1:11:35 and then have it do some kind of statistical crypto magic to create the output that we will then all spend the first coin, the funding transaction. 1:11:47 So the onboarding is huge. 1:11:48 Privacy is not good. 1:11:51 The payments fail. 1:11:52 The payments really can't fail. 1:11:53 It's not a big difference between. 1:11:55 So it's like the on-chain, the payment has trouble making it into a block, but it hits the mempool instantly. 1:12:07 And then once it's into a block, everyone knows that everyone can kind of estimate for themselves how likely it is to settle based on the number of confirmations. 1:12:17 So once you get one confirmation, most people in most contexts would just think, well, this seems really, really, really likely to be going through. 1:12:27 It seems legit. 1:12:28 And then even the exchanges only require three-ish confirmations these days. 1:12:36 And then if you want more certainty, wait for more confirmations. 1:12:38 Yeah, you could detect an attack as well. 1:12:41 They could just wait for more confirmations if everyone knows there's something happening. 1:12:46 But the DA problem, we classify it in the modular blockchain paradigm as a specific problem. 1:12:57 We understand that all blockchains, it's all they are is data and proving the data is mapped to some state. 1:13:05 But specifically when you're talking about blockchain, modular blockchain, where you're separating execution state and not as much with UTXO, but with the accounting model, you do have the roll-ups proving just a Merkle state or just a hash that gets put on the chain. 1:13:22 But for people to be able to know, not that 8 billion people are going to move back to the base chain, but just that they have immutable decentralized ability, should they choose to one day do it, they have the ability to do it. 1:13:35 The roll-ups can say they inherit the base layer security if that data availability exists so that someone can take that data and prove to the Merkle state that they are part of that state. 1:13:49 And to solve that problem, you have to take the data off of the blockchain, but you have to do it in such a way that the blockchain knows that the data has been posted and that's a specific problem once you get into modular blockchain. 1:14:03 And the problem would be that if you have a validator or a layer 2 block producer and he chooses to withhold the data or change one byte, it's called the fisherman's dilemma, then they could manipulate that byte and then post the state on the layer 1 and no one is able to move their money back off of the chain because the state has been updated with the bad Merkle route. 1:14:27 So in order to solve that problem, the validator must prove that the hash of the data exists on the layer 1 and then commit that hash into the state that they're updating. 1:14:38 Therefore, when the state gets updated, it checks, okay, does the hash of the data exist because the network would have validated it? Yes? Okay, then the state's accepted. 1:14:47 And with the ZK roll-ups, they actually go further and put the hash in the public verifier, we check the circuit, and it checks that the data actually was committed, it was on chain, and it's part of the circuit. 1:15:03 And so there's no fraud proof assumption there either. 1:15:07 So that's a really, really cool breakthrough that we've considered on the data availability. 1:15:13 Because you're only proving short term during the time of confirmation, you can actually delete the data after from the blockchain. 1:15:21 So not only do the miners only propagate raw data without checking it, just a simple hash, and that data could represent all the transactions, that raw data is really cheap. 1:15:33 It's bandwidth cost, but it's cost that goes away after the finality event. 1:15:39 That's why we have finality. 1:15:40 After the finality event, you remove the data, and all you get left is a hash of the data. 1:15:46 So when you replay the blockchain, you're just replaying hashes that represent any number of transactions that have happened off the chain. 1:15:55 So that's specifically, in the Ethereum or EVM world, when we talk about DA, the DA problem, it's a short term data store that gets removed from the base blockchain after some time. 1:16:07 And you're using that in your roll-ups to prove that data was stored to solve that problem, the withholding data problem. 1:16:15 Yeah, because of the pruning, it helps full nodes be able to run more easily and sync very easily with the network. 1:16:24 It keeps things lightweight. 1:16:30 Okay, we have some other people who are now speakers. 1:16:34 Grant and Klaus. 1:16:37 When you were talking about Nakamoto consensus and what it takes to allow for miners to do something but not have the energy so high, 1:16:52 I was thinking about the Nakamoto consensus that Bram Cohen created for Chia. 1:16:57 Have you looked into proof of space and time much in the hard drive utilization? 1:17:02 A very long time ago. 1:17:04 I think it was weird because I was looking into it in the context of something that would be an exception to my nothing is cheaper than proof of work essay. 1:17:16 Because I thought there's a trapped market for hard drive space, which is that you buy a hard drive and you start filling it up. 1:17:32 But until you fill it up, there's a lot of empty space on there. 1:17:42 So what I was thinking was, this is something where it is sort of a free resource to you, but to the network, it could be worth something. 1:17:52 And in that sense, it might be an exception. 1:17:55 But the paradox is resolved if as soon as someone invents like a SIA or a Filecoin or a storage, then that becomes a valuable resource again. 1:18:07 And then you start wasting it. 1:18:09 And I've heard it's pretty brutal. I don't know if this is to be the case, but I have heard that it shreds the hard drives really quickly. 1:18:17 I don't know if that's true. 1:18:19 So the way the consensus mechanism works is you take the ASIC style portion of what Bitcoin is doing. 1:18:27 You do that once and then you write it to the hard drive. 1:18:30 So that part, which is called plotting, is the part that has been referred to as shredding hard drive. 1:18:38 Because that is such an intensive process that when it was originally done, people did it on an SSD. 1:18:45 And there's a lot of read-write that happens during that process. 1:18:49 And so the early plotters on Chia, they would use a technology called NVMe or they would use RAM to do the plotting. 1:18:58 And that eliminated that issue. 1:19:00 And so what happens is you write a hard drive completely with data. 1:19:05 So you fill it up. 1:19:06 And then for the consensus mechanism, you're occasionally reading from the drive. 1:19:12 So there's no write load on the drive once you reach steady state. 1:19:16 So after you've done that first bit of work, the only thing that's happening in the consensus mechanism is you're doing occasional reads from the drive to see if you have a proof. 1:19:26 So the way that they describe it is it's like proof of proof of work. 1:19:31 And there's an attack that if you were just using proof of space, that there are certain ways that you can subvert that. 1:19:39 And so the next part that needed to be brought in was a verifiable delay function. 1:19:45 And so on Chia, there's actually two elements that are participating in the consensus mechanism. 1:19:51 It's the distributed network of people who have hashes on hard drives. 1:19:56 Operating in the same facility that ASICs are. And then you have these entities that exist on chain called time lords that are running a VDF. And that VDF has to verify that a certain amount of time has elapsed and then is responsible for emitting the signage points that are then distributed to the network. 1:20:16 So if you look at the Nakamoto coefficient for Chia, it's actually quite high. And there's a broad base of participation. So I think from a Nakamoto consensus mechanism, this one is fairly interesting from it being a novel. 1:20:34 The hardness of those coins are dependent on the hardware, the disk, Moore's law cost, right? So as the hardware gets more efficient, the hardness of the coins being output and the security of the coins through the verifiable delay is ensured, right? 1:20:52 And so my concerns there would be around the hard drive costs and the abundance of hard drives, especially with artificial intelligence being able to go into servers and attacking them and taking them over and using that for re-plotting. 1:21:07 And then also just dynamically on the VDF or the time lord being able to manipulate that because the hardness is based on hardware rather than a simple hash function, which even in the quantum realm, you cannot get around the XOR and the AND gates that are happening within the SHA-256. And so it's actually much more secure and much harder to manipulate that than go and get some hardware, some disk, and manipulate that to do your attack. 1:21:37 Well, I'm not sure about that. Like, if you're talking about the capability to plot to win, right? So this is getting down into some of the deep guts of how this stuff works. But there's a way that you can determine whether the files that you have on your hard drive could possibly contain a proof for the signage point that's been proposed to the network. 1:22:03 But let's stop for a second. So let's assume Chia has found a way to do modular scaling, right? I'm assuming there's a general purpose execution, right? Not just file storage, or is it just file storage? 1:22:17 It doesn't do file storage at all. Like, Chia is not doing SCIA or Filecoin type operations. 1:22:23 If it's general purpose execution, you're going to be doing modular execution, right? Then you end up having some sort of mechanism where the proofs get stored for a state transition of that modular chain. 1:22:36 The event of manipulating that just once is a fault tolerance on the system level that we're looking to try to prevent. If you can even do it once by rolling back or manipulating that state, that ends up creating a fault in the entire design, in my mind. 1:22:51 And that's why I was thinking, all it takes is someone to get ahead on the curve on the hardware for the disk to be able to plot an attack once, which undermines the system almost forever because you have to fork and manipulate the algorithm around it to prevent that scenario from happening. 1:23:08 Well, in the two years that Chia has been in operation, what you've described has never been demonstrated, so I don't believe that that type of attack exists, but I mean... 1:23:21 Yeah, I mean, of course it has. I'm just saying, theoretically, if you wanted to get ahead and try to do that, and if Chia scales up through a modular approach, like we're talking pushing 1,000 plus transactions per minute or per second around these sidechains, then you end up having this asynchronous... 1:23:42 Yeah, I really wasn't talking to the scaling capabilities in any way. That's, I think, something that's been yet to be demonstrated on chains. Chia has been operating without L2s, so I think with respect to how... 1:23:58 But you know that my normal commentary... 1:24:02 Like, I wrote that essay, Nothing is Cheaper than Proof of Work, because the proof of stake people were claiming, not only is this better or whatever than proof of work, but also it doesn't waste... 1:24:18 Proof of work is just wasting a bunch of resources. 1:24:22 And so I was like, well, you know, whatever you do, there's this pile of money coming in. There's the newly minted coins plus the transaction fees. This is coming in. It's coming into the blockchain, you know, it's coming, it's being paid to someone. 1:24:36 So people are going to fight. You have to build some clever way of people fighting over it that doesn't cost any resources. So that's why I was originally looking into the proof of space and time, because I was like, maybe if there's more hard drives out there that are just empty and just used by normal people versus the... 1:24:55 Although it's hard to tell, because you mentioned that one thing about AI hacking into a server or whatever, AWS. But I was also thinking of, doesn't the NSA have all these multiple giant facilities that just have absurd amounts of storage space, and they just log everything in the hopes that there'll be an encryption vulnerability one day, and they'll just decrypt everything? 1:25:17 Well, I mean, on the network, right. So it operates just like Bitcoin from the perspective of, like, what are you trying to do when you're making a new block, right? You're trying to have the correct answer to a particular math problem, and it's different on Chia. So it uses a different set of algorithms. And so if you have the correct input to that, then you can sign the block and you swoop up what's in the mempool. 1:25:41 So I think there's this misconception that Chia is somehow like storing data. It's like, what you're doing is you're storing proofs. So those same proofs that are being generated by Bitcoin are sitting there on your hard drive. So you're just looking them up. So from the perspective of, like, what does Chia look like? It looks like Bitcoin in so many different ways, like it's a UTXO... 1:26:01 No, no one's disputing that. I'm just saying, you're saying, what do you think of proof of space and time? I famously wrote this essay about, like, these are all just different versions of proof of work. And I mean, so one way, I think that if you really want to look at it from my perspective, I think, let's fast forward to the future. Everything happens on Chia. Chia is collecting $700 million a day in transaction fees from everything, which includes payments, 1:26:31 and smart contracts, and data storage, and whatever else. So just a billion, a billion a day is coming in. There's no US dollar, because we've had hyper Chia-ization. And there's only Chia, right? So, but whatever was a billion dollars worth of stuff, that's how much you can buy with the crypto that comes in each, each Chia block each day, or each day's worth. 1:27:00 So, in that scenario, someone out there is going to build the equivalent of an ASIC for, to mine Chia, or whatever. And they're going to build something, and it's going to have just the right amount of RAM to do that thing you mentioned before, and it's going to have the right amount of storage space, and it's going to have great heating. So they'll just build an ASIC version of that is my, is my extremely strong suspicion. 1:27:28 So yeah, so I'm kind of very happy with my, I just sit here on my little hill. I'm like the king of this little hill, where I just say, everything, it just is proof of work. And I've been living here a long time, and I don't really see why I'm going to move. 1:27:43 Yeah, that's, that's definitely an argument. And that's been demonstrated to some extent, because that's called plotting to win. And we've demonstrated that with GPUs already. It's not about whether you can do something like that. It's about whether it's more cost and energy efficient to do it with a hard drive, right? Because you can simulate a certain amount of space on Chia with a GPU. So just purely with a GPU, you can plot to win, you can keep your plot in memory. 1:28:13 And then you can use that. So that's already possible. Like, 1:28:17 yeah, but that's great. I'm saying, I'm saying your revenue is now gone up to $365 billion a year. So can't you go into the lab and design something that's going to be more efficient for you to just produce and run than what people are doing in their home? Maybe, maybe not. 1:28:40 There's intrinsic levers built into the protocol that caused that to not work. So like, one of the things that you have to do when you're proving that you have a certain amount of space is that the space you have is a certain size. And so ultimately, what you get down to is bus speeds and how fast you can move data across, you know, whatever it is that you're going to use to generate the data internally, and then make use of it. And so when you talk about an ASIC, 1:29:10 it's not exactly the special purpose kind of hardware that you would want to design. But you're talking about something like maybe an FPGA that would be able to do something like this. And then being able to move the data around on the bus. And so it's not like the team didn't think about this kind of stuff. Like they said, Hey, let's look out 100 years, like what, what is it going to take to keep this chain resistant to this type of stuff. 1:29:35 And so there's there's inherent levers within the consensus mechanism that say, the plot filter can require a larger size file. So right now use K 32, it could require K 33, or K 34. And if that occurs, then you wouldn't have consumer grade hardware that could do this. Now, if you're saying, well, the NSA could build something like that. And it's like, okay, maybe like you could go after that. But if the size of the network is growing proportionally, right, it still might not be cost efficient to even 1:30:06 try to go about, you know, simulating that amount of space. 1:30:09 That's civil defense, right? So you're, if AI takes multiple servers of small servers over, you're not able to detect those events, you can't differentiate those between regular users, there's going to be small plots, right? I had a question was, is there a longest chain rule? How's the fragility handled in that? 1:30:29 It's the heaviest chain, so longest and highest CLVM cost. 1:30:33 Yeah, so that helps. Because if someone was to attack, over time, you know, the longest chain or the heaviest chain is a ghost, it should prevail. So anyways, that does help. But yeah, I would lean to look into like the nitty gritty details, for sure. Obviously, it's really cool design. I, you know, I've gotten some feedback from what it's about. But um, yeah, I always the devil's in the details. 1:31:01 So for me, like the fault tolerance is the main thing. If you can cause one fault ever, I think the heaviest chain might prevent that if you can ever cause one fault, then when you're scaling up, and you're going modular, that's when those risks start to appear. But it seems like it's really, really good as a base layer as a monolithic design. I'd love to see how it scales up once once it starts to get going. 1:31:27 I think the other thing that's worth considering as far as like what Bitcoin's been attempting to do and sort of been listening to the Bitcoin VM guys bemoan the lack of certain types of opcodes that they wish they had. 1:31:39 So for example, we had Chia adopted, you know, a Lisp language, which is, you know, similar to the approach that Cardano took, which is having a functional language that's relatively complete in its ability to express ideas, and then keeping what's on chain pretty small by saying, we're not going to reveal what the smart contract is until you actually try to spend one of the coins that exists. 1:32:02 Starting from the hash and saying, each coin has a hash, and if you want to spend this coin, you have to reveal what the smart contract is, and what solves that smart contract in the spend bundle. And that's the point at which you pay those costs to be able to move that coin to some new owner or some new contract that secures it. 1:32:25 Yeah, so like a UTXO sort of model built into the smart contract. 1:32:30 Well, each coin has its own attached smart contract. 1:32:34 So just like how Bitcoin has Bitcoin script attached, and it's like almost what Chia did was say, pay to script hash is the primary thing that we want to do, as well as having, you know, something equivalent to partially signed Bitcoin transactions, which is Chia offer files to eliminate counterparty risk and allow you to trade assets back and forth. 1:32:56 There's some other interesting stuff that's happened with Chia with respect to, like, how do you actually do, you know, bigger, harder problems like making an AMM singleton spend? 1:33:16 We lost that person, I think. 1:33:20 I think everyone got muted. 1:33:22 Okay, we're back now, though. 1:33:25 The last thing you heard me say is like, I was just trying to say a few things about like, how the on-chain programming is working. 1:33:30 I think you said AMM. 1:33:32 Yeah, like just how you do more sophisticated style programming, like AMMs and like singleton spend aggregations. 1:33:39 And how do you actually do some of these more sophisticated vehicles that we've seen on Ethereum? 1:33:45 And how do you do that in like a UTXO model? 1:33:48 Yeah, Cardano is also tackling that as well. 1:33:50 Anyway, did we ever get to Klaus? 1:33:54 We should probably get to him. 1:33:56 I think Ron covered everything. 1:33:58 My mind is melting right now. 1:34:00 Well, yeah. 1:34:02 We'll post the recordings later so people can listen when they're in a very, very busy time. 1:34:10 We'll post the recordings later so people can listen when they're in a very, very quiet room with a pen and a piece of paper. 1:34:20 And listen a second time. 1:34:26 It's great to finally meet you guys. 1:34:29 Likewise. 1:34:30 Nice to talk to you on Spaces. 1:34:32 I think we've messaged like once or twice, I think, before over the years, maybe. 1:34:40 Paul, I'll definitely give you guys a follow. 1:34:42 I hope we can continue this discussion and further Spaces if you guys would like. 1:34:48 Cool, great. 1:34:50 Excellent. 1:34:52 All right, back to Klaus. 1:34:55 Yes, thanks for stopping by. 1:34:57 We'll just go back to our regularly scheduled rambling now, I suppose. 1:35:03 Oh, yeah, welcome. 1:35:04 I think we do Spaces as well. 1:35:06 We'd love to have you on as well. 1:35:08 Just chatting. 1:35:09 Absolutely. 1:35:11 All right, cool. 1:35:12 Thanks for having us. 1:35:13 Hope everyone enjoyed the discussion and I'm sure there will be more in the future. 1:35:19 Yeah, for sure. 1:35:20 Thanks a lot. 1:35:25 I had a comment to make about what Jagdeep was talking about scaling. 1:35:31 I feel like there is an inherent issue in both the UTXO and account model with the scaling aspect, 1:35:40 you know, at least for on-chain, because there's a lot of coordination that is involved. 1:35:46 And I'm not sure what he was referring to when, you know, these popular buzzword these days is 1:35:52 modular blockchain. 1:35:53 I mean, is that consensus separate? 1:35:57 Because, you know, UTXO has outputs and it has to coordinate as well to, 1:36:08 you know, you basically have downloaded the entire blockchain. 1:36:11 So it still has to, on a local machine is fine, but among other machines, 1:36:16 it still has to do the coordination part. 1:36:18 So does the account model. 1:36:20 So I'm not quite sure that, how these change address any kind of scalability, 1:36:26 because that's where the bottleneck is, is the coordination. 1:36:33 Well, I'm not an expert on that particular topic, but I do think the one advantage of the UTXO 1:36:39 model is that it is easier to chop into tiny pieces, you know, 1:36:45 easier to chop into tiny pieces, you know, 1:36:49 where you have your UTXO set, and then you bolt on a bunch of things. 1:36:55 Each one of those could usually be discarded or added back. 1:37:02 And so the damage is limited, whereas with the account model, 1:37:07 each thing could affect the state of like everything else. 1:37:10 And so it's just a little more interactive. 1:37:13 But I don't really know specifically exactly what that was referring to. 1:37:18 But I know the account model has numerous disadvantages. 1:37:22 Yeah, definitely. 1:37:23 I mean, UTXO is definitely better in that respect. 1:37:28 What I'm looking from the both point of view, 1:37:30 like even UTXO and account models still require two things to be changed, 1:37:39 like the bearer and the sort of amount. 1:37:43 And so it's still not like one-to-one like cash exactly, 1:37:49 where you just change the bearer, right? 1:37:53 And so it's interesting. 1:37:55 That's where I think most of the scaling issues come in these approaches. 1:38:03 And I'm not sure. 1:38:06 I was just trying to ask him if deep left, like how modular approach solves it, 1:38:11 but because it's still be based on either UTXO or account model. 1:38:15 So I'm just trying to wrap my head around. 1:38:20 The coordination problem doesn't go away in either, both of them, right? 1:38:27 Yeah, that is a good point. 1:38:28 I mean, my view, which is not his, so I don't know what he would say. 1:38:32 But I think it's like the, you know, you have like small blocker, 1:38:37 large blocker type people in every paradigm. 1:38:41 And so the large, some people want the node to be more expensive. 1:38:46 And as a result, they are already game for, 1:38:56 so they're already game for paying more. 1:38:58 And that's the modularity coming in to save the day. 1:39:01 But I don't know. 1:39:02 I don't know what they would mean. 1:39:04 It's a good question. 1:39:06 Yeah, because there's no inherent parallelism, right? 1:39:09 In the transaction, like the congestion will happen as a result of coordination. 1:39:19 So I think L2s will be required for sure. 1:39:23 Unless you have some completely different paradigm from UTXOs and account model. 1:39:31 Anyways, just some thoughts. 1:39:37 Yes, definitely. 1:39:39 Okay. 1:39:39 Anyone else have any, anyone else want to come up and speak or have a question or comment? 1:39:52 Anyone want to say, should we just talk about the, the boring topics of the day? Like you 1:40:02 want to talk about Sam Bakeman Fried or whatever everyone is talking about, whatever all the 1:40:08 cool kids are talking about. We try never to do that here. We only talk about the pure 1:40:13 alpha cutting edge. We are the, we're like the hidden jazz club, you know, that none 1:40:20 of the, none of the other people know about. We do the best, but you know, we can just 1:40:23 hang out and talk about fun things. 1:40:25 I'd like to ask you a question about what you think about what's going on with Bitcoin VM. Like 1:40:29 what the, are you paying attention to that at all? 1:40:32 You mean BitVM? 1:40:34 Yeah, yeah, yeah. 1:40:36 Well, I think it's a, it's a very good idea. In a way, I think it's related, it's kind of 1:40:41 what I was talking about before, where the whole community has gone in the wrong direction, I 1:40:46 think, which is that what I was saying before about the L2s is that there's a direction of 1:40:52 always requiring each person to get something that they could bring unilaterally to L1. 1:40:58 But I'm saying that's kind of pointless because there's no way everyone can use it, so 1:41:01 there's no way it's really been given. 1:41:04 And what I mean in this case is there's this whole like, ever since the block size war and 1:41:10 the frustration with scaling, the frustration with SegWit, and the frustration with Soft 1:41:18 Forks, that has led the community to evade the Soft Fork issue and delay Soft Fork. 1:41:23 Soft Forks went from being just get 51% of miners to upgrade to the software to like an 1:41:30 unjumpable hurdle that cannot be an impassable barrier. 1:41:35 And as a result, people are now interested in this type of thing with, can we do something 1:41:41 really, really complicated that the end user is not going to understand, so it really won't 1:41:47 be self-sovereign, it would be a black box, it might as well be closed source. 1:41:52 Should we do a very complicated thing in order to get, you know, basically Bitcoin, the EVM 1:41:59 onto Bitcoin? 1:42:01 And so I think it is actually, in a way, it may be part of the healing process, because 1:42:07 once people see that you can do it this way, they'll want to do it and then they'll want 1:42:12 to do it a better way. 1:42:13 And then they'll evolve back toward the correct answer of just doing it the efficient way 1:42:19 with modular sidechains. 1:42:22 So I think as I've started to look at this, it really does bring up the fact that Bitcoin 1:42:27 script is pretty limited in what it's capable of doing right now. 1:42:31 I'm wondering if this is, you know, a sea change moment where the development community 1:42:38 starts to really push back and say, look, you know, we have all these opcodes that are 1:42:42 disabled and they're causing us to be able to just not do fundamental, like fairly simple 1:42:48 operations. 1:42:49 And some of the stuff that BitVM is demonstrating here is it's like, all right, if you're not 1:42:54 going to give us, you know, the opcodes we really need to do what we want to do, then 1:42:59 we can go down and build them in digital circuits. 1:43:02 And then from there, we can do these very convoluted things that allow us to do whatever 1:43:06 we want. 1:43:07 So it might force the issue just from the perspective of like, we're going to do it 1:43:11 anyway. 1:43:12 So either you make it in programming language or we're going to compute all these pre-images 1:43:18 offline and use them to prove that these digital circuits are working correctly on Bitcoin. 1:43:23 Seems like we're trying to push Bitcoin script with this whole idea. 1:43:31 I think that's correct. 1:43:32 And I think that it is also, I'm not really sure how to go about saying this, but there 1:43:39 is a kind of, you have to equivocate if you're in Bitcoin, you have to take two different 1:43:47 things at once where you say, well, it's really cool when things are built on Bitcoin. 1:43:52 And then you also have to say, Bitcoin's perfect the way it is. 1:43:58 And so if you can't build on Bitcoin, then that's also good. 1:44:02 And those are just two different things. 1:44:04 So, but I think that you're right, that part of this would be, I mean, Adam Beck often, 1:44:11 very often tells a story about how people didn't want to add support. 1:44:15 They were worried about adding support for some kind of thing in, I don't remember what 1:44:20 it was. 1:44:20 I think it was like a C++ compiler or something like that. 1:44:24 Ancient history. 1:44:26 And people were worried about it. 1:44:28 And then someone demonstrated that you could emulate that, whatever the thing was, you 1:44:35 could emulate it already. 1:44:37 And then they just dropped the issue. 1:44:39 They added it because they said, well, you can already do it. 1:44:42 So, I feel like that's where we're at. 1:44:49 With this whole discussion and what's being demonstrated, it's like you can keep withholding 1:44:53 updates to the language, but you can't fundamentally stop computing. 1:44:57 And if you make us do it in this really convoluted way, it just makes us do a lot of offline 1:45:02 computation to be able to do the same thing that would be much more reasonably done. 1:45:08 It's just part of the programming language of the chain. 1:45:13 I had a quick thought on what I was talking earlier really quickly. 1:45:26 And I'm going to say, historically, all the blockchains with UTXOs and as well as account 1:45:36 model, they at least involve at least two transactional units. 1:45:42 Because you have this interconnected ledger and history of each asset in the ledger is 1:45:49 kind of dependent on the other assets. 1:45:52 So, I think there's the limitation of any performance improvements because you can get 1:46:05 this congestion, additional layers, such as with rollups. 1:46:13 So, you're looking at what AVAX or these folks do with federated consensus. 1:46:21 I mean, they do introduce compromises, of course, in security and performance. 1:46:26 So, I don't know where I'm going with this, but I'm saying if there is a way to have these 1:46:35 cross-dependencies reduced or these assets be independently updating and don't require 1:46:44 any sort of coordination. 1:46:45 I mean, that would be an example of a perfect sort of parallelism, analogous to, say, real 1:46:52 cash in the real world. 1:46:55 No, but that is the goal with sidechains is that they're all independent pieces of software. 1:47:00 And each person is a benevolent dictator of their own little realm. 1:47:05 And they do whatever they want over there. 1:47:07 Right. 1:47:08 But, yeah. 1:47:29 Come on up if anyone has a question or comments. 1:47:32 Happy Friday to everyone. 1:47:35 Um, what's coming up? 1:47:38 Anything good? 1:47:39 Nothing, really. 1:47:43 There's the Texas thing in the middle of the Thanksgiving is coming up. 1:47:47 Before that, there's that mining thing in Texas. 1:47:51 That could be fun. 1:47:52 Right now, isn't there something now in Dubai? 1:47:55 Not there, but whatever it is. 1:47:56 I think there's something in Amsterdam right now. 1:47:59 In April, there's a big one. 1:48:01 I went to the Singapore one. 1:48:03 The 2049, I was crowded. 1:48:05 10,000 people. 1:48:07 Oh, good. 1:48:08 Yeah, the Dubai one is going to be a bigger one because, you know, I'm in podcasting and 1:48:13 stuff. 1:48:13 So, yeah, it was pretty massive. 1:48:15 Singapore, a lot of companies showed up. 1:48:20 And Dubai one is going to be a bigger, supposedly bigger. 1:48:25 So, yeah, Asia is buzzing. 1:48:28 Actually, I compare US and Asia. 1:48:31 I'm so slow right now in the US. 1:48:35 Yeah, that was also my experience that in Asia, everyone was very optimistic, excited 1:48:42 all the time about everything. 1:48:44 And also US, that's when also I was like, USDT is taking over. 1:48:49 And this mirrors an experience I had before when it was like all the NFT stuff was doing 1:48:55 ETH. 1:48:55 Like, I'm not really active with NFTs at all, to say the least. 1:48:59 But I think people should be free to do it. 1:49:02 I don't consenting adults to do it. 1:49:04 And I know that they like the art market. 1:49:06 They people are using it for like all kinds of money laundering and tax evasion. 1:49:12 Yeah, which is fine for them. 1:49:13 Good for them. 1:49:14 But I was just trying to say that USDT is taking over in Asia. 1:49:17 But everyone, but people are loving. 1:49:19 They love everything. 1:49:20 They love every project. 1:49:21 They love every coin. 1:49:23 There's no Bitcoin maximalism over there, I think. 1:49:25 And now that I could tell. 1:49:27 Yeah, exactly. 1:49:28 That's what I felt. 1:49:29 And you know what? 1:49:30 I'm kind of scared of this naive optimism over there. 1:49:35 Actually, we don't want SBF 2.0 happening over there. 1:49:39 But from what I'm hearing, the regulators are pretty strict. 1:49:43 So maybe what we do see like this, the overly optimism in the space over there that every 1:49:51 project is cool. 1:49:52 It's probably people are catching up to the tech. 1:49:55 I mean, I had some hard questions from some folks on stage, and they literally would dodge 1:50:00 the questions. 1:50:00 And people wouldn't even worry about it because I think it's still new for a lot of folks 1:50:07 there. 1:50:08 But yeah, I kind of agree with you that it's like, what's going on here? 1:50:15 Like it's everybody is so positive about every project and nobody's like critical about how 1:50:22 this is going to scale. 1:50:23 What are the security assumptions? 1:50:25 Blah, blah, blah, this, that. 1:50:28 Yeah, I have the same feeling. 1:50:40 And a little bit of regulatory ping pong, so to speak, which was FTX collapsed last 1:50:49 November. 1:50:51 And that was bad, of course, for, I mean, you know, Bitcoin had nothing to do with it, 1:50:56 but we all suffer as a result. 1:50:59 And then regulators moved in at the same time that for all these other Asian markets, I 1:51:08 think, Bitcoin and everything had been banned. 1:51:11 Then it was like unbanned. 1:51:13 And then Coinbase, Binance, et cetera, people decided to fight the SEC in the U.S. 1:51:18 They sort of slowly won. 1:51:19 So we've kind of been trading places back and forth. 1:51:22 Very fast-paced industry. 1:51:27 Anyway, come up on, oh, we have Satoshi and Joyous here. 1:51:30 Hello. 1:51:31 Nice to see you again. 1:51:32 Hey, what's up? 1:51:33 Happy Friday, everybody. 1:51:36 So I have this weird thought. 1:51:38 I was listening earlier about the merged chains, and I'm wondering, maybe I'm overthinking 1:51:45 it, but I'm wondering, like, what's the actual purpose for a new token for a merged chain? 1:51:50 If it's merged anyways, it's like miningly merged with Bitcoin. 1:51:55 It's leveraging Bitcoin for security or like some sort of finality. 1:51:59 So it's like the new asset really just like trying to compete in the market, like for 1:52:04 its own niche. 1:52:04 I think you might know a little bit about that because HiveMind needs its own token, 1:52:09 but I still don't understand like where. 1:52:11 Ah, yes, that's a completely different thing, yeah. 1:52:13 So the way it started was with Namecoin, and Namecoin was an attempt to create a new project 1:52:21 for DNS and have a new coin that would float against BTC. 1:52:26 And it's a very intriguing case because anyone can go back and read the original Bitcoin 1:52:30 talk thread that has Satoshi and Joyous on it. 1:52:32 And he doesn't seem to be threatened at all by a new coin, and he helped invent Namecoin 1:52:39 and he helped invent merged mining so that each chain... 1:52:42 Because I think Satoshi thought that they might compete for mining strength, and he 1:52:49 thought it's better if they just cooperate and they can all merge the same. 1:52:54 All the transaction fees will go to the same group of people, increasing total strength. 1:52:59 That's his three words, increasing total strength is his quote. 1:53:04 So he wanted the hashrate to be very high, and he thought we can merge them all there 1:53:09 into one project, into one set of miners for free. 1:53:15 And that is what he did. 1:53:17 And there was no way of making it so that Bitcoin could be used on the Namecoin side 1:53:22 chain. 1:53:23 On the Namecoin sidechain, that is what, you know, that is what in drivechain, BIP300 1:53:32 aspires to do that we don't have. 1:53:34 And this is a long, long, long, long time ago. 1:53:37 And... 1:53:37 One second, let me stop you. 1:53:39 So if there was a way to move Bitcoin back and forth, do you think there's any purpose 1:53:43 for a new asset as Namecoin? 1:53:47 In the case of Namecoin, no, because I think the only reason they did it that way was because 1:53:50 they didn't know what else to do. 1:53:52 Whereas I would do it with BTC the whole way. 1:53:56 BTC paying the transaction fees. 1:53:58 So just in the case of that, you know, good mechanism of back and forth, that's the reason 1:54:02 why the other chain has its own asset. 1:54:04 But really, it's really not competing with Bitcoin since it's still giving all the all 1:54:09 the revenue fees to the miners anyways. 1:54:11 But in a way, it's still diluting the value because it has its own like floating price. 1:54:15 But you're saying it's only because there was no good mechanism to go back and forth. 1:54:18 Otherwise, it would just be useless to have another asset, right? 1:54:22 Yes. 1:54:24 In the case of Hivemind, there is something. 1:54:27 First of all, almost everything is done with the BTC asset, the cash asset, you know, 1:54:37 placing a bet, you know, betting in the markets, you know, going long or short, buying a share, 1:54:44 buying all the shares and opening your own little exchange with all the like opening 1:54:50 lightning channels with trades in them, creating a market, creating a question that you want 1:55:00 the oracle to answer. 1:55:01 So all of those things, basically everything is done with the BTC coin unit. 1:55:08 There is a different coin. 1:55:09 Is the fees for all these transactions also on BTC the unit? 1:55:12 Yes. 1:55:12 Nice. 1:55:13 Yes. 1:55:15 The reason why there's a second unit is because again, it's just was technically and economically 1:55:20 required in my opinion. 1:55:21 I may be incorrect in my, I'm like this peer to peer oracle, like mad scientist. 1:55:29 So I may be wrong about this, but the point of the, there's like an employees only section 1:55:35 that is sort of like a corporation where you buy in and then you're running the oracle, 1:55:40 you're answering the questions. 1:55:42 And we want, if that thing does a terrible job, we want basically the market capitalization 1:55:47 of that corporation to collapse. 1:55:51 So there must be an independently floating having the coin. 1:55:56 Exactly. 1:55:56 Yes. 1:55:57 Centralized type of company where people can come and go as, as pleased, right? 1:56:01 Right. 1:56:01 It has this very, very narrow, very narrow and well-defined purpose of doing the peer 1:56:07 to peer oracle. 1:56:08 And as a result, it's, it needed to have a floating thing. 1:56:13 And it also, it doesn't compete. 1:56:15 There's absolutely zero sense whatsoever in which it competes with BTC to be used as 1:56:20 money. 1:56:21 If that would be ridiculous. 1:56:22 Right. 1:56:22 So, and in fact, even on its own blockchain, it's not used for payment. 1:56:26 It's the same thing as like tokenization as far as just like just a company that has 1:56:30 shares. 1:56:31 It's not really competing as a base asset. 1:56:33 Another reason why I'm actually in favor, even though I don't, I don't, I haven't 1:56:36 really messed. 1:56:37 I, I don't really like participate in ERC20 or NFTs or whatever, but I think it's a 1:56:43 perfectly valid thing to want the blockchain to do because we do it all the time with 1:56:50 every stock, every bond, every IOU, every forward contract, every receipts, every, 1:56:57 when you buy concert tickets, when you catch the game ball at a baseball game, it's 1:57:03 like we have collectibles, you know, when they make only 2,000 of a certain type of 1:57:09 watch. 1:57:10 Yeah. 1:57:10 I agree as well that like, there's nothing inherently wrong with tokenization. 1:57:13 It's just, you know, like that, that's fine. 1:57:15 I think the issue that comes in is the misleading marketing that goes along with it. 1:57:19 Right. 1:57:20 That's really what everybody doesn't like. 1:57:22 And they conflate it together with just, you know, creating tokens. 1:57:24 But I think creating tokens is not an issue. 1:57:27 It's the marketing that. 1:57:28 So, yeah, back to the question is all these other chains that, you know, that were 1:57:32 previously speaking. 1:57:34 So, like, what's the point of the asset there? 1:57:36 Are they really just trying to compete as far as like just having new assets that's 1:57:40 competing in the market? 1:57:41 Because it seems like, you know, they are like, I don't know, what would their say 1:57:46 be? 1:57:47 Is the new asset required as far as, you know, in their protocols or it's more just, 1:57:52 you know, they want a new asset? 1:57:53 The CISCOIN people? 1:57:55 The CISCOIN, yeah, I guess the CISCOIN people, but also I think what's that other 1:58:02 protocol that Stacks as well? 1:58:05 Yeah, well, with Stacks, I think it was so that they could raise all that money they 1:58:09 want. 1:58:09 That's the problem. 1:58:10 And, you know, this is the corruption of incentives. 1:58:13 It's very it's and I say corruption because it goes both ways. 1:58:18 I think there's a mixture of legitimate and illegitimate incentives, which they each kind 1:58:25 of chip into each other, which is not so great. 1:58:29 But I think the lack of sidechains opens the door to this where people just say, well, 1:58:35 what do you want me to do instead? 1:58:37 I can't I don't want to just not launch the project and it has to have some, you know, 1:58:42 it has to have some currency because with no two way peg, your options are one way peg, 1:58:49 which one way peg just isn't that good. 1:58:51 One way peg has a lot of the disadvantages of having your own coin because the exchange 1:58:58 rate can still fluctuate. 1:59:00 And but it has none of the advantages of just giving yourself more money and rewarding the 1:59:05 people in your community who are helping make your your project a success. 1:59:10 So the so your options are like create a new coin or do like a one way peg. 1:59:17 The one way peg is not that's like what, you know, like RSK is kind of doing. 1:59:21 Well, not really because they have a two way peg with the. 1:59:24 But it's. 1:59:28 Yeah, like Liquid and RSK, they emphasize. 1:59:34 The two way peg that is like assisted by the federation. 1:59:39 But the. 1:59:42 It's it becomes a one way peg if people don't believe that it works, so. 1:59:47 So, yeah, I think the having a new coin obviously gives you a lot of benefits in the form of. 1:59:53 You have early adopters, you have. 1:59:56 People cheering for the project. 2:00:00 So that's good. I don't think there's. 2:00:04 A. Like a globally. 2:00:10 Beneficial purpose. 2:00:13 But it's hard to say because, you know, this is not like something where. 2:00:18 Everyone is going to know about every project, this is a very pretty complicated industry overall, so you never know you hear about some project and you think it's a scam, but then you hear about it, you know, it's actually not that much of a scam at all on a relative basis. 2:00:34 But I don't I don't see why we would if you can get rid of the new coin, then you always should. 2:00:42 I mean, when else would we do that? 2:00:43 You know what I mean? Like what if you went to Walmart and you had to buy a Walmart coin to shop there and the prices are fluctuating? 2:00:52 No one would want that. 2:00:53 So. Yeah, I just I don't know, maybe maybe maybe I don't fully understand because sometimes they pitch it as it's like it's actually required on a network level, but then they do have this quasi two way peg, right? 2:01:07 Like sacks or something like that. 2:01:09 So they still have Bitcoin that goes back and forth. 2:01:11 So it's like, well, we need a new asset to pay our our, you know, our validators or something. 2:01:17 But then it's like, all right, well, if you need if you need a new asset to pay for the validators, that means you you need money to pay for those validators. 2:01:24 Right. So it's either you have that money or you don't, because if you do have that money, then just pay them in Bitcoin. 2:01:29 And if you don't have that money, then like what are you paying them with? 2:01:33 You're paying them with money that you don't have. 2:01:34 So then then it just then it's you just like, you know, creating a new asset to speculate on. 2:01:39 So I kind of seem to me it seems like it's a little bit like misleading to like to say that, you know, this this token is absolutely needed on a protocol level. 2:01:49 Maybe that part I don't understand if they are needed or not, you know. 2:01:52 But I do understand the idea of of like, you know, crowd sale, raising the speculation also, you know, obviously people want to get into something early and it can help bootstrap network adoption to have like early bag holders and things like this, you know. 2:02:08 So from that perspective does make sense. 2:02:09 And that's fine if you're like transparent and honest about it. 2:02:12 But then the moment you start saying, well, this is needed on a protocol level, there's no other way to do it without actually creating an asset. 2:02:17 I have a trouble understanding that. 2:02:18 Well, but I could be wrong. 2:02:20 I don't know. 2:02:22 Yep, just wanted to share my thoughts on that. 2:02:31 Oh, well, come up and ask if you wish. 2:02:39 Or we will end the space. 2:02:44 I have a question for the Chia guys. 2:02:46 So they were talking about the plotting on the hard drive. 2:02:49 I'm wondering if you have like a one terabyte hard drive and you have, you know, maybe 16 gigs or 32 gigs of RAM, how much write like total write is needed on this one terabyte, like to plot it one time or something like that. 2:03:03 Like the initial plot, the initial plotting operation, you would need a separate thing to actually achieve the plotting. 2:03:12 You wouldn't do it on that drive. 2:03:14 So you would you would use like your graphics card to do it. 2:03:18 Nobody really plots with with like drive based plotting anymore. 2:03:23 So that's like way old tech. 2:03:25 So you would use like a leftover graphics card or something like that and use that to create your plot. 2:03:29 And then all the intensive operation would be done using your graphics card. 2:03:34 And one terabyte isn't going to get you anything on Chia these days. 2:03:36 You've got to have something like one hundred terabytes. 2:03:39 You know, I get that. 2:03:40 I'm just thinking, like, if it's a one terabyte drive, how much is actually being written on it? 2:03:44 Like to fully plot it, would it be about one terabyte, two terabytes, or is it actually going to be like 100 terabytes to fully plot it upright? 2:03:54 Oh, I see what you're asking, like how how much in total write operations does it take to produce the actual plot file? 2:04:02 Yeah, I think it's about I think it's about two or three X the size of the plot. 2:04:08 So you need about 300 gigs of space to make a 100 gig plot. 2:04:13 Got it. And how often like once you plot it, is that it or does it you have to replot it every week or like how does that work? 2:04:20 Nope. You only plot you only plot it once. 2:04:22 You never have to plot it. 2:04:23 And then from then on, you don't need any more. 2:04:24 Right. You just have like occasional read speed, right? 2:04:27 Read operations. Correct. 2:04:28 Got it. Is there is there also like an amount of read operations that you need or like per like time or how does that work? 2:04:37 So you have to respond to the signage points. 2:04:40 So if you're not part of a pool, then the only time you're reading from the disk is when you're responding to signage points. 2:04:45 So the network will broadcast and say, who's going to make the next block? 2:04:49 And then you need to go look at your plots. 2:04:51 And there's a thing that allows you to determine ahead of time whether you would need to look at a plot. 2:04:58 So it's called passing the plot filter. 2:05:00 And roughly if you have 512 plots, you would pass the plot filter once every signage point. 2:05:06 So just to break it down, if you have, let's say, you know, one terabyte drive, that's that's fully plotted. 2:05:13 How often do you actually read one terabyte? 2:05:16 Is that like a long time? 2:05:18 So you see, you don't read all of that one terabyte. 2:05:21 So that would you'd have like eight plot files, like eight to ten plot files on that disk. 2:05:27 And you would be passing the plot filter about once every maybe 50 times. 2:05:34 So you would you would be reading from that disk to see if you had a proof, like maybe once every 10 or 15 minutes. 2:05:45 And the amount of data you would read would be like a couple of megabytes out of that. 2:05:52 So a couple of megabytes every 10 or 15 minutes. 2:05:54 So how long would it take to read one terabyte for one terabyte drive is what I'm thinking. 2:05:59 Like, are you are you reading one terabyte a day? 2:06:01 Are you reading one terabyte a week? 2:06:02 Are you reading one terabyte a month from one terabyte drive? 2:06:06 Like on average, right? 2:06:10 You're asking for like a very specific number, and I don't I don't have that right in front of me. 2:06:14 I could I could I could point you to a place where you could reference this and I could look it up for you. 2:06:20 But I don't have that right at the top of my head of exactly how much data throughput goes across the drive, like per unit time, because it's different if you're pooled versus not pooled. 2:06:32 Because the way the pool works is it's sending you fake signage points. 2:06:35 So you're reading from the disk more often if you're pooled. 2:06:38 But if you're not pooled, then it's like way less. 2:06:40 So it's like two different styles of doing this if you're going to participate in a pool or not. 2:06:46 And usually I work with like the pool, not the not the base case, which is the 2:06:52 I brought it up because, you know, you guys are talking about shredding the drive. 2:06:56 So I was just wondering, like, what an actual, you know, how much read write on average? 2:07:00 Yeah, it's the shred part was the plotting that that that's the thing that people sort of got confused on early on. 2:07:06 But since you only do that once, like the reading from a drive with almost nowhere on the drive whatsoever, like it's a it's an operation that's fundamentally non destructive and non mutative on the drive. 2:07:18 So it puts almost no load on it to do a read from it. 2:07:22 I see. All right. Well, thanks for answering those. 2:07:31 If nobody has a topic, I wanted to come back to the horizontal scaling of blockchains. 2:07:40 I mean, is there an inherently architectural issue with the way we design public blockchains for scaling horizontally? 2:07:53 Could something better be done in that respect? 2:08:03 Yeah, I mean, any thoughts would be. 2:08:05 You mean horizontally, like by increasing block size or horizontally by like just adding a sidechain? 2:08:12 Yeah, see, that's the that's mostly where the discussion ends up, right? 2:08:17 I am going more fundamentally, like at a lower level, because, you know, you have this, you know, there is a cross dependency among assets, you know, they require sort of a global ordering of transactions, you know, to have some sort of deterministic execution. 2:08:41 And so I'm just wondering, can that dependency be minimized or? 2:08:51 I think the fundamental thing is the block size, because it's limited. 2:08:55 So no matter how much you optimize other parts, there's still a limited block size, right? 2:09:00 And the limited block size is for the decentralization of the network, less data being needed to be stored, less data that's needed to be verified, and, you know, running the compensation on that, verifying the signatures, whatever, passing that data around. 2:09:19 The less data is, you know, the argument for less data that's being passed around, the more decentralized it is, right? 2:09:28 So you still have a limit there. 2:09:29 So if you want to horizontally scale, it's either you increase that limit, which a lot of people are opposed to, because that's going to, you know, hurt decentralization, or you just have to just have another chain. 2:09:41 And the only way to have that other chain is if you're going to go back and forth, you need a mechanism to go back and forth, right? 2:09:47 Yeah, well, you know, I'm thinking how cash works, and I'm no proponent, like, I do crypto podcasting and stuff, so I'm not saying, yeah, I'm just trying to see, like, you know, you can settle, do a cash settlement independently, right? 2:10:05 see like, you know, can settle, do a cash settlement independently, right? 2:10:11 It's just the bearer changes, there is, that's just one. 2:10:19 What I'm saying is you don't need to look at prior transactions to 2:10:24 verify the current one in cash, sort of purely 2:10:30 cash kind of transaction. 2:10:33 I think that's what there is some parallelism there, but 2:10:38 in context of blockchain, trying to rephrase that and probably rethink 2:10:43 in terms of how a public blockchain can really 2:10:49 reduce those sorts 2:10:54 of interdependencies. I I don't see a way, but I'm just sort of 2:11:00 to brainstorm with the rest of you folks. 2:11:05 Yeah, but what was the question, you know, that was originally about horizontal scaling? 2:11:10 Yeah, yeah. So this ties to the horizontal scaling, right? Like, how do you scale horizontally? I mean, you know, with L2s as a solution, sidechains and all that, but 2:11:26 they come at the cost of like, performance, you know, all those trade offs that they come, you know, at the cost of. 2:11:40 So, I mean, there's no question, I mean, it's pretty open ended in terms of like, is, is there a way to rethink these 2:11:47 monetary schemes like UTXOs and account based money schemes like that lends itself to more sort of throughput? I don't know. 2:12:05 What about something like eCash? 2:12:08 Do you have a shiny cake has come up? Maybe we should let them ask their question or comment. 2:12:16 Yes, I have a question. One moment. Give me two minutes. I'm just preparing something to eat. But give me two minutes. 2:12:23 Okay, great. 2:12:25 You said eCash? Like Chaumian eCash? 2:12:29 Yeah, like Chaumian eCash. If you see like what Kali is building with the Nodcash and eCash stuff over there. Like, there's still like a Mint that is like centralized. But as far as like, if you do trust the Mint, I think that there is some more scalability. Not sure. 2:12:50 Yeah, I mean, once you remove the, yeah, yeah, that's not, not, that trust me bro model is the issue there. So yeah, that's not probably what I'm pointing to actually. 2:13:10 All right. 2:13:19 Is somebody still having a question or can I ask something? 2:13:24 Go for it, man. 2:13:25 Nice. Okay. Sorry to interrupt. It's not Bitcoin in this case. I have a question about SkyDodgeNet from Paul's perspective. Is somebody like the developer MLG, who's also known as one of the most involved testers of your idea. 2:13:51 Is he able to solve those tasks to make these sidechains available? What he's showing for, for example, Thunder sidechain inside. Is it really like a solvable task for one person alone? Or do you think he needs some more and some more people help him? 2:14:12 Well, I'm not sure. I think, so SkyDodge is an altcoin that implements BIP300, and which is great. And we also, we develop like LayerTwo Labs and myself and other people who collaborate with us, but mostly us. 2:14:33 We develop the sidechains, the sidechain full node software, which really anyone can use if they just point them at a different L1. So we do like a lot of the work and he is certainly a great tester and helper of, you know, he tries out our software and that's very good. I like that. 2:14:59 And, but we do an enormous, you know, percentage of the work, I can know whatever, 99% of the work. And he has turned it into like a funny little altcoin. And they didn't, they give themselves a huge amount of the coin. I don't know, but I would think they would probably want to hire someone to. 2:15:20 But they're very funny. I think the SkyDodge people are very funny. And they, I like that they are testing my idea, but I don't know. 2:15:29 At the moment, it's not really like profitable. It's more like helping the idea grow. He's believing in your idea, and he's 100% behind. But for me, it's like, not a professional coder. It's also sometimes really hard to judge about the ability to get something like this ready. 2:15:51 And I think he's just able to work with what you give him and then to add maybe some things and maybe fix some miner bugs. I don't know. Or some small bugs. I'm not sure. But I'm just questioning myself. What do you think about this idea? Is it like, possible to realize it like this? Also, like he's fighting alone at the moment for nothing. No money at the moment. 2:16:15 It's completely community-driven, the project. There's nobody getting paid. 2:16:20 paid. 2:16:23 In Skydosh, I mean, I don't know, I'm not paying anyone over there. 2:16:26 But I think they offered me like many, I don't know how many, I don't know what that's 2:16:30 worth, but they offered me 2:16:31 like a bunch of coins. 2:16:32 They told me take a bunch of these coins. 2:16:34 I haven't quite taken them yet, I think I eventually will or something I suppose. 2:16:39 I mean why not? 2:16:40 but I don't know. 2:16:42 think they, 2:16:45 I'm not sure what their plans are for that. But I think it's a funny name. It's a funny meme. It's got BIP300. What do you think? We should do something differently? Like we being LayerTwo Labs or myself or something? 2:17:03 I mean, it's already the first mainnet of your idea. But it's still not 100% ready to run. And he's fighting for this kind of idea. And I'm asking myself, will he be able to solve these issues that at the moment it's not easy to do? Alone with your help, I don't know. 2:17:28 Well, yeah, we have a different... It might be a good idea for them to start over. Because we have our own way. We have the launcher and stuff now to make it very easy for people to run all those different sidechains and find them. And keep everything very organized and relatively user-friendly. And we intend to continue to improve. I intend to improve everything. 2:17:56 We have more of an emphasis on front-end and back-end. I'm expanding the team often. And so I'm expanding it a lot. So maybe we should. Maybe LayerTwo Labs should just release... We should hard fork Sky Doge. We should partner with this. Presumably they have the domain and social accounts. And we should just do it as a joke. 2:18:22 And redo it with a pre-mine and hire those people. And maybe that would be funny enough. Why not? It's not like we're getting enough love from the actual Bitcoin maximalists. So that should be a funny idea, I think. Especially if it has an absurd name, an absurd logo. We could have a lot of fun with it maybe. I don't know. And really test it out. But there you go. 2:18:47 It's interesting. And what kind of time period would you say? How long would it take to get these things, what they have on their website, running? 2:18:56 What do they have on their Bit Assets? 2:19:00 We have already. 2:19:02 Some of them, I'm kind of perfectionist. 2:19:05 We had Bit Assets a long time ago, months ago. 2:19:10 I took it, sort of like... 2:19:12 I've become relatively 2:19:14 ashamed with some or another. 2:19:17 even like some, it's relatively easy to upgrade to the latest version. 2:19:21 We had the latest version of Zcash. 2:19:23 we're like, should we upgrade again to again the latest version? 2:19:26 The answer is sort of yes. 2:19:28 And yeah, I realize this is probably very frustrating and confusing for people on the 2:19:32 outside who 2:19:33 who just don't know what's going on, are... 2:19:41 BitAssets is... let me try to explain, you know, sometimes I have a lot of thoughts 2:19:45 bubbling in my head, you know, and I just want to convey them all at once. 2:19:50 And it doesn't go over very well, but 2:19:53 dump them all out now. So BitAssets is currently being rewritten in Rust, 2:19:59 because a lot of things are being rewritten in Rust. And the old sidechain template, 2:20:05 which is still available and still in use, is in C++ and Qt. Qt is the front-end GUI, 2:20:12 the graphical user interface. So Qt is okay. There's nothing wrong with it. It's nice. 2:20:20 I like it for mainchain L1. It's the same thing in Bitcoin Core. That's why it's the same. 2:20:27 And that's why we used it. But for the sidechains, it's not great. Because first of all, 2:20:33 we clone sometimes sidechains such as Ethereum and Zcash that have no front-end at all. 2:20:42 So if we were relying on the Qt front-end, now we just got deleted. So the other reason is that Qt 2:20:49 reads a lot of its data directly from... it's very tightly integrated with the code. 2:20:56 I'm not sure if this point is easy to convey, developer to non-developer, but other front-ends 2:21:03 will just ask for data with RPC or with something like that. But Qt doesn't work that way. So Qt 2:21:10 is tightly... it's like a Siamese twin. It's sewn in together. So we have decided to try to 2:21:18 separate out the front-end completely, which is its own project. But this is by far the right 2:21:24 decision, I think. Because when you separate the front-end, you cut the Siamese twin, 2:21:29 you also get competition among front-ends. So we just say, these are the RPCs you need to support. 2:21:35 And if you want to build your own front-end, which includes your own wallet, your own node, 2:21:40 your own whatever, your own way of viewing the node. Now, that just went from basically impossible 2:21:47 to very, very, very easy for anyone with any front-end experience to completely redo the 2:21:53 front-end of any sidechain. So I was kind of redoing that. And then while we were at it, 2:22:02 we have a lot of brilliant people over here, and one of them rewrote the sidechain template in Rust 2:22:08 because everyone loves Rust these days. And they got it down to a very high performance, 2:22:13 seven megabytes. So we had bitnames, the namecoin sidechain that was in C++ Qt, but now 2:22:20 that's been rewritten in Rust. And now that bitnames happens to be very similar to bitassets. 2:22:26 So we were just like, well, when bitnames is really, really good, we'll just redo bitassets 2:22:35 in Rust and all with different front-end. So on one hand, you could say this is evidence of 2:22:41 Paul being scatterbrained and having no focus, and this is all just mission creep, 2:22:46 and it's all chaos. But it's quite difficult. It's quite difficult to get something that has 2:22:54 a good user experience for Mac, Windows, and Linux, and has a front-end that works on everything, 2:23:01 which includes the desktop, laptop, phone. So it's a very unfortunate growing pain. 2:23:09 But we have actually done a lot of work already. The bitasset sidechain, I specced that out 2:23:17 in 2018 or something. And I get messages from people all the time asking me about it now. 2:23:25 And they like the idea. They think I was right about everything that I said. 2:23:31 And so, yes, it's true. Developers are always going to want to take time, rewrite everything 2:23:37 from scratch, redo everything from the general case, solve the general case, whatever. 2:23:44 And that is definitely frustrating for the end user, who wants everything to just be finished 2:23:52 now. But a lot of stuff also fell through the cracks temporarily. So we had the latest Zcash 2:24:02 and latest Ethereum. But those don't have any GUI. So we either build one from scratch 2:24:10 ourselves, or we modify something else, modify Metamask or something. 2:24:16 So what we're going to have instead is this unified modular front-end experience. And 2:24:24 then we're going to have the back-end node experience. And this is all going to be organized 2:24:30 in the launcher. And the launcher will know what operating system you are running. And then it will 2:24:36 download the node software or point you to the node source code. Just download the binaries or 2:24:42 download the nodes or stuff like that. So that is my plan, because it actually becomes 2:24:49 a bigger task when you say, instead of just shipping one piece of software, 2:24:56 you've got to ship Bitcoin Core. You have to ship Mac, Windows, Linux version. 2:25:01 Now we have to ship one. We have basically three columns, Mac, Windows, Linux. Three columns, 2:25:08 and then as many rows as there are sidechains. And it's difficult. And the decisions that I 2:25:18 have made in working with this problem of how to do this, and I've tried to reshrink the problem. 2:25:25 And as a result, we've done a lot of work that no one has seen yet. 2:25:29 But I think if you plot it on a graph, it's gone very well. LayerTwo Labs was only created, 2:25:36 you know, we only started to, I don't even remember when I made the slides to ask for 2:25:42 money. I think that was August 25th of last year. We got funding in November. 2:25:47 And we have large block sidechain. We have latest Zcash sidechain. We have Ethereum, 2:25:53 latest Ethereum sidechain. We have, they said large blocks. I don't remember what I have to 2:25:58 put them in the right order. So I don't repeat or forget any. But we have BitNames sidechain. 2:26:03 We have, we had BitAssets, but it just doesn't work for every, it's just not as good. It works, 2:26:09 but it's just not as good as it could be. So I kind of like pulled the plug on it temporarily. 2:26:15 If anyone, you know, if you want, like, if you're a great developer and you want a job, 2:26:18 then there's plenty of stuff that we're working on if you're very good. 2:26:23 And we have the latest core sidechain. It's the sidechain that is the latest version of Bitcoin 2:26:28 core that has 118, 119, you know, it has CTV in it. It has one, it's basically Bitcoin Inquisition, 2:26:35 but it has a sidechain. And it has OP_VAULT. And it may have other fun arc-shaped opcodes soon. 2:26:47 We have the Rust rewrite large box sidechain, which is different from the original 2:26:52 large box sidechain. We're working on the Bitcoin Hivemind as a sidechain. 2:26:58 So we have quite a bit. We have a blank version. We have a sidechain that is very blank 2:27:03 in terms of the test. The C++ template is just a blank sidechain that has the GUI, C++ Qt. 2:27:13 And we have, that's just like where the starting point would be. We have that and we have the Rust 2:27:19 starting point one, which we call plain, the LAIN. So we have a lot that we have already made. 2:27:28 And stuff, it's, I realize it's annoying because the end user, they either just see something that 2:27:34 works or they just kind of see nothing. But we are, we're in the arena trying things. 2:27:43 No, I just mean like we are, 2:27:50 I don't, you only see like the tip of the iceberg really. 2:27:55 It's, and yeah, if we had more, sometimes we got like whatever Zcash just got stuck on this very 2:28:02 obscure linking error for a tiny, for like a little while. So sometimes just a tiny little 2:28:09 thing trips us up. And this switch to unified frontend, I think will be key because 2:28:18 we want the frontend and backend to be separate. And when they are separate, 2:28:26 we will have a frontend that works for every sidechain that does not, 2:28:31 because the Qt one is too tightly integrated. So you can't, you couldn't use the Qt frontend 2:28:37 in something that was, that had like a very different backend. So they just didn't fit 2:28:42 together. And this is Bitcoin Core has just had Qt for a very long time. Well, the Bitcoin Core 2:28:47 is also trying to get rid of Qt. So, so I don't know, that's kind of a rambling answer, but we 2:28:52 plan on, we're doing a lot of software development over here. We do much more actual real work than 2:29:00 I think. I think by, you know, certainly if you actually plotted it on a graph, 2:29:09 like the actual raw capacity of what we can accomplish for the user in a user-friendly 2:29:18 way in terms of transaction throughput, privacy, usability, extensibility, reliability, 2:29:28 just overall ability. Like I would say that we have blown way past the entire 2:29:33 Lightning Network like a long time ago. We're like 10x or 100x superior. 2:29:41 If or slash when BIP300 activates on Bitcoin Core. So it was already 100x 2:29:51 superior to the entire state of the art of the industry. And BitAssets, the old Qt version, 2:29:58 that's in C++, it still exists. The old versions are all there for download. It's just, 2:30:03 it's very, very difficult to like, right now, we're changing it, but it's very difficult to 2:30:10 just release something because you, the end user, you'd have to download it, install it. 2:30:16 You have to download it as, you know, if it's just a binary, you have to download it and run it. 2:30:20 Then you have to like edit the conf files so that it can connect to L1 main chain. And 2:30:30 we just have a different way of doing it now that will make it much easier for you. 2:30:35 So it's just like, I don't know, it's a work in progress. 2:30:38 Hey Paul, I wanted to know, you have considered building on top of Chialisp? 2:30:45 On top of what? 2:30:47 Chialisp. 2:30:50 What is that? Chialisp? What is that? 2:30:54 Chialisp. Chia, basically. 2:30:57 Oh, Chia. You mean coin? Well, no, I don't know why I would. Should I? 2:31:05 I think you should. You should definitely look into the code. 2:31:09 But the code really doesn't make any difference, does it? 2:31:12 In what form? In what sense? 2:31:14 Well, it's all about the community and the market cap. And so the code is kind of irrelevant, I 2:31:24 think. Do you disagree or no? I mean, is that a fair thing to say? I mean, someone launches a 2:31:30 project and it's number 500 on coin market cap. The code could be perfect, but that doesn't 2:31:38 necessarily matter. Especially if it's for a while, because then people will just say, 2:31:44 though this code, they'll either say this code is bad. Here's the proof. Or they'll say, 2:31:52 here's this idea. It's number 500 on coin market cap, but the code is good. And then they'll say, 2:31:58 well, okay, we'll copy paste the code into our own coin. And then we'll launch a new coin. 2:32:05 So I think that's what, so I think the, I don't think the code matters. 2:32:22 I still think you should look at it. Trust me. 2:32:25 Okay, cool. 2:32:27 I have a last question. 2:32:29 If you would go to mainnet in the next year, what would be your plan? How would you arrange 2:32:36 the project to get as time efficient as possible to some kind of product? 2:32:42 For BIP300? 2:32:45 For example, you have to go like the Skydodge, sorry to name it. But for example, you have to 2:32:50 go the same name like this one, that's the same way like them. And you have to go to mainnet 2:32:55 and to create a usable product. What kind of strategy would you suggest? 2:33:03 Ah, well, I mean, if you're saying if I activate on Bitcoin, the easiest way would be to just 2:33:11 do as a miner activated software to go to launch a completely separate thing. I mean, I don't know, 2:33:18 that would be, that's a very different thing that involves a lot of 2:33:21 strategy, I would think. So it is a, but I think the, I don't know, I'm not sure. 2:33:34 Okay, because you were talking about a funny idea to launch an on thing with Skydodge Corporation. 2:33:40 I just wondering what kind of plan you would go with and if you have to. 2:33:45 In that particular case, I would say maximum. 2:33:48 If you have to. 2:33:49 In that particular case, I would say maximize like the irony and humor of the situation 2:33:54 and just do it with a lot of self-awareness and just Skydodge is a funny name. So if we could, 2:34:04 that would be funny to partner with. So that would be a funny, I think that would be kind of 2:34:09 fun and the irreverence of it, like the idea that I have been in the Bitcoin community for like much 2:34:14 more than 10 years and I have like never done a thing and now to do it, I think that would be 2:34:20 funny actually. So I think that would be a neat idea. But I think that would be like to just 2:34:28 demonstrate the product and just kind of have fun with it. What I would say would be the strategy 2:34:35 to, because I think the Bitcoin's biggest weakness now is that everyone is, 2:34:44 Bitcoin is not succeeding at its core goal to replace the banks. 2:34:48 And so everyone is the conflict in their own mind with how upset they are over that. 2:34:56 And this is causing people to be annoyed and we have this cult also. 2:35:03 The cult cannot tolerate anyone having fun, which is why it doesn't like people like Udi or whatever. 2:35:09 So if you could just bring back the fun and the humor of it and target the end user, 2:35:14 just don't try to evade the world of developers, especially developers in 2:35:20 BTC and try to evade every cultural leader in BTC, just try to avoid all that and then just 2:35:28 do something that the end user thinks is fun and just completely ignore 2:35:33 the BTC orthodoxy. That would be probably the strategy in that case. 2:35:39 Yeah, it's good that they support each other. And it's a little bit of pity. It's a big pity 2:35:44 that such less core developers help you developing the project. And it's great to 2:35:50 have such people like MLG also around that really try to help you. 2:35:56 Yes, it is. And yeah, having help is good. And Bitcoin Core, unfortunately, there has been a 2:36:04 prestige economy has been created around Bitcoin Core where it matters a lot who contributes. 2:36:14 And as a result, there's gatekeeping and they don't want certain people contributing because 2:36:20 it would increase their prestige, which is very bad for the end Bitcoin user. 2:36:26 But the gap between the end Bitcoin user and the insider technical person is enormously vast. 2:36:35 Anyway, I think we had Ethan come up first and then Morgan, I think. 2:36:42 Hey, can you hear me? 2:36:44 Hello. 2:36:46 Hey, thanks for having the space. So I've looked at a lot of your tweets going back and forth 2:36:51 with people on about the Lightning Network and why and some disadvantages of the way that it's 2:36:59 implemented. And it seems that that any scaling mechanism for a small blockchain is going to be 2:37:10 bottlenecked by the very low write throughput to the global state. 2:37:17 And that a lot of the B-cashers, I guess I really did not ever approach Bitcoin Cash with an open 2:37:28 mind. And as I know that the large blocks threaten the consensus because not as many people can 2:37:39 consume them. It's more difficult to mine. But how will the low write throughput to 2:37:49 the current Bitcoin blockchain be overcome then? 2:37:56 Yeah, well, I have a there is a way of scaling Bitcoin with sidechains that and there's also 2:38:05 there's quite a there are a lot of there are more different there are more L2s than just the 2:38:10 Lightning Network, which is part of why I named the company LayerTwo Labs, because 2:38:16 it's the obsession with the Lightning Network is has 2:38:22 shut off interest in other L2s and even the biggest. 2:38:28 Well, it's a long story. But basically, we had a similar question earlier today where 2:38:34 it I think it's not a good idea to have the L2 give every single user something that they can 2:38:40 broadcast on L1 to get their coins back. Because if 8 billion people won't fit on L1, then it 2:38:45 doesn't matter that 8 billion people are on an L2 or each of them could go because it's just not 2:38:50 the case. So I think that that's actually dead end. And I think that's how both Ellen and 2:38:56 and ARK insist on that. And it's a long story short, but other sidechains have wrestled with 2:39:05 that concept. And I think that's not headed in the right direction. I wrote a post. 2:39:14 The difference with scaling with sidechains is that you can both onboard on L1 and off, 2:39:20 excuse me, on L2. And you can actually offboard on L2 also. So you can someone on L1, like a rich 2:39:29 guy, Michael Saylor, or whatever, Roger Ver, I used to use the example, they send like 50,000 2:39:36 coins to the sidechain with one L1 transaction. And then only periodically there's L1 transactions. 2:39:47 You know, to maintain the sidechain. But on the sidechain, the 50,000 coins can you can make a new 2:39:53 address, make your new sidechain address and collect the coins over there, send them to other 2:39:58 people send them back and forth. And, and then actually, if you say that someone panics, and 2:40:05 they say, Oh, I want to go to L1, but there's not enough space. What do I do? The L1 ARK method is 2:40:11 to give everyone something they can broadcast on L1. But as I've explained, and as you asked in 2:40:16 your question, that idea is dead, actually, it's not going to work. So because everyone can't 8 2:40:23 billion people can't use it. So what difference does it make if they have it or not, you might 2:40:26 as well just raffle them off and only let a few people get them get the unilateral exit transaction, 2:40:32 because that's the only people who are going to get it in time. So the what you do instead is you 2:40:37 say that on L2, you sell your right to go back to someone else to anyone who will buy. And in that 2:40:46 way, the transaction, the coins can actually collapse back, they can fan out. And they can 2:40:51 also collapse on L2. You say, Listen, I have eight coins on this network, but I want to have L1 coins. 2:41:00 And you say, Well, okay, I'll give you something on either the different L2 network or on L1. 2:41:08 And that's still both of those things require block space, getting on their respective networks. 2:41:13 So there, the difference is small, but I think it's big enough to make a difference because 2:41:21 the end user signs off the coins on L2 or for an exchange for whatever else they want. 2:41:27 So it could be a car could be a coins on some different L2s and different sidechain. 2:41:34 And then you end up with people on the L2 who have, you know, at the end of the day, there's 2:41:40 only like, one guy with 10,000 coins, one guy was 15,000 coins, and then a third guy was 15,000 2:41:46 coins. And then those three people withdraw to L1. And those three people could be like foundry, 2:41:53 marathon, whatever. You know, I see. The thing is, I wouldn't. Yeah, okay, my bad. 2:42:02 So, so I wrote a post about this called thunder sidechains for scaling. 2:42:08 Thunder is like a little joke. And the, there's certainly no perfection. But I do think that 2:42:17 lightning, my problem with the lightning network, I would like to reiterate and make this very, 2:42:22 very clear. The fact that the lightning network basically doesn't work. That doesn't concern me at 2:42:28 all. I think that's fine. Nothing works at first. You know, Elon Musk's rockets explode, etc. 2:42:37 Before they work. The issue is instead, that the dishonesty and just everyone losing their brain, 2:42:47 basically. And that is the kiss of death. Now, I think nothing will, you see what I mean? Like, 2:42:53 what if it was, there's a huge difference between Elon Musk is building reusable rockets, 2:43:00 and they explode. And they're, they're spending a lot of the running out of money. But every time 2:43:06 the rocket explodes, they've got cameras, they have the black box, they have sensors, they have 2:43:10 debates, that people with blueprints and, you know, AutoCAD and everything. So they're changing 2:43:18 it around. And they get a little closer each time. And they think, I really think that we'll get it, 2:43:25 they try. And they launched the rocket. And I think we know that this is going to fail. 2:43:29 But we want to collect the data. We need to try if we're ever going to collect real data. 2:43:36 Now imagine if SpaceX work differently. And every time the rocket exploded, they said stuff like, 2:43:43 well, actually, we want the rocket to explode. Or they said, or they don't, there's no actual 2:43:48 rocket, like there's just like a custodial rocket, that's like a fake toy model rocket. 2:43:53 And everyone's using that. And, you know, and people seriously believe that with the toy model 2:44:00 rocket, we're going to make it to Mars or something. And so the self delusion is what 2:44:08 proves that the project is doomed, because you can't fix a problem that you are in denial about. 2:44:16 So that, that is the problem. And the fact that so many people have reacted recently to my 2:44:22 very, like, tame, little tweet, it proves really just how insecure. You know what I mean? Like, 2:44:30 if someone said, Oh, SpaceX, they'll never, you know, they won't, they're not going to make it. 2:44:35 If you work at SpaceX, and you think you're going to make it. 2:44:39 What do you do? You just ignore that. And you just say, Yeah, we have to work hard. 2:44:43 We're going to ignore that person. A lot of people think we're not going to make it. 2:44:46 But then if to like, to react in such a 2:44:52 paranoid and sort of defensive way, that's because what lightning network really is, 2:45:00 is not a protocol for scaling BTC. The lightning network is just a medium. It's a figurehead. 2:45:08 It is a kind of tribal mask or something. It's a ritual. Lightning network is like a 2:45:15 performance. It's something that people say is good for Bitcoin. But and it's something that we 2:45:21 kind of in a sense, there's this very real sense in which the lightning network is just something 2:45:24 that everyone pretended that they believed in, in order to win the scaling war. 2:45:30 Right, the so so the the L2 that you're referencing, 2:45:34 which I kind of understand better how you're you're thinking that. 2:45:39 That right, like the I like the analogy you made about the HTLC is about like how 2:45:44 someone steals you, steals less than it costs to bring them to court, then you're basically just 2:45:50 SOL. And it's the same deal for getting like scammed on a lightning micropayment. Is it so 2:45:56 expensive to actually go and submit the redemption? The L2 that you're referencing, would that be just 2:46:03 a large block sidechain then? And you have the same problems as Bitcoin Cash? 2:46:11 It does. But the key is, this is, I think, the key to understanding it all, which is that 2:46:18 what was the problem that the large block nodes had is that the runaway cost of running a full 2:46:28 node. So, for example, one of the many problems with Bitcoin Cash was there are enormous number 2:46:35 and I've tweeted a big table. If you guys think I'm hard on the lightning network, 2:46:38 I like crush Bitcoin Cash and I go on their podcasts and don't mean to them. And I have this 2:46:45 big table of like the Bitcoin Cash mistakes. And I think you can find it. And I put them all on a 2:46:50 huge list, just like I did with the lightning network. And I was just just that kind of guy. 2:46:56 And they had lots of problems. But one is that they never really explained. 2:47:01 Like they said, OK, one megabyte is too small, but then it's kind of like what is so they wanted 2:47:08 to change the eight megabytes. But it's kind of like that's just the difference of degree. 2:47:12 They're evading the issue of principle, right? Like what is what should the number be? 2:47:18 And they said, well, they think the line should be at eight instead of at one. 2:47:20 I don't know. 2:47:21 They never really explained. And they also never really explained like an obvious question. 2:47:29 I mean, it's insulting to use that word, but I really feel it was obvious, 2:47:33 which is that what do we do when the blocks fill up at eight megabytes? 2:47:39 Want to just be the same thing, the same problem, just double now because we already 2:47:45 had some people like in the reality where we hard forked eight megabytes. 2:47:49 We already did a hard fork to go from one to eight. So when we go from eight to 64 and then 2:47:54 64 to infinite. And if we're going to go to infinite, then why don't we just go to infinite 2:48:02 like today? You know, like I think that would have been the smarter thing. But even the Bitcoin 2:48:06 cash community was very the large blocker community was very divided. I remember Gavin 2:48:10 Andreessen saying no one wants no one supports an unlimited block size increase. But then look 2:48:17 at exactly what happened to them as they created their large block version and then they got 2:48:22 immediately outflanked, not immediately, but like one year later by a Bitcoin SV. 2:48:28 And now they're trapped because they have to say, well, small blocks are bad. And then people 2:48:32 say, well, what about what about Bitcoin Cash? Your blocks are smaller than Bitcoin SVs. And 2:48:38 then they have to say, well, Bitcoin SVs are too big. So is this starting to make sense about now? 2:48:43 That problem doesn't even with so there's two already we've hit two things, two huge differences 2:48:49 between scaling via sidechains and by a large block sidechain. 2:48:55 Two huge differences between that and Bitcoin Cash is one is no one pays the cost except the 2:49:00 people who want to run the large block node, whereas in the naive L1 large block is in 2:49:07 Bitcoin Cash world. It's a one size fits all. So everyone has to run the more expensive node. 2:49:13 Inside chains, you have everyone run the small, tiny node. And then whoever wants 2:49:18 the minority who wants to run the more expensive node. 2:49:25 They do so. So that's one huge difference. But the second difference is that naturally, 2:49:29 the block size limit is going to be changing around now because you make it too big on the 2:49:35 large block sidechain, you might kill the network and then it'll all be that network will die out. 2:49:41 But the regular main chain L1 network will survive. And if you make it too small, 2:49:46 then someone will make a new large block sidechain somewhere else. So now I have three networks, 2:49:50 one small one that's mandatory for everyone, a large block sidechain that's kind of full and 2:49:55 then one that's newer and is empty. So naturally, more block space would be coming in and out of 2:50:01 existence. There would be something to regulate it. And that thing would be adding and removing 2:50:07 sidechains on L1. So this answers the question because it says it has the potential to be 2:50:15 unlimited, but at any given time, it would be limited. And I want to mention a third thing, 2:50:21 which is that there is a big difference in having one chain for everything, Bitcoin SV. 2:50:28 The scaling via sidechains allows you to have a lot of heterogeneity, first of all. 2:50:34 So you can have some sidechains that are very, very pushing the envelope. 2:50:38 Maybe they don't have a long run future, like a Solana style sidechain that is huge, 2:50:44 that's not sustainable, but has huge blocks, huge transaction throughput. And then you'd have maybe 2:50:49 some that are different. You have a whole range, you know, we have a whole spectrum of like 2:50:53 decentralization and security and fees. So we hit the whole region. But in particular, you'd have 2:51:02 geographically distributed sidechains, since most of the trade in America is like you have 2:51:10 North America sidechain, you have Europe sidechain, you have Southeast Asia sidechain, 2:51:16 you have China sidechain, and you have the rest of the world sidechain. And just like 2:51:22 almost all my transactions are in US dollars since I live in the United States. 2:51:27 And then when I go to Europe, I make one big transaction swapping some of my cash for euros. 2:51:34 And then I make transactions over there. So it's just the way it is, it just happens to be the 2:51:38 economic reality that people are clumped up already into big clusters of and that within 2:51:45 cluster trade is enormous, much 10, 50, 100x, the inter cluster trade. And so you already get some 2:51:53 efficiency there. It's actually the same principles of lightning network that if even if I 2:51:59 this is superior to Bitcoin SV, because in my universe, I would run, even as a full node user, 2:52:06 I would run main chain L1. And I would run the US large block sidechain only I don't need to run 2:52:14 the Southeast Asia, I don't really care. No one's paying me over there. I get paid on the 2:52:19 North America chain, because I work at whatever North America company. So only rarely do people 2:52:26 have to go in between. And so it's a similar principle to lightning network where you ignore 2:52:32 things that you don't see just on a slightly different, you know, slightly different trade 2:52:36 off. So those are some of my lists is my reasons why I want to emphasize the large block scaling 2:52:42 via sidechains is very different from naive L1 large block ism. 2:52:48 I see. So you're so like a BIP300 is, in essence, going to be increasing the amount of block space 2:52:58 there is, but then regulating exactly how much through market forces, I guess that like, it's 2:53:04 just if it's way too big, then people just won't run it because that's annoying and expensive. And 2:53:13 that it doesn't, I guess, inflated on L1. It just uses L1 is sort of like a peg for all whatever 2:53:22 ideas people might have. Okay, that makes a lot more sense. I appreciate it. Thank you for doing 2:53:27 the right. Yeah, of course. The issue with the block size is that is global is like a, you know, 2:53:34 tragedy of the commons type problem. So once you have it so that a sidechain is free to fail, 2:53:40 that is very healthy, because sidechains free to fail. And as a result, you know, 2:53:47 like it's it's completely the absolute reverse night and day where it would be with L1. Because 2:53:53 we no one would say, Oh, we'll just set the block size to something else. And if Bitcoin fails, 2:53:56 then whatever, it's not a big deal. Everyone L1 fail at Bitcoin fails, that's a big deal that 2:54:02 everyone's gonna be really pissed off about that. But if it's an optional sidechain that starts 2:54:08 with no coins, and people have to voluntarily send the coins to that, then they have to attract 2:54:14 users. And then if they die, then we won't shed a tear. They sort of have their fate is in their 2:54:20 own hands. So that's the idea. So now maybe we'll go to Morgan, I suppose. Hello. Hello. 2:54:27 Hey, Paul. I haven't talked with you before, but I'm a big fan of all of your pontificating and 2:54:35 harassment of Bitcoin Core. I salute you for your work. One of the things I wanted to talk to you 2:54:42 about was this user interface, the front ends, either using or not using BIP300. And I'm 2:54:50 wondering if you could answer a question maybe. In an example, like how samurai wallet uses BIP 2:54:59 47. And it doesn't seem to be a very popular thing and other wallets. Is there a way today, 2:55:09 a wallet could implement BIP 30 without any approval from other nodes and, and basically 2:55:19 do some some testnet stuff on a user interface level with BIP300. And, and like building user 2:55:27 interfaces that are, you know, showing off these sidechains and drivechains. Is that being done 2:55:33 in a way or can it not be done in a way? No, it's being, we do it now with our test 2:55:39 software, which is not actual real Bitcoin. So I don't think it can be the way. 2:55:46 BIP 47 is a great example of, I can give you more pontification, especially BIP 39, 2:55:53 where everyone, people for some reason hate BIP 39. Everyone in the technical community, 2:56:00 where I should say is a small minority of the elite technical community 2:56:05 had some kind of problem with BIP 39 and BIP 47. And I'm going to get into that in a second. 2:56:11 But, you know, who didn't have a problem with it was the end user who was like, oh, this is great. 2:56:16 I just have to write down 12 words. And then I don't need to like, print out this thing of 2:56:21 gibberish where I have no idea. I have no idea if I printed it outright or if someone switches 2:56:28 the piece of paper for a different one, I won't even notice. But now I just write down 12 words 2:56:32 and I just, you know, stick them in a book or something. So people love BIP 39, which is like 2:56:41 taking over. BIP 47 is, you know, people thought it was wasting block space 2:56:52 because the first transaction between two parties is 20% larger. 2:56:59 And then we would have never, we could have gotten rid of Bitcoin addresses. 2:57:04 You just like friend request someone on the blockchain. And then you have an encrypted 2:57:08 tunnel back and forth. You can send them money. You could send them messages. 2:57:13 And this is what we all would have had a long time ago. Like, I don't even know how long, 2:57:19 like 2015 or something. And yeah, this is because of people, you know, and I love Luke Dashjr. 2:57:26 He's great, but he's also crazy. And he just thought this is wasting block space. 2:57:32 The friend request thing, which is like 40 bytes. Again, it's like 20. 2:57:37 The first transaction is like 20% larger, which is nothing. And so then they fight this war 2:57:45 to keep this type of thing out. And then they get a big, you know, 2:57:51 when ordinals and stuff, inscriptions, when that finally broke out, I mean, 2:58:00 now they have to deal with it. It's like, it's a perfect example of like, 2:58:04 not solving a problem and kicking the can down the road. 2:58:08 And then having to pay for it later. And yeah, bit 47 is another thing where 2:58:15 it's super arcane reason why it was like, unanimously, quote, unanimously discouraged 2:58:24 for implementation. But again, the end user, I think they would love nothing more than to get 2:58:30 rid of all, no one likes these Bitcoin addresses. They're weird. People, they're definitely not 2:58:36 going to last, you know, like in 30 years, people are going to look back and they're going to be 2:58:39 like, oh, yeah, kind of like how we look at like, maybe IP addresses, or I don't know, maybe like 2:58:48 usenet or something where it just, it's going to look old. And we're gonna be like, yeah, 2:58:52 that's what Bitcoin was like. Yeah, we got to get rid of these stupid Bitcoin addresses. 2:58:57 But anyway, that's just some of my pontification. So let me let me continue on that, then, like, 2:59:02 why is, or how can BIP 30 be implemented in multiple wallets now, as an example, 2:59:14 you know, to go against the common grain where BIP 39 and 47, 39 is used everywhere, 47 is used in 2:59:22 Samurai and some other things. Why can't BIP 30 be used right now? 2:59:29 As an example, I work at Edge, I'm QA engineer over here. And I in the past, when I, my company 2:59:38 made the first Bitcoin ATM long ago, and I white labeled air bits and said, Hey, can you guys help 2:59:43 me build a point of sale? And I changed, you know, air bits into Bitcoin pay, it basically put some 2:59:50 lipstick on the make, you know, on the pig and tried to get it put into these casinos. And so I 2:59:57 took someone else's wallet and said, Hey, let's make a user interface for the casinos. And they 3:00:03 helped me build that. So my thought process, knowing that that could be done is why is someone 3:00:09 not making a wallet that's like BIP300 friendly or throwing BIP300 in like a test version of a 3:00:16 wallet now? And trying it out? Is that able to be done? Or? Well, there's courage for a reason. 3:00:23 The Well, I would incur I mean, the thing, the reason why it wouldn't, it won't, the actual rules of BIP300 will not be enforced until 51% hash rate runs software that enforces them. So that is the soft fork. And we have testnet software already that we're trying to make. And we have a test network with, you know, with fake coins. And we already we have that 3:00:53 and we we want to continue to make that. And in fact, I think the right thing to do next would be to make a signet version, but who has time, but we have we have it because we have now we have just open in reg test. So that it's very easy for each per I think signet is actually probably better. We'll have a common signet and we'll have like a faucet. And so we kind of go back and forth between just like turning the network loose. And just letting people reg test only, which is the most convenient. 3:01:23 Because, you know, they can find the programmer to just find a block every two seconds or something. But I think signet is the way to go next. But but we have wallets, we and we do need, we have like desktop node software with desktop wallet. But we do need people to make to add support to whatever wallet it is that they, you know, people like certain wallet, then it would have to support this type of thing. If each wallet is kind of like the each sidechain is a 3:01:53 piece of software. So each wallet is kind of be like the the jacks multi currency wallet or something like that, where it would have to be a little like that, to be a mobile wallet. But, but yeah, I think all that should be done, but they won't, you understand that they won't actually be able to send Bitcoin to like the Zcash sidechain until BIP300 activates, they won't, they'll actually be able to send the coins, but they won't be safe. The coins will be vulnerable to just 3:02:21 Okay, so when, when, when we have, say, a user interface, like, say, we make a desktop version of a user interface that is playing with the BIP300 test coins, are we able to make test sidechains like a test Zcash and a test Ethereum and and test the interoperability in that way? 3:02:46 Yeah, exactly. Yes, that is the case. 3:02:51 Okay, so because I've always wanted to do stuff like that on top of Bitcoin, and I used to use counterparty, like religiously, and I think the biggest problem with all this cool shit that was done on top of Bitcoin just didn't have good user interfaces, you know, and I think that's a big issue and getting it to people, you know? 3:03:14 Yeah, I agree. I mean, you know, Steve Jobs would repeat that over and over and over again, everyone would say, Oh, Steve Jobs is so smart, business genius, and they'd say, what's the secret? And he's like, repeating it over and over and over again. He's like, just focus on the user focus on the customer focus on the user experience. So certainly, that should be number one priority. 3:03:36 Well, you've been around as long as me, but you may know this, and I don't know this. What was the threshold that got 39 and 47 put in play? Like, 3:03:47 right, 39 and 47 do not require, whether or not you use them does not affect if it's not affected by if other people are using them. In fact, the more people using 39, the more the better it is, whereas 300 requires 51% of miners to run the software first. So that's the that's the stumbling block. Otherwise, this all would have happened long ago. 3:04:17 Is there a specific technical reason that you could kind of explain to a layman why that is different? 3:04:25 Yeah, the coins the way the design of 300 is that the coins and must the coins can easily be you track which coins are in which sidechain and then they return through this kind of gauntlet, the sort of like slow hallway with a bunch of locked doors. And that is a very slow withdrawal system that is designed to make it infeasible for the coins to be withdrawn to any mistaken 3:04:53 destination. And if you if everyone, the BIP300 code enforces that, that like hallway thing, but if no one runs, if miners aren't running the code, then they can just mine a block that just takes the coins in the very next block, or someone else can just take the coins and pay a high fee and the miners won't really know that what they're doing is breaking any rules. So 3:05:24 they that is why this the rules of BIP300 will not be enforced, unless 51% of miners upgrade, which is the used to really not be a big deal. We did, we did many of these such soft fork things 16 or 17. In the first seven years of bitcoins history, but then came this SegWit disaster and everyone 3:05:50 panicked on doing them further to our detriment. 3:05:53 Is that why 301 is there as a sister bit because that encourages this hallway action from happening in a faster way? 3:06:05 I'm afraid they're actually completely different things. And 301 is a much tamer smaller problem that is already solved in many different ways. And a BIP300/301 301 itself is easily emulated by 3:06:20 other things like Ruben Somsen, and Super Testnet have already built space chains that use line merge mining. So they are they already kind of exists. It's not a very efficient way, but it does exist. And so 3:06:37 everyone got muted. 3:06:41 Okay, I'm back. So they 3:06:44 really ask, I wanted to kind of get an idea from it. You know, I'm a developer, but I wanted to get an idea from a user interface user experiences. Why can't we just shove new features in everyone's face? Why do we have to have the Illuminati of Bitcoin like approve it when these other BIPs just don't work anymore? 3:07:03 interface user experiences? Why can't we just shove new features in everyone's face? Why do we have to have the Illuminati of Bitcoin like approve it when these other BIPs kind of went through? But that answers my question on that. 3:07:17 If we had BIP300, then even things that are soft forks or hard forks, they would also be in the same category as BIP 39 or BIP 47. Where you could just launch it without getting approval from the community. And in fact, the point of sidechains is actually if you disagree with the community, you think they're doing something wrong. So actually, the point, you do have to get 51% miners to 3:07:46 approve you taking the slot, so to speak. But again, the slot is supposed to be for dissent. 3:07:54 So, 3:07:55 so this may be the last BIP we ever need if we can kind of get it. 3:08:00 Yes, that would be the hope, or even maybe this BIP would just make it so that extremely contentious ideas move to the L2. And as a result, there's just so much less politics and less unproductive chatter on L1 so that the only L1 changes are like, very, very tame, very, very conservative, very, very good ideas that just are made based on merit and not on 3:08:28 Is there a metric, a way of measuring right now approval? I mean, do we have a percentage threshold of, I mean, we don't have our user authenticated, you know, soft fork flags and stuff right now. But is there a way to measure how much approval BIP300 has right now? And when it comes to nodes? 3:08:47 Well, not when it comes to nodes. I think we have, yeah, we should probably try to do that somehow. I mean, for now, I'm actually focused on getting out the, an actual pull request to Bitcoin Core that activates both 300 and 301 that is like, rebased and like good enough to actually use. Because until you have a real proposal, 3:09:13 people won't really know like, what it is, I think. You could ask, like, the smart thing would be, I suppose, to ask the heads of mining pools, you know, if they would support activating it. So we've had the one public yes from Luxor pool. So that's whatever, 3% or so. So that's three down 47 to go. 3:09:43 But that's just I wasn't me who asked some random other person. So I think that would be a good 90% though. 3:09:49 Well, I suppose, if to actually add the sidechain, you would need 90%. But maybe it could be the case that a miner has a belief of the form. I don't want to get involved in soft fork activation. But then if BIP300 activated, they might be like, okay, whether I wanted it or not, BIP300 is activated. So now that it is activated, should we, 3:10:19 should we try out the slot? And then they could so their willingness to participate? 3:10:26 Is there any way to know it's activated? Unless there's like actually already a slot with 90% 3:10:31 approval? I really know, right? 3:10:35 It is like all things, it exists between a state of certainty and doubt. But it, there is a way of 3:10:45 activating it where everyone knows and there's a way of activating it where people don't know 3:10:49 yet until after it had activated. There's always a way of finding out eventually that it did activate, 3:10:54 which is to try and break the BIP300 rules and notice that the block never is the block is always 3:11:01 getting orphaned. The but what I was we were just talking about moments ago was like having 50% of 3:11:07 the pools say they would support it and then have them run the software. And then in that context, 3:11:17 people would know that it had activated. 3:11:23 Well, that that's all I wanted. I'm going to step down but I will I'll join the discord and I want 3:11:28 a job so whenever you need someone you call me and I fully support you and I think we need some 3:11:34 dramatic changes to Bitcoin. And I think technical options like this are great. So keep 3:11:40 keep kicking ass, man. Go ahead. Thanks. 3:11:46 I just want to say BIP 47 is amazing. I know the Samurai users really, really love it. You made a 3:11:52 really good point that it would just be so much easier for like the user, you know, if you just 3:11:56 share the, you know, like like the payment that is derived from the BIP 47. I think Sparrow also 3:12:03 implemented it. I mean, I know that it has, which is great. Yes, it's not. It's like a wall. Any 3:12:09 wallet can just do it also. So even though core so it's an example of how core, you know, there's 3:12:17 it's often thrown around this this great lie that Bitcoin Core is like an open system and that 3:12:24 anyone can work on it and that it's a meritocracy. But but basically, you know, one person's merit 3:12:30 is another person's. Terrible idea, but but I think BIP 47 in particular proves that Bitcoin 3:12:40 core often makes mistakes and they don't take the idea with the most merit. Well, 3:12:44 39 as well, of course, still doesn't have 39, right? Bitcoin Core does not implement 39. 3:12:51 No, there's no you get no passphrase. And if for a while you back it up, if you generate 3:12:58 more keys, you still have to back it up every time. No, I think there's HD wallets. But I think 3:13:03 but you have to for a very long time, you could not 3:13:11 switch wallets if you were using the QT client. I think it was only version 20 that allowed you to 3:13:17 like save the wallet to like a file and then like open a different one or something. You could not 3:13:22 like oh, I think it was maybe set the seed randomness. That's what it was. You could not 3:13:27 set the seed randomness until version 20 or something like that. But so it's just not the 3:13:35 case that Bitcoin Core is a meritocracy. And it's not the case that if it was, then no one would 3:13:41 ever complain about anything. All the complaining would be illegitimate. So I guess if someone wants 3:13:45 to live in that fantasy world where they believe two lies at once that are mutually reinforcing, 3:13:51 then that's great. Good for you. But there is no such thing. Nothing is infallible. And there's 3:13:58 no process that is without error. And the key is to how to correct the error. The number one 3:14:05 question is always how can Bitcoin survive mistakes made by Bitcoin Core, which are inevitable. 3:14:12 And so yeah. 3:14:15 Do you think that there's benefits to alternate implementations being more widely adopted, 3:14:21 like Bitcoin Knots and like whatever other alternate implementations about following 3:14:26 the same consensus? The alternate implementations, 3:14:30 I think Satoshi was strongly against them. He said they would always be a menace to the network 3:14:36 because if they ever disagree, it may fork the network. And Corey Fields is a brilliant guy. He 3:14:42 gave a talk about 2 plus 2 equals 3 about saying whatever Bitcoin Core does is what everyone else 3:14:49 has to do. Although I think that that position is a little bit overstated because it's not like 3:14:56 literally the end of the world. You could run Bitcoin Knots and you could also know that you're 3:15:00 running a minority piece of software. And so you could know that, hey, I'm in danger of forking 3:15:06 off. And certainly, I think you don't want any miners. You want all the hash rate to be running 3:15:11 the exact same piece of consensus, the exact same consensus library. But other than that, 3:15:18 an end user could run whatever to process the node, however they like. The blockchain process, 3:15:24 the blockchain, however they like. But these don't tend to be very popular. 3:15:33 Bitcoin, the JavaScript one, and Bitcoin Knots, Luke S. Jr.'s thing. But Bitcoin, 3:15:41 the JavaScript node, it has excellent comments. It's very easy to read. It's been said, 3:15:47 when I was in San Francisco in 2019, it was often said that if you want to actually learn how 3:15:52 Bitcoin Core worked, you should just read Bitcoin, the JavaScript Bitcoin Core node because it had 3:16:01 a layout and it had comments and it had like a consistent style. 3:16:06 So it was easier to read. And yeah, there's not that many comments in Bitcoin Core, 3:16:12 which is very shameful. And also very, very, very, very much against the spirit of self-sovereignty. 3:16:18 It's a very hypocritical Bitcoin Core too. We talk a big talk about your node protects you 3:16:25 and trust don't verify. But really, it should be like every single line of code should have like 3:16:33 more comments than code should be explaining every single thing so that 3:16:39 anyone who tried to do something nefarious would be caught easily. 3:16:47 I don't know. Sam Parker has a question. He has written it in the form of a tweet 3:16:53 and he has not come up to ask it. And I don't know why, but the question is, 3:16:58 was this larger TX optional if you didn't intend an ongoing relationship with a party? 3:17:05 I have to admit that I'm not totally sure what that means. 3:17:07 You're talking about BIP 47. And in that case, you don't really do that transaction unless you want it. 3:17:15 Yes. Was this larger transaction optional? Yes. You only did it if you sent the BIP 47 style 3:17:24 transaction. So it's not like every transaction would become 20% bigger, but if you wanted to 3:17:30 use a BIP 47 and create the tunnel. And one intriguing thing also is that the recipient 3:17:35 gets it for free. So if I want to send money to Edward Snowden, I look up his friend code or 3:17:41 whatever. Then I friend him on the blockchain. And then he and I have an encrypted tunnel and 3:17:47 he didn't even have to do anything. So BIP 47 was fine to me. BIP 47 has also been improved. 3:17:55 There's a version 3 and a version 4. The guy, Justice Ranvir, he had some differences of 3:18:03 opinion with Bitcoin Core and he kind of just left to do his own thing, many such cases. 3:18:08 But he has a different GitHub library where he does version 3, version 4 3:18:14 of BIP 47. And Ruben Sampson has further tinkered with this idea to do the 3:18:23 silent payments thing, which is basically a sort of a version of BIP 47. 3:18:30 And so the larger transaction is optional. Yes, the answer is yes. 3:18:38 And so one of the benefits is that all other transactions 3:18:42 in this channel of sorts that you've opened is much, much smaller on-chain? 3:18:49 No, they are regular size, but you get an endless, each of you gets an endless vector 3:18:55 array, so to speak, of like, I get a list. You and I, I make the BIP 47 transaction to you, 3:19:03 and it causes both of our wallets to become aware of each other's friend code. We store it. 3:19:10 It's this one blockchain event. We store it and then I can look up, there's like a Paul-Sam 3:19:15 list where Paul pays Sam. And then it's like, imagine like a Microsoft Excel or something. 3:19:21 We have like a column that's like Paul pays Sam. And then you have a second column that's 3:19:26 Sam pays Paul. And then the rows are just Bitcoin addresses. So we have now, each of us has an 3:19:31 unlimited stream of new Bitcoin addresses to use. That's the key is they're all different. 3:19:38 So there's never any address reuse, but we know now we have an infinite number. 3:19:42 We can just keep going down the list over and over and over again. And whenever I want to send 3:19:47 you money, I know what public key to use next. And then once you receive money there, you think, 3:19:52 okay, now not only am I going to check everything on this list, but I'm going to check the next one 3:19:57 on the list because Paul just used one, two, three, four, five, six. So now I'm going to check 3:20:01 number seven. And you're going to add that as like a watching only thing, as one of your, 3:20:07 is mine equals true. And then you, when I send you money on seven, then you switch to eight. 3:20:13 So it's like, they all have their own gap limit or something. 3:20:17 And, but also key is that we derive a, not only do we have an endless series of. 3:20:24 Bitcoin addresses and ECDSA public keys, and, you know, it could be any kind of, any kind of public key. 3:20:32 We have a shared secret also. So just between all these things, we have basically the ability to send messages to each other. 3:20:41 So that can include me sending you like a text message or something on Telegram that's encrypted, like through some rail, but your wallet could open it. 3:20:52 Cause we have, you and I, we have this tunnel already. And since you can do that, then the message could be something like. 3:20:59 It could be, I could literally send you a message that has you like, you know, like curl CURL something and decode that you and I can have like a whole encrypted video chat or something. 3:21:11 We can disguise the limit because we, we know, know each other's key. And so it's also replacing like PGP, it's replacing web of trusts, replacing everything. 3:21:20 It wouldn't literally replace all that overnight, of course, but I'm just trying to give an idea of. 3:21:25 Yeah, I just want to give one more example. The Bitcoin company, it's a gift card website, like similar to BitRefill. 3:21:32 It actually implemented paying in BIP-47 logins. So you can just like, you know, have an account that you don't have an email address or anything. 3:21:40 You just, you know, you register your account with the BIP-47 and you log in with BIP-47. 3:21:45 So that's an example of what you're talking about. You can use this for, you know, authentication, logging in, talking, whatever you want to do. 3:21:52 That's a, excuse me, that's a really cool idea. It's like you're entangling your, your HD wallet generation with this other guy. 3:22:01 But yeah, like I, it seems like something, excuse me, it seems like something like you can, you don't necessarily need to do anything on the core level for that. 3:22:12 You could push this up a layer. And I'm assuming that that's, you said that Ruben Somsen was doing something like this for silent payments. 3:22:18 I'm assuming that's what he's doing. 3:22:22 Yes, that's right. 3:22:25 Interesting. 3:22:26 And I just want to, yeah, so look into silent payments if you want to see. That was kind of like what became of BIP-47. 3:22:33 But what I want to stress is it was, this idea was killed because it was wasting block space quotes, quote wasting. 3:22:41 And there was this notification transaction on Chain that people thought was inelegant. 3:22:47 But it turns out, in my opinion, they're certainly right. 3:22:52 I mean, if you look at the technical details, it does feel inelegant. 3:22:56 However, later attempts to replicate this functionality without the inelegant part have faced disastrous trade-offs. 3:23:06 And what this does with the, with the one quote inelegant part, which again can be, can be shrunk down and compacted so that the only cost you pay is the first transaction between two people is 20% larger. 3:23:21 It goes from being 220 bytes to like 200 and whatever, 48 bytes or something. 3:23:25 So, so that's not that much. 3:23:29 And the different versions, either they introduced a requirement that in order to receive money, you have to scan the entire blockchain again. 3:23:42 Like, so whenever you get a new, like a new person wants to friend request you or something. 3:23:49 So the, you either have to scan everything or you have to do some other stuff. 3:23:54 So it's something that looked kind of inelegant at first. 3:23:58 My view now is that this is just an example of how something can really, really, really be misleading and how it looks. 3:24:07 Because I, now with hindsight, I now look back on this and I see it, I still see it as inelegant, not elegant. 3:24:15 I see it as this creepy kludge, like friend requesting someone on the blockchain. 3:24:21 And it's just this, this pointless OP_RETURN bare multisig junk is in the blockchain. 3:24:25 It does look like it's spamming the blockchain. 3:24:27 But now I, I see that as, even though I still see it, I see it as an illusion. 3:24:33 It's like when you see an optical illusion that you already know. 3:24:36 And you say, you see the two lines and you know that they're the same length, but they still look like a different length. 3:24:42 That's how I see it now. 3:24:44 I look back on bit 47 and I think that was a really, really good idea. 3:24:49 We could have had that idea in like 2014 or 2015 or something. 3:24:53 It wasn't that long after stealth addresses. 3:24:57 And that was like 2012, 2013, 2014, something like that. 3:25:00 So, so we could have had this idea the whole time. 3:25:03 No addresses in the UI, just the friend code. 3:25:09 Imagine that, that would have been amazing. 3:25:11 Would have been way better. 3:25:12 Wow. 3:25:13 It's quite soon also to go from, people would have also hated this, but it would have been, 3:25:18 I think this is like if history had taken a different turn, the idea of putting the name system on the blockchain also. 3:25:25 Which again, this would have been like wasting space because you would broadcast a message that says like, 3:25:30 I want to be truth coin on the Bitcoin world. 3:25:35 I want to be truth coin, you know, dot bit or something. 3:25:38 And then if two people, you know, you do like a hash reveal and whoever gets it first, 3:25:43 whoever is first in the blockchain gets it, gets the name. 3:25:47 But if we had done that, like Namecoin, like Namecoin as counterparty on Bitcoin, 3:25:54 you know, if that makes any sense to anyone. 3:25:56 But if we had done that, we would be wasting more block space, but then it would be even, 3:26:02 it would be even more user friendly. 3:26:04 It would be like very difficult because now you just have, 3:26:07 we would just be truth coin and then there'd be Satoshi Enjoyer and then there'd be ZK Chesterton. 3:26:13 And that we would just be sending you, there would not only be no addresses, 3:26:18 but the friend code would just be an actual name. 3:26:21 It would not even be a, and there would have been, 3:26:23 we would have long blown past the point where like Silk Road, 3:26:28 Silk Road would have been like a name on the blockchain. 3:26:31 And that would have done all the authentication. 3:26:33 There would still need to be like a site and stuff. 3:26:35 But then with Telegram bots and stuff, it would have been a very different world. 3:26:38 But because people were so, well, you know, I'm just going to say that this is like, 3:26:44 you know, this is the decision that was made to, it's right there in the BIP. 3:26:47 It says unanimously discouraged for implementation. 3:26:50 So decide for yourselves, of course, a meritocracy after that. 3:27:00 I think there was other reasons why they might have not thought it's a good idea. 3:27:05 Because like if you want to say it's like privacy preserving, 3:27:09 if on chain I'm opening up this tunnel between me and you, 3:27:14 and then I go ahead and do a transaction, 3:27:16 like chances are I'm probably just sending you those funds, right? 3:27:19 Because I just opened the channel with you. 3:27:21 But I think the BIP had that, it had some ways of obscuring that. 3:27:27 So like when A, if I open the tunnel with you, 3:27:31 I use your friend code and everyone sees that someone is friending you, 3:27:37 but they don't know it's me. 3:27:39 I spend some BTC and they don't know. 3:27:44 They don't know it's a, there's three outputs. 3:27:49 So I spent some inputs and then one of them I, one of the, 3:27:57 okay, let me re-explain it. 3:27:59 I select one input and then there's two outputs. 3:28:01 One of those outputs has a bare multisig with three keys. 3:28:06 One of those three keys is your friend code. 3:28:10 So everyone knows that someone's friending you, 3:28:13 but they don't know. 3:28:14 And then the other part is a, something that you will be able to calculate a, 3:28:20 I give them a code that enables you to decrypt my friend code. 3:28:24 So my friend code is never given away. 3:28:27 So you don't know, everyone knows that Soshi and Joy are getting a new friend, 3:28:31 but no one knows who it is. 3:28:32 Only you know that it's me. 3:28:34 And then the other script of the three is an output, 3:28:38 which would normally have been changed, 3:28:40 because I'm going to select, you know, something that has four Bitcoin. 3:28:43 I'm going to pay 0.01 in fee and then pay 3.9999 back to myself. 3:28:49 And, but however, that does not need to be, that part you can flip around. 3:28:56 That could be my first payment to you, or it could be the change. 3:29:00 And then the change could be a different output or my first payment to you 3:29:02 could be the other output in that transaction. 3:29:05 So you can scramble all those. 3:29:07 So they don't really know how much you're being paid in the first payment either, 3:29:12 if anything, you know what I mean? 3:29:14 Like I can select 3.99, pay some to myself, pay some to you. 3:29:17 Yeah, yeah. 3:29:18 Like it would look like a page join kind of, or not a page join. 3:29:21 It's more. 3:29:22 They just don't know which one. 3:29:24 You can't really know which one is which, yeah. 3:29:26 Yeah, they don't know which one's the change, which one's to you necessarily. 3:29:29 So, so it's not, it's not great, but it's not, it's also not broken either. 3:29:36 For most of the people, most of the time it would have worked. 3:29:40 I agree, because you can open a channel, like open this tunnel with you today 3:29:44 and with somebody else tomorrow and somebody else the third day, 3:29:46 and then I could just pay one of you and nobody knows which one I paid, right? 3:29:50 Like automatically. 3:29:52 Right, no one knows who wants. 3:29:54 Someone knows. 3:29:55 So on day one, I opened the tunnel with you. 3:29:58 So someone knows that you have a tunnel. 3:30:03 And then I open a tunnel with Sam on Tuesday, 3:30:06 and then on Wednesday I open a channel with Moon. 3:30:09 And then from then on, no one knows if I'm using any of the tunnels. 3:30:12 Everything just looks like a normal Bitcoin transaction. 3:30:14 The tunnels are only for so we can agree on a list of Bitcoin addresses 3:30:18 without interacting. 3:30:23 And so, because the problem with the interaction is I have to – 3:30:28 well, interaction is bad for privacy, 3:30:30 but the problem with interaction is like I interact with Sam 3:30:33 and he gives me a Bitcoin address. 3:30:37 Is that one like burned forever? 3:30:40 Because if he doesn't use it, like then what do I do? 3:30:43 So I have to keep track of a lot of state, actually, 3:30:48 because if I give one to Sam and one to Satoshi and Joy and one to Moonsettler, 3:30:53 then I don't – I've given three off of my list. 3:30:59 So I have to keep track of how many I've given out. 3:31:03 In this scheme, everyone gets their own list that starts at one, starts at zero. 3:31:08 So I can do this all day, and I don't have to be online to do it either. 3:31:13 Just someone needs to find my friend code. 3:31:15 So I can be like Edward Snowden. 3:31:16 I can be like on the run, and I have no internet. 3:31:19 But then when I get to my internet, I can find it. 3:31:21 All of a sudden I have 40,000 new friends, 3:31:23 and they all want to send me money and messages. 3:31:27 Send me a little thank you email or something. 3:31:32 You know how it is. 3:31:36 So that was – again, it looked ugly, but it still does. 3:31:40 But now I see it as an illusion. 3:31:42 When you would see – you look at those two lines, 3:31:44 you know the lines with the arrows pointing in and out on the edges, 3:31:48 and the one line looks bigger than the other, 3:31:52 even though you know they're the same length. 3:31:54 So, again, I see it for the ugliness that it is, 3:31:59 but at the same time, it was by far a superior idea 3:32:04 to whatever else we thought was good at the time or even since. 3:32:09 Like it's a better idea than probably Taproot, 547 is. 3:32:13 So there you go. 3:32:18 Just – excuse me. 3:32:20 Going back to something you were saying earlier. 3:32:21 I've been thinking recently that you were talking about the problem 3:32:25 with Lightning basically is, yeah, there's a hard cap 3:32:29 on how small of an amount you can basically go to court for. 3:32:39 And that really – like that – the problem of kind of guaranteeing 3:32:45 settlement of fraud proofs in a very general way 3:32:48 is something I've been thinking about a lot recently. 3:32:50 And basically, yeah, if that problem actually is unsolvable 3:32:57 without either going full sidechain or increasing block size, 3:33:03 like if we're like really bumping up against like information theory 3:33:08 type stuff there, which is possible, but I suspect we aren't, 3:33:13 then yeah, that's – we absolutely need to do that. 3:33:17 But if there is a mechanism by which we can guarantee, you know, 3:33:27 asterisk settlement of fraud proofs in like extremely high volume 3:33:34 and for arbitrarily small amounts, you know, I'm thinking something 3:33:40 by some mechanism along the lines of congestion control 3:33:45 or some kind of batching like that, then yeah, the whole – 3:33:50 basically the small block dream is dead. 3:33:58 Well, I think certainly a lot of stuff could be invented. 3:34:02 And who knows? Like I think Ark is a good example of like fixing onboarding. 3:34:09 But Ark, of course, has its own weird tradeoff with the – 3:34:17 every transaction there must be funds that are locked 3:34:20 for like the whole period of time. 3:34:22 But that's not really the point. 3:34:24 I think the point is like the small block dream – I don't know. 3:34:32 I have a very different way of just looking at the whole thing. 3:34:35 Like the way I would see it was it's a dispute between two groups of people. 3:34:42 Like these two types of people are very – they have a different type of person 3:34:47 and different types of transactions. 3:34:49 So I see the world as very heterogeneous. 3:34:54 You know, it's very different. 3:34:56 So like – so for example, Roger Ver was a famous large blocker, 3:35:02 and he was rich, and he lived in Japan. 3:35:06 So he could easily afford – to him, running in the full node was nothing. 3:35:11 And running one that was 10 or 100 or 1,000 times larger would have been nothing. 3:35:17 He would have been thinking to himself, you know, I have whatever, billions of dollars. 3:35:22 B-I-L, billions of dollars. 3:35:26 He's thinking like what? Like I'll just buy nodes like all over the world. 3:35:30 So he had like a very different point of view. 3:35:33 You know what I mean? 3:35:34 And then like you had Cypherpunk people were thinking like, okay, it's happened. 3:35:40 They're going to arrest everyone. 3:35:42 We're all on the – it's like the end of hackers. 3:35:44 We have to use the payphone to hack into whatever, the mainframe. 3:35:51 And so we need – you know, every second is going to count, 3:35:58 and we need to be able to hide our location behind Tor, 3:36:03 and we need the network to be – we need the network to be there. 3:36:09 We can't have it die. 3:36:11 You know, it's like a little fire, like a little candle, so we've got to protect it. 3:36:15 And so just – I don't think one of the two groups was wrong or right. 3:36:20 I think it depended on the end use case, really. 3:36:25 So when Ross Albright is running Silk Road, he – 3:36:33 the biggest priority for him would have been privacy. 3:36:37 That would have outweighed by far like, you know, the decentralization. 3:36:44 But decentralization would also have been a priority because he has to hide his node. 3:36:49 So he would have liked small block, you know, Zcash or whatever. 3:36:54 So let's just assume that Silk Road large block would have – he would have been a small blocker. 3:37:00 However, he was leading a double life, so he, you know, 3:37:09 he would be running Silk Road as Dread Pirate Roberts, but then he would – 3:37:13 as Ross Albright, he would walk down the street of San Francisco and buy coffee or something. 3:37:21 And my point is, when he buys coffee, he switches from being a small blocker to a large blocker. 3:37:26 He doesn't have a type, you know. It's the transactions that have a type. 3:37:30 So he switches, and then he switches and becomes a large blocker for a couple moments, 3:37:36 and then he switches back. 3:37:38 So my view is all the transactions are very different, like all these USDT, these Tron transactions. 3:37:44 They are different for a different type of person. 3:37:48 The Zcash transactions are for a different type of person. 3:37:51 The Monero transactions are for a different type of person. 3:37:55 So I don't have this view that, you know, like the small block dream is – 3:38:04 to me, the small block dream is just that L1 is there, you know, as a small block version, 3:38:09 and then the large block version is – can also be there. 3:38:14 So I don't see it – I think trying to make it one size fits all is doomed kind of no matter what, 3:38:20 because I think there's just too much of a difference in people and in their circumstances around the world. 3:38:27 You know, there's 8 billion people. 3:38:30 Some are very poor. Some are very rich. 3:38:34 Some have great Internet. Some have terrible Internet. 3:38:37 Some can easily afford transaction fees. Some cannot. 3:38:41 Some can easily afford to wait for block inclusion. 3:38:45 Some need it instantly. Some need, you know, some need privacy. 3:38:49 Some don't. Some need decentralization. Some don't. 3:38:52 So I think there's just too much of a difference in the world is my guess. 3:38:57 Excuse me. Interesting. To me, the quote-unquote small block dream, like what I like about it is it is a – 3:39:08 it's a hyper-conservatist take on the kind of stateless blockchain idea, 3:39:17 where the idea is basically you have a – like basically you have no blockchain. 3:39:26 Like all you have is the UTXO, something equivalent to the UTXO set, the active state, 3:39:33 and you have a recursive zero-knowledge proof that has recursively verified the complete history of the chain 3:39:41 up to and including that particular snapshot of the active state, and then that is your blockchain. 3:39:49 And so the size of your blocks at that point doesn't really matter for anything other than like bandwidth requirements 3:40:01 for running a node and like latency finality stuff. 3:40:06 And this is – in my opinion, this is like the endgame for blockchains, 3:40:11 and this is eventually where every blockchain is going to converge into. 3:40:14 In fact, you can see even though there's only a small number of non-Bitcoin blockchains that do this, 3:40:20 we have through the zero-sync guys like Bitcoin is actually going to. 3:40:24 be getting this, albeit like at a layer up, and that they're – what they're doing is like recursively verifying it. 3:40:30 But if you don't want to trust the – your execution history of your chain to put that under 3:40:42 cryptographic assumptions in the way you have to do to do that, well then the next – 3:40:50 the – so if you – yeah, so then the next best thing to do would be something like SmallBlock. 3:40:56 I do – I already disagree though with a lot of what you said, or I'm not even sure that you realize what you're saying. 3:41:01 You're saying the large – see, because the zero-sync – let's start with the zero-sync guys, which, you know, I love the zero-sync guys. They're great. 3:41:08 Okay, we got Robin Linus. He's a big PIV300 supporter, by the way, but he's also great for other reasons. 3:41:17 So he's got a great accent. No, no, the zero-sync guys are great, but what the zero-sync solves is a problem where you run an SPV node and they lie to you about it being a valid blockchain history, but that problem is – that's a non-existent – that has never happened, so it's a non-existent problem. 3:41:36 So it's nice to get that extra cherry on top certainty, but just the way that SPV works already, you have to put in all the thermodynamic work to get the header, and having the zero-sync proof, I think, doesn't really add – you know, it doesn't really add that – I'm all for this research being done. I think it's great, but it doesn't add very much, so that's the first point I want to mention. I'm curious to see your reaction. 3:42:04 They're also proving the hash chains as well, so you don't need to download the complete hash chain. They're just basically summing the cumulative hash weight across all block headers. 3:42:14 Yeah, but the headers are 4.4 megabytes per year, so it might as well be zero, right? I mean, I'm not really – what it would take to download the ZK Prover would be larger than – 3:42:30 That's probably true, yes. 3:42:32 Okay, so that's not really an issue, and it speaks to maybe a misunderstanding of nodes, but I don't know. We should probably evade that for now, because the large-block vision – that was the large-block vision – was that everyone will run SPV, and no one needs to run a full node, because we can trust the miners, so to speak. 3:42:54 Now, with ZeroSync, trust the miners is basically replaced by trust the ZeroSync Prover, but I think that is a small improvement, but you understand what I'm trying to say, that that has never actually happened. 3:43:11 The other thing, though, is that neither SPV nor ZeroSync solved the problem that we discussed in this space a few hours ago, which is the data availability problem. 3:43:24 You always need to have someone who's out there who's able to serve block 350,425, because if no one has that block, then no one knows if everything in it is valid, and this is a horrible problem, because now you have to assume that it's invalid, and so now you have to re-org. 3:43:48 So if anyone ever loses a block, then the blockchain must go back to that spot. So this is a horrendously difficult problem, which is not good at all. 3:43:58 It's the only blockchain, I think, that has really solved it, and that's Arui. They're, in my opinion, the only ones that have actually solved this problem. 3:44:05 Yeah, well, it's very difficult, because you need to... Well, yeah, in Bitcoin, that's part of the reason for small blocks, though, is to say we'll keep the cost of serving all this data limited, which will prevent it from growing to an unlimited degree, and then the higher the cost, the fewer people are going to do it, of course, other things being equal. 3:44:29 So that's another point to make. Third, you mentioned... Actually, I have also solved it myself, and you mentioned fraud proofs. I don't know if you read, I wrote a little post on truthcoin.info about fraud proofs, including a solution to data availability where you randomly query the block, and I think it was in 2016, a long time ago, and I wrote about this. 3:44:55 That's interesting. That's what Arweave does, although they base it... Maybe... Were you the guy that inspired the Permacoin proposal? 3:45:04 Oh, I don't... I'm not aware of that. 3:45:05 Maybe this is your idea. 3:45:06 But it probably was my idea, because I do get lots of questions often about my ideas, and then later I find out that... And, you know, I think... Isn't Eric Wall involved with Arweave or something? 3:45:17 I don't know. The NFT people seem to like Arweave. That would be cool if that was your idea, because I agree, because that precise idea is... 3:45:25 Oh, you can look it up on my... 3:45:27 Yeah. This is the only thing that I've stumbled upon... 3:45:32 My idea requires that you open a channel. It doesn't have to be a Lightning channel, but it can be any bi-directional channel with whoever is claiming to have the block. 3:45:45 So that is kind of a funny thing about it, is it sort of required the Lightning network or something like it. It's a very neat idea. 3:45:54 And, yeah, the fraud proofs are very cool. A very, very interesting thing. 3:46:00 It used to be said in the past that... People used to say certain things that I think are very suspect now. 3:46:07 So, for example, they said, oh, we could have large blocks if only we had fraud proofs. 3:46:12 I kind of didn't believe that. I still don't. 3:46:15 They could have large blocks in Layer 1 if we had fraud proofs. 3:46:18 I don't actually believe that, but it was just kind of like... 3:46:22 Because then they would say... Right after that, they said, we can't have fraud proofs because there's no way to check if a block is... 3:46:32 There's no way to check if the block is following the block size limit or not without downloading the whole block. 3:46:40 And I thought, that's not true at all. 3:46:44 That's not actually a good reason, because you just break the block into tiny little pieces and have each piece keep track of how much of the space it's using. 3:46:53 And then, since the sum has to equal the total thing, each incremental part, you have a fraud proof for each increment, any mistaken increment. 3:47:03 So I wrote about that in the fraud proofs post. 3:47:07 But what really got me on the whole thing was that it's clear that everyone's kind of being a little dishonest, as usual. 3:47:16 Where they were saying, like, X is why we can't have Y and Y is why we can't have Z. 3:47:23 And so just for fun, I invented X and Y, but it was just to show people that it's not the real reason for Z. 3:47:29 The real reason for Z is the full node cost. You have to serve the data. 3:47:33 That the blockchain is harder, is more effort to process, store, validate the blockchain, the larger the block size. 3:47:40 And so if you buy that frame, then everything can have some kind of fraud proof already and will. 3:47:51 And it could have ZK proofs. 3:47:53 And then the real problem, my point of view, is that the entire problem is just people disagree over how expensive the node should be. 3:48:03 And so that's the idea, is to try to get them to cooperate despite that disagreement. 3:48:09 So that's kind of my take on the whole thing. 3:48:12 Sure. I just looked it up here. Andrew Miller published the Permacoin paper in 2014. 3:48:19 So he scooped you by a couple of years. 3:48:21 But the cool thing about Arweave is that they actually use this. 3:48:25 And actually, I think Greg Maxwell actually, of course, had a post years ago where he was like, 3:48:31 an idea for an altcoin would be a coin where the proof of work algorithm was doing random querying into the UTXO set. 3:48:39 So that I wouldn't be surprised if that predates this paper. 3:48:43 But the cool thing about Arweave and the cool thing about using this random data querying system as your consensus mechanism is if you construct it right, 3:48:53 the chain doesn't advance on blocks that have not achieved a sufficient threshold of data availability that has been proven on them. 3:49:06 So if the data hasn't been disseminated, there is no data withholding attack in this model. 3:49:11 Because if the data hasn't been disseminated to a sufficient number of miners and a sufficient amount of the hash has mined on it, 3:49:19 the block doesn't get produced at all, which is a very, very cool thing. 3:49:25 And it's the only thing I've seen that I think solves the data withholding, data availability problem sufficiently. 3:49:32 Because I don't even know whether or not we know to a sufficient degree how we prevent data withholding attacks on Bitcoin now where the miner hits a block. 3:49:45 If someone announces that they have found a block and then they don't give you the whole block, then they are kicked off. They are banned or whatever. 3:49:54 Okay, so this is basic data. 3:49:57 Yeah, exactly. Which is not ideal. 3:50:01 That seems unsatisfactory. 3:50:03 Well, you know, I think it's like the network is robust to not having found a block. You know what I mean? 3:50:13 It's not like if we go 11 minutes without finding a block, your computer bursts into flames. 3:50:18 So it's kind of like saying, if you say that you have found a block, then give us the block. 3:50:23 You know, and otherwise, don't bother us. 3:50:25 So but yeah, it's not. 3:50:28 There's in particular, not like I was asking you before. 3:50:32 How do we know that someone has blocked 300,000? 3:50:38 Like, I mean, you know, we we just kind of hope that no one loses it. 3:50:47 Yeah. And actually, the interesting thing about Bitcoin not having a recursive ZKP is that basically every block is part of its recursive proof. 3:50:58 And the reason why we're trusting that someone is going to be keeping the data is because you can't actually verify the tip of the chain unless you have that data. 3:51:08 It's part of the proof. 3:51:11 And so, you know, by like this weird dichotomy whereby like, quote unquote, modern blockchain design is all about stripping away the need to store the full proof of the chain. 3:51:23 But in doing so, you're also dramatically reducing the data availability requirements for the data that you do want to store. 3:51:30 So anyway, yeah, I just thought I would mention that about how. 3:51:41 I don't think zero sink. 3:51:46 Zero sink is not is totally irrelevant. 3:51:50 Zero sink is like a slightly improved, modified SPV. 3:51:54 But SPV already has like a 100 percent success rate. 3:51:59 And for SPV not to work, you need to have someone mine a whole block. 3:52:07 That is invalid. And then. 3:52:11 You know. Propose it to the network, like send it around as if it were valid. 3:52:19 And the network should rely on SPV, that would be a bad idea to rely on it. 3:52:24 But the network. I mean, SPV, it would cost them about an opportunity cost. 3:52:33 It costs them whatever that is, seven BTC to make that. 3:52:37 So this is already a very expensive attack just to do something that won't work. 3:52:42 And that's why it has never been done. 3:52:45 Because it costs you whatever that is, like. 3:52:48 Whatever that is, like $200,000, $250,000. 3:52:55 So it's expensive and. 3:53:02 But the I think the problem is just full node costs, and that is why. 3:53:08 Sidechain solve it because there is heterogeneity. 3:53:13 The distribution is different. 3:53:28 Anyway. Nice to hear from you. I see you on Twitter sometimes. 3:53:33 We seem to be dwindling down. We had many more people before. 3:53:36 But now we have. What do we have here? 18. 3:53:42 So maybe we should begin to wind it down. 3:53:47 Let's. 3:53:50 If anyone wants to ask anything. 3:53:56 I just want to give a suggestion. 3:53:58 Great. 3:54:00 Chris Black was part of the audience like a few weeks ago. 3:54:03 He never came up. 3:54:05 He has like a podcast that he started recently. 3:54:08 And he started having like some, I guess, controversial figures on there. 3:54:12 He had Uri on there yesterday or something like that. 3:54:14 Might be a good idea to go on there and like explain BIP300. 3:54:22 I almost always accept. 3:54:25 I'm not aware of having really rejected anyone when they said come on my podcast. 3:54:33 Oftentimes people invite me to their conference and then they're like, just get on this nine hour flight. 3:54:38 And I was like, oh my God, I'm not going to do that. 3:54:41 I feel bad for everyone that I say no to, though. 3:54:45 But yeah, I don't. 3:54:47 But yeah, I'd love to go on. 3:54:50 I'll go on whatever podcasts people think so. 3:54:57 Chris Black. 3:54:59 Hopefully he'll read it. 3:55:00 Yeah, that'd be great. 3:55:02 Yeah, I think it's a much more effective if someone else asks. 3:55:06 Oh, you should. 3:55:07 They say, oh, you should have this person on. 3:55:09 Then the one person. 3:55:15 I kind of have a very full schedule anyway, so it's not as though I'm desperate to go on any podcast anytime soon. 3:55:22 But if someone wants to have me on, then that's good. 3:55:25 Probably a better interview, too. 3:55:31 OK, well, this has been fun. 3:55:33 I don't see any new people. 3:55:35 No new questions. 3:55:37 So. 3:55:39 Thanks for a great space, everyone. 3:55:41 Until next time. 3:55:46 Until next time. 3:55:48 Happy Friday. 3:55:49 Happy weekend. 3:55:50 Every Friday.
