Peter Rizun: A Bitcoin Fee Market Without A Blocksize Limit (Episode 172)

Transcript

[Music] this episode of epicenter is brought to you by Jax Jax is a userfriendly wallet that works across all your devices and handles both Bitcoin and ether go to jx. and embrace the future of cryptocurrency wallets and by The Ledger Nano s the hardware wallet which sets the new standard in security and usability get it today at Ledger wallet.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 faan Crane and I'm me Roy today we are joined by Dr Peter Ryon who is associated with the Bitcoin unlimited movement let's say uh uh Dr Peter has uh 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 uh fee Market in Bitcoin and

what it means for the block size debate so before we we start let's have an introduction from from Dr run so Peter could could you give us bit about your background and have and how you got to be interested in the Bitcoin space sure so I come from the uh the physics engineering side of things and uh I I I work in the high-tech sector right now but I've always been a bit of a hobbyist or or armchair Economist so before I learned about Bitcoin I was reading the various logs I was a bit of a a gold bug you might say and I really felt that we needed some solution to make a new money system for the the

Modern Age a money system that would be more fair than our our inflated Fiat currencies and more useful than something like gold that can't be transmitted over the Internet so um I guess it was early March of 2013 when on the radio I heard uh that the price of Bitcoin had hit some new all-time high because of the Cypress confiscations of bank accounts and I'm like I keep hearing this thing Bitcoin I I had never really taken it seriously so I'm like that's that's interesting I've heard that a couple times so I'm going to go home and and learn about it so so I did and I hit uh upon satoshi's white paper and 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 uh to solving major problems that were actually thought impossible so for about two weeks I did nothing but read everything I could on bitcoin and and ever since then i' I've been I've been hooked awesome yeah I think we a lot of us had that experience it's interesting that you mentioned before you were looking for a fair way of mining so did you can can you explain that I mean I I understand that people say like okay I have an issue with F money or with uh quantitative easing and the way the financial system works but

but why did you think of mining in that context did I say mining I I think I meant a fair money system yeah okay there no problem so another thing I wanted to to ask about brief so you're a physicist by background and I remember reading quite a bit uh with speculations about Satoshi that he might have been a physicist as well these people were quite uh most people I think think he he is not a computer scientist right cuz his uh programming was often not so clean but but people would say okay he for for various reasons he he might be a physicist so I'm I'm curious when you look at Bitcoin you uh from the eyes of a physicist do you does that make sense

to you do you feel like it is built the way a physicist would approach this or I I haven't thought about that uh until you answered asked me that right now but I I I 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 the natural world uh but by 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 um so yeah I I I think maybe he was he was a

physicist I mean it clearly has you know like you said examples of computer science but there's also economics mixed in and a whole bunch of fields that had to come attempt together to create Bitcoin and maybe maybe a physicist has has a better higher higher level view to to combine all those ideas uh 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 like 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 I got to call the time uh history uncertainty principle so a block was recently mined and then more and more transactions are being added to the me pools of all the notes and the state of The Ledger becomes less and less certain As Time grows on eventually a minor finds a block and in a certain way that collapses the wave function and uh we get more certainty of the of the history of the 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 is built so you can so back to the time history uncertainty princip you can know the

time of transaction very quickly but you don't really know for sure what'll be con confirmed but if you want to know the whether it's confirmed to a high degree of certainty then you have to wait a whole lot of time to pass so so I I think that's a cool analogy to physics yeah I mean uh in like I think in quantum physics uh 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 is a point at which you can argue that the electron is like moving through two slits it's in a state where it's it's physically present at at both slits and you in order to describe it states you have to consider that it

might be taking both parts at the at the same time right and so you 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 is the next block is M that you know uh it reduces from that point of uncertainty confirmed unconfirmed 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 that's exactly right so uh let's let's probably move on to uh the topic

of our conversation uh that is the fee Market of Bitcoin right so uh like your paper is is is very interesting so what your your paper says in in essence is um that even if we had no block size limit there would be a transaction fee Market with with transactions uh with there being some kind of transaction fee levied on on transactions tell us why you spent so much maybe so many hours writing this paper like what 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 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 uh 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 1 Megabyte the average block size was significantly less than one megabyte uh when Satoshi put the limit in place the average block size was something like 5 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 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 uh from Gavin

Anderson who is talking about um if a miner includes a bit more transaction in his 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 minor will find a competing block and now you have two blocks of the same height and as you know only 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 uh gin and at the Bitcoin conference in maybe 2014 or something

like that and asked him exactly that question uh and especially I asked him about what he what he thought the cost was of uh including an additional transaction in terms of increasing the orphan rate and and back then he said he told me he thought it was 8 cents I mean I guess today it maybe maybe the Dynamics have changed with full blocks uh is there still a way to say um or or maybe not I don't know maybe you know 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 you know 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 in my very simple paper is that the uh the the fee in uh like Satoshi per bite is proportional to the inflation rate of Bitcoin like how how big the blocker word is multipli by what I something I call the transaction and P 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 the cost of including an extra transaction due to orphaning risk at about 8 cents uh per transaction and that makes a lot of sense uh and if you look at some of the

the the the numbers in my paper they align with that as well and what's happened since then uh 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 one or two cents uh for the cost of inclusion of a transaction uh but like I said earlier as we can improve the rate at which information can be propagated the 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 cuz cuz you know today have

this limit of 1 Megabyte and what is the cost of transactions on average like 20 30 cents right it's gotten very high uh so if if you say that kind of you know the supply curve in a way of a trans producing transactions like 1 cent then I I guess somewhere that intersection would be I don't know do you have an estimate there 3 4 megabyte today or or 2 megabytes or 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 around 1.5 to 2 megabytes in size and the average transaction fee would probably be between one and three cents would be

my guess so the the core of the argument here is that for a minor so when I'm a minor uh and I am receiving transactions that are waiting to be confirmed that that are sitting in my M Pool and I'm generating a block uh I have to decide how many transactions to put in my block and out out here I am situ I'm facing a situation of both risk and reward in a sense right so exactly so the the reward is if I put more trans 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 a block and then there's a risk side the risk side is the more transactions I put in my block the uh the greater the risk that I broadcast the block and my block moves slowly around the network therefore there's a there's a risk that some other minor is going to generate a block let's say 5 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 uh and his block is the one that will get included in the chain so so basic basically I I'm facing a situation of like Risk and reward and and there is some point there is some point at which uh 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 minor this is this size I should create optimally in order to balance both my risk and reward is is that a right way to frame it yeah that you did a you did a very job very good job probably better better than I could do um yeah exactly if you if a minor publishes an empty blog 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 he could let's assume there was no block ASM it for a second he could make a 20 megab block and full

full of all this lowcost fees and maybe the fees plus the block instead of being 12.5 Bitcoins is now equals 14 Bitcoins so you know he's up by 1.5 Bitcoin but if the block is so big that certain nodes uh AR aren't relaying it uh other miners take a long time to validate it there's like you said there's a good chance that another minor 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 uh analytically describe how that equilibrium works and how there will be

a point on the curve where by mining a smaller block The Miner is leaving fees on the table but by mining a bigger block he's risking too much uh a chance that his block will be orphaned so at that sweet spot there's a certain block size sweet spot where the Minor's profitability is a maximum [Music] let's take a short break to talk about Jax Jax is your wallet your complete user interface to cover all your blockchain needs I've been using it and I've been loving it now Jack supports a lot of different cryptocurrencies suppos Bitcoin ether Litecoin ethereum classic zcash or rep and they're adding many more keep responding to users needs now

with jackon 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 word seed that you can use to back up 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 you can there are even browser extensions for Chrome and Firefox and on top of that in Jack 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 cryptocurrencies accessible for the masses easy to use for the masses make sure sure to get your own Jax wallet at Jax iio or you can get it from any of the app stores you are using we' like to thank Jax for their support of EP Center one of the key dynamics that is required I mean we we touched on it briefly here for this model to work is that creating bigger blocks has a cost right so we we we've talked about that there have been a few efforts right there's like Xin which we talked about in a previous episode with Bitcoin unlimited which is

essentially a way to like propagate blocks faster and and uh propagate less information for blocks to to go around and then there's the Bitcoin relay network uh and there there's some other efforts too that all go towards yeah propagating it faster and it it's not I think inconceivable to imagine that one will figure out ways so that there is actually no increased orphan risk for bigger blocks do you think that's a real risk and and do you think that would be a problem for the fee Market you've described here well I I got a whole bunch of things to say on

that topic so you you mentioned exen first which I think I think exen is is a fantastic technology which allows miners to propagate their blocks more quickly and the testing that that that we did showed that Exon uh 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 eight so that uh has the effect of decrease the propagation impedance that I talked about that the drive is a price of block space so yes an innovation like Xin does um make it faster to propagate blocks

and therefore uh make fees under an orphan RIS 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 are are free to to use Bitcoin for for making transactions uh so right right now I I I mentioned earlier that I figured the equilibrium fee without a block saym it would be between 1 and 3 cents um but but but now imagine that the price of Bitcoin jumps from $1,000 where it is today to $10,000 because fees are denominated in Bitcoin the average transaction fee would therefore increase from 1 cent to three cents up to 10 to 30 cents and now you're you're

getting pretty pricey again so all these Innovations like Xin that allow miners to more efficiently propagate their blocks and reduce orphan RIS are good because they they drive fees down while the increasing price of Bitcoin is driving fees up so so so that was uh my 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 uh there's been a lot of debate about U between myself and Gregory well in particular and to answer that question more definitively I wrote a paper called subchains and it's basically analyzing

how the fee Market would change if miners cooperate to build blocks layer by layer over the 10minute block interval uh using weak blocks so basically these an the idea of prec consensus where the miners are agreeing what's going to go into the block before they actually find it so when they find the proof of work most of the transactions they uh that are including in the block have already been propagated so what was interesting from that work is that even with those prec 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

got to you 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 week block the the higher the orphaning risk as well so 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 uh which says uh we should do away with the block size limit uh I think any any proposal uh needs to answer for at least two SE 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 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 fix fair so okay what do I mean by Fair here so so ideally what you want is a system where no matter if you are a small or a large minor so case one might be I'm a minor and I have only half a percent of the hashing power I'm a small minor case two might be I'm a large

minor 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 would be mine in in in expectation over a long time time frame 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 right so and and you you want you want this game to remain fair for for 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 does it make no difference or does it make some difference do you have any insight there yes I I have I have some ins so so the first thing I'll say is um personally and uh Bitcoin unlimited we we're not trying to remove the block size limit completely uh we just want to make it a market driven block size limit uh but academically we can argue what would happen if there if there was no block size limit which I think is the question your answer asking and the question was H how would the fairness in the Bitcoin mining game change and and that's an important question and it would change I mean 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 minor and I have 10% of the hash rate and I propagate a block and let's say that block takes 6 seconds to propagate so all the other miners they're going to start mining on that block 6 seconds later but me I'm going to start mining on the next block right away so I have a sixc head start on all the other Miners and 6 seconds is 1% of the uh 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% of the hash rate the chances of me

solving the next block is 10% so I have so 10% of the times I have this 1% Advantage a smaller minor who only has 1% of the hash rate he has that 1% Advantage only 1% 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 spread across the network uh and ultimately what the average orphan rate is so to to me it's it's not a big deal like Gavin Anderson ran the numbers and the the advantage to a 10% minor over a 1% minor in like a one or two% uh

orphan rate environment was Tiny like it was it was insignificant compared to the advantage that some miners have in terms of lower electricity costs and especially in terms of of just the random Revenue fluctuations due to some miners being more lucky than others so so yes there is a small theoretical Advantage but until orphan rates are like 25 or 30% which I don't think we'll ever see I don't think that Advantage is ever really significant enough to affect the health of the mining ecosystem you you mentioned in your paper that one of the assumptions uh you have 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 to zero at the moment the inflation is 4% it's going to go to 2% by 2021 to and under 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 or you just need a nonzero inflation what does that Dynamic look like yeah so that's a 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 uh 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 like when we do have a block reward so I assumed a priority that there was a block reward and that was one of the parameters in my model um 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 you 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 to to answer that that question uh so 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 um Assurance contracts like what my K talked about I'm not sure uh 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 end game 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 know I I very much agree with that sentiment I think it's very important that one you know solves the problems that exist today and doesn't like not solve the problem that exist today for fear of creating other problems that might only actually be problems 20 years down the line exactly we we we right now we do not know whether that will be a problem at all I think there's there's still the question

though which I I think is one of the point of view informing people critical of bigger blocks which is that the block is going down uh and so the minor revenues do decrease from that side I mean the Bitcoin security model kind of game theoretically it you we do have a relationship between uh 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 the miners and is that a good thing for security do you think like with um no uh block size limit that you know bigger blocks would

lead to lower revenues of Miners and less security what what's your point of you that that well I think there's two competing phenomenon like if we uh 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 agregate than 1 million people using Bitcoin paying higher fees uh 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 uh what the optimal block size limit is

I don't know but 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 um from your statement like you're in favor of having the market kind of decide or give mechanisms for 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 unlimit approach Bitcoin

unlimited sorry is is the project that's taking this this idea and putting it into practice so explain to us like what Bitcoin unlimited is is doing sure so Bitcoin unlimited I would say is about two main things and one is uh peer-to-peer electronic cash we feel that uh in the last couple years the the Bitcoin has been steered in the wrong direction away from electronic cash and focusing more on digital gold or or as a settlement only Network so one of our goals with Bitcoin unlimited is to uh bring back Bitcoin to the thing we all got excited about when we read that

white paper uh I told you guys about reading that at the very beginning of this talk so we want more on capacity to uh to allow more users more transactions uh 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 uh confirmed and we want to uh get rid of all this idea of replace 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 uh for many use cases and uh we'd like to bring that back and in fact enhance the security and usability of a zero from transactions so so that's Bitcoin unlimited goal in terms of PE

peer-to-peer electronic cash uh the other goal of Bitcoin unlimited is decentralization and decentralization is seems to be a buzzword and it seems to mean different things to different people uh 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 want more key holders 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 node implementation to run and ultimately more say or more vote in how

the network evolves so those are the goals of pick one unlimited so explain to us this word unlimited like uh so so when I hear the word unlimited um the thing that comes to my mind is hey there was like Bitcoin classic with a 2 MB limit there was I don't know some other Bitcoin with an 8 MB limit and unlimited means no limit so does 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 uh they they they hear Bitcoin AIT and they scream oh my God we're going to have gigabyte blocks in 2018 but 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 to evolve uh in a more decentralized bottom up fashion as opposed to being dictated from above by core Dev um so I how do we do that um so so 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 their node will accept uh with core if you wanted to do that you'd have to recompile the code which a lot of people can't do with bitco unlimited it's just a it's just right there in the in the gooey it's very easy to increase your nodes block size limit today so how will the limit be practically set so we so so so imagine like okay there's there's this network in which a lot of people are running the Bitcoin unlimited code each of them is individually setting their own block size limits so I run a full node and I say I'm willing to take 4 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 H how does the network actually settle on a particular particular number well I think it'll ultimately be the miners that settle on the numbers um I think the miners will look and think well most of the network seems to be happy with up to 2 megabyte blocks so let's miners agree to uh to make 2 megab blocks our block size limit uh from the perspective of nodes uh it doesn't matter like my node has a 16 megabyte block size limit and it's had that limit since uh since the summer time and it's

it's tracting 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 um so this concerned that different nodes are setting different values of block size is is kind of a 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 the consensus of the blockchain through something we call the U the excessive block gate that's what we call it but but that's that's that's getting kind of kind of technical um I I I think what people need

to think about when they uh 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 coordinator find some way to make that happen uh and we were looking towards cordev to help that but it doesn't seem like they're doing so so so now it's like how do we Empower users to

to you know to make a difference themselves so I like to imagine you know you got a couple minors all over the world and and instead of looking at nodes they're looking at a bunch of people and imagine these people all holding signs and right now 90% of those people which are note operators are holding a big sign that says 1 Megabyte I'm only going to accept 1 Megabyte so 10% are already holding something that says 4 megabytes or 8 megabytes or 16 megabytes those are big unlimited node operators but they're still in the vast minority right now so when a minor 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 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 they're going to 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 minor 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 and the problem is largely solved we have come actually to what I think is uh a good way to think about this problem right so so uh so let's let's let's let's take this this imagination further so so the idea here is you imagine like a room full of node operators and each of them is holding a sign one one megabyte 2 megabyte 4 megabyte and I as a minor in some senses I have a camera with which I can I can see this room full of people and I can read uh what they're willing to accept M right and now 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 m megabytes I I'll be willing to produce some up to 8 megabytes right if if producing up to 8 megabytes maximizes my profit which is like which is defined in in your paper right yes yes of course you can always produce smaller blocks I can always produce smaller blocks now see but like one challenge with this is so in Bitcoin what is ultimately secure what is ultimately secure is the uh veracity of trans of a transaction that is buried under a lot of blocks everything else in the Bitcoin Network can be attacked so if you 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 right so 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 the these puppets can hold end up holding uh placards of any value nothing in the Bitcoin system prevents this right right so why 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 a willful attacker can spawn nodes

easily okay so that's a that's a 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 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 the bus to hold these signs to to get the the bigger blocks blocks that their their Puppet Master wants so so yeah there is this sort of fundamental uncertainty but but if you imagine uh this this picture with actual real people holding signs and you have

this rally and everybody's there they all seem passionate uh they're all holding up signs for8 megabytes um you you the miners can can talk to them they can email them they can call them the phone they can shake their hand if they're 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 strange they don't seem to know what they're talking about and and other odd behavior well it's pretty convincing that well maybe those weren't real note operators so I think in practice because there's many ways that miners can get the information

of what the majority of the network wants um by talking to people going to conferences reading redddit uh reading academic papers uh and and also the the signaling I'm talking about uh uh in in the user agent string of the Bitcoin unlimited nodes uh 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 uh der risked enough that a minor will say well you know even if I'm wrong and I highly doubt I'm wrong I'm only risking 12.5 megabytes so heck let's try this 1.5 megab block and see if it gets accepted and that cannot be fix so once

that block is accepted in the blockchain so now it's done what you said and now it's it's eventually gets built on by by other blocks and now we have proof that cannot be Cil attacked that the network is is ready to accept blocks larger than one megabyte yeah and of course we also besides node signaling we also have minor signaling right with which version they support and currently uh Bitcoin unlimit is is around 20% and uh Bitcoin uh SE core with seg is also around 20% or a bit over I'm curious actually on on that point of the people running unlimited notes how many is there default block

size and 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 uh for node operators if you're a minor 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 attack that that would be possible on miners it's not possible for node operators yeah yeah um how many people have changed it uh I don't know the numbers off hand I think the majority are sticking with the default of 16 megabytes but um there's a website called node counter.com

um that talks about statistics for segwood and Bin unlimited and classic and recently the author of that site has updated it to include graphs of Bitcoin unlimited node statistics so listeners can can go there to take a look at the current uh distribution of block size limits of the Bitcoin unlimited nodes let's take a break to talk about The Ledger Nano s the new flagship Hardware wallet by Ledger I'll let Ledger's CEO Eric larc tell you all about how simple the Nano s makes it to securely store all your private Keys The Ledger Nano s is our latest generation Hardware wallet this is a multicurrency hardware wallet it has a screen and

buttons to manage everything on screen you can generate a new seed restore your seed or set up your pin on the device your seed will never be exposed to the host computer on the Nan you have different apps you have the Bitcoin app you have the ethereum app and you have the pho UF 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 ethereum Chrome app with the Nanos all your Bitcoin and eum addresses are derived from one unique seed with one seed you can have in the same time

Bitcoin ethereum ethereum classic balances and also if you restore your seed you will also recover all the keys Associated to other apps such as Pho 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 Ledger wallet.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 uh Bitcoin unlimited is is going to continue to gain traction or are you worried that Bitcoin unlimited you know

might just stay where it is seg might stay where it is and we're at this kind of point where there is a a set of different views that all have a significant minority but none of them really has uh 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 by the end of 2018 we will have blocks larger than one megabyte so so I mean I I if if if that doesn't happen I think that's an example of a of a failure case for Bitcoin uh I don't like to think about that I I think

we will prevail um but yeah but yeah nobody knows the future until it happens still the the the way this um happens right so 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 gonna I'm gonna mine this bigger block now I mean another 40% don't have that right so they might uh declared that an inval Block M somewhere else so it seems like that the chance of a network split at

that point uh and a fork is very high do you see that differently yeah I I I see it see it quite a bit differently um if we look at all the different hard forking changes that have been implemented in various altcoins I think the only example of where you've actually where a hard foring change has ever caused a persistent blockchain split is the example of ethereum and ethereum classic um dogcoin Monero uh and there's several other examples have hard forked many times and they've always kept uh the network effect intact and um there there's a huge incentive to be part of

the like to all come to consensus on a common history so I I I I think people underestimate how strong that effect is when the rubber actually meets the road and it's time for people to make real decisions I I don't think we'll have this point where 60% of the nodes are signaling for 4 megabytes and 40% are still not I think some point you reach a a Tipping Point and it goes from you know 30% signaling for for larger blocks all the way to like 80 90% quite quickly and then by the time the miners actually publish those larger blocks and the 10% of nodes that haven't upgraded are partition 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 as always done but this time with a higher block size limit and like kind of the beauty of this proposal is that because the market itself is handling it um we don't have this debate in the future right right right like yeah this is I mean I this debate has caus me lots and lots of of stress I 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 the network just evolves in some sensible fashion and we

we we we trust that Nakamoto consensus works this is actually the ENT potential difference between something like Bitcoin classic and Bitcoin unlimited right so when somebody favors a solution which has a block size limit but just a Higher One than 1 MB which is probably like some some people some node operators some miners uh that solution has the potential to be to create a similar debate a few years down the line whereas if the Bitcoin unlimited solution is is like shown to work and adopted then then you get rid of this debate Al together right right Bitcoin unlimited assuming it's accepted and it

works and 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 a compromise made in good faith just to try to do something to alleviate the congestion that we were facing uh 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 altcoin Network would have a system like Bitcoin unlimited and show that it

works you know under under all scenarios and make the decision process very easy for the Bitcoin network but do you see like a test case appearing in in some way or another where uh where some other network might demonstrate Behavior like this first and make it easy for Bitcoin as a whole to to adopt it well that's a tricky question to answer because like what what is bitcoin unlimited proposal like in a way it's sort of a meta proposal it's not proposing anything what it's 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 uh converges on on a new block size limit so to me it's it's Bitcoin unlimited is just Bitcoin how it's always been meant to be and how it really how 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 uh from Roger Murdoch so I should credit him for that the Bitcoin core guys right are very much against Bitcoin unlimited can you run me through why do you think

that is like what why is there so much such a big division and why is there so much resistance against these ideas I I have no idea I it just lows my mind the things that some of the some from the small block Camp are are are are are saying against bitco limited and the idea that somehow having larger blocks and more transaction capacity and more users is is is somehow a bad thing I I don't understand so you know this is something that we discuss in our internal circles quite a bit and uh there's basically three theories Theory number one is that they view the Bitcoin Network um

you know it's just like a bunch of automatons running their software in a way that you know if if the software is not perfect and everything's not perfectly tweaked that 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 uh so that its evolution is follows a curated score I would say I think Gregory Maxwell has actually used the word curated before to to to describe uh how Bitcoin might evolve so that's Theory number one Theory number two is that they have conflict of interest that's blinded them uh somewhat and um like for instance blockstream's business plan

that a lot of core developers work for blockstream is to uh monetize side chains uh 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 offchain solution works and your paycheck will continue and maybe your stock options go up I can see how that can make you biased uh to keeping the block size slow to put more pressure on those offchain Solutions so that's Theory number two Theory number three is that for some reason they're or or not for some reason that that they're actually actively

being paid by the bank cartel to bitcoin crippled to uh prevent Bitcoin from growing and uh they know what they're doing is is causing problems but that's the point of of doing it so I I don't know if I really lean towards any one of those theories more than the other maybe maybe maybe they're all a bit true maybe maybe n are quite true maybe it's something else but uh those are the three theories that are discussed most often in my circles and and they might also be one to just in terms of um uh keeping power know uh yeah yeah Beyond any further motive right yeah that's that's actually

a great Point too and that's probably Theory number four I've heard lot people mention that as well yeah that they just enjoy 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 course slack group can you talk about what happened there and and what has the kind of debate or the attempts at debate and conversation between you and the people around you and biton core look like okay well uh they have recently let me back in so I'm no longer banned from Bitcoin course Lo but yes I I was banned

and it was kind of odd really because so so I decided to uh to join Bitcoin course lock and just you know try to talk to people and just open up a dialogue hoping that we could uh 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 that's what I got banned like it wasn't like these angry debates I was actually banned during like the time when uh you know there was all these productive discussions going on we were joking around having having a good time coming to compromis as I thought and then suddenly I'm bad I'm like what the heck but um but anyway So eventually

they they did let me back in and and I'm back in but uh but it seems like since that time the The Divide between the small blockers and the big blockers is uh is even even more uh is is even broader and uh the communication I think is is worse than it's ever been that's that's not good to to hear yeah um I recently also heard that uh there's been there was a block created by Bon unlimited that was uh ended up being invalid because it was too big can you share about what happened there sure so when we we recently released Bitcoin

unlimited 1.0.0 and unfortunately there was a bug and um it allowed uh when a minor used a custom uh coinbase transaction for the total block to be slightly bigger than one megabyte and bitcoin.com Mining pool was mining with 1.0.0 and uh indeed one of their blocks was 1 Point 000023 megabytes 23 bytes uh bigger than the 1 Megabyte limit and yes it was uh 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 it was an unfortunate thing for us just from a PR standpoint uh 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 un opportune time uh but the damage was was was was very minimal like in a way it kind of showed how strong Bitcoin was like we had this bigger block uh because the majority doesn't presently accept bigger blocks it was just orphaned and it didn't cause anybody any profit and I mean any money uh and and nobody lost money except for the mining pool that

produced the the big block uh just like it it should should be yeah as you said that was a that was a perhaps a peer incident that Bitcoin unlimited didn't need in in these circumstances when probably it's fighting for bigger principles but yeah yeah it was but it's actually funny though cuz um you know it was definitely uh a downer now that a lot of people can say oh look Bitcoin un limited 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 if you plot our node count it's shut up right after that event so so maybe what they say in in Hollywood is true that that all all publicity is

good publicity but perhaps but but it we take that event very seriously we've uh published an incident report and we're uh having an internal meeting uh on Wednesday to discuss the incident in more detail and see how our processes can be improved so we're more likely to catch buges like that before we actually release the next version of software sure that's that's good to hear so apart from this Bitcoin unlimited uh sorry apart from Bitcoin unlimited you you're also in involved in another um another project and that is the project of Ledger um so tell us what what Ledger is and what you're doing there so Ledger is

the world's first peer-reviewed academic journal for cryptocurrency research uh it started as just kind of a a half joke in the Bitcoin talk Forum uh there's a professor named Chris Chris Wilmer um from the University of Pittsburgh and we just got sort of chatting one day and and we had the idea oh what Bitcoin really needs is a peer review journal so we can actually you know have authors publish their ideas have those ideas go through peer review and uh the journal published nice polished versions of these refined ideas that other people can use to build uh new research upon so we worked on that

for a long time and we launched the the journal through the University of Pittsburgh um I guess it was September 15 2015 so maybe a year and four months ago U the reception at the time was really positive on our editorial board we have uh lots of uh representation from great universities we have professors from Oxford uh Cornell uh Stanford Duke uh we even have a central Banker from the bank of England uh who joins us that ledger so um yeah so that's what what Ledger is about so people uh can write articles on new research in Bitcoin and

cryptocurrency submit them to Ledger we send it out for peer riew um the author gets a feedback uh with the decision you know accept reject or can like conditionally accept uh if if they can address the critiques of the reviewer and eventually we we we published those articles U our first issue was published in December of of 2016 so bit more than a month ago and we are working on the next issue at the moment so what were 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 uh uh cryptography uh there was a paper on the the ring signature confidential transaction technique that's been recently implemented at Monero uh so something very technical we had a um an economist from France published a paper on the Bitcoin mining game uh He was discussing kind of a different take on on the paper that uh my paper on the transaction fee Market that we discussed earlier um so that's kind of in the middle a bit technical uh and a bit economics and then we even had uh a philosophy paper uh talking about about uh governance and decentralization and and all sorts of great philosophy topics so so I was happy with the with the broad spectrum of of uh

submissions I looked through Ledger uh the past month I must say uh I I personally found it a potentially very valuable resource in uh scheduling the next interviews for epicenter cuz uh I I looked at Ledger and I was like wow like like we have like these 10 of papers and I can you know just interview all of all of these people one of the papers I I really like to see in that issue was uh there was a paper that was talking about how you could build virtual worlds on a blockchain and I and I and I found that like like this this theme hasn't been covered anywhere else and I I I really like that you know um that this

journal would accept something so radical oh I'm glad to hear that uh internally it was it was actually a bit of a a discussion because some we were slightly concerned that the topic might not resonate with enough of our readership because like you said it is 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 was great great to see you publish that paper you know I'm super interested in in the idea 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 um involved in Bitcoin unlimited contribute to it uh what's what's the best uh 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 doin so bit co.in uh 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 uh give you an invitation to our slack channel uh where we discuss uh more more more 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 for listeners for joining us if you want to support the show then you can do that by leaving us iTunes interiew 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 uh on lin.com so thanks so much and we look forward to being back next [Music] week [Music]

no