Epicenter - Learn about Crypto, Blockchain, Ethereum, Bitcoin and Distributed Technologies - Peter Rizun: A Bitcoin Fee Market Without A Blocksize Limit

Episode Date: February 28, 2017

With both the Bitcoin Unlimited and Segregated Witness efforts far from reaching majority support and exploding transaction fees, the debate around how to scale Bitcoin continues on. One of the key arguments against bigger blocks and Bitcoin Unlimited is that a blocksize restriction is needed to create a healthy fee market. Dr Peter Rizun has been researching the economics of transaction fees in Bitcoin extensively and joined us to discuss what dynamics affect fees and why he thinks the blocksize limit will eventually fall. Topics covered in this episode: Bitcoin seen through the eyes of a physicist The dynamics that determine transaction fees in Bitcoin How orphaning risks drive the fee market economics The relationship between the block reward and the fee market Why the blocksize limit will eventually fall Why Peter supports Bitcoin Unlimited The state of discussion between Bitcoin Core and Bitcoin Unlimited Episode links: Dr. Peter Rizun's talk at Scaling Bitcoin Montreal 2015 Dr. Peter Rizun: "A Transaction Fee Market Exists Without a Block Size Limit" XT Nodes - Bitcoin Hashrate Distribution Bitco.in Forum Ledger - Peer-Reviewed Journal on Cryptocurrency and Blockchain This episode is hosted by Brian Fabian Crain and Meher Roy. Show notes and listening options: epicenter.tv/172

This is Epicenter Episode 172 with guest Peter Ryzen. This episode of Epicenter is brought you by Jax. Jax is the user-friendly wallet that works across all your devices and handles both Bitcoin and Ether. Go to J-A-A-W-X.I-O and embrace the future of cryptocurrency wallets. And by the Ledger NanoS, the hardware wallet which sets the new standard in security and usability. Get it today at ledgerWallet.com and use the offer code Epicenter to get 10% off your order. Welcome to Epicenter, the show which talks about the technologies, projects and startups driving decentralization and the global blockchain revolution.

My name is Brian Fahian Crane. And I'm Meheroy. Today we are joined by Dr. Peter Ryzen, who is associated with the Bitcoin Unlimited Movement, let's say. Dr. Peter has written papers on two very interesting topics. The first is that a transaction fee exists in the Bitcoin system without a block- size limit and second he has a proposal called subchains which gives a way to improve the user experience and scaling of the Bitcoin system. So we are going to talk about mostly his

economic analysis on the fee market in Bitcoin and what it means for the block size debate. So before we start, let's have an introduction from Dr. Eisen. So Peter, could you give us a bit about your background and how you got to be interested in the Bitcoin space? Sure. So I come from the physics engineering side of things. And I work in the high tech sector right now, but I've always been a bit of a hobbyist or armchair economist. So before I learned about Bitcoin, I was reading the various blogs. I was a bit of a gold bug, you might say. And I really felt that we needed some solution to make a new money system. for the modern age, a money system that would be more fair than our inflated fiat currencies and more useful than something like gold that can't be transmitted over the Internet. So I guess it was early March of 2013 when on the radio I heard that the price of Bitcoin had hit some new all-time high because of the Cyprus confiscations of bank accounts.

And I'm like, I keep hearing this thing, Bitcoin. I never really taken it seriously. So I'm like, that's interesting. I've heard that a couple times. So I'm going to go home and learn about it. So I did. And I hit upon Satoshi's white paper. And that's when I was really blowing away.

I realized that Bitcoin wasn't some little toy internet money. It was actually a breakthrough in computer science to solving major problems. They were actually thought impossible. So for about two weeks, I did nothing. but read everything I could on Bitcoin. And ever since then, I've been hooked. Awesome. Yeah, I think a lot of us had that experience.

It's interesting that you mentioned before you were looking for a fair way of mining. So can you explain that? I mean, I understand that people say, like, okay, I have an issue with Fiat money or with quantitative easing and the way the financial system works. But why did you think of mining in that context? Did I say, I think I meant a fair money system. Yeah, okay, there's no problem. So another thing I wanted to ask about brief is, so you're a physicist by background.

And I remember reading quite a bit with speculations about Satoshi that he might have been a physicist as well. These people were quite, most people, I think, is not a computer scientist, right? because his programming was often not so clean, but people would say, okay, for various reasons, he might be a physicist. So I'm curious, when you look at Bitcoin, from the eyes of a physicist, does that make sense to you?

Do you feel like it is built the way a physicist would approach this? I haven't thought about that until you answered, asked me that right now, but I think it makes a lot of sense because, you know, physics takes a very broad view of looking at things as a system. We're normally looking at things in the natural world, but Bitcoin, in a way, is kind of this organism that's come into being and has taken on a life of its own. There's all these neat emergent properties that would excite a physicist seeing how this complex economic computer system works. in practice.

So yeah, I think maybe he was a physicist. I mean, it clearly has, you know, like he said, examples of computer science, but there's also economics mixed in and a whole bunch of fields that have to come together to create Bitcoin. And maybe a physicist has a better, higher level view to combine all those ideas.

One thing I've always thought was cool is, like the blockchain itself and the process of confirming transactions, to me, sort of reminds me of collapsing the wave function in quantum mechanics. So in quantum mechanics, we have something called the time, energy, uncertainty principle. And I think something similar applies to the blockchain. I got to call it the time history uncertainty principle. So a block was recently mined, and then more and more transactions are being added to the mempools of all the nodes. And the state of the ledger becomes less and less certain as time grows on. Eventually, a miner finds a block and in a certain way that collapses the wave function

and we get more certainty of the history of the ledger. So I imagine these blocks is all these collapsing of the wave function, and that's how this consensus of the transaction history is built. So back to the time history and certainty principle, you can know the time of transaction very quickly, but you don't really know for sure what it'll be confirmed. But if you want to know whether it's confirmed to a high degree of certainty, then you have to wait a whole lot of time to pass.

So I think that's a cool analogy to physics. Yeah, I mean, in quantum physics, you have this famous like Young's double-slit experiment where there's like two slits in a wall and you throw an electron at it and there's a point at which you can argue that the electron is like moving through two slits. It's in a state where it's physically present at both slits

and in order to describe its states you have to consider that it might be taking both paths at the same time, right? And so you're... So this is like you're kind of saying that when I do a transaction when I pay for coffee, when I broadcast that transaction immediately to the network, it's in this state that it's both confirmed and unconfirmed. And it's only when the wave function collapses that the next block is mine,

that it reduces from that point of uncertainty, confirmed and confirmed, and it takes either one of two values, confirmed if it was included in the block and not confirmed it if it was excluded from the block mine. Yes, that's exactly right. So let's probably move on to the topic of our conversation, that is the fee market of Bitcoin. So your paper is very interesting. So what your paper says in essence is that even if we had no block size limit, there would be a transaction fee market with transactions, with there being some kind of transaction field

they read on transactions. Tell us why you spent so much, maybe so many hours, writing this paper. Like what motivated you to make this point and build a mathematical model around it? So what motivated me was the block size limit debate. So people were saying, oh, we really need a block size limit.

Otherwise, the blocks are going to become infinitely big because there's zero marginal cost for a transaction and miners will just, you know, even if someone pays one Satoshi for a transaction, miners will include them in their blocks. And since who wouldn't want to write to the blockchain for only one Satoshi that the blocks will become arbitrarily large, I thought that was an odd thing to say because it just didn't feel intuitively right to me when I heard it. And empirically, we knew it was false true just by looking at the history of the size of blocks since 2009. And what we've seen is, so the whole time,

the block size limit has been in place at one megabyte. The average block size was significantly less than one megabyte. When Satoshi put the limit in place, the average block size was something like five kilobytes. So the block size limit was about 200 times bigger than what was required. But still, for those first seven years, Bitcoin wasn't suffering from these spam attacks and blocks weren't right up to the limit. It wasn't until the summer of 2015, as I was working on that paper, that we actually started to hit the limit.

So I wanted to have a better idea of what was going on. And as a starting point, I remember I had read something really interesting from Gavin Anderson, who was talking about if a mitre includes a bit more transaction in this block, his block will propagate a bit slower, and it will thus be more likely to be orphaned, because during that propagation time, there's a good chance another miner will find a competing block, and now you have two blocks at the same height, and as you know, the blockchain can only have one block at each height. So the purpose of the research I did for that paper was try to get a better analytical understanding of what's actually going on there, and if it made sense that orphan risk could drive a fee market.

That's interesting because actually I met Gavin and Dries at the Bitcoin conference in maybe 2014 or something like that, and I asked him exactly that question. And especially I asked him about what he thought the cost was of including an additional transaction in terms of increasing the orphan rate. And back then he told me it was eight cents. I mean, I guess today maybe the dynamics have changed with full block. Is there still a way to say, or maybe not? I don't know. Maybe if somebody has a block that's 900 kilobyte,

do you know how much it costs them in terms of increasing the orphan rate or like including an additional transaction? Yes, definitely. And that's what my paper is all about, is trying to apply numbers to this concept. So the inflation rate, according to the model in my very simple paper, is that the fee in like Satoshi's per byte is proportional to the inflation rate of Bitcoin,

like how big the block award is, multiplied by something I call the transaction appeals, and that's how many seconds it takes to communicate a megabyte of block information. Now, you mentioned that when you spoke to Gavin, he estimated the cost of including an extra transaction due to orphaning risk at about $0.8 per transaction. And that makes a lot of sense

And if you look at some of the numbers in my paper, they align with that as well. And what's happened since then in 2014 is there's been a bunch of efforts to improve how quickly information can propagate across the Bitcoin network. So when I wrote my paper, I had estimated it more at one or two cents for the cost of inclusion of a transaction. But like I said earlier, as we can improve the rate at which information can be propagated, that reduces orphaning risk and drives the cost of including a transaction downwards. Okay, great, very interesting. So I suppose that also kind of answers the question, because today we have this limit of one megabyte and what is the cost of transactions

on average, like 20, 30 cents, right? It's gotten very high. So if you say that kind of, you know, the supply curve in a way of producing transactions like one cent, then I guess somewhere that intersection would be, I don't know, do you have an estimate there, three, four megabyte today or two megabytes? Where do you think that intersection would be? My guess is if you were to suddenly remove the block size limit, we'd probably have blocks from 1.5 to 2 megabytes inside.

and the average transaction fee would probably be between 1 and 3 cents would be my guess. So the core of the argument here is that for a minor, so when I'm a minor, and I am receiving transactions that are waiting to be confirmed that are sitting in a M.M. pool and I'm generating a block, I have to decide how many transactions to put in my block. and out here I'm facing a situation of both risk and reward in a sense, right? Exactly. So the reward is if I put more transaction, if I stuff more transactions into my block,

I'm going to make all the transaction fees. Okay. So that's the reward side of putting more transactions in the block. And then there's a risk side. The risk side is the more transactions I put in my block, the greater the risk that I broadcast the block and my block moves slowly around the network therefore there's a risk that

some other miner is going to generate a block let's say five seconds after me but generate a smaller block and his block will propagate faster in the network than me and he will end up winning and his block is the one that will get included in the chain so so basically I am facing a situation of like risk and reward

and there is some point, there is some point at which these two kind of forces balance out, and when these forces balance out, that gives me sort of my optimal block size that I should create. So I, as a miner, this is this size I should create optimally in order to balance both my risk and reward. Is that right way to frame it? Yeah, you did a very good job, probably better than I could do. Yeah, exactly. If a miner publishes an empty block, well, that block is going to propagate pretty fast, and he'll earn 12.5 Bitcoin if it's included in the blockchain.

But of course, he's missing out on all the fees. So on the other hand, let's assume there was no blocks for a second. He could make a 20 megabyte block and full of all this low-cost fees, and maybe the fees plus the block award instead of being 12.1. 5 Bitcoins, it's now equals 14 Bitcoins. So, you know, he's out by 1.5 Bitcoin, but if the block is so big that certain nodes aren't relaying it, other miners take a long time to validate it, there's, like you said, there's a good chance that another miner mines a faster block in the meantime, and it's that faster block that gets accepted, and the minor not only loses the extra transaction fees he was trying to claim, he also loses his block award. So what I worked out in my paper was to analytically describe how that equilibrium works and how there will be a point on the curve where by mining a smaller block, the miners leaving fees on the table, but by mining a bigger block, he is risking too much a chance that his block will be orphaned.

So at that sweet spot, there's a certain block size sweet spot where the miner's profitability is a maximum. Let's take a short break to talk about Jacks. Jacks is your wallet, your complete user interface to cover all your blockchain needs. I've been using it and I've been loving it. Jack supports a lot of different cryptocurrencies. I suppose Bitcoin, Ether, Litecoin, Ethereum Classic, Zcash, Augur Rep, and they're adding many more, keep responding to users' needs. Now with Jacks, the nice thing is that you can manage all of those coins within a single wallet

and you are in control of your own private keys, they're not on their server, and there's a single 12-Vird seed that you can use to backup your wallet, all your coins, and sync them across different devices. Talking about devices, they're on pretty much any device that you can think of. You can get it on PC, Mac, Linux,

you can get it on smartphones like Android and Apple and iPhone. You can get it on tablets, or there are even browser extensions for Chrome and Firefox. And on top of that, in Jax, You can actually exchange different cryptocurrencies for each other because they've integrated a shape shift. And more partnerships and integrations are coming down the line in 2017 that are going to make Jax even better.

So Jax is really making blockchain and cryptocurrency successful for the masses, it's easy to use for the masses. Make sure to get your own Jax wallet at Jax.I.O. Or you can get it from any of the app stores you are using. We'd like to thank Jax for their supportive epicenter. One of the key dynamics that is required, I mean, we touched on it briefly here for this model to work, is that creating bigger blocks has a cost, right? So we've talked about that there have been a few efforts, right?

There's like X-Thin, which we talked about in a previous episode of Bitcoin Unlimited, which is essentially a way to propagate blocks faster and propagate less information for blocks to, go around and then there's the Bitcoin relay network and there's some other efforts too that all go towards you are propagating it faster and it's not I think inconceivable to imagine that one will figure out ways so that there's actually no increased orphan risk for bigger blocks do you think that's a real risk and and you think that would be a problem for the fee market you've described here? Well, I got a whole bunch of things to say on that topic.

So you mentioned Ex-in first, which I think X-N is a fantastic technology, which allows miners to propagate their blocks more quickly. And the testing that we did showed that X-DIN reduced the amount of bytes required to propagate a block by around a factor of 24, and it improved the time it takes for blocks to be transmitted between nodes by around a factor of 8. So that has the effect of decreasing the propagation impedance that I talked about, that the drive is a price of block space. So yes, an innovation like X-Thin does make it faster to propagate blocks and therefore make fees under an orphanous model cheaper. But I think that's a very good thing because we want to have cheaper fees.

We want fees to be low, so people are free to use Bitcoin for making transactions. So right now, I mentioned earlier that I figured the equilibrium fee without a block is, and it would be between one and three cents. But now imagine that the price of Bitcoin jumps from $1,000 worth this today to $10,000. Because fees are denominated in Bitcoin, the average transaction fee would therefore increase from one cent to $3,000, up to $0,000. 10 to 30 cents and now you're getting pretty pricey again. So all these innovations like X-Thin that allow miners to more efficiently propagate the blocks

and reduce orphan risks are good because they drive fees down while the increasing price of Bitcoin is driving fees up. So that was my one comment on that topic. The second comment was about a possible future scenario where there is no marginal cost to include transactions. And this is a question that there's been a lot of debate about between myself and Gregory Maxwell in particular. And to answer that question more definitively, I wrote a paper called subchains, and it's basically analyzing how the few market would change if miners cooperate to build blocks layer by layer over the 10-minute block interval using weak blocks.

So basically, the idea of pre-consensus where the miners are agreeing what's, it's going to go into the block before they actually find it. So when they find the proof of work, most of the transactions that are including the block have already been propagated. So what was interesting from that work is that even with those pre-consensus techniques, the fee market still didn't fall to zero because miners still need to include new transactions in their block. And there's an orphan risk of including those new transactions because at some point you've You've got to tell the other miners what's going to be in this block that's coming. And the more new transactions you include in your weak blocks, the higher the orphaning risk as well.

So I think we can do lots to reduce orphaning risk and drive down fees. But no, I think it's fundamental that there will always exist fees based on orphaning risk in a free competitive mining market. whenever I see a particular proposal which says we should do away with the block size limit I think any any proposal needs to answer for at least two set two different questions right so question number one that must be answered is if you increase the block size limit then will there be a transaction free market so that is I think something that your paper really nails down Yes, there will be a transaction fee market.

The other question, though, is, is it going to be the case that the Bitcoin mining game remains fair? So, okay, what do I mean by fair here? So ideally what you want is a system where no matter if you are a small or a large miner, so case one might be, I'm a minor and I have only half a percent of the hashing power, I'm a small miner. Case two might be, I'm a large miner, I have 20% of the hashing power. So a fair game is when, if I have half a percent of the hashing power, then every 200 blocks, one block would be mine in expectation over a long time. An unfair game is I have half a percent of the hashing power, but less than one in 200 blocks is mine for some reason.

So, and you want this game to remain fair for four large, and small miners, because if it becomes unfair in some way, then it might be the case that you drive away the small miners. Now, does your paper have anything to say about whether removing the block, how removing the block size can impact the fairness of the game? Does it make no difference or does it make some difference? Do you have any insight there? Yes, I have some insight.

So the first thing I'll say is, personally, and Bitcoin Unlimited, we're not trying to remove the block size limit completely. We just want to make it a market-driven block size limit. But academically, we can argue what would happen if there was no block size limit, which I think is the question your answer asking. And the question was, how would the fairness in the Bitcoin mining game change? and that's an important question and it would change it could change let's say so let's talk about where the unfairness

comes from so if like you said if I'm a large miner and I have 10% a hash rate and I propagate a block and let's say that block takes six seconds to propagate so all the other miners they're going to start mining on that block six seconds later

but me I'm going to start mining on the next block right away so I have a six second head start on all the other miners and six seconds is one percent of the of the 600 second 10 minute block time so I have kind of in a way I have like a 1% time advantage on that particular block but because I have 10% the hash rate the chances of me solving the next block is 10% so I have so 10% of times I have this 1% advantage or a smaller minor who only has one percent of the hash rate, he has that one percent advantage only one percent of the time,

if that makes sense. So that's where this unfairness argument comes from. And it's not, the unfairness doesn't depend on how big blocks are. It depends on how quickly blocks are spreading across the network and ultimately what the average orphan rate is. So to me, it's not a big deal like Gavin Anderson ran the numbers and the advantage to, a 10% minor over a 1% minor in like a 1% or 2% orphan rate environment, it was tiny. Like it was insignificant compared to the advantage that some miners have in terms of lower electricity costs and especially in terms of just the random revenue fluctuations due to some miners being more lucky than others.

So yes, there is a small theoretical advantage, but until orphan rates are like 25 or 30%, which I don't will ever see, I don't think that advantage is ever really significant enough to affect the health of the mining ecosystem. You mentioned in your paper that one of the assumptions you have is that there needs to be an inflation, so a block reward. Now, of course, with Bitcoin, the block reward over time goes to zero. At the moment, the inflation is 4%. It's going to go to 2% by 2021 to another 1% 2025.

So I'm curious, I mean, your analysis, does it fall apart then? So do you need an inflation of a certain size and you just need a non-zero inflation? What does that dynamic look like? Yeah, so that's another great question. And it's been a criticism of my fee market paper. So the way I would answer that is when we're doing. analytical modeling like I did in my fee market paper. We're not trying to model all the complexities of the Bitcoin ecosystem. We're just taking, you know, we're making a very simple

model to try to understand one phenomenon. And the simple model I focused on was what the transaction fee market looks like when we do have a block reward. So I assumed a priori that there was a block reward. And that was one of the parameters in my model. If you, the way the math works out, if you take that parameter and you let it go to zero, then the model breaks down. It doesn't really give you an answer. So I wouldn't say that the fee market will necessarily break down in 100 years when there's no inflation. I would say that we don't know what will happen.

So I've shown that there is a transaction fee market based on orphaning risk without a block size limit today and for the next many, many, many, many years, but I think it requires a model specifically focused on the no block reward end game to answer that question. So right now, I don't know how security will be paid for in the future. Maybe transaction fees will be sufficient. Maybe they won't. Maybe we'll need assurance contracts like what my current talked about. I'm not sure. But I don't think it's really something we should be worried about at this point in time because that's a long ways down the road and I don't see how the security endgame really affects the decisions we're making right now. I think right now what we want to do is we want to increase the number of people using

Bitcoin so we have a bigger economy that could pay for security when that need arises. Yeah, no, I very much agree with that sentiment. I think it's very important that one, you know, solves the problems. that exists today and doesn't like not solve the problem that exists today for fear of creating other problems that might only actually be problems 20 years down the line. Exactly. Right now we do not know whether that will be a problem at all. I think there's still the question though, which I think is one of the points of view informing people critical of bigger blocks, which is that the block reward is going down.

and so the miners revenues do decrease from that side. I mean, the Bitcoin security model kind of game theoretically, we do have a relationship between the revenues of miners and the security of the network, right? So the question is, if the block reward is artificially limited, doesn't that increase the revenues for the miners? And is that a good thing for security? Do you think with no block size limit, that bigger blocks would lead to lower revenues of miners and less security?

What's your point of view that? Well, I think there's two competing phenomenon. Like, on the one hand, if we remove or increase the block size limit, we make room for more transactions and more users. And those users can pay more fees. If we have 10 million people using Bitcoin, they can afford a lot more fees in aggregate than 1 million people using Bitcoin paying higher fees. But on the other side, yeah, if you keep block space highly restricted, the limited amount of block space the miners do sell will fetch a higher price. What the optimal block size limit is, I don't know, but neither does the Bitcoin core developers, which is, I think, another argument that we want to allow the block size limit to evolve through some free

market process as opposed to being centrally planned by a group of developers. Because I think the market can answer questions like this much better than any of us can. Let's move on to Bitcoin Limited because from your statement like you're in favor of having the market kind of decide or give mechanisms for digital. deciding on what the block size limit should be, right? And I think Bitcoin Limited is like the project that is kind of taking this approach. Bitcoin Unlimited, sorry, is the project that's taking this idea and putting it into practice. So explain to us like what Bitcoin Unlimited is doing.

Sure. So Bitcoin Unlimited, I would say, is about two main things. And one is peer-to-peer electronic cash. We feel that in the last couple of years, the Bitcoin has been steered in the wrong direction away from electronic cash and focusing more on digital gold or as a settlement-only network. So one of our goals with Bitcoin Unlimited is to bring back Bitcoin

to the thing we all got excited about. When we read that white paper, I told you guys about reading that at the very beginning of this talk, So we want more on-chain capacity to allow more users, more transactions, make a better user experience. Right now, users are frustrated because they don't know how much fees are going to be charged. They don't know when their transactions will be confirmed. And we want to get rid of all this idea of replaced by fees and things that make zero confirmation transactions less secure. We think zero transaction confirmations are though not perfectly secure.

We're working pretty darn well for many use cases. and we'd like to bring that back and, in fact, enhance the security and usability of zero-confir-confir-confir-prom transactions. So that's Bitcoin Unlimited's goal in terms of peer-to-pay electronic cash. The other goal of Bitcoin Unlimited is decentralization. And decentralization seems to be a buzzword and it seems to mean different things to different people. The core supporters will probably disagree with what I'm saying now. But when I think of decentralization, how do we get that? Well, we want more users, we have more keyholders, we want more miners, we want just a bigger economy.

So it's decentralized over a broader group of people. And in particular, we want more development teams working on Bitcoin so that the users and the community have more choice in what note implementation to run and ultimately more say or more vote in how the network evolves. So those are the goals of Bitcoin unlimited. So explain to us this word unlimited. So when I hear the word unlimited, the thing that comes to my mind is, hey, there was like Bitcoin classic with a 2MB limit. There was, I don't know, some other Bitcoin with an 8mb limit. And unlimited means no limit. So do you really mean when you say unlimited that there's no limit at all?

Yeah, another good question. a lot of people think that they hear Bitcoin unlimited and they scream, oh, my God, we're going to have gigabyte blocks in 2018. But no, Bitcoin Unlimited is about unlimited choice for the users, node operators and miners to shape Bitcoin as they see fit through a market-driven process. So we do not want to eliminate the block size limit completely. We just want to allow the decision for what the block size limit should be. be to evolve in a more decentralized bottom up fashion as opposed to being dictated from above by core dev. So how do we do that? So basically all Bitcoin unlimited has done is we've made it

easier for users to directly adjust the block size or the maximum size of blocks that their node will accept. With Core, if you wanted to do that, you'd have to recompile the code, which a lot of people can't do. With Bitcoin Unlimited, it's just a, it's just right there in the GUI. It's very easy to increase your node's block size limit today. So how will the limit be practically set? So we, so imagine like, okay, there's this network in which a lot of people are running the Bitcoin Unlimited code. Each of them is individual. setting their own block size limits. So I run a full node and I say,

I'm willing to take four megabyte. Brian's running some node somewhere else and he's like, I won't go beyond one megabyte. And maybe there's some, maybe there's you who is accepting 16 megabyte blocks. So all of us have different numbers. So when all of us have different numbers, how does the network actually settle on a particular number?

Well, I think it'll ultimately be the miners that settle on the numbers. I think the miners will look and think, well, most of the network seems to be happy with up to two megabyte blocks. So let's miners agree to make two megabobobob blocks, our block size limit. From the perspective of nodes, it doesn't matter. Like my node has a 16 megabyte block size limit, and it's had that limit since the summertime, and it's tracking the blockchain just fine. So from the nose perspective, there's no security risk by setting a block size limit above whatever the miners set. So this concern that different nodes are setting different values of block size is kind of a fake concern. It doesn't make a difference.

As long as your block size limit isn't below the size of blocks of the miners you're producing, then you're fine. And even if it is, Bitcoin Unlimited allows you to still track the consensus of the blockchain through something we call the excessive block gate. That's what we call it. But that's getting kind of technical. I think what people need to think about when they imagine how we might get larger blocks is they have to realize that the Bitcoin network is ultimately made up of people that are operating these nodes and these miners.

So right now we have a bunch of people, and those people, I think for the most part, want larger blocks. They think that one megabyte is hurting Bitcoin's ability to grow, and they're just trying to coordinate or find some way to make that happen, and we were looking towards core dev to help that, but it doesn't seem like they're doing so.

So now it's like how do we empower users to, to make a difference themselves. So I like to imagine, you know, you've got a couple miners all over the world, and instead of looking at nodes, they're looking at a bunch of people. And imagine these people are all holding signs. And right now, 90% of those people, which are node operators, are holding a big sign that says one megabyte. I'm only going to accept one megabyte.

So 10% are already holding something that says 4 megabytes or 8 megabytes or 6 megabytes. Those are the big kind of unlimited node operators, but they're still in the vast minority right now. So when a miner looks around and he sees all these people holding these one megabyte signs, he's not going to dare produce a 1.2 megabyte block because he thinks that the market doesn't want it. But what we want to do with Bitcoin Unlimited is just convince all those people holding one megabyte signs to, you know, erase the one and put a two or a four or whatever it is that you think is a reasonable block size limit. And eventually, when the miners look around, there's going to come a point where most people are holding signs that say, let's say 8 megabytes.

And now the miners can get together and say, well, I think, you know, it's pretty clear that the nodes are ready to accept larger blocks. We can start producing them. And eventually a miner produces a 1.1 megabyte block. The nodes accept it. There's no event. There's no fork. There's no panic. There's no drama.

It becomes part of the blockchain. And from that point on, we have bigger blocks. And the problem is largely solved. We have come actually to what I think is a good way to think about this problem. Right. So let's take this imagination further. So the idea here is you imagine like a room full of node operators. And each of them is holding a sign, one megabyte, two megabyte, four megabyte.

and I as a miner in some senses I have a camera with which I can see this room full of people and I can read what they're willing to accept and now your contention is that I'm going to see what they're willing to accept and if a lot of them are willing to accept

8 megabytes I'll be willing to produce up to 8 megabytes right if producing up to 8 megabytes maximizes my profit, which is defined in your paper, right? Yes, yes, of course, you can always produce smaller blocks. I can always produce smaller blocks. But one challenge with this is, so in Bitcoin, what is ultimately secure? What is ultimately secure is the veracity of a transaction that is buried under a lot of blocks.

Everything else in the Bitcoin network can be attacked. So if you take some metric, like what are the nodes willing to accept? For an attacker, it is possible to spawn thousands of nodes advertising different values. So imagine you have this room of people, and for an attacker, it is possible to push in as many people as he wants to. These people are just his puppets, and these puppets can end up holding placards of any value. nothing in the Bitcoin system prevents this, right? So why do you think that will work as a signaling mechanism for miners when attackers can spawn nodes at zero cost,

or not at zero cost, but at some cost, but a willful attacker can spawn nodes easily? Okay, so that's a great question. And I think it comes down to the difference about, you know, what's possible in theory, and then what actually happens in practice. So in theory, you're right. There's no way to know with 100% certainty that those people holding signs are actually legitimate Bitcoin users and that they're not just people that were paid

for brought in on the bus to hold these signs to get the bigger blocks that their puppet master wants. So, yeah, there is this sort of fundamental uncertainty. But if you imagine in this picture with actual real people holding signs. And you have this rally, and everybody's there. They all seem passionate. They're all holding up signs for 8 megabytes. The miners can talk to them. They can email them.

They can call on the phone. They can shake their hand if they're in person there. And then suddenly, this bus comes in with all these people holding one megabyte signs or holding 100 megabyte signs, and they all jump off the bus together. They stick around for a couple hours and leave. and when you talk to them, they act strains, they don't seem to know what they're talking about, and other odd behavior. Well, it's pretty convincing that, well, maybe those weren't real node operators.

So I think in practice, because there's many ways that minors can get the information of what the majority of the network wants, by talking to people, going to conferences, reading Reddit, reading academic papers, and also the signaling I'm talking about in the user-agent string of the Bitcoin and Limited nodes, that although they will never know for sure that their block will be accepted when they try to push above one megabyte,

at a certain point it will become de-risk enough that a minor will say, well, you know, even if I'm wrong, and I highly don't I'm wrong, I'm only risking 12.5 megabytes. So heck, let's try this 1.5 megabyte block and see if it gets accepted. And that cannot be fixed. So once that block is accepted in the blockchain, so now it's done what you said. And now it's eventually gets built on by by other blocks. And now we have proof that cannot be civil attacked that the network is, is ready to accept blocks larger than one megabyte.

Yeah. And of course, we also, besides nodes signaling, we also have minor signaling, right, with which version they support. And currently, Bitcoin Unlimited is around 20%. And Bitcoin SecWit, core with Segwit, is also around 20%. A bit over. I'm curious, actually, on that point, of the people running unlimited nodes, is there a default block size? And what is that? And how many people have changed that to some other value?

Yes, the default block size is 16 megabytes. And that's what we recommend for node operators. If you're a miner and you're running Bitcoin unlimited, we recommend that you adjust that to a one megabyte block size limit. And that just kind of prevents this very unlikely a tactic that would be possible on minors. It's not possible for node operators. Yeah, yeah. How many people have changed it?

I don't know the numbers offhand. I think the majority are sticking with the default of 16 megabytes, but there's a website called nodecounter.com that talks about statistics for Segwit and Bitcoin Unlimited and the classic. And recently, the author of that site has updated it to include graphs of Bitcoin and Limited No statistics. So listeners can go there to take a look at the current. distribution of blocks as limits of the Bitcoin and limited nodes. Let's take a break to talk about the Ledger NanoS, the new flagship hardware wallet by Ledger.

I'll let Ledger's CEO, Eric Larchavec, tell you all about how simple the NanoS makes it to securely store all your private keys. The Ledger NanoS is our latest generation hardware wallet. This is a multi-currency hardware wallet. It has a screen and buttons to manage everything on screen. You can generate a new seed, restore a seed, or set up your pin on the device. Your seed will never be exposed to the host computer. On the nano-s you have different apps.

You have the Bitcoin app, you have the Ethereum app, and you have the Fido U2F app for strong authentication, for instance with Google, Dropbox or GitHub. You can manage your cryptocurrencies with the ledger wallet Bitcoin Chrome app, or the ledger wallet It's a Ethereum Chrome app. With the nanoS, all your Bitcoin and Ethereum addresses are derived from one unique seed. With one seed, you can have in the same time Bitcoin, Ethereum, ICAMC classic balances, and also if you restore your seed, you will also recover all the keys associated to other apps such as Fido U2F, SSH, GPG.

So it's very simple, just one seed and multiple applications. The nano-s sets the new standard in hardware wallet security and usability. You can get yours today at ledgerwallat.com. And when you do, be sure to use the offer code Epicenter to get 10% off your first order. We'd like to thank Ledger for their support of Epicenter. So what makes you think that Bitcoin Unlimited is going to continue to gain traction? Are you worried that Bitcoin Unlimited might just stay where it is, Segret might stay where it is, And we're at this kind of point where there is a set of different views that all have a significant minority,

but none of them really has support enough to change the protocol in their direction. I mean, nobody knows for sure, but I strongly suspect that, you know, if not by the end of 2017, surely about the end of 2018, we will have blocked larger than one mega. byte. So, so I mean, if that doesn't happen, I think that's an example of a failure case for Bitcoin. I don't like to think about that. I think we will prevail. But yeah, but yeah, nobody knows the future until it happens. Still, the way this happens, right? So we have this room full of people signaling different stuff. Let's say now Bitcoin unlimited, 60% of people are running Bitcoin unlimited nodes and they are signaling 16 megabytes on average. So a miner says, okay, I'm going to mine this bigger block now.

I mean, another 40% don't have that, right? So they might declare that an invalid block, mine somewhere else. So it seems like the chance of a network split at that point and a fork is very high. Do you see that differently? Yeah, I see it quite a bit differently. If we look at all the different hard-forking changes that have been implemented in the various alt-coins, I think the only example of where a hard-forking change

has ever caused a persistent blockchain split is the example of Ethereum and Ethereum Classic. Dog coin, Manero, and there are several other examples have hard forked many times, and they've always kept the network effect intact. And there's a huge incentive to be part of the, like to all come to consensus on a common history. So I think people underestimate how strong that effect is when the rubber actually meets the road in this time for people to make real decisions. I don't think we'll have this point where 60% of the nodes are signaling for,

or 4 megabytes and 40% are still not. I think at some point you reach a tipping point and it goes from 30% signaling for larger blocks all the way to like 80, 90% quite quickly. And then by the time the miners actually publish those larger blogs and the 10% of nodes that have an upgrade are partitioned from the network, I think they just say, well, I guess we're getting larger blocks

and they download a client that is compatible and they join the Bitcoin blockchain once again. And Bitcoin proceeds. Just the same was always done, but this time with a higher block size limit. And the beauty of this proposal is that because the market itself is handling it, we don't have this debate in the future.

Right, right. Yeah, this is, I mean, this debate has caused me lots and lots of stress. I really don't want to go through it again in a couple years. So I'm hoping that once we move to a system more like Bitcoin Unlimited, it never happens again and the network just evolves in some sensible fashion. And we trust that Nakamoto consensus works. This is actually the essential difference between something like Bitcoin Classic and Bitcoin Unlimited. So when somebody favors a solution which has a block size limit,

but just a higher one than 1MB, which is probably like... some people, some node operators, some miners, that solution has the potential to create a similar debate a few years down the line, whereas if the Bitcoin Unlimited solution is like shown to work and adopted, then you get rid of this debate altogether. Right, right. Bitcoin Unlimited, assuming it's accepted and it works, and I think it will be accepted and I think it does work,

is a permanent solution to the block size limit debate. Classic was a can-kick solution. And I don't think anyone, at least anyone from the big block side, really was 100% behind Classic. I think it was more of a compromise made in good faith just to try to do something to alleviate the congestion that we were facing. But since Classic wasn't adopted, I think most people aren't interested in it anymore,

and they would like to see a permanent solution, such as the one offered by Bitcoin Unlimited. So what would be ideal is like some alt-coin network would have a system like Bitcoin Unlimited and show that it works under all scenarios and make the decision process very easy for the Bitcoin network. But do you see like a test case appearing in some way or another where some other network might demonstrate behavior like this first and make it easy for Bitcoin? as a whole to adopt it? Well, that's a tricky question to answer because, like, what is Bitcoin a limitus proposal? Like, in a way, it's sort of a meta-proposal.

It's not proposing anything. What is doing is it's just allowing people to change their nodes block size limits more easily and hoping that through that ability and through some signaling that the network converges on a new block size limit. it. So to me, it's Bitcoin Unlimited. It is just Bitcoin how it's always been meant to be and how it really, how it, how it always was. It's just people for some reason didn't realize the power that they've always had. Bitcoin has always been defined by the code people are willing to run and the version of the ledger they want to value. I think that's a quote from Roger Murdoch, so I should credit him for that. Bitcoin core guys, right, are very much against Bitcoin Unlimited. Can you run me through? Why do you think that is?

Like, why is there so much such a big division and why is there so much resistance against these ideas? I have no idea. It just blows my mind the things that some of the, some from the small block camp, are saying against Bitcoin Limited and the idea. that somehow having larger blocks and more transaction capacity and more users is somehow a bad thing. I don't understand. So, you know, this is something that we discuss in our internal circles quite a bit. And there's basically three theories.

Theory number one is that they view the Bitcoin network, you know, as just like a bunch of automaton's running their software in a way that, you know, If the software is not perfect and everything is not perfectly tweaked, the network will just break. So theory number one is they view Bitcoin as something that's very fragile that needs to be managed from the top down so that its evolution follows a curated score, I would say. I think Gregory Maxwell has actually used the word curated before to describe how Bitcoin might evolve. So that's theory number one. Theory number two is that they have conflict of interest,

that's blinded them somewhat. And like for instance, Blockstream's business plan that a lot of core developers work for Blockstream is to monetize side chains, possibly Lightning Network, and other ways of transacting in Bitcoin the currency without using Bitcoin the blockchain. And if you work for a company that will succeed if those off-chain solution works and your paycheck will continue and maybe your stock options go up, I can see how that can make you biased to keeping the block size slow to put more pressure on those off-chain solutions.

So that's theory number two. Theory number three is that for some reason, or not for some reason, that they're actually actively being paid by the banking cartel to Bitcoin crippled. to prevent Bitcoin from growing. And they know what they're doing is causing problems, but that's the point of doing it. So I don't know if I really lean towards any one of those theories more than the other. Maybe they're all a bit true. Maybe that are quite true.

Maybe it's something else. But those are the three theories that are discussed most often in my circles. And there might also be one theory just in terms of, keeping power, you know, beyond any further motive, right? Yeah, that's actually a great point, too. And that's probably theory number four, for the last people mentioned that as well, that they just enjoyed the position of there only being one implementation of Bitcoin

and the power that comes with that, and they don't want to give that up. So I read that you recently got kicked out of the Bitcoin Core Slack Group. Can you talk about what happened there? And what has the kind of debate or the attempts at debate and conversation between you and the people around you and Bitcoin Core look like? Okay, well, they have recently let me back in. So I'm no longer banned from Bitcoin Core Slack. But yes, I was banned.

And it was kind of odd, really, because so I decided to join Bitcoin Core Slack. and just, you know, try to talk to people and just open up a dialogue, hoping that we could come to some kind of compromise. And I feel I was really starting to make headway. You know, we were starting to joke around, and we were seeing eye-to-eye and a lot of things. And then suddenly, when it seemed like my efforts were really starting to work, that's what I got bad.

Like, it wasn't like these angry debates. I was actually banned during, like, the time when, you know, there was all these productive discussions going on. We were joking around, having a good time, coming to compromises I thought, and then suddenly I'm bad. I'm like, what the heck? But anyway, so eventually they did let me back in, and I'm back in. But it seems like since that time, the divide between the small blockers and the big blockers is even more, is even broader. And the communication, I think, is worse than it's ever been.

That's not good to hear. I recently also heard that there was a block created by Bitcoin Unlimited that ended up being invalid because it was too big. Can you share about what happened there? Sure. So when we recently released Bitcoin Unlimited 1.0.0. and unfortunately there was a bug and it allowed when a miner used a custom coin-based transaction for the total block to be slightly bigger than one megabyte and bitcoin.com mining pool was mining with 1.0.0 and indeed one of their blocks was

1.0003 megabytes 23 bytes bigger than the one megabyte limit. And yes, it was deemed invalid by core nodes, and it was deemed excessive by some Bitcoin unlimited nodes, and it was orphaned by other Bitcoin unlimited nodes. So it was an unfortunate thing for us, just from a PR standpoint. You know, we had been working really hard to get that release out.

We had done lots of testing and lots of checks, and it's just something that slipped by at an unopportune time. but the damage was very minimal. Like in a way, it kind of showed how strong Bitcoin was. Like we had this bigger block because the majority doesn't presently accept bigger blocks. It was just orphaned, and it didn't cause anybody, any profit, I mean, any money. And nobody lost money except for the mining pool that produced the big block, just like it should be. Yeah, as you said,

That was perhaps a PR incident that Bitcoin Unlimited didn't need in these circumstances when probably it's fighting for bigger principles. But yeah. Yeah, it was actually funny, though, because, you know, it was definitely a downer now that a lot of people can say, oh, look, Bitcoin Unlimited. They don't know how to code. They produced a block that was bigger than what they meant to. But on the other side, just because it created so much publicity, like if you plot our node count, It's shot up right after that event. So maybe what they say in Hollywood is true,

that all publicity is good publicity. Perhaps, perhaps. But we take that event very seriously. We've published an incident report, and we're having an internal meeting on Wednesday to discuss the incident in more detail and see how our processes can be improved, so we're more likely to catch bugs like that

before we actually release the next version of software. Sure. That's good to hear. So apart from this Bitcoin Unlimited, sorry, apart from Bitcoin Unlimited, you're also involved in another project, and that is the project of Ledger. So tell us what Ledger is and what you're doing there. So Ledger is the world's first peer-reviewed academic journal for cryptocurrency research. It started as just kind of a half joke in the Bitcoin talk forum. There's a professor named Chris Wilmer from the University of Pittsburgh, and we just got sort of chatting one day, and we had the idea, oh, what Bitcoin really needs is a peer-review journal, so we can actually have authors publish their ideas, have those ideas go through peer-review,

and the journal published nice polished versions of these refined ideas that other people can use to build new research upon. So we worked on that for a long time, and we launched the journal through the University of Pittsburgh. I guess it was September 15, 2015, so maybe a year and four months ago. The reception at the time was really positive. On our editorial board, we have lots of... representation from great universities. We have professors from Oxford, Cornell, Stanford, Duke. We even have a central banker from the Bank of England who joins us at ledger. So, yeah, so that's what ledger is about. So people can write articles on new research in Bitcoin

and cryptocurrency, submit them to ledger. We send it out for PURG. review, the author gets a feedback with the decision, you know, accept, reject, or can, like, conditionally accept if they can address the critiques of the reviewer. And eventually, we, we published those articles. Our first issue was published in December of 2016, so a bit more than a month ago, and we are working on the next issue at the moment. So what was some of the highlights of the first issue? A highlight for me is that we had a really nice ballot spectrum of different topics. We had papers from all sectors of what I call Bitcoin knowledge.

So we have papers on cryptography. There was a paper on the ring signature confidential transaction technique that's been recently implemented in Monero. So something very technical. we had an economist from France published a paper on the Bitcoin mining game. He was discussing kind of a different take on the paper that my paper on the transaction fee market that we discussed earlier. So that's kind of in the middle, a bit technical and a bit economics. And then we even had a philosophy paper talking about governance and decentralization

and also is a great philosophy topic. So I was happy with the broad spectrum of. of submissions. I looked through Ledger the past month. I must say I personally found it potentially very valuable resource in scheduling the next interviews for Epicenter. Because I looked at Ledger and I was like, wow, like we have like these 10 or 15 papers and I can just interview all of all of these people.

One of the papers I really liked to see in that issue was, There was a paper that was talking about how you could build virtual worlds on a blockchain. And I found that, like, this team hasn't been covered anywhere else. And I really like that, you know, that this journal would accept something so radical. Oh, I'm glad to hear that. Internally, it was actually a bit of a discussion because we were slightly concerned that the topic might not resonate with. of our readership, because like he said, it is a bit obscure. But I'm super glad that we published it because the comment I just heard from you, I've heard from several people now, and they're like,

oh, it's great to see you publish that paper. You know, I'm super interested in the idea of virtual worlds tied to blockchains. So thank you. I'm glad to hear that. Cool. Well, Peter, thanks so much for coming on. It was a pleasure talking with you. Now, of course, we're going to link to Ledger and a few other resources like your academic papers. But if people want to get involved in the Bitcoin Unlimited, contribute to it. What's the best way they can do that? The best way to get involved with Bitcoin Unlimited is to join the Bitcoin Forum and the address for that forum is

Bitco.in. So B-I-T-O-I-N. That's a forum where a lot of Bitcoin Unlimited proponents hang out and discuss things and from there, If you get more involved, we can also give you an invitation to our Slack channel where we discuss more development, hardcore development related topics. Cool, excellent. Well, Peter, thanks so much for coming on. It was a pleasure. It was great to be here. Thanks for the opportunity, guys. And thanks so much to listeners for joining us. If you want to support the show,

then you can do that by leaving as iTunes to you. that helps new people find the show and is very much appreciated. And of course, you find this show and other shows on the Let's Talk Bitcoin Network on ListstockBitcoin.com. So thanks so much. And we look forward to being back next week.