Drivechain is a proposal from 2015 which has had a lot of community debate recently. Joining us today to debate this are Paul Sztorc (CEO of Layer Two Labs) and Peter Todd. We discuss a range of points:
- Drivechain and miner centralisation
- Whether you can delegate running a node
- Legal risks
- Fees and Blocksize limits
Links:
- Earlier episode with Shinobi: Will Drivechain Centralise Bitcoin? With Shinobi
- X: @peterktodd
- Peter Todd’s post: Drivechains: A Detailed Analysis
- Peter’s Site:petertodd.org
- X: @Truthcoin
- Paul’s site: Truthcoin.info
- Paul’s response to Peter’s post: https://www.drivechain.info/peer-review/peter-todd-2023.png
Sponsors:
- Swan.com (code LIVERA)
- CoinKite.com (code LIVERA)
- Mempool.space
Stephan Livera links:
- Follow me on X @stephanlivera
- Subscribe to the podcast
Podcast Transcript:
Stephan (00:02.737)
Okay, Paul and Peter, welcome. We are gonna do the great drive chain debate, so welcome to the show, both of you.
Peter Todd (00:11.15)
Thanks for having me on.
Paul Sztorc (00:12.604)
Thanks for having me.
Stephan (00:14.633)
Great, okay, so we’re just gonna talk through a little bit around the structure. So the topic is drive chains should be implemented in Bitcoin, obviously Paul is taking positive, Peter will be taking negative. We’re gonna start with some timed segments. So it’ll be a 10 minute opening argument from Paul, 10 minute opening argument against from Peter. We’ll have a few rounds of rebuttals, and then we’ll have a segment where you can both challenge each other. I may throw in some questions here and there, but I’m gonna…
I’m going to try to do my best to be impartial and let the two debaters make their points. Listeners, you are big boys and girls. You can make up your own mind. And so that’s the format. If we’re all ready to go, I’m going to start my timer when you start talking, Paul. You’ve got 10 minutes on your clock when you start.
Paul Sztorc (01:04.503)
Okay, great. Thanks. I think, you know, this conversation has gone a long time, so I’ll give a tiny overview, but then we’ll sprint right to the cutting edge because that’s what people really deserve is something that can’t get anywhere else. And we want to, we want to be at that cutting edge. So this, this idea is very old. It says the origin of Blockstream and their fundraising is in this side chain idea. And, um, it
It has enormous potential because it allows, for example, 8 billion people to be onboarded to Bitcoin more or less immediately, and they can obtain Zcash level privacy. And we can have, for example, an EVM chain and we can just anything that an altcoin can do, we can just copy. So we no longer have to fight this kind of like media battle against which coin is better. You just have everything at your disposal and then the user just chooses what they want. And we have a competition for developers. So it’s a very old idea.
And Peter has always been like the chief critic of this idea, which has always fascinated me and long ago he did a, let’s talk Bitcoin podcast on his tree chains idea in which he critiqued the side chain idea. And so, so he’s, he’s always been kind of the chief critic of this idea, which has always fascinated me and I’ve always been trying to figure out what is it about this idea that Peter doesn’t like. And I have to say, to some extent, I still don’t know.
But we hired Peter in the summer to write it down. What is it that he doesn’t like? And then he produced this piece of writing and I have replied to it in this giant image, which is sort of an unconventional way to reply to something. But I go through like line by line and I just reply to the whole thing because I didn’t want anyone to say that I missed something or that something’s more important than somebody. I just reply to the entire thing. And then at the bottom, I have some questions. So I’m just gonna read the questions.
And then that’ll be the end of the opening statement because the questions I think are just, that’s the, that takes us to the cutting edge that I mentioned before. So the question number one is a two-parter. I snuck two in at once. It’s kind of unfair, but it’s that on his two most important points, I think Peter contradicts himself completely. So the first is that can miners defer to someone else’s full node? At first,
Paul Sztorc (03:28.155)
Peter says that the answer is yes, and that the pool can pay this fixed cost and then distribute the cost downstream to a number of different clients who all reuse the same node. And so the cost is shared and divided among them. But then later he says that he has a whole section of his piece where he makes a big deal about claiming that it’s impossible and that everyone needs their own full node. So can people defer to someone else’s full node?
I think the answer is clearly yes, but the point is he says both things. So we should just figure out what he’s trying to say. And then the second part of part one is that Peter gives, he makes a big deal about mining centralization, which is the entire critique against this idea. But he proposes, as far as I could tell, eight different definitions and measurements of this idea, none of which are that consistent with each other. Some of them overlap a little bit, but some have absolutely nothing to do with each other at all.
More than half of those have nothing to do with what Bit 300 does. So there are things about the distribution of heat or distribution of resources on the geography of the planet Earth that have nothing to do whatsoever with Bit 300 activating. So we should just get what is the actual definition of mining decentralization or centralization? When does it go up and down? What are the examples? Question two is his section about how drive chain affects Bitcoin.
which would be the only basis for rejecting it from Bitcoin is that it had a negative effect. He only mentioned miners’ costs. So my question is, is it really limited to that or intentionally or is there something else that’s not in the piece of writing? Or is he tacitly agreeing with me that the regular user who doesn’t mine and who just runs a full node or is a Bitcoin user, they are not negatively affected by BIP 300? He’s only saying that it has some mysterious effect on miner costs. So that’s question two.
Question three, at Baltic Honey Badger in September, Peter said, quote, no one has written code, unquote, to do this, but the software, including a fully functional Zcash sidechain has been on drivechain.info for more than two and a half years. So the download button is right at the top of drivechain.info. So the question is, Peter, have you actually visited the website drivechain.info ever? And if not, isn’t that sort of negligent to be?
Paul Sztorc (05:54.611)
commenting on it after we even hired you to write this critique. You are on stage having not visited the website one time. Question number four, Peter complains that there’s no math at the very end of his post. Makes this complaint. But in the original 2015 article, there is an entire section labeled some math and it contains mathematical arguments. So has Peter not read that or is he not interested in that or does he reject it for some other reason?
Peter complaints almost all the complaint in his writing centers around the terms fixed costs and economies of scale But Peter doesn’t use these phrases the way they’re Correctly because they always are Inherent to one single decision-making context and they are not something that they’re not something that you just have They depend on the time frame and they depend on the current scale and they depend on that sort of a given scenario They’re like choosing between a and b
but there’s not something that you just have. And then question number six, I would like to know if Peter agrees with me that miners long run total costs equal their long run total revenues, which in the field of economics is uncontroversial. So if Peter is complaining about fixed costs rising, which is most of the complaint, then he’s just complaining that Bitcoin is very successful and that the miners are earning a lot of money and transaction fees and that the network is doing really well.
And so shouldn’t we just, isn’t that a illegitimate complaint? Uh, and then a question seven, I would like to know if Peter really doesn’t understand why he complains about this in his piece, why does the sidechain kicks things off by committing to the hash or the full node software that is not that. It doesn’t even, it’s not even that wasteful, all things considered. And it solves an immensely important problem very easily. So, but he complains about it in the piece. I don’t know why.
He says it’s like an amateur mistake or something like that. And then question eight, I’d just like to know if Peter actually believes that if side chains are a good idea, like does he actually see any benefit on the other side of this? Like, does he agree that about the planetary scale and does he agree with of achieving Zcash level privacy? And does he agree that we can just let people use different pieces of software, which if we’re, we just have people competing over which software they run.
Paul Sztorc (08:23.112)
And yet all of these projects are Bitcoin projects that use the same 21 million coins. Like, does he actually believe that has any benefits? Because mostly I would say probably a block stream prototype of a person would have said something like, this would have been amazing benefits if we had achieved this, but we hadn’t achieved it yet. But maybe Peter doesn’t think that there is any point in pursuing this goal.
Nine I’m actually I am curious Peter can he’s free not to answer this at all I mean, it’s free not to answer anything, of course But I would like to know more about I just personally would like to know more about what? What the convert may be conversations Peter had with block stream in 2015 or if Peter just has any idea or any guess? About why it is that because he does talk about this in his piece He talks about why it is block stream has sort of abandoned the original sidechains idea and has kind of retreated to liquid
which is a lot like Ethereum doing these L2s and retreating to just something that’s just a multi-sig or something. This is a big regression in my view, and a very significant event in Bitcoin’s history. So I’m just honestly, I’m very curious to know what he’s up with that. And then that question 10 is, does Peter think that his own arguments are what stalled the side chain idea from moving forward? And does he think that…
As a result of what he said, we have had all these altcoins proliferate and we’ve had all these other things happen indirectly, such as the rise of the exchanges and the rise in activation drama and the toxicity and the paralysis that has plagued our project and I think set us back on the task of taking on all the banks. So I don’t know how much time that was, but I just read my 10 questions. Well I think then I surrender the minute. I don’t think I…
Stephan (10:07.757)
Yeah, you’ve got one minute left, Paul. One minute left.
Okay. Fair enough.
Paul Sztorc (10:15.099)
Just say that that’s just this conversation has been going a long time and that’s just where it left off as far as I’m concerned
Stephan (10:22.613)
Great. OK, well, thanks for that. And Peter, let’s hear your, I guess, you don’t necessarily have to respond to all of Paul’s points here. You can make your own opening argument. And then if you want to have the rebuttal time afterwards, where you guys are going to go back and forth on this. So Peter, I’ll let you take your 10 minutes. I’ll hit Start when you start talking.
Paul Sztorc (10:30.203)
Yeah, whatever.
Peter Todd (10:48.33)
Let’s go. So, you know, I think like the way I’ll go and talk about this is, you know, start at really the big picture, which really comes down to, you know, well, what kind of thing does this really bring to Bitcoin? And I’ll actually start at the very last thing Paul mentioned there, which was the sort of idea of like toxicity and, you know, figuring out how to go and negotiate changes and all this. I mean, in the reality is like,
So much of this toxicity we go see is driven by financial motives. And this is not something that drive chains can fix. I mean, the fact is that people create all coins because they want to go make money. And how do you make money? Well, the easiest way to go do it in the crypto space seems to be creating a new currency, pre-mining that currency, or at least in, you know, issuing it, investing it and hoping the price goes up and then dumping it on other people. Like you just cannot get away from that.
Nothing us Bitcoiners will ever be able to do will go change that other than us maybe like cozying up to the SEC and like giving them tips or something. I mean, you know, it’s just inherent to the nature of cryptocurrency and it’s, you know, it’s just not relevant to Bitcoin. Now what is relevant to Bitcoin is, well, what do we use Bitcoin for? I mean, of course, some people use Bitcoin to go and upload JPEGs and, you know, go trade them. But
Mostly what Bitcoin gets used for is sending and receiving money. And secondary to that, or potentially, you know, potentially on how you look at it, first of that is just holding money. I mean, that’s really what it comes down to. And the applications that have really caught on for that is good ways to send and receive money. And, you know, that’s all well and cool. So then the question is, well, what is a good way to go send and receive money? And
You know, I think there was this idea with Blockstream initially that sidechains would be cool, we’d have all these interesting new features. And, you know, I can speak to that in a couple of ways, but say like the number one thing that happened there was Lightning went and like killed all these ideas because it’s just so incredibly effective. You know, Lightning isn’t perfect, but like the fact that you could just send money around the internet instantly really took the wind out of a lot of sales. You know, that’s just a fact.
Peter Todd (13:11.67)
But then the bigger issue for like Blockstream was, well, what was the original idea? Well, it was a fairly complex proposal of merge mine side chains with some kind of undefined fraud proof, some way of making that miners were not fully in control of the money on a side chain. And they thought they could do this. And the fact is that they failed. And eventually Blockstream realized institutionally as well as like individually the people there that this idea just didn’t work.
And that’s why they abandoned it. And so I gave up and decided to go do liquid instead with functionaries. And, you know, liquid is a compromise between like what people wanted to do and what’s actually possible. But, you know, that’s kind of how that ended up. Well, and then, of course, finally, you got to go and go and talk about, well, why is like, why is this so hard? And what it really comes down to is in Bitcoin, full nodes are the things that prevent money from being stolen.
You know, miners go and create an ordering, miners go and create some kind of consensus, but the most important thing is that the
your full node actually goes and validate signatures. You know, once you have like some kind of proof of work, that’s really what keeps money from getting stolen. And a lot of people would love to go find clever ways to avoid that. You know, drive chains is an example of someone trying to go find a clever way to go avoid that kind of system. But, you know, there’s no getting around the fact that you need, you know, you need people validating signatures in some way to be able to go and pull off that kind of independence.
And then you can then get into the more technical weeds of, well, all right, how exactly does all this fail? But I’ve got to admit, I don’t know I had my timer up, but I think I might be running into 10 minutes. So let me go see now that Stephan’s back.
Stephan (15:11.795)
Okay. Okay, I’m just back now. Apologies for some reason, my internet just dropped out there. I know we have your.
Peter Todd (15:18.734)
Ha ha.
Paul Sztorc (15:18.807)
Oh, I thought you did a, I honestly thought you did like a producer move and like put us side by side. Cause no, we didn’t even notice that you weren’t there. But we do, we do know we were wondering about, Fido was asking about the time.
Stephan (15:26.748)
Okay.
Stephan (15:30.567)
Okay, well, I think you’ve still got another four or five minutes, I think, so if you want to carry on.
Paul Sztorc (15:35.373)
Oh, okay.
Peter Todd (15:35.526)
Okay, yeah, okay, good. So, all right, so I can’t go carry on to some of the more tech stuff. Okay, so, all right, so what is this minor centralization thing we don’t care about? I mean, you know, we can go back up and say, well, why don’t we just increase the block size, right? That’s an obvious thing to go do if you want more transactions per second. And you know, that was really what the block size wars were all about and increasing the block size all well. Cool, but even if you ignore
Paul Sztorc (15:47.047)
Yes.
Peter Todd (16:04.75)
people’s ability to write a full note, even if that problem is magically solved, you’ve run into this issue that the more data miners have to go deal with, the harder it is to be a miner. And funny enough, since we last started talking about this, Ocean is a new mining pool now getting started. Unfortunately, Ocean just faces so many issues trying to get their mining pool off the ground. And…
This is even when running a Bitcoin node is relatively simple these days. You can easily run one with high enough performance and so on. But the problem with things like drive chains really comes down to the fact that they make running mining more difficult. And I was trying to take some notes earlier, Paul went and mentioned fixed costs and overhead and so on. And this is really what it comes down to. It’s like…
When you think of you trying to run a mining pool, you have a certain, you have a budget, right? Some of that budget is gonna go to fixed costs. The things that don’t, that, you know, more ties over how many hashes you have at your mining pool. The higher those fixed costs are, the bigger barrier to entry it is to run a mining pool at all. That’s really as simple as that. And most of the things that I think Paul is gonna talk about with mining costs are things that are not like that.
In fact, mining is inherently decentralized at the hashing level for a lot of reasons. And one of the big ones is that the access to cheap energy is inherently decentralized. Solar power, flare gas, like all these things that are cheap energy, stranded hydro, these are all things that are not in one place in the world. They’re inherently in many different places around the world. So we got very lucky with that. But the mining pool side of thing, everything we add to mining pools to make them more complex
That increases fixed costs, makes it harder to get into the mining business. You know, if you don’t like that, like say, uh, you know, F2 pool, as an example, started censoring OFAC transactions. Well, one of the ways you respond to that is by spinning up a new mining pool and not censoring those transactions. And the harder we make it spin up a new mining pool, the harder it is that we, you know, the, the job that we have to go keep Bitcoin decentralized. That’s really what it comes down to. And drive chains, because they act like a block size increase.
Peter Todd (18:28.626)
inherently they have this problem of increasing the fixed costs for all miners. Of course, they’re even worse than a block size increase because drive chains are different pieces of software. Like at least with a block size increase, I can throw hardware at the problem. You know, I can get a faster computer, I can get a faster internet connection. You know, it’s not a great thing to go have, but at least it tends to be things where, you know, money kind of solves the issue. But when you’re talking about a drive chain,
And now there could be like a hundred different drive chains. We’ll now have to go and hire guys, go run a hundred different pieces of software to try to eco-profit, to make my mining pool competitive with a much bigger one that happens to be censoring transactions. You know, we don’t want Bitcoin to be in that place. We want running new mining pools to be easy. You know, that’s how we are in it. We get ourselves in a position where the response to censorship can be, well, screw you, we’re going to start up our own thing, not well, oh shit, how am I going to get, you know, $10 million to go set this up?
Notably, I mean, Ocean, apparently the initial budget, including software developments, was like a $6 million investment. You know, that’s already kind of worrying. And hopefully that investment will actually get open sourced and it will be easier for the next guy, but we’ll see. You know, last thing we want to do is make this problem even harder. So I think, uh,
Stephan (19:45.427)
Yeah, one minute left, Pedro.
Peter Todd (19:49.696)
I think that kind of sums up my opening argument, if you will. Like, that’s a, I’d say, that’s an important thing.
Stephan (20:00.371)
Okay, so then that’s the opening arguments from each side. So let’s now have, we’re gonna do five minutes of rebuttals and we’ll start with you, Paul.
Paul Sztorc (20:13.311)
Okay, sure. So he started by saying like, that drive chain wouldn’t necessarily stop people from issuing their own altcoin. And I agree with that. Although I think it’s a strong deterrent, because it gives you it makes it so that the altcoin person isn’t adding anything except a their own investment, they’re not adding any new ideas. So you see what I mean? Mostly what an altcoin or will say is, oh, I would have wished I would like to do this on BTC, but I couldn’t because of such and such. So that’s what for example,
Vitalik said that is what whatever Roger Vera all these people David Vork at SIA name coin people sense all the people with a good The Manero people often in writing they say so but this is a very minor point Really the ten questions were in order from sort of most important to least important. So the mining centralization one Peter did repeat that he was saying that sometimes it’s energy costs distributed around the globe and then sometimes it’s the fixed costs
The question of whether or not drive chain significantly increases fixed costs or makes it more difficult to start a pool, I think maybe we should just focus on that, because I’m absolutely certain that the answer is no. And in fact, it has a negative effect. So for example, the pool operator, just as the individual clients would defer to the pool’s node, the pool can defer to someone else’s node. And that is exactly what BIP-301 does. And even without BIP-301,
If through one just makes this is like a trustless relationship Even if there was bit no bit 301 they would do that anyway just for reasons of economics We’re more or less the exact reasons that Peter has outlined just in reverse Which is that no one wants to pay a cost that they don’t have to pay Again Peter repeated this thing about the fixed costs where he’s right that a fixed cost is something that you can amortize over people but
To some extent, like an ASIC, it depends on the context and the timeframe. So an individual ASIC is a fixed cost in a certain time, on a timeline that most people would just assume that Peter and I are talking about. And then on a long enough timeline, all of the fixed costs become variable costs. So if you just take a huge view and you say you’re the dictator of the continent and you’re gonna rule for 200 years, that whether or not you build eight hydroelectric dams or 10.
Paul Sztorc (22:31.483)
That’s a variable cost to you. You just think how much is it going to cost to build each dam? How much electricity are we going to get? But of course, when you zoom in, the hydroelectric dam is almost everyone would say a fixed cost. So it depends on the context. But even in any context, the whole idea that the fixed cost of starting a new pool would be significantly increased by drive chain is incorrect because the pool doesn’t need to run a side chain node at all. So.
I don’t know if we’re ever going to walk it. This is like the most important contradiction of all. And so I said it first. The fact that Ocean is having trouble competing, I think Peter and I both know there was a bunch of other reasons, uh, tacked on with that, but even if Peter were right, it would just say, demonstrate that it’s very difficult to start a new pool if none of the other pools are doing anything wrong and that, um, despite not having drive chain is still difficult. And so.
drive chain doesn’t have any marginal effect on this type of thing. But a pool is proportional. You know what I mean? Like the pools add up to a hundred percent. So a pool is a special kind of thing where you won’t automatically be able to judge. If a 100% of the pools that currently exist are behaving optimally, then there’s no reason for you to be able to break in. You know? So, um, I don’t know. I think those are the important things. Like,
I did ask about the history. I’m very curious about exactly what Blockstream decided. I mean, it is really as true as Peter says that people just thought Lightning was a lot better than I think with Anthony Riard and other people, a lot of people, Rene Picard, these people, author of Mastering Lightning, they are now trying to kind of walk a lot of things back. So I suppose that’s kind of good news for me as I think Lightning, the sun is starting to set on it.
Stephan (24:23.895)
You got one minute left on this round, Paul.
Paul Sztorc (24:27.095)
Well, I think I just would like to emphasize that the questions I didn’t I should have mentioned this, but I didn’t. But the questions are sort of in order from how important they are. So the question of what is mining centralization and do people need to run? Can you defer to someone else’s full node? The answer is certainly yes, that you can defer to someone else’s full node. And that is why the effect on mining pool costs is basically zero, because they can just use a side chain node that someone else, someone out there is running.
And the reason why it may actually improve mining decentralization is that unlike Lightning and unlike Ocean, you can just pay people out. The transaction that nets out all the hashers pay per share every 10 minutes, that is just a normal transaction on a BIF300L2. It’s no different than buying a BIF300L2.
cheese or whatever. So you just pay them out every 10 minutes and that is what Ocean aspires to do but that is what we could do with Bit 300 tomorrow if we had.
Stephan (25:30.991)
Okay, great. So that’s the first round of, well, first rebuttal from Paul. And now Peter, you’ve got five minutes to have a round of rebuttals.
Peter Todd (25:42.09)
Yeah, I mean, like, if you’re going to go for most important, I mean, the simple fact is your claims both, you know, bit 301 and bit 300 are just wrong. Like, I mean, it’s, you know, and I, if I’m going to go and speak to an audience, I mean, I think the thing I need to go speak to is really what why and, you know, bit 301 and bit 300, they really don’t have very much to do with each other.
BIP-301 is sort of this idea of, well, somehow we’ll kind of go do blind merge mining and we’ll go pay, you know, to go create a consensus on a side chain and so on. And you know, it’s a way to think about it. It’s kind of like an auction per block per chain. And you go pay money and you go get a hash and that’s considered to be consensus. But none of that has anything to do with the money on the chain. You know, that’s what BIP-300 is about.
You know, I personally made this mistake when I read about side chains. You know, for the longest time, I just assumed, oh, yeah, obviously the way that we decide how this pot of money assigned to the chain gets split up and withdrawn and so on. Obviously, you go and have that related to this bit 301 idea of blind merge mining. But the reality is it purely is miners are in control. You know, that is the drive chains proposal. Miners are in control. You have to have a majority of miners to do anything to money on a chain. And the only way of majority of miners
And really the miners themselves can actually validate what’s happening, you just go to run and drive chain nodes. And if they don’t validate what’s happening, they can be participants in theft, which poses enormous problems. And equally, if they want to be malicious, well, they can just go take the money. And that’s that. Like there’s just no getting around that problem. And I think like where this kind of comes from is people have this dream that surely we can create some easy way to go and, you know,
Some have new consensus rules without having new consensus rules, but people just haven’t figured this stuff out. Now, obviously, I think with MoonMath, as we like to go call it with, you know, recursive zk-SNARKS and all this other clever stuff. Yeah, yeah, we can absolutely do things like drive chains that actually work properly and go and have this kind of a trawl, but that’s not what the drive chains proposal is. It just takes the shortcut of saying, let’s just assume miners are trustworthy and honest. And the honest, trustworthy miners, the only thing they can do
Peter Todd (28:01.13)
which case money just gets frozen, or they run a drive chain node, which creates all these fixed cost problems. I mean, you know, and like talking about like dams and all this, I mean, the definition of fixed costs is simple. It is a cost that is not related to how much hashing power you have. Like this is a pretty simple thing. The node that I could run for mining pool, that’s a fixed cost, because that’s unrelated to how many hashers you have.
Like it’s a, it’s a pretty simple thing talking about like Hydra dams, all that. That’s just a bunch of nonsense. I mean, you know, talk about like an ASIC. I mean, an ASIC is obviously a variable cost. An ASIC produces certain amount of hashing power and the cost is proportional to the hashing power. Like, I don’t know why we’re arguing about the silly stuff. Like these are very basic economic terms and you know, I would hope that we would be in agreement on this. So we could move on to a substantial thing to go talk about. So.
Stephan (28:59.763)
Okay, you’ve still got another one minute 45 on this if you have anything else you want to get to here.
Peter Todd (29:05.226)
I mean, I guess the one thing I could quickly say is like, you know, in terms of all this stuff, but, you know, altcoins having new ideas and so on. I mean, again, the biggest, you know, altcoin drama that’s happening right now is about putting pictures like low resolution pictures on Bitcoin. I mean, you know, there are altcoins out there that have interesting ideas, but they are very, very rare. I mean, the vast majority of the stuff is just marketing and that’s just the nature of it. I mean, it’s a scammy thing.
You want to go make money, you scam people into buying into an idea that doesn’t quite work, but it’s different from the other idea. That’s just the nature of altcoins. If a couple of altcoins exist like Monero where they actually have an interesting idea in a different model, that’s fine. It’s probably fine if they aren’t actually the same monetary base as Bitcoin. After all, Monero involves different trade-offs. Monero is fundamentally more vulnerable to losing everything due to crypto mistakes than Bitcoin.
And it’s fine if those people choose to have their own currency to go do that. I mean, I don’t, I’m not, I’m okay with that. Like if anything, there’s arguments that this actually good thing, that they’re firewall off Bitcoin and that their failures don’t affect us and vice versa. But certainly the vast majority of the stuff is just scammy nonsense and, you know, drive chains aren’t going to change that.
Stephan (30:26.863)
Okay. All right. Well, that’s sort of concluding the more timed, more formally timed segments. And now we’re going to have more of a, let’s call it gentlemanly back and forth. But let’s start with Paul. And Paul, you can socratically question or challenge Peter on something, perhaps one of the questions that you maybe you believe it was not answered or was not answered. Well, perhaps you want to ask a question now and challenge Peter.
Paul Sztorc (30:56.283)
Yes, so if someone wants to profit from running the side is their minor and they want to profit from Take advantage of the revenues generated by the sidechain They don’t need to run a node. They can just get information from someone else’s node and the fact that nodes Again, it’s the exact same time contradiction as always which is you it can’t be a fixed cost that is split among multiple people unless it’s possible for
one full node to be used by more than one person if it were the case that I have to run a full node and then if someone else if a friend of mine walks into the room If they don’t run their own full node, then they just like burst into flames or something Then it would be the case that running the full nodes is mandatory But it’s because it’s split you really are you’re determined to have it both ways You’re determined to say this is a fixed cost that uh is split among multiple people And then you also have to say everyone has to run the
the full node. So I don’t know how to resolve this contradiction. If I, but I would say if someone is, if a client of the pool is deferring to the pool’s full node, then what happens if a pool just decides to defer to a different pool’s full node?
Peter Todd (32:10.799)
I mean, this is the basis of the block size debate. I mean, that’s really what it comes down to. The block size debate was about saying, hey, we should go set block size low enough that running a full node is not a significant overhead for anyone. You know, that is really what it comes down to. And the, well, but what you, but what, but.
Paul Sztorc (32:25.847)
I agree with that completely. Right, but that’s not what this is. You understand that the block size debate was about was that different people wanted to pay different costs. They wanted the full, Roger Veer really wanted the full note to cost more. I mean, I think you understand that perfectly.
Peter Todd (32:36.054)
Look, what?
Peter Todd (32:40.598)
Look, what you want to do with drive chains, especially the scaling argument, which is really like how drive chains competes with Lightning is to go have big blocks. I mean, that is the reality of that. And this is how drive chains compete with Lightning. Well, look, as long as people go and actually go validate rather than just blindly trusting other people to run the system, you have to go and run a full node. Like there’s no shortcuts to this.
Paul Sztorc (32:53.723)
But, uh, not for everyone.
Right, yes.
Peter Todd (33:07.67)
And when people stop doing that, things fail in very bad ways. You know, you can’t get around that. Well, the great example being if you’re trusting someone else to go run a full node, those people are in control of the money on the drive chain because they can go lie and they can go approve withdrawals or they can be pressured to go approve withdrawals and money gets taken. I mean, like in Bitcoin, we try to keep, in Bitcoin, we always try to keep so far away from this by keeping the cost go run fully validating nodes low.
Paul Sztorc (33:11.975)
Okay, why don’t you give me an example.
Paul Sztorc (33:27.751)
Yeah, but you understand that the… You understand that this…
Paul Sztorc (33:37.539)
I fully agree with that. We want every node to be as cheap as possible, but some people, not everyone agrees on how much it should cost, which is the whole point, is how to make it cost more expensive for some and less expensive for others, and thus keep everyone in the same.
Peter Todd (33:48.642)
Look, if you don’t agree, I mean, you’re welcome to go and, you know, sell Bitcoin and go buy a different currency. Like, I think that’s a very good. Well, look, we, it, it would, it would be lovely if we could solve that problem, but the technology doesn’t exist. I mean, you know, my, my tree chains hang on.
Paul Sztorc (33:54.267)
But that’s exactly the problem that sidechains are supposed to solve, though, is that someone who disagrees doesn’t have to do that. I mean, Roger really did not want to create Bitcoin cash.
Yes. But Peter, you’re the only one out of everyone I could find. I’ve searched high and low for the reason why this technology can’t exist. And I think everyone is deferring to you. And all you have is this contradiction on whether or not you can defer to someone else’s phone number. Yeah.
Peter Todd (34:18.406)
People are deferring to me. I’m just some dude who has the patience to go sit through the stuff. I mean, the vast majority of devs have work to do. Yeah, because I am the original guy who went and had the patience to sit down and write arguments about this. The majority of devs are not interested in going on social media or writing blog posts explaining to the general audience. The majority of devs want to actually get real work done. I specialize in this kind of communication.
Paul Sztorc (34:23.279)
Well, you say in your post that you’re the original.
Paul Sztorc (34:31.399)
Yes.
Paul Sztorc (34:34.747)
Yes.
Paul Sztorc (34:43.727)
I know, but Peter, what is the actual reason? Well, you know that, you know, we have a list, layer2labs.com slash friends, and it’s full of a lot of very technical people, including Renee Picard, author of Mastering Lightning, and Robin Linus, who invented BitVM, Adam Back, who are big supporters.
Peter Todd (34:59.766)
Your quotes are very optimistic how much support you actually have. I’ve read that page. It’s optimistic.
Paul Sztorc (35:05.363)
Yeah. But I’m just saying though that we have a big list. Yes. Well, you know, I’m just saying that a lot of people seem to think, yeah. Yeah.
Stephan (35:06.155)
Yeah. Well, so guys, let’s, okay. So, okay guys, let me just jump in here. So guys, let me just jump in here. I think this is, we’re sort of veering off into who supports Rivechain, who doesn’t. Let’s try to keep this more to this. Let’s keep this to the specific point, which.
Peter Todd (35:15.698)
Yeah, yeah, it’s not a good argument. Yeah.
Paul Sztorc (35:19.771)
But he is the one who of course said though, he says like, I’m trying to ask him about why, why doesn’t he like this idea and he’s saying, well, you know, I’m the only one who’s bringing this up or something. He’s the one who’s taking us in this direction.
Peter Todd (35:30.282)
No, I didn’t say that I’m the only one who says he’s against Drive Chains. I’m saying I’m the only one who has the patience to go do things, like sit through 35 minute long interviews about this silly topic. The fact is, so many devs are just tired of this subject and don’t want to have anything to do with it because it’s dumb. That’s what you’re up against. Well, I was explaining it.
Paul Sztorc (35:43.989)
Yes.
Paul Sztorc (35:49.547)
Yeah. But Peter, why is it dumb though? Because you…
Stephan (35:50.287)
So look guys, let me try to, let me try to, so Paul and Peter, guys, let me just jump in here. I think what we’re trying to get to the bottom of on this particular question is do miners have to run, so as the concept of drive chains, there are 256 drive chain slots, and there would be Z coin and some of these other coins. And I think the criticism, as I’m hearing, as I’m understanding from Peter and others is that
you would create this additional, let’s say burden or fixed cost as we’ve been talking about. And I think Paul’s answer to this point is, no, the miners do not have to run these side chain or side drive chain nodes. They can delegate that capability. And as I understand, Peter, your counter argument here is more like, no, part of the whole point of Bitcoin is that you need to be able to validate everything because otherwise the pools…
can lie to you. Is that sort of a fair?
Peter Todd (36:52.363)
I think you’re missing my point a little bit. So what it comes down to is that if drive chains are to work properly, you need miners to be running drive chain nodes and properly validate it. Because otherwise you don’t actually have separate miners. What you really have is a very small number, potentially just one dude, you know, running nodes that actually go do stuff. And then the drive chains start working properly. So…
You don’t like I think thing is here is Paul’s trying to get it both ways. He’s trying to go say, well, you go have this thing that works properly, but you don’t actually need to do the hard work of validating anything. That’s okay. You know, you can just trust the other guy to go do it. And.
Paul Sztorc (37:21.467)
Yes.
Paul Sztorc (37:27.923)
Yeah. But Peter, I do, I have it both ways. You just don’t, for some reason you just don’t understand why. Yes, because the people, because Roger Veer wanted, he wanted to run a more expensive node, but Luke Dash Jr. did not. So the question is just, how can we exploit the fact that Roger was actually willing to run the more expensive node and thus only have them synchronize very rarely.
Peter Todd (37:32.182)
Well, you don’t have it both ways. Like…
Stephan (37:35.453)
Okay, so Paul, let’s hear from you.
Paul Sztorc (37:52.303)
with this tiny hash once every three months, thus making it very unbelievably impractical for anyone to lie, because everyone who’s running the sidechain node is gonna report the same hash for three months. So it’s gonna be extremely impractical for anyone to get away with the lie. The slow synchronization, which would be way too slow for anyone to actually make use of as an actual feature, but since it’s synchronizing an entire three months worth of blockchain history, no matter what the sidechain’s block size is.
That is what would allow anyone to have whatever Bitcoin SV node or Zcash node or Monero node or whatever.
Peter Todd (38:22.314)
Well, I mean, for starters, that’s for starters, I mean, that’s not even what you’re what the thing is. Like you have to go put in these hashes every block for anything to go happen, you know, like that’s, and you know, of course you’re in, and also hang on, hang on, hang on. And also you remember, we’re not talking about one side chain node, we’re talking about hundreds of them potentially, you know, like you’ve allowed for that. So hundreds, even in your hypothetical example, still a lot of work.
Paul Sztorc (38:34.56)
The withdrawal hash, this is again the difference between QNR 2.1.
Paul Sztorc (38:44.912)
Yes.
Peter Todd (38:51.074)
constantly that miners have to be doing to be updating this. That’s just not feasible in a decentralized environment.
Paul Sztorc (38:54.555)
But for each one, it is the for each one, people have to make a decision on whether or not they want to try to lie. And for each one, it takes three to six months worth of lying. It’s too impractical. It’s too much of a deterrent. You know what I mean? Like
Peter Todd (39:04.758)
Look.
Peter Todd (39:08.67)
You, you, there is no deterrent here. Like there isn’t a deterrent. There is nothing in drive chains that, there’s nothing in drive chains that stops this. Like.
Paul Sztorc (39:13.019)
Yes, how can anyone get away? How would imagine we play the game? Well, why do you say that though? Because there is, of course, the fee, the fee revenue to the miners is enormous. So they have an insummon center to get it right. Yeah. And then they will, it will be like a hundred billion dollars a year. It’d be enormous. But I admit though, that this is the
Peter Todd (39:24.714)
No, it’s not!
Peter Todd (39:28.574)
Look, free revenue in drive chains is not enormous because there is no block size limit.
Paul Sztorc (39:34.555)
This is the security model of drive chain, though. If the fee revenue is not low, then the chain will not be secure or viable. So this is fully admitted by me, 100%. But the good news is that every reason to suspect the fee revenue will be 10,000 times higher than what is currently available on PTC. Yeah.
Peter Todd (39:41.154)
The fee.
The… well…
Peter Todd (39:50.178)
I mean, it’s clearly not because the fee revenue of drive chains doesn’t have a block size limit. And that just like Bo’s arguments with Bitcoin itself, there will obviously be a race to the bottom and there’s not going to be fee revenue. I mean, it’s like there’s no getting around that.
Paul Sztorc (39:58.556)
Yes.
Paul Sztorc (40:02.607)
Yeah, but Peter, do you actually, I thought this, but Peter, you don’t understand, it’s a basic math question of height versus area. Do you really not know the difference between price and revenue?
Peter Todd (40:12.586)
Look, I am well aware of all these arguments, but the fact is that in drive chains, there isn’t a block size limit, so that there is no incentive for there to be any fees at all. You can always undercut the next guy. Like that’s just the fact of how drive chains go to work.
Paul Sztorc (40:15.123)
Apparently not, though.
Paul Sztorc (40:19.579)
But Peter…
Paul Sztorc (40:27.867)
Each, this is not true, each individual chain has its own limit. So it’s no different than, it’s no different between Ethereum and BTC today. So how is it that Ethereum can pay…
Peter Todd (40:33.871)
No, there is nothing in drive chains that actually goes and makes it happen. I mean, I can undercut different…
Paul Sztorc (40:42.711)
If, if it’s here, yes, of course, Peter, there is because the designer wants to, their side chain to be popular. So they want their side chain full node to be a reasonable cost. I see every, every single argument that applies to Ethereum today. That makes it have 7 million. Right. And we guess that’s exactly why we have something like a Ethereum and. Yeah, but, but Peter, no, I was, we have to, we have to emphasize this point though, about the.
Peter Todd (40:53.12)
Well, if you want your drive chain to be popular, you go have low fees, which pushes the race to the bottom. You can’t get away from that.
Stephan (41:04.835)
So just let me just ask a question so I can understand something here as well. Okay, Paul, make your point and then I’ll go.
Paul Sztorc (41:10.683)
Peter is saying race to the bottom. He’s referring to the fee rate, the dollar, like a dollar per apple or something. And this doesn’t matter if you sell a trillion apples to Peter, but this is a basic elementary school math that he is, yeah, that’s what I’m saying. It does matter, Peter. The revenue is the total area of the rectangle. This is a very simple point. So do you understand that the race to the bottom of the fees can be what maximizes the total? The fee, the price can go down and the revenue can go up, which is exactly what we’re having.
Peter Todd (41:19.551)
It certainly does.
Peter Todd (41:28.831)
Look.
Peter Todd (41:36.926)
And, but look, if you, if you’re going to have a large amount in total, you need to go have some mechanism where fees do not end up at zero. Like you don’t have that in drive chains.
Paul Sztorc (41:46.255)
But why aren’t these zero? Why aren’t these zero on something like Tron or whatever? Ethereum, anything, any other alt network? They’re not zero.
Peter Todd (41:53.762)
because Tron is centralized systems so that the people who run it can just decide what the fee will be to set the revenue.
Paul Sztorc (41:58.183)
It doesn’t make any difference. The sidechain can be fully centralized or decentralized. It can have an expensive node or an instance. It makes no difference. Absolutely no difference to my argument at all.
Peter Todd (42:05.494)
Look, Tron is able to have non-zero fees because they can go pick an amount. And that because Tron is something that people have to go buy into rather than competing like drive chains would with each other, they’re able to go sell that. Now, a better example is actually, hang on, a better example is actually something like Ethereum where that was the idea. But because Ethereum has technical limits, because it does actually have a block size limit,
Paul Sztorc (42:09.453)
is saying it’s not zero.
Paul Sztorc (42:21.011)
But they all come in compete with each other to some extent.
Peter Todd (42:33.77)
It’s still wound up with non-zero fees. But that’s the same reason why Ethereum doesn’t go work properly. Like drive chains can’t get away from that.
Paul Sztorc (42:36.919)
Of course, yes. The side chain of Ethereum.
Paul Sztorc (42:45.345)
A side chain of Ethereum will also have its own block size limit that will just be similar to what Ethereum’s is. And then when that one fills up, someone will just create a second one that has a different block size limit and then maybe two. The fee rates will be going down all the time, yes, but the total revenue will be going up.
Peter Todd (42:51.262)
Yeah, and well, which, hang on, which is the process by which these, that’s super, I mean, it’s just such a bizarre argument to go try to go make. Like, yeah, it’s like, oh yeah, I have something where there is no cost to it, but I’ll go make it up in volume. Like, you can’t.
Paul Sztorc (43:06.543)
It’s not bizarre at all. You know perfectly well that many people have a small margin of business.
Paul Sztorc (43:16.183)
Exactly, yes But Peter the argument is literally that It’s one chain versus one plus N So it must go up But I’m already admitting to you that if the if the revenue isn’t there then the chain won’t be secure and then this idea Will have been a failure
Peter Todd (43:31.722)
But the problem is the chain can easily be sufficiently secure to go and screw over Bitcoin I mean remember like all this stuff ends up competing with Lightning in you know in a lot of the visions of the drive chain for Proponents in lightning at least has hang on. Hang on in lightning at least has a Easy to understand fee model which does at some point go back to miners now of course, I argue that on top of all this we should have a built-in suspenders option with
Paul Sztorc (43:45.459)
Well, unfortunately for me, yes. I wish that weren’t the case.
Peter Todd (44:00.382)
some kind of security tax, like either influence with deflation or inflation. Either way, it doesn’t really matter. But this idea that drive chains will just bring magical fever of used to miners just doesn’t work. I’m sorry, but it just doesn’t. And you have the overhead problem.
Stephan (44:15.639)
Okay.
Paul Sztorc (44:17.715)
Well, Peter, you know, you go to like crypto fees that info or some sites. This just this is just the reality of what is happening every day. And you can just you can just see here’s a bunch of projects that pay these fees. Maybe the data is not reliable, maybe whatever. But it’s just, you know, it’s certainly for many, many years, Ethereum has been collecting like 10 million dollars a day in fees. And I think that the Ethereum version is more unpopular than would be a unified thing.
where we have one coin behind all these different blockchains. I think people would like that even more. They would, they would say, I don’t, you know, I’m a lot of people I think are waiting on the sidelines saying, I don’t know. You know, I give you talk to a normal person at Thanksgiving or something. They have no idea. I had a cousin who had no idea. You guys will all think this is funny. I brought a Bitcoin. And then he asked about Bitcoin cash. And then he was clear that he thought that they were the same thing. And he was like, Oh, I didn’t even know that. And then he’s like, isn’t that
Peter Todd (45:12.534)
Yeah, that’s called fraud.
Paul Sztorc (45:13.279)
It’s like I said that fraud or something. Well, that’s what he that’s what he said. And he is actually a lawyer so but he’s a smart guy and So this is what the layperson is very confused by all this and I think that there would actually be more popular if we just Had our completely the act together 100% and we said no Bitcoin can literally do any software operation that anyone wants
Peter Todd (45:16.67)
It is fraud.
Peter Todd (45:30.658)
Look, and that is a total pie in the sky because you cannot make those people go away. You will always go have that problem without government interference. Like that’s just the reality of this.
Stephan (45:32.022)
Okay.
Stephan (45:36.928)
Okay.
Paul Sztorc (45:38.724)
Sure, sure, sure.
Stephan (45:40.887)
All right, Jen, so let’s stop that segment there. Now, I think let’s count that as Paul’s chance of questioning Peter. Peter, now you have a chance to challenge Paul. You can ask a question and use a Socratic approach or approach to challenge him there.
Paul Sztorc (45:47.431)
Sure.
Peter Todd (45:59.35)
I mean, I’ll be honest. I look at this and think like, oh, I think we’ve covered all, like drive chain’s not a complex thing. We’ve covered all of the interesting points there. You know, like there really isn’t much to drive chains. It is two BIPs that are extremely simple. And the simplicity, unfortunately, is in such a way that doesn’t work. Like…
Paul Sztorc (46:14.291)
Okay, I’ve got more than… Yes. They are simple, I’m- what?
Peter Todd (46:25.386)
I mean, Bitcoin happens to be something that’s extremely simple at the core, but fortunately it works. Drive chains went a little bit too simple and it doesn’t work.
Paul Sztorc (46:25.862)
I think.
Paul Sztorc (46:31.301)
Yes.
bit too simple and it doesn’t work. So, okay, good.
Stephan (46:35.731)
Okay, so let me ask one or two questions just for the sake of the listeners as well, so that people can, you know, because they may have these questions as well. As an example, one of the, to some extent, you gentlemen, you both touched on this idea, this question of if miners were to try and steal, now Paul, I believe your argument is that they would not have an incentive to do so unless the chain was insecure. But I suppose this is one area that I’ve been curious about as well,
seen it be said that, okay, imagine if the miners were trying to steal. And the response I’ve seen is this idea that somehow the network would try to do a soft fork to try to stop that miner or that group of miners from stealing. Would that not create a lot of soft fork, you know, cacophony and craziness of people saying, well, I need to run this node or that node because I want to reject this particular, you know, stealing attempt?
And maybe there’ll be, you know, in the future that could get more complicated. I’m curious, how are you both seeing that? Is that a, you know, is that a problem for this drive chains approach? Would that create more politics and drama where currently it’s not there?
Paul Sztorc (47:45.424)
Yeah.
Paul Sztorc (47:50.415)
I’m glad you asked that because this is unfortunately a misunderstanding where it is another case where we actually do have it both ways, but people just kind of don’t really see why. But what I mean by that is the security model of drive chain does not rely on people from L1 taking some sympathetic action, UASF or something like that. If it did…
Just as a kind of a side note, if it did, then notice that those people could already, the sympathetic L1 people, they could already UASF to force drive chain into existence and then force the specific drive chain into. So you see what I mean? If you had this group of people who are like willing to take extraordinary action, then it would already.
Stephan (48:30.011)
Right. But I guess the point would be it’s very hard to coordinate that kind of action, right?
Paul Sztorc (48:33.439)
Indeed, right. And that is so the model of drive chain is that the fees will outweigh the The what can be removed? And so for example, let me just give you a tiny example and all the math in the original november 2015 post That peter said didn’t exist or didn’t enjoy or whatever it is. Whatever the case may be That was all that math is about this question of what will the miners decide to take the funds?
And just on a completely separate tangent, the miners can basically steal from anything. If you’d hold liquid funds hostage or UTXO sausage, or even the Lightning Network hostage, and I’m putting people have admitted that that’s the case, but that’s sort of a tangent. The math, like, so for example, if you took like the, what, what ETH is earning each day, which is like, it depends on the day, but I did one example recently with the $7.2 million per day math. You can go to crypto fees that influence, see if it’s a good day or bad day today.
or whenever you visit the site. But that is $216 million a month. And if you just discount that with the NPV, net present value, then it depends on kind of like how many months do you wanna go out before you just truncate and say that the life of a tech project ends after 18 months or whatever. But that’s like $4 billion. So that’s like, you know, tens of thousands or a hundred thousand BTC depends on the price any given day, but.
that the point is you could fit like 50,000 BTC in something like that. And that’s just what Ethereum is currently doing today. And if you do that, then they, the miners will kill the goose that lays the golden eggs by stealing from the chain. And that is exactly why they won’t have any motivation to do it. And this is actually far superior security model than what people are trying to do is get some kind of technical thing where it’s impossible for the miners to steal. But that always means if there’s ever some kind of bug,
or if there’s any some kind of glitch or misunderstanding or whatever, some kind of like Psyop campaign or something. If there’s ever some way, then it means the miners are always waiting to steal. Whereas if the miners have an incentive not to steal and they just want to grow the network and what they’re really thinking about is how can we get more people paying a Bitcoin transaction fee and grow the size of the Bitcoin economy and the network, that’s always better than to have them not want to steal.
Paul Sztorc (50:55.675)
Now, of course, the steel, the reason why it’s threaded is many reasons why it’s threaded into this one withdrawal every three to six months, which is unbearably slow. But that’s in part because it makes it impossible for anyone to lie about what the withdrawal is supposed to be, since it’s just this one hash. And so it’s just anyone who would try to lie is just, you know, not going to succeed and they will only embarrass themselves. And but another reason is that the mining, who the composition of
of hashing and which hash belongs to which pool, those are things that can change enormously in three months. They could even react, you know, the pool operator may be doing something that the individual hashes don’t know about or don’t want to have happen. And so three months is more than enough time for them to stab the pool in the back. So this is what I was saying before about the pool being proportional.
Peter Todd (51:45.658)
I mean, I think you’re kind of getting a little far from like where Stephan started here and I think first of all, with regard to the UASF stuff, I mean, I think, you know, the simple answer what Stephan said is absolutely, yeah, this can create enormous amounts of very, very disruptive drama. You know, we did, well, at least arguably we did a UASF to get SIGWIT activated, and that was very, very risky.
We could have, in the Bitcoin community, we could have easily been in a position where the UASF had failed, but not clearly failed, or had partially succeeded, but not clearly succeeded. And all hell would have broken loose because you would have different parts of the Bitcoin network out of consensus with other parts, without a clear way to go deal with that. If we’d been in a position where some big exchanges had decided, you know what, we do want SACWIT to happen, some miners had gone along with that and created SACWIT blocks and some other miners hadn’t.
It would have been a real disaster for Bitcoin. And, you know, I could see the arguments that for sacred, this was worth it. I would make the arguments that was sacred would actually happen to us. People threatened that scenario, said, hey, if you don’t go along with this, miners, this may actually go happen. You would be much better off just activating Segwit. You know, I think that’s actually the more accurate interpretation of what happened there. But the bigger picture there is that’s dangerous as hell.
And you really want to be doing this for a very, very good reason. Now, inviting that kind of failure to happen more often just because some, so I hang on, hang on a second. Hang on a second. So inviting that failure to hang on a second. So inviting that failure to happen more often just because some drive chain potentially failed is really ugly. And one of the ugly things drive chains does is because miners are now in control of money on the drive chain. It is actually feasible to do things.
Paul Sztorc (53:17.147)
Yeah, but I don’t invite that failure at all. I’m just saying I don’t invite it though.
Peter Todd (53:37.226)
Like give them court orders and say, hey, you know, you damn well better go comply and take this money and reassign it to someone else. You know, of course, I personally am the target of a lawsuit from Craig Wright trying to go do basically that and redefine the Bitcoin protocol to go give him a couple billion dollars that’s well, he’s never proven that he actually owns. And right now it’s very, hang on a second, hang on a second, hang on a second, hang on a second and hang on a second.
Paul Sztorc (53:57.895)
But Peter, doesn’t it prove that we should ignore frivolous lawsuits and follow our own moral compass?
Peter Todd (54:04.39)
And right now it’s very easy to go and say, look, this lawsuit is nonsense because I personally do not have the ability to go do this. You know, from miners point of view, it’s very easy for them to go say, Hey, we do not have the ability to go do this. Drive chains fundamentally changes that drive chains means that miners can go seize money. They can go move it to other people. And that’s a very sketchy thing to be involved in.
Paul Sztorc (54:26.591)
Well, who exactly would be sued though? Foundry or like every miner, you know? And then Foundry would have to say, no, right. Peter, you know perfectly well that that’s a good example because of how bad it is though. Because think about it, like think about what would really happen in that scenario. Foundry would just be able to say, well, we think we’ll probably lose all of our clients if we do this. And even if they were going to keep the clients, they could say,
Peter Todd (54:33.35)
Oh, it’d be very easy to sue Foundry and Edpool and give them a court order to go do that. And the problem is that, hang on.
Well, but hang on, the pro-
Peter Todd (54:50.634)
Nope, I don’t think that would happen at all.
Paul Sztorc (54:55.023)
We gain this amount of Bitcoin by moving. Think about what it means, Peter. If I’m saying that, could you let me finish my point, which is that, because this is essential to the math, this is essential to the security model of drive chain. So I’m saying that the miners earn a certain amount of money each period, such that the net present value is like $100 billion. And maybe there’s $20 billion in the chain.
Peter Todd (54:58.098)
Look, this is not, with drive chains, it is not a disaster for Bitcoin. It’s just one drive chain failing.
Peter Todd (55:09.694)
No, it’s hot!
Peter Todd (55:21.702)
It, a hundred billion is nuts, come on.
Paul Sztorc (55:22.323)
The court order is going to be but think about what it means be it just pick anything pick say it’s a hundred dollars Then and twenty dollars total there’s twenty dollars total in the drive chain and then it’s worth a hundred dollars It’s worth, you know Like a dollar a day and that comes out to a net present value of a hundred dollars or whatever This is made up numbers. The point is the miners are getting a certain amount of money. It’s a magic money tree This is a goose that lays the golden egg. That’s worth a hundred dollars to them The court order says you have to take that twenty dollars out and give it to some other guy
then the miner can just say, well, you’re asking me to just delete $100 for no reason. And I don’t know if we want to comply with that.
Peter Todd (55:54.486)
Look, you are giving a very lovely, everything went exactly the way you want the scenario. The much more likely scenario, hang on, look, no, that wasn’t my example. No, no, the example, I would go, listen, can I go talk? Are you going to let me talk? Thank you. So the example that’s actually much more likely is that if drive chains activate, they don’t really catch on. You know, you get a bunch of people interested in some drive chains. They put a bunch of money in.
Paul Sztorc (56:00.995)
I’m taking your own example. You set this example.
I’m telling you that this drive chain only applies for the chains where the keys are. True. Yes.
Peter Todd (56:24.254)
something more like liquid, wait a second, something more like liquid where there’s some transactions. People basically, because well, it’s kind of sketchy to go and steal money, don’t steal the money even though there really isn’t that much incentive not to. And some court order comes along and says, oh yeah, we got to go and change this. And something like Ant Pool and Foundry are probably going to say, well, you know, we can go to jail for failing to comply. We can comply because we’ve changed the Bitcoin protocol to allow this to happen in the first place.
Paul Sztorc (56:24.452)
Maybe not.
Peter Todd (56:53.67)
And we might as well go with it because it’s really only like, you know, $50 million and this is tiny part of our actual business. And now you get the whole mess of people then going and saying, no, we got a UAS effort or worse. Like you get another quarter or saying you got a UAS effort in a different way. You know, this all invites an enormous amount of drama for frankly, very little benefit. You know, like drive chains is a model where
Paul Sztorc (57:16.803)
Yeah, but Peter, this is the security model is that the fees will be enough to dissuade the miners from not wanting to. If you’re just going to say that the miners can just become a victim to various lawsuits demanding that they change this and that, I mean, you yourself gave an example of yourself being hit with a totally frivolous lawsuit that’s based on nothing. So it’s really, it’s completely unrelated to drive chain.
Peter Todd (57:36.859)
Only because the Bitcoin protocol is such that I genuinely can say I can’t do that. Right now for miners, mining pools can genuinely say we are not in a position to go seize money. That is not something we can do. But you want to change the Bitcoin protocol to let them go do that. That is why this is so sketchy. And this is why I think it’s in miners own interest to definitely not allow drive chains to go happen.
Paul Sztorc (57:43.337)
Exactly.
Peter Todd (58:01.23)
because they don’t want to be in a position where they can go steal money. Like if you, if you’re in that position, you can be told to go do it. You can be coerced to go do it.
Paul Sztorc (58:07.403)
They’re really very limited because they have to declare the it’s like being buzzed through a gate that takes Three months to get through so they’re very they have a very limited ability to direct the eventual destination of the phone I have to do with what
Peter Todd (58:22.374)
All they have to do is go put a number by quarter order into their coin basis and the money will get stolen. Like there is nothing in drive chains that actually disincentivizes this. You’ve got a bunch of stuff about saying about fee revenue, but like, there’s not.
Paul Sztorc (58:36.212)
No. Well, I think that would be fine, honestly. If that’s what you see, that’s the point is if the side chains have their own fate, where they rise or fall on their own merits. So if something about a side chain attracts a court order that destroys it, then that’s the side chains business. I make no claim. I don’t even really, you know, I don’t make any claim about what people should
do with their own money. I think if people want to use a Monero sidechain, then they should, and if they don’t want to use that, then they shouldn’t. And if they want to use Zcash sidechain, they should use that.
Peter Todd (59:06.858)
As long as you’ve involved Bitcoin miners and the Bitcoin consensus, you have gone involved the rest of us. Like you would have a good argument if you were talking about creating a new proof or coin. Monero itself is independent to Bitcoin. You know, they can go do their own thing and it’s really not very relevant to us.
Paul Sztorc (59:13.039)
Yeah, of course.
Paul Sztorc (59:22.243)
Yes, but that’s the problem though, is that it’s its own coin. We’d rather have all the flexibility of launching new software, but with all the same coin. That’s the whole idea.
Peter Todd (59:29.066)
Well, that’s all well and good, but you have not come up with technology that lets that happen without affecting Bitcoin. You know, that’s really what this comes down to. Like I agree with your goals. I agree with your goals, but you haven’t succeeded.
Paul Sztorc (59:35.631)
Yeah, but when you say affecting Bitcoin, Peter, you only mean… But when you say affecting Bitcoin, you just mean… You only mean that the… Like even in… You’re even wrong about this. You’re not right about almost anything that you say, but even this one thing that you’re wrong about of the fixed code… So of the idea that miners might have an incentive to pay the full node cost.
Even that is absurd because you know full well that must be one billionth of what the eventual mining costs will be as Bitcoin continues to become more popular and difficulty adjusts upward. The marathon alone, you can look this up, they have like a selling general administrative cost or something that was like $57 million a year. So the full nodes already are utterly minuscule and they are not significant.
cost to anyone, even the regular people who are running the full nodes just as end users who have 0% hash rate and derive no income from whatsoever. Full nodes is not a significant cost at all under any circumstances. So even if it were the case that the miners were required to run full nodes, even that would make no difference whatsoever. The costs of mining Peter are in electricity and the ASICs and stuff. Someone has to pay $3,000 for a Solana
Peter Todd (01:00:33.262)
The what matters is how easy it can you say what matters.
Peter Todd (01:00:45.173)
Look, look, I mean, hang on.
Peter Todd (01:00:51.374)
Call.
Paul Sztorc (01:00:55.719)
Only in the event of a withdrawal dispute also, of which there won’t be any.
Peter Todd (01:01:00.957)
Paul, what you are arguing is that it’s okay to go and increase the block size. That is your argument. Yes, it is. Everything you just said…
Paul Sztorc (01:01:06.327)
Absolutely not. The whole point of the whole point of drive chain is to differentially increase it for some people and not others So that this small blocker is in this way
Peter Todd (01:01:17.059)
Everything you said is an argument that because mining costs money, it’s okay to go increase fixed costs. That is your argument there. And the simple fact is, we are trying hard to go keep fixed costs down. I mean, in particular to allow things like P2, Paul, let me finish. Paul, let me finish. Let me finish. So
Paul Sztorc (01:01:30.779)
The box size increase has nothing to do with the box size for mining cars. Sure.
Peter Todd (01:01:39.13)
Why we want to go keep these fixed costs low is because we want to go decentralized mining. We want actual hashers, the people, you know, pointing hash power at things like Oceanpool, to be able to go run actual full nodes and actually go validate what they’re mining. You know, that is why we’re trying so hard to do this. Unfortunately, it certainly does because if you want to go and validate what you are mining, you have to either trust someone else to go do that,
Paul Sztorc (01:01:57.147)
Bit.300 has no effect, zero effect on that.
as a zero.
Peter Todd (01:02:08.342)
Potentially exposing yourself to all kinds of problems when that person lies, fails or whatever, or run into a notes. You can’t get away from that.
Paul Sztorc (01:02:11.751)
So there’s what though? No, what are the problems? No, I choose the first option. So like, let’s say the sidechain nodes are all designed so that the header of the sidechain has the withdrawal hash in them for three months so that you can just look at the head SPV mode or you can just look at anyone who’s on the network, any block explorer or anything like that. And again, because it’s so difficult to get away with it.
Peter Todd (01:02:38.207)
Look, your argument is that it is okay for people to go and sit there manually going and copying, pasting hashes, hoping it’s correct. That is just not something we want to go add to the Bitcoin protocol. That is something miners shouldn’t want. Yes, you are. You are.
Paul Sztorc (01:02:40.027)
Or they just take copying past that hash.
Paul Sztorc (01:02:46.363)
Yeah. Right. Yes.
Paul Sztorc (01:02:51.887)
It’s not part of the Bitcoin protocol. No, of course not. The BIP 300 rules are just counting to 13,000 and that’s the only thing that’s not part of the Bitcoin protocol.
Peter Todd (01:02:59.85)
You’re advocating for a soft fork. You are advocating to add this to Bitcoin vertical.
Paul Sztorc (01:03:04.015)
Or to take OpNop 5 so that the people who want to spend lock coins to this, the Hash Rate Escrow can do so.
Peter Todd (01:03:09.238)
That’s it’s a soft fork. If you do not follow those rules, your blocks are invalid. Like you are advocating for change to become protocol to add this kind of voting, creating a lot of hands.
Paul Sztorc (01:03:14.671)
Yes, of course, yes. Well, I actually don’t really care if people don’t. But you’re just evading the issue completely, which is that we’re OK. Fair enough. But let’s say I’m a miner and I want to start up a new mining operation and I don’t run. We have a bit Solana and I don’t want to run a bit Solana node. But I just.
Peter Todd (01:03:23.698)
I think you’re evading the issue. This is why this is such a crazy argument.
Paul Sztorc (01:03:43.795)
use the withdraw hash that I think my other minor friends are using. What is the, what’s the disaster scenario in that case?
Peter Todd (01:03:50.19)
Well, I mean, the fact is, if that is what’s happening, you’re in a situation where, if this is what’s happening, you’re in a situation where drive chains aren’t gonna go work properly. On the other hand, if they’re actually doing their job, because people start going and copying, pasting hashes and something goes and fails, because people aren’t validating. Like, you can’t get away from the fact that people need to go validate if this is gonna work properly.
Paul Sztorc (01:03:54.415)
What is the disaster?
Paul Sztorc (01:04:01.243)
Not true, they are, because how do they not work properly?
Paul Sztorc (01:04:09.847)
What is going to be done? No, Peter. Peter, I have gotten away with it. It’s just you that just doesn’t understand how it works, Peter, which is that the sidechain, the end users, those are the people, the Roger Veer type people, they wanted the block size, the Bitpay Coinbase, they wanted to pay. They wanted to pay, and they have a full, they have a completely full node, and that node gives them the ability to find out anything they want on the block.
Peter Todd (01:04:18.125)
Look.
Look, this-
Peter Todd (01:04:27.245)
You are not making a technical argument now.
Peter Todd (01:04:35.05)
Look, you are not making a technical argument now. Like you are making a, I hope things go work this way. Well, in fact, you’re talking about what Roger Ver, what I wanted. Like, look.
Paul Sztorc (01:04:40.292)
What’s not technical about it?
Paul Sztorc (01:04:46.319)
No, you don’t get it, Peter. Some people want to pay the higher node cost. So the question is, how can you have some people pay small cost and some people pay large costs? That’s the whole idea. That is the idea.
Peter Todd (01:04:57.358)
and yeah and your idea didn’t work. I mean you know a good counterexample here, a good counterexample.
Paul Sztorc (01:05:00.955)
but what about it doesn’t work? You don’t even realize that a lot of the things you say are contradicting yourself.
Peter Todd (01:05:07.562)
I mean, I think you’re just making shit up. Like you saying all this stuff contradicts. Like there is nothing I can go say that will make you not say that. I mean, this is just silly. Now hang on. Paul, Paul. Paul,
Paul Sztorc (01:05:10.867)
Thank you. Yes.
Can you defer? How can it be a fixed cost that’s amortized if people can’t defer to? What do you think is happening when the mining pool runs a full node and then they have all the other people not running a full node? They’re trusting the mining pool, but they have reasons for trusting the mining pool. They have reason to believe that the process that gives them information from the mining pool is accurate. If we could apply the exact same argument in an inverse sense where you say,
Peter Todd (01:05:42.867)
What are you even talking about there? I mean, like, a mining pool needs to, a mining, Paul, a mining.
Paul Sztorc (01:05:45.295)
I’m Peter we have been discussing this one topic the whole time of whether or not you can defer to someone else’s full node And what does it even mean to run a full node because you could run a full node and then you could you could actually Conclude, you know what actually this full node is bugged or something wrong with my computer And so even running the full node is not enough to gain a full amount of confidence That you are actually looking at the blockchain the right way conversely
Peter Todd (01:06:10.256)
Now we’re just getting into metaphysics. Like this is very silly. Yes it is.
Paul Sztorc (01:06:13.167)
There’s nothing metaphysical at all. We are talking about epistemology, Peter, because we are talking about epistemology because what a full node does is give you information. But this is exactly what a… You are just trying to evade answering this one question because this contradiction obliterates your argument. And I’ve been very polite by letting you meander around and around, but now it’s time for you to just either admit that you’re wrong about this, or say…
Peter Todd (01:06:20.174)
God, what a god. You are getting into metaphysics around the nature of full nodes. This is…
Stephan (01:06:24.407)
So.
Peter Todd (01:06:38.75)
I think out of the two of us, there’s one of us as meandering and not the other.
Paul Sztorc (01:06:44.751)
Yes, because it’s you. I don’t know, Stephan. What do you think? I’d like to ask Stephan, do you think that you can defer to someone else’s full node or not to get information from it? What happens when you visit a block explorer? Does the world just universe just burst into flames when that happens, or what?
Stephan (01:06:45.535)
Alright, so I guess I’m not sure if we’re making a lot of progress here.
Stephan (01:07:02.051)
So the question I think you’re getting at is, can you defer to someone’s node? Obviously, yes, you can. But the question, but I think the point, the counter argument is more about how easy it for people, is it for people to self-verify and to do it themselves, right? And I think as an example, you know, there’s a lot of progress now in the ecosystem around stratum V2, as an example, this idea that you can sort of separate the mining pools from the individual miners and let those individual miners, you know,
Paul Sztorc (01:07:06.735)
Yes, right.
Paul Sztorc (01:07:23.379)
Sure, yes.
Stephan (01:07:31.715)
create their own block template and select which transactions go into the block. And I, I guess, you know, people could have a concern that drive chains could cut against some of that progress or potentially in the future, we’re going to get maybe braid pool or some of these other ideas. Yeah.
Paul Sztorc (01:07:43.983)
Oh, no, it’s far from it. I think, in fact, this is the only way that idea will succeed, is if you can pay out the miners on L2. And in fact, the miners getting paid out every 10 minutes on L2 is fantastic deal for them, because this reduces the leverage of the pools, and it also pays them…
Stephan (01:08:00.195)
So I guess, so hang on Paul, if I’m understanding you here, you’re sort of saying you would have like a big block drive chain and this would allow the miners to get paid out onto that. But I guess that still doesn’t, that still doesn’t sort of.
Paul Sztorc (01:08:09.423)
Yes. Because think about it, the miners, they’re very confident that they’ll eventually get paid if they get paid on a BIF300 sidechain. So they feel happy about that.
Stephan (01:08:17.011)
Yeah, but I think that to me, that doesn’t quite answer the question of how easy is it for people to verify?
Peter Todd (01:08:19.502)
BIP-300 doesn’t involve payments.
Paul Sztorc (01:08:24.335)
No, but I’m saying what it does, it does what stratum V2 and ocean wish they could do, which is it pay, you can do pay per share and you can get paid out every 10 minutes. And in that way, it reduces
Peter Todd (01:08:33.218)
Yeah, lightning light. This is why ocean is working towards lightning. I mean, that’s not that’s not like anything necessary there. I mean, you’re not.
Paul Sztorc (01:08:40.943)
No, but lightning is far worse. Peter, do you really not understand that the lightning has tremendous onboarding cost and liquidity cost and tremendous channel risk, which each person, if you want to onboard someone to lightning, that is.
Peter Todd (01:08:52.054)
Sorry, that’s just nonsense. I’m sorry, but there are not tremendous liquidity costs around lightning. Like ocean, Paul, open, ocean adding lightning is a very simple thing for them to do. Miners.
Paul Sztorc (01:08:55.323)
What is nonsense about it?
Paul Sztorc (01:08:59.847)
You have to pay them, you have to open a channel up front. Yes.
Paul Sztorc (01:09:08.667)
They have first of all Peter, I’m not sure you should get your stories checked with them because people who are there asked them and they said running LN infrastructure is complicated and we have no intention of doing it. So I’m not saying that maybe I misunderstood something that happened, but.
Peter Todd (01:09:19.818)
Well, I mean, like Ocean is doing a lot of very weird stuff. So I wouldn’t be surprised if they don’t understand what they’re doing. But the fact is for Ocean to go pay miners out with Lightning is very easy. They have coins, they open new channels. They go pay the miner.
Paul Sztorc (01:09:33.147)
but that costs money. You have to load, you have to have the money that you plan on paying someone. This is an anonymous hasher that you don’t, I mean, you don’t need to understand the basics of lightning either.
Peter Todd (01:09:38.51)
Oh my God, Paul, Ocean has money coming in whenever they find a new block, right? They get, oh, Paul, let me, Paul, let me speak. Paul, let me speak. Ocean, whenever they go find a block, they have a pile of money that comes in. They can use that money to open channels and they can go pay miners over those channels. If they choose to go set up the economics.
Paul Sztorc (01:09:46.927)
Yeah, but when someone is a new person joins, this is a fixed cost. Okay.
Peter Todd (01:10:06.018)
such that payouts still happen when they find blocks. That’s exactly how they can go do it. They find a block, they open a channel. Once the 100 block, you know, unlock period has come up, they go use that money to go pay miners. And then they can just close the channels and all that money will go flow to miners. I mean, this is not a hard problem. And for, oh, Paul, let me, let me speak.
Paul Sztorc (01:10:23.839)
strict cost of that man far exceed what could be ever spent on the 300.
Peter Todd (01:10:32.142)
And on the incoming side for miners, all they have to go do is have a channel to them and money will flow in on that channel and they can go spend that money out on that channel. When there’s nothing difficult about this. And this is much easier than trying to go and say, oh, by the way, are which of the 256 possible drive chains you want to go have money on? Oh, by the way, you also go need to run a token. And if you want to go verify it, of course it’s going to be a big blocker thing. Say you got to go run a full node and all this stuff. I mean, like,
Lightning is a very simple solution here, you know, and I’ve never understood why people try to go add so much complexity to this. It’s not a hard problem. Lightning is a way of amortizing multiple transactions over one transaction. It’s really a pretty simple thing in this context.
Paul Sztorc (01:11:18.511)
It wishes it can do that, but it really can’t though. So if we can easily compare the two scenarios where in one scenario we have my proposal of just paying them out, pay per share every 10 minutes on L2 side chain, just pick whichever one is the biggest. And so there will only be one. And then you compare that to Lightning. In the case of a new miner joining, which is this is exactly what you wanted, Peter. This is the low barrier to entry, low fixed costs, little guy, he’s joining the new pool.
Peter Todd (01:11:44.682)
Nope, you are talking about something different right now. You are talking about hasher is not mining pools.
Stephan (01:11:49.326)
So, Peter, let’s have Paul finish that answer, and then you can go back.
Paul Sztorc (01:11:51.643)
Yeah. This new person wants to join. In my scenario, they just point their hash at the pool and they can get paid in 10 minutes, pay per share. No matter how small they are.
Peter Todd (01:12:00.366)
Paul, you are talking about hashes, not miners here. I am talking about mining pools. You went and changed the topic to go mislead people. You know full well, you know full well, I am talking, Paul, you know full well I’m talking about mining pools here for barriers to entry. You now changed the topic to hashes.
Paul Sztorc (01:12:08.763)
Well, I honestly have no idea what you’re talking about. No, we’re talking about here. I thought that what we’re talking about is. Well.
Paul Sztorc (01:12:20.02)
I don’t know. I honestly was not trying to mislead anyone, but I thought we were talking about the idea of Ocean to pay the hasher, the pool paying the hasher out every 10 minutes. So is that what we’re talking about or not?
Peter Todd (01:12:33.462)
When you went into, we’re talking about overheads there. Now for Ocean to go pay miners, all they have to go do is pay, can I listen to you a second? If Ocean wants to go pay miners over Lightning, all they have to do is do Lightning transactions to these miners. Like if they choose, Paul, if they choose to do Lightning transactions on a pay per share basis, well, that is their choice. Currently Ocean doesn’t do that.
Paul Sztorc (01:12:39.207)
Are we talking about that or not?
Paul Sztorc (01:12:46.939)
But Peter, that’s exactly what I’m saying is more expensive.
Peter Todd (01:12:59.726)
because they know that they do not have the financial resources to go deal with full paper share. I mean, a lot of mining, look, a lot of mining fools choose not to do it.
Paul Sztorc (01:13:04.783)
Yeah, blah, Peter. I mean, why don’t you want to go back to what I was saying with just comparing the two ideas? I was just you’re the one who cut in and said I was, now you’re saying I was not at all being misleading. And I was talking exactly about what we were trying to talk about. So I’m going to continue now. I’m going to continue to just, cause for the poor people in the audience, I would like the poor people in the audience to at least learn something from this. And I don’t know what you’re going to say after I’m done, but I would like to just give everyone something that they can take home at least, which is the,
Peter Todd (01:13:17.246)
No, you tried to sneak in overheads over here.
Stephan (01:13:20.972)
Okay, let’s have Paul finish your point. Yeah.
Paul Sztorc (01:13:34.435)
In the scenario I outlined, the people connect. They do some hashing. They don’t find a block because their hash rate is very, very small. And of course, statistically, it’s very unlikely that if you have a tiny amount of hash, you’ll actually find the block. But they find some shares at the pool. And the pool pays per share. That’s what I’ve been saying, pay per share. So they can just pay those people out every 10 minutes, taking the absolute theoretical minimum custody possible. They just pay them out on the L2 chain. It’s no different than buying.
Cheese or alpaca socks or whatever so they do that and in the scenario with lightning What the pool has to do is they have to make a decision when someone connects to them They have to say if we want to pay this person we must broadcast a transaction on L1. Oh with a new two of two multi-sig output paying the L1 fee and We have to fund this channel with about as much money as we think we are going to be paying them over the next Time horizon of whatever this is
Peter Todd (01:14:31.566)
Paul, you know that is not how lightning works. No it is.
Stephan (01:14:32.511)
Let me just jump in at this point. No, let me just jump in and explain something here. I think that’s not quite, I kind of, I think I agree a little bit more with Peter on this point. It’s not that the mining pool has to have direct channels with every participating hasher or miner in this case. Somewhat, yeah, but remember, you could put it on, you could put that obligation on the individual hasher or miner in this case and say, hey, it’s your responsibility to have inbound liquidity and…
Paul Sztorc (01:14:34.343)
That is exactly how it works.
Paul Sztorc (01:14:47.099)
But someone has to open those channels. When some of the person is new.
Stephan (01:14:59.431)
Because remember, Lightning has multi-hop, so it’s not like you need direct channels between the mining pool and the miner in this case. They could just have…
Paul Sztorc (01:15:01.851)
Well, of course, yes.
But anyone who wants to join the Lightning, you’re right about that, that’s true, but anyone who wants to join the Lightning Network at all must do that with someone, I guess is what I’m saying. I’m just having a zooming out and I’m saying, someone has to, whereas in my scenario, they don’t have to do that. So I think that is a better.
Stephan (01:15:13.539)
Correct, that you need at least a channel, but I think it’s fair to say, yeah.
Peter Todd (01:15:16.078)
Call…
Paul, you know that Lightning is a routed network. I know you know this. Don’t try to obfuscate it for our audience here. The fact is, if I was to go join Ocean and start getting paid, Paul, if I was to go join Ocean and start getting money in, I would go tell them, all right, which Lightning account you wanna go pay out to, and they can go do it on the wallet I would go have. Like, and if I choose to go and set up a wallet,
Paul Sztorc (01:15:27.283)
But didn’t we disagree that someone must do it? It is literally a fixed cost.
Peter Todd (01:15:47.83)
with some inbound liquidity that’s running 24 seven because I might want to do it. That’s fine. I do not have to get a channel open with ocean specifically. I can use that same wallet for anyone like, and.
Paul Sztorc (01:15:58.333)
That’s true. The point is you have to open some panel somewhere that has some inbound liquidity for this project and you have to hope that you have enough threaded through the whole route and etc. But that is more expensive than what I propose.
Peter Todd (01:16:03.671)
which is
Stephan (01:16:06.767)
I mean, ultimately this minor, I mean, yeah.
Peter Todd (01:16:06.954)
And there’s nothing hard about that. Lightning solves that problem.
Peter Todd (01:16:13.302)
Look, what you’re proposing to, like, we… No, you’re not!
Paul Sztorc (01:16:14.515)
So I’m making mining more decentralized. And also just the fact that a Zcatch sidechain exists and it means that probably people like Foundry would not want to touch it. So I’m further decentralizing mining by offering them this set of revenues.
Stephan (01:16:27.747)
I’m not sure people would agree there because ultimately, if this person is a miner, they are likely to want to use Bitcoin and they may be earning and spending over Lightning anyway. So it may not really be a big deal for them to like earn over mining. In this example from Ocean or whoever. It’s not.
Paul Sztorc (01:16:37.872)
Well, that’s true.
Paul Sztorc (01:16:43.035)
Well, yeah, but I’m just saying it is better. My scenario is actually more advantageous to the small miner in every scenario, because they have less total setup cost. I just think it’s an irony that it happens to be, it’s not essential to my argument, but it’s just ironic that it happens to be better in a way that is a fixed cost.
Stephan (01:16:58.747)
Yeah. So anyway, look, I think this is maybe one, okay, so here’s how I’m thinking about this. What we’re talking about here is, the advent of stratumv2 is, we’re talking about this idea because it’s going to help mining decentralization. It’s going to help more individual hashes actually be miners instead of merely hashes. But part of what’s predicating there is that person being able to run a Bitcoin full node. And in the…
drive chain context, that now means running a Bitcoin full node and up to 256 drive chain nodes if they want to be validating everything. Right?
Paul Sztorc (01:17:32.753)
It does not though.
Not to me, because they could still only run the Bitcoin core node if they only wanted to run that. And what people are still imagining that there’s some enormous negative consequence of them not getting the node. It’s because the node synchronizes so slowly that the you can rely on you either than think about it like this. The network is either going to be this huge successful network like Ethereum, where it’s going to have seven million dollars a day in fees being paid and has huge amounts of users.
And that’s really the only case that I’m interested in. Because if something else happens, then it won’t be secure. But I don’t really care. I’m kind of like, I’m dropping that. So I skim the vital few. So I’m saying like, most new restaurants fail, you know? So I’m saying, okay, maybe a lot of these ideas will fail, but I don’t care. We’re talking, if it’s really big, then all those people, you know, there’s gonna be a huge number of users and some percentage of them will have full nodes and some percentage of them will have big projects. And it will just be very, very implausible since the node…
calculates for everyone the same hash every three to six months. It’s just going to be available information. It’ll be much easier to get that information than it will be to do other things that the miner has to do, such as keep up to date on which ASIC technology is being invented or something like that. There’s no fundamental difference between the other ways that miners compete, which involve a huge mixture of fixed and variable costs. They compete over getting the ERCOT demand management
Peter Todd (01:18:50.198)
Miners do not have to do that.
Paul Sztorc (01:19:03.995)
Weird mining cooling ideas are invented. So they compete on millions of dimensions. Probably it’s not an understatement to say millions of dimensions in some form or another. And this is just some random thing. And there’s just a prejudice against this dimension, I think, because it would be so good for Bitcoin that people can have trouble wrapping their heads around it about how it’s better.
Peter Todd (01:19:25.862)
We care because you are adding fixed costs that miners go have to go in and add. All this stuff you talk about like Epcot, Paul, when you go talk about things like, you know, they got to go pee for the latest AC and all, that’s just not true. You know, if I go have a hashing setup and I go set up with Stratum V2 and I go to that bit of effort, I can just let the thing sit there indefinitely and it’ll continue to work. I mean, I’ve done this before, you know, back in the day when, hang on, back in, back.
Paul Sztorc (01:19:31.192)
I don’t think anyone even agrees with you.
Paul Sztorc (01:19:50.978)
I don’t know, but Peter, like, it’s not the point.
Peter Todd (01:19:52.546)
Paul, let me finish back in the day when I was mining myself. I set up with P2Pool, which was decentralized hashing, and P2Pool, other than one or two technical reasons, would work just fine these days. And all I had to do was set it up and just leave it there. I mean, that is the world we want to be in. We do not want to be in a world where hashers like me now have to go track down websites and block explorers and copy and paste hashes for drive chains to go work and for fees to keep coming in. Like all that’s nonsense. You know, we want to…
Paul Sztorc (01:20:19.891)
Well, we’re not living in a world of hobbyist miners ever because the difficulty adjustment hires the bottom half of performers. And so it will get more and more specialized over time. And these days, every miner has a business plan and things like that. If you’re really trying to say, it’s about whether or not the miner can compete profitably. We don’t want the mine, is someone who’s sitting there and they have the ASIC plugged in and they’re losing money every month because they spend a certain amount on electricity and what they get in is less than that.
Peter Todd (01:20:23.126)
Well, we are.
Peter Todd (01:20:33.989)
That’s not what’s happened.
Paul Sztorc (01:20:47.835)
That is not setting it and forgetting it. They’re lighting their own money on fire when they pay their electricity bill. That’s irrational. Doesn’t make any sense.
Peter Todd (01:20:55.306)
I mean, you’re ignoring the actual model of mining, which is you go find an opportunity you have, such as a cheap source of power. Paul, let me finish.
Stephan (01:20:55.341)
It’s probably fair to say, gone.
Paul Sztorc (01:21:04.967)
But you have to find the opportunity. That’s the fixed cost. You have to go look for the opportunity. But is it not true that finding the opportunity, searching for the opportunity is a fixed cost, Peter? Is that the case or not?
Peter Todd (01:21:15.326)
No, it’s not because opportunities go in very by amount of hashing power. You know, and again, the Paul, Paul can’t, no, it is not. No, it is not. That is fundamentally wrong. Paul, your efforts go find these things scales with amount of hashing power. I mean, you’re, you’re just throwing out nonsense here. Now,
Paul Sztorc (01:21:20.57)
But searching for it up front is a fixed cost because you don’t know what you’ll find. Well, how do you, how are you searching for it if you don’t find it?
Paul Sztorc (01:21:34.191)
Yeah, but the fact that it co-varies with the scale doesn’t change the fact that it’s a fixed cost.
Peter Todd (01:21:39.186)
Yeah, you’re just talking nonsense now. Anyway, so what we go see miners go do is they go find an opportunity which tends to be a source of cheap power and or a thing to go do with the heat. Now, as we know, those things inherently are decentralized because they come in small quantities. And you see things such as flare gas where people go run a small number of miners on a location with some flare gas. We want those people to be able to actually participate
Paul Sztorc (01:21:43.852)
the audience.
Peter Todd (01:22:09.034)
an actual block creation. That’s really what all this comes down to. We do not want there to be a high barrier to entry for that person to go do that. You know, and again, in my case, let’s just say I had access to cheap power up to a certain amount. There’s a lot of people in that same position as I was for the same reason as me. And we want those people to be able to go without very much effort, set up something like P2 pool, or at least stratum V2, do their setup once and just let it run. That is just incompatible with drive chains.
Paul Sztorc (01:22:25.809)
Yes.
Peter Todd (01:22:38.95)
And all you go not, oh, because you are asking for a system where that person like me would have to go cut and paste drive chain assets for the system to work. Like.
Paul Sztorc (01:22:40.106)
How is it incompatible?
Paul Sztorc (01:22:47.707)
Well, you have to like download pizza pool and you have to like poke in every now and then to make sure it didn’t like crash or something. You know, you have to do you have to do some work. Yeah. Yes.
Peter Todd (01:22:54.078)
And all of these things are software which can be, which we can make reliable in a way that we cannot make a bunch of drive changes automatically to do this.
Paul Sztorc (01:23:01.575)
But they’re not complete. Nothing is completely software though, as you know, Peter, because this does cut back to what I was getting at before, which is that even if you run the software, you could still think, oh, I’ve run the software. And then you would think, wait a minute, did I run the right software? Or is there a bug in the software? Or do I have to update? And this has happened in Bitcoin’s history also where they’ve been made. So I’m just trying to sketch out that there’s no, you don’t have this like fundamental.
Peter Todd (01:23:15.287)
Your.
Paul Sztorc (01:23:26.923)
thing rock solid. What is fundamental though is that the miner puts in effort and they get paid money. That is fundamental. In fact, the effort takes different flavors.
Peter Todd (01:23:26.958)
call you.
Peter Todd (01:23:33.006)
Paul, you are arguing that because the world isn’t perfect, we should make it a lot worse. That is what your argument boiled down to there. Because the world isn’t perfect, because occasionally we screw up and people have to… Paul, Paul.
Paul Sztorc (01:23:38.524)
Absolutely not.
Paul Sztorc (01:23:43.599)
I’m saying that you don’t what the miner must do. So let’s take the flare gas thing, they have to have to research opportunities to take advantage of the flare gas, they have to measure the flare gas and all this other stuff, they have to get equipment, and then they have to pay for the ongoing costs like physical security, you know, this, as I’m saying, the miner has enormous number of costs on different dimensions, and all the miners will compete on each other.
Peter Todd (01:24:08.294)
All of everything you mentioned there scales with hashing power. It is not a fixed cost. It is a marginal cost.
Paul Sztorc (01:24:15.427)
everything there, even searching for the opportunity and buying this machine. Yeah. But there’s no sense in which, you know, you’re really trying to say that one, if there’s ever any cost, no matter how small it is in absolute term, like it could be like 10 cents a year that will, that will irrevocably centralized mining. I mean, it’s absurd.
Peter Todd (01:24:17.598)
Yes, because the opportunity is scales relative to how much search effort you go into.
Peter Todd (01:24:37.63)
Nope, that’s not my argument. My argument is that we want to go keep these costs as minimal as possible. Running the fixed node is a fixed cost, but we do not want that to be an expensive fixed cost. We want to keep these things small.
Paul Sztorc (01:24:43.753)
Yeah, but the man doesn’t need to run.
Paul Sztorc (01:24:51.303)
The idea that miners’ costs will be low is incompatible with proof of work, which forces the cost to be high. We want to see how this works.
Peter Todd (01:24:58.134)
Sorry, that is nothing to do with my argument, and you know it. No, it is not. You the idea that miners’ costs will be low. I am talking about fixed costs here, not variable costs. Obviously, proof of work will scale to variable costs, and we do not want the fixed costs be high to allow things to just end up purely on variable costs.
Paul Sztorc (01:25:01.287)
They just, you know, the exact arguments. The difference between proof, the difference between, yeah, but Peter, they, every.
Paul Sztorc (01:25:18.427)
Do you think that when Marathon pays $57 million a year in general and administrative costs, which you can look up on the site, is that a fixed cost?
Peter Todd (01:25:27.754)
Well, chances are the majority of that is not because it’s related to the scale of their gigantic hashing system. Like Marathon has a…
Paul Sztorc (01:25:35.246)
But this is the point is, you just don’t include anything. But isn’t it also the case that Marathon could just say that, oh, what if they misplaced the run-aside chain tool node, which is here a disaster scenario, and they accidentally put it in that line and they say, oh, we put it in here, the selling general in this.
Peter Todd (01:25:43.594)
In fact, Paul, we know that it is not, Paul, we know that it is not a, Paul, let me finish. We know that Marathon’s 57 million is a variable cost, not a fixed cost because I myself have been a miner creating blocks on the same level as Marathon. And I did not pay 57 million because the cost to me scaled with the amount of hash power. I had a tiny percentage.
Paul Sztorc (01:25:59.377)
Good.
Paul Sztorc (01:26:10.379)
they must have paid. So you see, no matter if Marathon lights a briefcase full of $100,000 on fire, then that was a mandatory cost. It was inherent to Marathon.
Peter Todd (01:26:19.798)
Nothing in the Bitcoin protocol says Marathon needs to light briefcases on fire.
Paul Sztorc (01:26:22.151)
I don’t know. I mean, I don’t know what to say, Peter. I think you just you really hate this idea. And I think you kind of just won’t explain why. And I would like to know why. But I don’t I don’t this is I wish that I wish that maybe Stephan can translate for me, but I don’t this is not real this idea that mining costs may increase hypothetically by the need to run some free open source software sometime that other people are running and the way the producers extremely slow moving information is readily available on the internet.
Stephan (01:26:29.811)
Okay guys, well look I think we’ve gone for a while here guys.
Peter Todd (01:26:30.91)
I’ve spent an hour and 30 minutes explaining why.
Paul Sztorc (01:26:51.335)
that is much easier to obtain than even stuff like troubleshooting P2Pool or like other basic stuff like how to make sure that the investment you put in mining pays off, which involves some kind of planning and independent business modeling and running forecasts of costs and things. The idea that would be definitive enough to block a technology that would transform Bitcoin from something that’s very niche and reliant on…
bunch of other stuff like the lightning network, which is not going anywhere near as well as people thought, including myself over the last eight years. And that would give us tomorrow global scale and privacy and unlimited flexibility and immediately humiliate all of our rival coins. So by allowing us to copy them and um, I mean you laugh, but what is your alternative? I just like to say, what is the alternative to global 8 billion people?
Peter Todd (01:27:48.727)
Paul, my alternative is to go continue to work on technology that actually works. Like, you just gave a bunch of goals, and that’s great to have goals, but your tech doesn’t work. So, you know, like it’s not really relevant to what your goals are if your tech doesn’t
Paul Sztorc (01:27:57.303)
Yes. Yeah. But you don’t, again, you don’t literally claim that it doesn’t work. You say it will work and it will change, affect mining costs in some way. So you actually say that it does work.
Peter Todd (01:28:07.318)
And in the trade-offs that we are willing to make in Bitcoin, it doesn’t work. I mean, in SQL database, Paul, putting everything in one giant SQL database would go achieve all this stuff that you’re talking about, but without achieving the goal of decentralization censorship resistance. You know, like, I mean.
Paul Sztorc (01:28:12.823)
Isn’t the case though that each increase in every time Yeah. Sure, yes.
Paul Sztorc (01:28:25.507)
Well, I think actually a full node that’s slightly more expensive than Bitcoin Core is not. Yeah. Yes.
Stephan (01:28:25.871)
Okay, so guys, let me, let me, okay, so we’ve, guys, we’ve both, we’ve all been going for a while, so I think it’s probably a good spot here to have, you know, closing thoughts from each side. We’ve covered a lot of different aspects, but ultimately, you know, let the listeners decide, but let’s have, you know, each of you just take a few minutes and give sort of your closing thought or your closing argument if there’s, you know, one key argument that you want listeners to take away from this.
Paul Sztorc (01:28:35.238)
Okay.
Stephan (01:28:52.843)
What would that argument be? So Paul, do you want to start off there? And Peter, you’ll go after Paul.
Paul Sztorc (01:28:59.119)
OK, I’d just like to talk a little bit about why I’m doing this, which is that I think that it’s not that I actually want any of the individual sidechains, with the exception of one that Peter already knows about, which is that I’m very interested in one. But I would launch that as an altcoin if I absolutely had to. I would prefer it as a Bitcoin sidechain, of course. But that’s not really the point. The point is that people, I believe in the individual sovereignty and freedom of different people.
And I felt sorry for the large blockers and they wanted something. And I thought, can we, what can we do to give them something that’s almost everything that they want? They have this, they have to deal with the withdrawal, the three month delay. And I said, we can give them something that’s almost what they want. And it will not have any costs to the L one participants. And this is why I’m kind of frankly obsessed with, because Peter says that there is a cost to L one.
But I really think that there is not. And that is why I poured so much research into Peter’s ideas. And I wrote this post, The Mirage of Mining Centralization, in January 2017. And why I follow him on Twitter and why we hired him to write these thoughts. Because he really says that he has found some way that this affects L1 in a bad way.
Yeah in some kind of net way also because I don’t forget them it’s also part of my claim that all the hard fork drama will also go away when you have this because Roger would never have wanted to campaign if italic and all these people wouldn’t campaign for Changes to Bitcoin if they just had this as an alternative They would just do it over there. So I’m very Motivated to try to figure out, you know leaving no stone unturned as to what
what is the negative impact that this idea has on L1? Because I really don’t think that there is any. And now I just kind of think that, I don’t know, Peter has started this argument, and now he just kind of doesn’t want to admit that he was wrong or whatever, and instead he has produced this piece of writing that it has these contradictions in it that are immense, including the big one about whether or not you can defer to someone else’s full notes. So that’s just.
Paul Sztorc (01:31:15.591)
That’s just my side of the story, for lack of better words.
Stephan (01:31:20.883)
Okay, great. Well, thank you, Paul and Peter. A couple minutes from you on your closing arguments.
Peter Todd (01:31:26.23)
Well, I mean, I’ve made a bunch of sort of more technical arguments and so on. But, you know, I think like maybe what I should go close on is, you know, with all this tech, I mean, people love to go talk about what their goals are, what their vision is, et cetera, et cetera. But tech has to actually work. And, you know, I myself, I had an idea for something that’s frankly, lightning does a lot better.
And my idea was to Fidelity-Bonded Banks, and I had a bunch of cool ways to go do things. And the fact is, Lightning just out-competed me in every single way. And my response, which I think was the same response, was say, yep, they solved it. My Fidelity-Bonded Bank thing is basically entirely obsolete. It was never even given a chance to really prove itself. But great, Lightning can go do its thing. And the goals may have been the same, but the reality is I lost, because my tech didn’t.
You know, and I think this is really what comes down with drive chains is that Paul has been unwilling to admit that his tech, you know, doesn’t really work. It’s something that’s very simple, but it’s not sufficiently complex to actually solve the big problems it goes and has. And, you know, people got to go look at it, not with like starry eyes, like what Bitcoin could be, but with the hard reality of, well, does this thing actually work? You know, and I also think this gets back to all these ideas like competing with altcoins and stuff. It’s that look.
The tech has to work well enough to compete with altcoins. You know, you have to work well enough to compete with something like Ethereum. You have to work well enough to compete with something like Lightning. And if you’re talking about, well, we’ll just put money on this drive chain, which miners absolutely can go steal, or potentially even more likely, miners will just fail to do anything about it because, well, shit, now a majority of hashing power has to go and move money around on this drive chain. Well, it’s not even sure that’s really gonna happen. Now the price crashes. I mean…
There’s a lot of ways these things can go fail. And it’s much more likely that if somehow the soft fork does get activated, nothing much will really happen in terms of working and we open up Bitcoin to a bunch of ugly attacks. Whereas, you know, things like Lightning genuinely are independent to Bitcoin and they’ve proven themselves. And finally, I mean, maybe I can go say, you know, Paul Storick, he owes me like $1,250 and I should send him a Lightning invoice.
Peter Todd (01:33:46.542)
Because last time he went and paid me, he went and paid me on chain. I’m not sure Paul’s ever actually gonna use lightning. So this would be a good opportunity for him to actually get a wallet and go and try this out.
Paul Sztorc (01:33:47.012)
Yes
Paul Sztorc (01:33:55.059)
has. I paid him from a legacy Bitcoin address too.
Peter Todd (01:33:59.282)
Yeah, yeah, I noticed that. Obviously you had a very old wallet and it might be time for you to actually use some of this tech. You know, like the biggest Lightning transaction I’ve ever done, I think was like about four or $5,000 US and it worked. I mean, this tech works pretty well. I mean, I think there’s edge cases that we’d love to go solve. I think there’s definitely things where we’d have to improve to go scale better, but.
Paul Sztorc (01:34:03.431)
Yeah. That’s the best.
Paul Sztorc (01:34:16.019)
Thank you.
Peter Todd (01:34:25.238)
You know, to go, it’s like present all these images of, well, drive chains will do this, drive chains will do that. I mean, you got to go prove this tech in ways that are, are actually convincing. And I just, the fact that it’s like drive chains hasn’t done that. And frankly, my advice to Paul would be go move on.
Paul Sztorc (01:34:38.939)
Well, can I ask you this, though? Is there anything that would convince you?
Peter Todd (01:34:43.794)
With drive chains right now, probably not, because the idea itself is just broken. Like, you’ve tried very hard and you’ve just failed. You know, whereas as a counter example, lightning was something where I remember when I first looked at it, I wasn’t totally clear where to go work. But you know, I’ve read more about the tech and thought through it. Yeah, yeah, this obviously could go work, you know. And of course, it did go work pretty well. It’s not perfect. Can’t necessarily scale to the entire world.
but there are plausible ways to go fix this. And we see this with Arc as an example right now, which is a plausible way that Lightning could scale to a much bigger venue. I’m not entirely clear that works, but it’s plausible enough. And these things convinced me because they’re good technical ideas. They may not be perfect, but they’re reasonable.
Paul Sztorc (01:35:29.523)
Well, you know, there’s one thing, not to keep going off, but the creator of ARK, he has said that lightning is, he created it because he thought that lightning is broken and he says that in the beginning of his ARK Masterclass whiteboard video.
Stephan (01:35:30.039)
All right, well, I think that’s.
Stephan (01:35:38.059)
Yeah, but to be clear, even in the case of Ark, Baraka’s also said that ASPs would also be LSPs, and it’s also in the context that Lightning would also work. So, yeah.
Paul Sztorc (01:35:45.219)
Yeah, he wants to transform lightning kind of into that. But he’s not like so super sanguine on lightning. He’s created it to fix the problems in lightning.
Stephan (01:35:52.083)
Well, I think that remains to be seen. I think there’s a range of different views on how far lightning can scale and potentially with other soft fork ideas or other ideas.
Peter Todd (01:35:52.435)
I mean…
Paul Sztorc (01:36:00.243)
I mean, he’s got the lightning light right behind him, so. You can’t like.
Stephan (01:36:03.807)
Well, of course. Yeah. So look, I think it’s been a good debate between you two. Hopefully listeners have had a chance to hear both sides. I’ve done my best to be impartial, even if I have a view, but obviously listeners, you’re big boys and big girls out there. You can hear both sides. As in the show notes, I’ll put the links to obviously to Peter Todd’s post as well as to Paul’s response.
Paul Sztorc (01:36:11.759)
I agree.
Stephan (01:36:32.124)
and listeners, you can make up your own mind. Paul and Peter, thank you for joining me today.
Peter Todd (01:36:36.93)
Thank you.
Paul Sztorc (01:36:37.756)
Thank you.