Rust-Bitcoin, Threats to Bitcoin & Bitcoin Data Science

Hello and welcome to the Stephan Livera podcast focused on Bitcoin and Austrian economics. I'm your host, Stephan Livera. This is episode seventy-three with Tamas Blummer, and he has a really interesting and varied background. I'll let him get into it as part of the interview, and, it's just a really wide-ranging conversation talking about Bitcoin scaling, Rust, Bitcoin, threats To Bitcoin and also Bitcoin data science. Here's the interview. Thomas, welcome to the show. I've, I've heard a lot about you, and obviously, I know you've got quite a storied and long history in Bitcoin, so thanks for coming on.

Thank you very much for the invitation. I'm very glad to be with you.

Excellent. So, look, I, I know some of my listeners might not know as much about you, and I know you've recently done a lot of work with, Ademant Capital and data science, but I know you've got

well, I started quite early with, with cryptography, I had a very, had an interest in it and, and worked in, on BGP long back in the nineties, and, then I worked in, investment banking, doing, pri- creating pricing models, for credit derivatives and similar, which later on, crossed again my career, and then, did trading automation there, portfolio risk management and- Very literally, I, worked on a ledger system of a big investment bank, and that's where I, I, I basically, met Bitcoin and recognized that that's, that's actually the ledger that, that really sparks my interest. so I, I quit there and, built, Bitcoin Enterprises. The very first one was Bits of Proof, which was a, an implementation of the Bitcoin protocol, in a manner, that was more suitable to integrate Creating an enterprise system, I had the idea at that time that this could be offered as a service. It certainly didn't work out, but, it helped me to, to deploy a few interesting services, like I created an exchange and, and, payment system And also the, the very first backend of the Trezor hardware wallet was, was, basically powered by BitsoPro. Then, then I used the software to, this software to, to run, one of the biggest, mining operations at Cointera, which was quite an experience, from, not from, from the, from that industry. unfortunately, this, this, This got, got ba-busted by the, by the low Bitcoin price in 2014. then I, joined, Digital Asset, which was, at that time led by Blythe Masters, who I knew from the very early days of, of investment banking. and the digital asset, I worked, mostly on, how to,

how to, combine the, the, the Bitcoin, type of ledger, what they call the blockchain, with, with a smart contract language that we were developing a digital asset that I think is really outstanding, to deploy this in, in, in traditional financial services, like at the ASX exchange, which is, the, the digital- Because it's ESX product project is probably one of the, or even the biggest, real life, deployment of the technology. then I, quit a digital asset for, for, m- Freely for health reasons, and that is also the reason why I'm only working on, on open source projects at the moment,

this And, and, and work on economic models, with, Edelman Capital, where I'm a, a advisor of. that's, that's basically my short story.

Excellent. And I can see from that you have a very varied background, right? You came from this banking world, but then you've also developed software for, you know, for, you know, using and interacting with Bitcoin, and then in, in, in the mining world as well. So quite a varied experience there. but I think it might be interesting as well. Just to chat a little bit on your thoughts around Rust Bitcoin, 'cause I know you're working on that alongside some others, such as Matt Corallo and Andrew Polster and some others.

Yeah. Yeah, since, since I'm no longer, working at a digital asset and, and retired basically to, to work on open source projects on, on my, as for fun, I focused mostly on building, a library type of Bitcoin implementation, with, Rust and, The, the projects I, I, I mostly contributed to is, is the Stratis Bitcoin implementation, which is, which is very, very, interesting because of the, the security and quality of the language and, people who are working on it, like Andrew Postrel, Matt, Matt Corallo and others, are, are quite serious about, about these projects and quite serious about the, the security promises it gives us. But started a, a Lightning implementation, which is also a li-library like implementation. So, I would like to en-emphasize that all these projects aren't really an alternative, for, to the Bitcoin node, but they are re-really, built to To help you to build, to, to create tools that interact with the network.

so my team at Corel started the Lightning implementation, and, I worked, started to work on the, mostly on the, on the underpinning of, of this Lightning implementation with a light, lightweight node. this, I, I call it more, more mobiles project. this is basically, an, an SPV like, mode, mode, but is, meant to be, using the BIP one hundred and fifty-eight standard, so it, it should, not have the privacy limitations that the current, SPV nodes have, but i-in, instead use the, the very new, block filtering method that is-- that will be offered by Bitcoin Core. There are pull requests also in Bitcoin Core that intro-introduce this service, and this Moomoo library even-- will, is already prepared to use it. And while, while I was building this library, I, I recognized that there is a, there is probably a, a need for, database to store, the Bitcoin blockchain, if you look at the, the current core, software, it is basically, storing the Bitcoin blockchain in flat files and then storing indices to it, in a LevelDB database. so it is a, it is a An approach of combining two different technologies, whereas, the features that LevelDB offers, are much more than, than those, actually needed by the software, actually needed to, to store the blockchain, since, for example, the blockchain cannot be edited, so there is no need for a delete operation. and if you have a software which, which have, features that aren't needed, then, it's very likely that this software isn't as optimal and as quick as it, it could be if you wouldn't have these features. So I started to, to create a database, I call it Hammersbald, which is a, a database that is only, able to insert, records into, into basically a, a, a persistent hash table and store, the underlying data, underlying blockchain data. And this, this, so this is the, these are the, these are the projects I'm currently working on in the Rust Bitcoin, community. If you, if you go to GitHub Rust Bitcoin, then you see quite a lot of other projects which I also

Excellent. And, just with the Merkle, could you outline a little bit there? So my understanding there is that, there's BIP one fifty-seven and one fifty-eight. One relates to the compact block filtering, and I think the other one doesn't relate to the serving of them, or could you articulate there how Merkle interacts with that?

one fifty-seven, fifty-eight, one is describing basically the, the network protocol and another, another is, is describing the, the actual construction of the filter. so they are, they are quite, tied to each other. And what, Mormail is doing is using this network protocol to communicate with, Bitcoin node, as soon as the Bitcoin nodes will be able to, to serve these compact block filters, and also is, implementing the, the, the reading and writing of these compact block filters. the reading because it's needed to be able to filter them, and the writing because as I develop The software, I, I needed something to test against, since the Bitcoin core node isn't yet serving these block filters, I, I, I also added the capability to, to normal to serve them just for test purposes. So it's basically running against itself.

Yeah, okay. And then how does some of these projects play into your vision of how Bitcoin will scale?

Well, I, I, I, I do think that, that Bitcoin's, s-- will scale into smaller and smaller- And, and more per-special purpose devices. so I, I pretty much understand why it is very important that we have, lots of full nodes and that we have the, and, a community, an enthusiastic community that ensures that, the forest of full nodes enforces all the rules of the network. On the other side, it is, it is quite likely that we will, we will have to be able to build, software for, for restricted devices which are, which would also like to con- connect to the network and do it in a, in a manner that is, even if it's not as trust, trustless like communicating with a Bitcoin node, but it's still, not depending on, central services or an, like the, like, like the current, infrastructure which is mostly using Electrum servers, servers, for example. there are people who think that connecting to an Electrum service is, is secure, but that's, that's quite-- I, I think it's quite a misunderstanding. It's, it's probably even worse than, than, doing SPV. the most, Most frequent critical on SPV node like that is normal is that it might, just basically blindly follow the miner, by, by, not, not checking other rules. that-- this is, this is also not quite true, because what, what set of rules, such a node really proves is, is, is, is not limited to the proof of work. So this, this, S- for example, normal in-includes the, the, Bitcoin Consensus Library from, from Bitcoin Core node. So it is basically sharing the s- shares the same Code base, checking transactions than the code that is checking transactions in a Bitcoin core node. So it is, it is capable of doing additional, checks on, on transactions that aren't just not covered by, by the SPV, but SPV, proof that is the first originally proposed by, by Satoshi, but it's, it's going much further than that.

Right. So let me just maybe just paraphrase that for some of the listeners who might not be as familiar. So in the Bitcoin community recently, there was some debate around this whole SPV versus Neutrino, and, one example actually, Nicola Doria came out with this idea that saying, "Oh, look, I actually don't mind if you, you know, hold your kit, like so to speak, from his own selfish point People held their coins out in exchange because then at least the risk isn't to the system, it was localized to that user. Whereas he was saying, "Look, if you just use-- if you use a wallet that blindly follows SPV, then there is a risk that, let's say, if SegWit three x happens, that your, wallet will just blindly follow what the miners do, creating that risk there that the miners would try to, let's say, create some fork of Bitcoin, and if people aren't, let's, you know, doing the correct-- running their own..." own full node, then the SPV wallet will sort of blindly follow. Now, as I understand you, Tamas, you're saying essentially one, that might not hold true in every case that, using certain libraries or using Mempool, as you mentioned, that there might be certain validation checks there as well, so it won't just blindly follow the most proof of work chain.

basically this, this, this decision whether to run a full node or run on SPV node or, or run, run a- Server, testing node, these are, quite black and white categories that, aren't, aren't really, I think, aren't really justified going forward. There is, there is really a scale between, between the, just, just trusting the, the work of the miner and doing the full validation, and this scale can be filled, with additional checks, provided you have, powerful libraries that are, that are That are adding, adding checking features that are in the core node. Certainly there are limits to that, but this is exactly what, what the, the Rust Bitcoin project is about, to, to create a library of, of functionality that, that, that, exactly, exactly follows or, the, the, the checks, done in the Bitcoin node, so, so you can build, build software that is, is as close as possible to the Bitcoin core node's validation. But it still probably makes compromises in some areas that are, that are not avoidable for your use case. So I, I think we will, we will, with time, and this is, this is certainly, there is some- there is some advantage of having a homogeneous network run by the single software, and, and certainly this is, this gives a very, very high, s-guarantee of consensus, and, and there is a risk of, of, of having, non-- having, having, agents running with other software, but- The, the, the fact is that we already live in, in this world, so we, we already have lots of different other softwares, connecting to the Bitcoin network using its protocol, which aren't core nodes. they are, they are, and, and these, and these software, has to make, h-have to make some, some trade-offs on, in which way they, they ensure that they, they, Stay in contact or stay in consensus with the network. I, I, I built such a system, such systems myself, and, and then usually what people do in the first step is, is they run a, run a core node, connected to the network, and then connect their software to the core node so they, they ensure the basic, the their software is also following, always following the core nodes,

consensus, but this, this setup is certainly very, very, resource intensive because you have then, the Anta Core node and additionally some data, data built by your implementation, which is Which is adding some, u-usual, us- some, some useful features. so I, I, I think we will, we will, get to the, to an environment with more hetero-heterogeneous, implementations, and, Well, the, the, the, the aim of this Rust Bitcoin, library is to make sure that, that these heterogeneous implementations still are very much aligned with each other, so they, they are, they're clu-- they are, this library will be easy to use to build your services, but they will be close to the Bitcoin core node

Excellent points, and I've, I've got two, points or questions to ring up on that, Tamas. firstly, maybe it might be good for the listeners just to understand what are some examples of these, let's say, might it be for low- Why would we need these sorts of things? An example might be lower power devices, such as mobile phones, or let's say that, P'tarigan, example with the new Nauta, understand, they were making a device. Do you have any examples there that you can share?

Yeah, lower Mobile devices or lower, resource restricted devices are certainly a, a, a good, example. but even if you, let's say you just need a A, an indexing service which is, which is just faster or in some other, in some way,

more intelligent than, what is available using the core indices. Even then, it would make sense to build a, a, a particular service to it, or a software that doing it instead of, as I said, doing this layered approach of, of, duplicating data between, the, the core node and then your, your indexing service.

Right, and so I guess that's an example where a Bitcoin business might use that kind of model as well. so another example I was just thinking as well, there was some discussion rat- similar to this time with all the SPV versus Neutrino versus full node, there was also some commentary which I thought was quite intelligent, was that maybe we also need to think of making it easier for an individual to pair their mobile device back to their full node back at home or to maybe- If they're not willing to run their own full node, to maybe their friend or so it's sort of like a semi-trusted scenario where maybe I trust my friend in the community, and he's the technical guy, maybe I'm a newbie, I trust him to run the node, and my phone pairs to his, device. Do you have any thoughts on that idea?

well, this, this is, these are certainly, scenarios which there are, like many wallets use this kind of setup where they, they promise to, or allow you to, to run your own service at home and then connect to it or to your friends. This is certainly also a path of, of, scaling the trust, within Bitcoin. But still, as I said, this is really a huge range, and we aren't, we aren't currently not exploring this range really. We, we, we are, many people are thinking in this community in, very black and white terms, and I, I think, yeah. We ha-we just have to ex-explore the possibilities and, and see what's, what works for certain use cases.

Okay, great. so let's now, talk a little bit about threats to Bitcoin. Now, you've been around for a while, I think, you might have some interesting insights to share. Did you have any thoughts on, potential, you know, soft forks that could happen or re-orgs that could happen as threats to Bitcoin?

Yeah, in, in our previous topic, we already, discussed

And is, is breaking the consensus, splitting the network. And, I think that there are, there are, a few misconceptions there too, like people, currently, consider, hard forks, evil and hard forks, demonstrably failing, like the Bitcoin project and others. So there is a consensus that this is probably a wrong way to do, or wrong, wrong way to go, and, and we, we want to avoid that. but there is a-- I think there's a misunderstanding, as if, soft forks would be much safer. that's not, no, that's not really true. So soft forks can, just as well split the networks as hard, hard forks. we had basically a few, If a soft fork is deployed, then nodes that are active, that are supporting a soft fork, have to actively reject a block that isn't complying the soft fork rules. And Rejecting it means, there is a, there is a chance of splitting the network. the, the, the only soft fork that we, we, we have seen until, until the, until now is, is the, the SegWit, im-deployment, which, where we actually didn't see this happening, because the, the miner were so, were really, so threatened, so, so they're faced with such a credible threat of the community that they, Unanimously moved and all upgraded, so we actually didn't even, even see the soft fork actually being Triggered by a block that is offending the soft fork. There wasn't, not, not a single block mined after, after that point that, that was basically offending the soft fork rules, and, and therefore we didn't even see that the software could have split the network. So we, and, and, and this, this missing experience created a, a, a perception in the, in the community as if soft forks would be safe, they wouldn't split the network, and, and that's not correct. They can just as well split the network as hard forks. And, this is certainly a, a very, very, very, pretty much the biggest threat I think to Bitcoin. It is the biggest threat because it is a technical possibility to split the network, and on the other side, it is also a technical necessity to do soft forks, because, there are very, pretty, pretty important things to fix in Bitcoin, which need a soft fork, just like in, in increasing the privacy of Bitcoin transactions, is, is Probably the best example. And at some point, at, at some point in the very near future, I hope these features will be, rolled out with the software and they, they will be, they'll be Hopefully going well, but I also think that this is, this capability of deploying features with software is also something that at some point needs to be challenged. so or have to be challenged. So at some point there will be a soft fork, which will not enjoy the same unanimous support of the entire community, or it will be just not well enough communicated so that some nodes will not be aware of what they have to reject. So, so at some point we will see, that the soft fork is splitting the network. And, a-at the point as we see it, the reaction to that will be probably very similar to than the reaction to the hard forks that we see, that the, the, this will also split the community, this will also, create, A technical appeal and, and at some point it will, basically lead to the, to the,

inside That, just as the hard fork window closed at some point by seeing that this, this isn't a viable, not a viable way to upgrade Bitcoin, the same way at some point the soft fork window will also close, and at some point, thereby, Bitcoin will basically ossify, and, the first layer will be then just us Impossible to change, then it is impossible to change the TCP/IP stack nowadays.

Excellent insight, and I think that's a, you know, maybe that's when we'll really know it's ossified when the first soft fork fails, right?

Yeah, yeah, that's ex- I think that, that will be the point. It's unfortunately, it's, it will be an ugly point, but this is how we, we have seen Bitcoin, developing, that The, the, the biggest, or the, the, the biggest cathartic events like, where, where we, we recognize that something isn't working the way it is, were always linked with some disaster. And, and this will be the same, it's said this time too.

Yeah, I'm just trying to think of what might be, an example of like a soft fork that, you know, could-- this sort of scenario could occur. I mean, it would it just be, Let's say people don't correctly coordinate and maybe ninety percent of the nodes in the, in the network upgraded or changed, and then those other ten percent just get lost?

Yeah, m-maybe, maybe some people don't, I, I don't care simply. Let's say we, let, let's take an example at the Casa nodes. Let's say, there are, there's a, there's a company, deploying lots of devices with full nodes implemented in them, and, and at some point Point, I, I don't imply this to Kazo, but I mean, such a company could go out of business and, upgrading of these nodes could be stopping and, so, so the customer, all, all the retail customer who aren't really Bitcoin savvy, are no longer, do no longer have an easy upgrade path, the, the automatic upgrade isn't working and these, these, these, these nodes are just stucking at a certain version, which is then, then not rejecting the blocks that should have been rejected by, If they would follow the software. so this,

I, I think this, this will just happen by, by the scale of the network. It is just not feasible to assume that, this network at any s- any scale will be still able to s- as syn- nearly synchronously upgrade. The upgrade cycles will be longer and longer, and at some point it w- they will be practically impossible.

Yeah, interesting to think about. I guess, yeah, that's a good example. I just, I hadn't thought of something like that, but it, it totally-- you know, not cast obviously, but you know, just the idea that some node product company could go out, yeah. Any other, risks that you see or threats to Bitcoin that you see potentially, like an incentivized reorg?

Yeah, there, there was a recent discussion about this topic and, and, I, I also, I also think that the idea of Binance to do the reorg at the time point they considered it was certainly, not feasible and it was a, it was just crazy to- I assume that they could do, they could do that. but, but the, the discussion whether this could be done in general, was sparked in the community and, there are,

on one side it's, it's quite clear that, that there is a, there's a high immunity of the, the network, due to,

Financial incentives and also due to the, the community's morale to, to do something like that. on the other side, we shouldn't, shouldn't never, never, never say never to something that is technically possible. And, a reorg of, of a short distance is, is very, very technically possible. It is ac-- it is in fact happening all the time, like a reorg of, of one block depth, depth is, is a reg-regular event on the, on the Bitcoin network. And if you let's, if you consider the following scenario that there would be an exchange that has a cold wallet, and the cold wallet address is very well known, and somebody is for some, in some way, able to, to get the key of the cold wallet Maybe by physical force and, moving out the Bitcoin out of the cold wallet, then, this, this movement, could have-- could be recognized by the exchange at the instant because they could see the transaction on the network that oh, way, money is leaving your cold wallet and something is, something is very wrong, or at latest it would be recognized as it is in the first block. So basically the alarms that there is something like this happening could go off very fast. Very quickly they could, it could go off within minutes. And then if you, if you have a scenario where you have just one or two blocks of advance of the network, and, and, and you have The, the, the, the, you, you can inst-- offer monetary incentive to some miner to attempt a reorg. Then, what I say, it is, it is technically feasible, and we shouldn't say never, and in, in such a short term time frame, it is, I think, even practically feasible, because, the timeframe would be too short to trigger the, the immune system of the community. It would be too short to, to, i-it would be easy to offer for a very short timeframe a very high financial reward, to those who are participating. And I, I actually wrote an article about how this could be-- this financial, reward could be, offered in a, in a way that, that is basically, Forming, a coalition, to do this, just by offering the, the reward on the blockchain. So my entire point is, yes, it is-- it wasn't feasible as Binance considered it, but on the other side, being a technician, I just say, hey, we shouldn't not say never to something that's technically possible.

Right, and I, I guess the other thing there is, let's say you were some exchange and you, you detected, as you said, this, transaction, and within like one or two blocks, you notice straight away, would they then have to go and coordinate with a miner within the space of, you know, because we're talking like twenty minutes here, they would have to quickly like pick up the phone, call, you know, a few miners and- No,

they don't. The, the, the point is they don't even necessarily have to pick up the phone because they can

Offer,

right? The, the minor fee.

They, they could, they could, a, a, a, a exchange could even have, all this, as an emergency me-measure set up in advance, so they could have on their, on their website, that in case we would, we would have a breach in of, of, of our cold wallet, we will do the following, we will issue a transaction on this page which is offering part of the, the cold wallet's content, as a financial incen- to do, incentivize re-go, re-org. so I think the, the feasibility if that something like this could work, mostly depends on the, the preparedness of the actors, the exchange and the miner to act on such an offer So if, if, if basically there is a, there is a standard for this offer, there is a procedure that people know how it works, then you could, you could communicate on the blockchain, your offer very quickly and, and miner would know without calling them, they could have, again, they could act on it. So the, the, the point is, you know, the, the, the security of Bitcoin is guaranteed by financial incentives. And, we, we, we currently, we currently have, a development of the network where the, the miners are following, Likely following very simple rules. They are, they are just using the Bitcoin core node and, and, using the, the in-- the incentives that are, or the decisions that are very obvious, like building on, on the, on the highest block is, is, is financially more sane than building on another one. But as the, you know, as the industry matures and the sophistication of actors improves, the, the optimizations, built into the mining companies, how they decide on which block they, they actually build, could be, could, could, could take considerations into this, into this decision, that are, that aren't Not obvious to now to take, but, but that are entirely financially sensible. So I, I, I think we shouldn't assume that, that miners-- since this, this is an industry that is, that is running hundreds of millions of dollars for each, each, each bigger miner, this industry will, will in-develop the same kind of advanced intelligence on decision, on how to, how to act like any other industry.

Right. I mean, it's probably parallel to like when- And banks develop all this sort of fraud technology to, like, block your transaction, that sort of thing.

Yeah, the, the same, same way like banks, basically employ, thousands of, of, of computer scientists just to optimize, trading, by observing the news from any, so they, they, they, they, they- The linkage of information and decisions is, is just getting more complex as these actors are more sophisticated.

Yeah, and it's a good point that, you know, we can't just rely on sort of everyone doing the, quote unquote, trying to do the right thing for Bitcoin. We have to consider seriously the risks. I think at this point it's fair to say it's theoretically possible, but quite unlikely in practice. But as you're pointing out, in the future, if things develop in certain pathways, it could become more likely on certain shorter time frames.

Yeah, yeah, you- You can see that, certainly as the industry matures and, and, and, and miners are more like, current corporations who have, responsibility to, to produce a profit in a very short timeframe and so on, they will have different incentives and different considerations than they have now.

and I suppose there is also, Matt Corallo's, I understand it's soon to be renamed, but BetterHash, that different, mining protocol, and I suppose that is one way that the, disconnection of the miners from the pools might somewhat, make it more difficult to coordinate that sort of, behavior.

Yeah, this is, this is an example of this arm race between, the intelligence, that is following some, some certain incentives and, and the software that we try to provide to align the incentives with what we think is better, better for the network. So I, I pretty much admire the work Mark Kurlow is doing there, but it's, it is something that isn't yet, really in action.

Right. Yeah. Excellent. look, let's switch now to another threat. I know you've written an article, so it might be good to from August 2018, it's called "The Case for Central Bank Controlled Digital Currencies." So do you wanna just talk a little bit about the threat that that might pose?

Yeah, that this is, this is an economic threat to the Bitcoin, not, not a technical one. I, I think that, It is quite feasible, that, central banks or one central bank will offer, a digital currency, for the main reason, that they, they need to, overcome the current-- their current limitation that they cannot impose negative interest rates to, basically to cash and, we, we, we see this theme all over, the, the, the, the central banks and banks in general talk about cashless society, they are trying to push the customer towards, towards these, payment methods, and, the, the, the, the main motivation, the, the real motivation behind this isn't, not like money laundering and, and tax evasion and so on, but the main motivation is, to be able to impose negative interest rates, and this is, this is very important. the, the entire system, because it is, it is the only way, central banks currently can press money into the market by, offering, by offering it as a loan, but, this, this a-and, and the, the interest that, that basically is im-is imposed on the m-on money is, is basically the price of money, okay? And since they are trying to press more money into the market to, to, sustain the current economic system, they drive down the price, so they drive down the interest, and the zero line is, um-imposing a, a, a limit there because, if, if, if they- Would go below zero on, on customer deposits, then people would just take out the money of the banks and, and keep it in cash, and that would basically equal a bank run, and all banks would become, bankrupt because they have fractional reserve. So they have to create a cash system which is able to impose negative interest rates And, this is where their, their blockchain technology, basically, some, some, some kind of fork of Bitcoin's technology, could help them, to create a payment system, that is able to impose negative interest rate, m-m-m- On, on the blockchains, like the transactions, the transaction fees of a transaction are actually nothing else than some kind of an interest rate. and they could do the same and, leverage that. And that's, I think this, this, this is already recognized by them, and, and, and I think it is very plausible that in a very short term frame, I mean, one or two years, Bitcoin will be contended by, central bank digital currency.

Some economists kind of have this crazy idea of, hey, let's go to negative rates and Keep things sustainable, but the other question then is how resistant will societies be? So for example, I was just recently in Japan, and I've noticed they're very much a cash-based society, but other societies are very cash-less. So how do you think that sort of plays into it, just culturally?

I, I think people don't really make, much thoughts about the, the payment system until it really fails, so,

the, the cashless society is, is, is currently accepted by most of the people because, the, the, the systems providing them are, aren't failing very frequently. but we, we already had this, this kind of events with Mastercard, very recently, and, I think, I, I think that, this is what basically determines, the cultural acceptance. people get used to something that is work, that is working and isn't asking question until it, it is failing. Maybe there are also differences in, experiences to, with regards to, to sta-- to state control or, I, I lived in, in Eastern Europe for a while and, there people are, are, using more frequently cash Cash, because they are, they are more sensibilized again, towards, state control of their finances, as it used to be in, in, in, in, communist times. So th-these are the differences I think culturally.

Yeah. And then, I guess the next question then as well is, why do you still believe Bitcoin would overcome some sort of central bank digital currency?

as I said, I think central digital currencies will come, because they need it. it is, the question if, where, where Bitcoin will find its place in this, new world of, of competing offers. I, I think that people are generally lazy, taking the, the simplest, The, the simplest to use approach, simplest to use offer, and, and therefore if, they don't, if they do it right, then the central bank digital currency might, might also get a wide usage, between, in, in, in ordinary daily, daily transactions. but still, people who would want to, That, that, like there's a theory of good money and bad money, and that bad money drives out good money. so people would, might, it, it, might definitely still use Bitcoin as a savings method, even if, the central bank of a digital currency is very convenient to, to pay, pay at some stores.

I, I, so, so this is, this is about people, but where I, I think there is, the central digital currency can't probably compete with Bitcoin is, is more in the rearm of, of the, of the technology. The, the, the s-- the bank system, central bank's thinking is always human centered. They think about users who are, paying and users who are responsible for everything. but, use cases that are, that are, that are automating economic interactions between processes and between devices, will more likely use, a system like Bitcoin, that isn't, not linked to identities, not linked to persons. So, so these, these are the, these, these are the areas where I think the competition will, will, will run and, and, and areas where maybe, Bitcoin will have a, a head start, head Ahead of the digital currencies of central banks, but it's hard to tell. It's, it, it, at least we, we should be, we should be aware that there will be, there will be a competition.

Yeah, excellent insights there, Tamas. let's talk a little bit. I know you've also done, a lot of work around Bitcoin data science, and I recently interviewed, Tua Demisto as well, so we talked about some of the work there, but let's get some of your insights around, some of this And, you know, unrealized P&L. Do you wanna go into that a little bit?

Yeah, well, I, I, I looked at, at Bitcoin, blockchain also in a, from an angle which, which is, which I was taught in, on the trading floor, looking at Bitcoin as basically a collection of transactions that are buying and selling, bitcoins, between, basically what, what we see on the Bitcoin blockchain is a record of Of Bitcoin transfers, we don't see the other leg, we don't see what people get for the Bitcoin, but we see that they are transferring it. And, my, my, my idea was there to, impose, the same, ideas to Bitcoin transactions than you would do to, to regular trading transactions, since Bitcoin is not only recording, where, when a transaction takes place, but it is also recording, what is, what is the path it If you have Bitcoin moved, so when was this particular Bitcoin last moved, you can actually calculate, an indi- a P&L individually by every transaction, since you can, you can value the Bitcoin at the time point it was last moved and at the time point now where it is moving, and you can calculate, a P&L, a profit and loss, that this transaction means to whoever did this transaction. If you aggregate all these profit or loss together, then you basically end up with a market cap, because that's a sum of all the Bitcoin movements and individual P&L sums up to the market cap. but if you compare the market cap with the The valuation of the Bitcoin as of their last, ac-acquisition, then you, can determine basically the, the, the unrealized P&L, the, the P&L that people could realize by selling it now. And, this unrealized P&L, tells, tells us a lot about, the, the emotional, state of the Bitcoin holder, because you can, you can basically say, you can basically see if, if in, in total, the Bitcoin holder are currently sitting on an unrealized loss or an unrealized profit. And, this is a very important indicator that I proposed and then, was used by Edmond Capital's latest, newsletter or latest, you know Paper, it's called, Bitcoin Heavy Accumulation Phase. so this, this really helps us to, to, it helps them, and I'm an advisor of them, so it helps us to, to time the market.

and another concept that this is an interesting concept, another concept I developed is the, I called it the liveliness. The Bitcoin liveliness, is basically a ratio of,

Bitcoin days burned in a, in a transac- Bitcoin days burned until a point with all Bitcoin days ever created. So you could, you could say, if you, you could calculate what would be the total number of Bitcoin days if they, if, if there was never ever been a Bitcoin transaction, and you can calculate what is the burn until now And if you, if you, calculate the ratio of these two measures, then basically you come up with something that I call a liveliness. I call it liveliness because it's a measure between zero and one, and, liveliness of zero would, would be a blockchain where you have an initial coin offering and never ever a transaction. And the liveliness of one would be a blockchain where in a single block all coins are moving. So this basically gives you, gives you a, a, a measure of how active a certain blockchain is, how economically active a certain blockchain is. And it's also a measure that is very difficult to forge, because if you, you, you cannot just take the same coin and move million times, that doesn't really add up to liveliness, because the same coin cannot acquire a new data Days to burn, and therefore the, the sub-subsequent washed transactions don't really change the liveliness measure. So the liveliness measure is, is very, very useful to, first look at the Bitcoin, but it's also very useful to look at other currencies. I, I personally proposed this, measure, but I'm not really interested into looking into other, other, other coins. I hope somebody does it and, and implements the liveliness measure for Bitcoin Cash and other shit coins. we, we will We would, we would definitely see a huge difference to Bitcoin. But the liveliness measure is very, very useful within Bitcoin already, because, you can also take the, the reciprocal meaning of liveliness. You could say, "What isn't live is something that is huddled." So you can take the liveliness measure and calculate a kind of, measure of, coins that aren't moving, that are huddled by long-time investors or, or is- Or have been lost. And if you look at, at these changes of liveliness within a short period of time, then you could, by, by taking differentials between the liveliness, you can basically calculate the movements of the hodled coins. And this is also a, a method that I proposed and now is using, used by Admon Capital to, to time the market so we, we can observe with the liveliness measure, the movement of the hodler with who are probably most influential ba- actors in the Bitcoin, economy. So, so it helps us to, to, to see how they are basically, acquiring or accumulating or liquidating positions.

Okay, thanks for that explanation around liveliness, and I think particularly interesting was this point you made around measuring the individual P&L of a specific output as it moves on the blockchain. and then another complexity that I think you touch on is this complexity of dealing with large custodian moves, because that can give a so-called false signal.

yes, the, the liveliness is certainly, flagging, movements, of, of custodians, considering them as, being basically hodlers, the cold wallets look like hodlers. and these moves can, and I, I think I, I proved in a, in a certain, case, actually trigger, market reaction. I, I'm quite sure that I'm not the first one, who thought of, measuring, the movement of lo- of, of hodlers, eventually not with aliveness, but with some similar measure. And therefore, I, I, I think that, a large movements were already, were spotted by traders who are observing the blockchain, and as in, as, and, and their observation were dr- was driving their trading decisions. And this is what we could see, in my opinion, in, last December, where, Coinbase is, was reshuffling their cold wallet, and this, cold, cold wallet reshuffling, was quite, quite obviously triggering, a sell-off in the market. I think this is, this is mainly because the, the, the fact of the reshuffling was communicated after the fact, and eve- even though Coinbase was, basically bragging that they are, they executed this, th- this, huge- Each move in a, in a very, very coordinated and, and, and stealthy fash- fashion, that wasn't actually the case because the, the blockchain does record all these moves and, and, and if, if somebody was observing a measure like the Divinest, he would have probably traded on it. So this, this basically me-teaches us that, movements of, of large stashes of Bitcoin, on the blockchain can have, market-moving effects and therefore they should be communicated

Interesting stuff, because then it's, it's sort of this, stewardship argument that it's this com-- you know, blockchain as a commons and certain aspects of transparency that an exchange should, undergo. But then at the same time, there might be certain risks that, that poses as well to their own hot wallet and their private, you know, their own security setup as well. yeah. I think another interesting thing that comes up with this whole discussion around liveliness, and I think, this actually came up for the listeners, this came up- On my discussion with Tua Demister as well, but I'd be interested to hear your take on this as well. Was the, you know, if we get into the psyche of the investor or the hodler and they are underwater for a few years, or maybe not underwater, but they're down from their all-time high, and then what was it that you noticed that as the price crept back up to the, what was the previous all-time high, was that you noticed that some of those hodlers were now kind of cashing in or locking in some gains? Can you articulate Yeah, the,

the liveliness, measure also gives us information of their, shift of, investment behavior and it, it actually, shows us what is, what was intuitively assumed beforehand, that most of the holders were ac-- accumulating bitcoins during the, the bear market phase of two thousand and fourteen and fifteen and sixteen, and, they actually began, liquidating their positions once, the, the new market high was reached. what is bitin-- what is interesting to see that the, the liquid- Condition was actually triggered quite early, so there, there were quite a few hodlers who were really riding the, the bull, market to the end, but most of the liquidations happened actually early in the bull market phase. and then after, the, after the, the crash in begin or end of two thousand and seventeen, we again saw, that accumulation is taking place and, this is what is happening. articulated also in this report, be basically we think that Bitcoin's currently in a, is in an heavy accumulation phase.

Fantastic. And I think one corollary there, so again Past performance, et cetera, isn't the same as future performance, but if we were to assume that same kind of, psyche were to apply, then if we see the price of, right now, s-sitting around eight thousand USD as we record this in May two thousand and nineteen, if the price were to now come back closer to twenty thousand, which was the previous all-time high, then we might anticipate another, another round of selling by some of the, OG or the hodlers.

Well, if, if history is a, is a guide there, then we could assume that. But fortunately, we have the, the liveliness measure, so we can basically observe this in, in real time. So we can see, we, we will see very, pretty much, if it is really happening or not.

Yeah, excellent insights there. And, I think it might be interesting now if you have any, anything to share on perhaps any future Bitcoin blockchain data science that you're, you're looking into at the moment.

well, I'm, I'm, I'm, I'm currently looking into the, the wealth distribution of, Bitcoin. this is a topic that, lots of people are interested in it and also, creates lots of criticism around Bitcoin. the question I would like to answer is how, how unequal, is the wealth distributed with Bitcoin? It's certainly in-unequal, just like in any other economy, But, the question is, is it really worse or, or better in the sense of equality if compared with a fiat, with fiat currencies or, or, or, well-known economies? To answer these questions, I'm currently working on a, on a methodology to, to cluster, the, unspent coins. And, this isn't, not really about de-anonymizing, it's really about clustering, figuring out which coins are probably closer to each other. So what, basically creating a, a bell distribution or a histogram of bell distribution.

Yeah, it's fascinating how some of these works, they just sort of build on top of each other, right? So at the start it was Bitcoin days destroyed, then it was Huddle Waves, then it's Liveliness, and now it's sort of getting to the level of trying to ascertain what is the genie coefficient. Now, I think many Bitcoiners, there-- and some commentators have made this point as well, that I think the intuition is that Bitcoin is spreading out over time, right? It started out very, very unequal, but it is spreading out over time. But I suppose that's question you're trying to prove out or answer, is that true or is that false?

Yes, exactly. that's, that's-- I think it was a very interesting question. And it's also probably also telling us about, Bitcoin's pro- future prospect if it isn't If it wouldn't spread out, that would be a bad news for this economy.

right. I suppose, I think maybe some level of inequality will always be there, but, you're right, I think it may, it may sort of still spread out from what was an extreme inequality.

Yeah, yeah, sure.

Yeah. L-look, I think, that's pretty much all we've got time for, Tamas, but I, I have to say, it's been a really, really interesting, honestly, very interesting conversation, and just so wide ranging as well

Bitcoin data science. Just also lastly, before we let you go, if you've got any, closing thoughts for the listeners, and also obviously before we let you go, make sure you tell them where they can find you online.

Well, I'm, my, my recent thoughts you find online, so you find me with my real name on Twitter, on GitHub, and then Medium. and I would love if, the listeners would, would check out these sources.

Excellent. Well, look, thank you very much for coming on the show again.

Thank you much.

I hope you guys enjoyed that interview and as always, if you enjoyed it, make sure you rate, review, and share the podcast. You can find the show notes on my website, stephanelivera dot com. This is episode seventy-three. That's it from me. Thanks guys, see you next time.