Read_856 - It's the Settlement Assurances Stupid

[00:00:00] Speaker A: This is also an interesting case study in how bitcoin resists duplication. You can create something which looks cosmetically similar to bitcoin, but you cannot replicate the settlement assurances which derive from the costliness of the ledger. The best in Bitcoin made Audible. I am Guy Swan and this is Bitcoin Audible. What is up, guys? [00:00:43] Speaker B: Welcome back to Bitcoin Audible. This is where you learn everything about Bitcoin, open source and the technology of liberty. [00:00:52] Speaker A: And I am Guy Swann, your host, the guy who has read more about. [00:00:56] Speaker B: Bitcoin than anybody else you know. And we have got a correction to make on the show today. [00:01:04] Speaker A: Not that I was wrong. [00:01:05] Speaker B: I'm never wrong about anything, so that's definitely not it. But Anil said so who is a great follow by the way on Twitter. [00:01:15] Speaker A: And social is Anil. Is Anil on Nostr. Definitely should be. I'll find his Noster pub key. [00:01:23] Speaker B: And we've read a few pieces by Anil on this show actually, but he put together in response to a thread or something. Oh, no, no, it was just, it was just a post, I think on like, if you were just learning about. [00:01:38] Speaker A: Or just getting into bitcoin, these are the fundamentals. [00:01:41] Speaker B: And I think he put like 13 or 15, some, some list of kind of like bitcoin essentials, Bitcoin fundamentals articles and papers and that sort of thing. [00:01:54] Speaker A: In order to read. [00:01:56] Speaker B: And I clicked on it, as I always do for those, and I was like, okay, well, let's see, let's see what he recommends. And I figured Anil would have some really great recommendations. [00:02:04] Speaker A: And lo and behold, I had done all of them in audio. [00:02:09] Speaker B: They were all like extremely solid pieces for getting a foundation and understanding Bitcoin. [00:02:15] Speaker A: And that's what Bitcoin Audible has been all about. [00:02:17] Speaker B: So of course we have them all in audio except for one. And I know, I know I have done this piece. [00:02:25] Speaker A: I know I have read this piece before. [00:02:27] Speaker B: I cannot find the audio anywhere. It's not in the feed, unless I. [00:02:31] Speaker A: Have accidentally changed the name. I did do that very early on in the show. [00:02:37] Speaker B: If I didn't, if I didn't like a title, I just named it like the topic rather than the article title itself. Because obviously we have the commentary and we have like just general discussion about it. So maybe that's what happened. But I cannot find it. And I've got way too many episodes to search through and I couldn't find it on my computer. But of course I've got a lot of this archived and kind of stashed away and I'm traveling so I don't have my Linux machine. So all of that said, I'm fixing it. I'm putting this piece on the one piece that I did not have a direct audio for. [00:03:10] Speaker A: I did have Unpacking Bitcoin Assurance Bitcoin's Assurances, which is another piece by Nick. [00:03:15] Speaker B: Carter on the same topic and it's. [00:03:17] Speaker A: Actually linked to in this article, so I will have the link to that one as well as this entire list. [00:03:23] Speaker B: Of bitcoin fundamentals if you want to. [00:03:25] Speaker A: Get a full grasp, like you want. [00:03:26] Speaker B: To get a good picture in your mind. Honestly, those like 12 or 13 or so episodes are a phenomenal base to build for yourself and understanding Bitcoin. [00:03:40] Speaker A: And this piece in particular by Nick Carter is also a really good one. [00:03:45] Speaker B: Because it hits on a particular topic. [00:03:48] Speaker A: That I think is still to this day misunderstood. [00:03:53] Speaker B: What is the value of Bitcoin and exactly what security does it provide? [00:03:59] Speaker A: And what is a confirmation on Bitcoin. [00:04:02] Speaker B: In comparison to a confirmation on Ethereum or Bitcoin Cash? [00:04:06] Speaker A: What do these things even mean? And we will be we will be. [00:04:10] Speaker B: Talking about it and expanding on it after the read. [00:04:13] Speaker A: Right now let's go ahead and get. [00:04:14] Speaker B: Into our article from Nick Carter and. [00:04:18] Speaker A: It'S titled it's the Settlement Assurances Stupid by Nick Carter. How to evaluate Blockchains what is the time to finality on major blockchains? How long should I wait before considering a Bitcoin transaction settled? What are the risk factors which might cause me to demand additional confirmations? How do confirmations affect settlement? Surprisingly, none of these questions have good answers. Even in 2019, over 10 years after the first bitcoin block was mined, rigorous investigation into the properties of proof of work has been hampered, both due to a perception that it's just a temporary staging ground for some future superior consensus or sibyl resistance mechanism, and due to a belief among bitcoiners that its quality is inviolate. But these questions are fundamental. If you believe that public blockchains with open validator sets and distributed convergence mechanisms will persist and mediate value transfer for the foreseeable future, they are worth pondering. And if you are an exchange and your livelihood depends on correctly assessing the number of required confirmations on a variety of blockchains, these questions are critical. First, let me explain why I think settlement assurances are the primary thing worth contemplating about any public blockchain. What's the interesting thing about Bitcoin? This is a surprisingly difficult question to answer ask 10 different Bitcoiners and you'll get a dozen different responses. Disagreements about what Bitcoin is for its teleology nearly tore the community asunder in the 2014-17 period. Hasu and I tried to chronicle these competing visions in a piece a while back. Others have noticed this and have covered it in detail. I particularly like Murad Mamedov's and Adam Tach's take. Daniel Krawitz covered the topic ably in 2014. Links to these articles will be in the show. Notes In Crowitz's piece, he posits that Bitcoin is understood very differently by two major tribes, the investors and the entrepreneurs. The investors, he posits, believe that Bitcoin is a new form of high powered money which primarily upholds the sovereignty of the individual. The investors tend to believe that Bitcoin will catch on because of the innate strength of its monetary properties. For them, evangelism is pointless. Price is the best evangelist. The entrepreneurs, as he dubs them, are more interested in Bitcoin as a global payments system and emphasize its use in commerce. As anyone who paid attention in 2015-17 knows, these two sides fought a bitter civil war over Bitcoin's telos or purpose, with the block size being the main battleground. Perhaps these views can be harmonized. I tend to believe that the interesting thing about Bitcoin is its capacity to facilitate the transfer of value through a communications medium with extremely strong assurances. I made an effort to disentangle and evaluate those assurances. Here he provides a link to the article Unpacking Bitcoin's Assurances, which we also have on the Bitcoin audible feedback. The link will be in the show. Notes I think that Bitcoin is a novel institutional technology high assurance, wealth storage and transfer without reliance on the state or a financial system, which will unlock new modes of human organization and will enable productive commerce in places where property rights are poorly enforced. So if the assurances you get around settlement are the most interesting thing about the system, how can we evaluate them? And how do we make consistent comparisons between Bitcoin and other systems with open validation evaluating settlement so what are settlement assurances exactly? They refer to a system's ability to grant recipients confidence that an inbound transaction will not be reversed. Wire transfers using a messaging system like Swift are popular in part because they are practically impossible to reverse. They are considered safe for recipients because originating banks will only release the funds if they are fully present in the sender's account. This is why the thieves behind the $1 billion Bangladesh bank robbery used Swift and bank wires. They wanted to leverage their settlement assurances. In other words, they chose to use a system for the theft, which they knew would be hard to reverse. Ultimately, $61 million from that heist remains unaccounted for. Far from being evidence of a failure of Swift in bank transfers, this demonstrates the system's strengths. Even in this case, where virtually everyone involved wanted to reverse the transaction, they could not. The system is resistant to rollbacks, discretion and post hoc edits. This doesn't make it a bad system. This makes it a system that gives counterparties a good deal of reassurance that a transaction will be final. In a similar manner, Bitcoin is a useful system because it provides users powerful settlement assurances. Just how good we don't know exactly, laurent MT wrote, probably the most scientific exploration in his excellent Gravity series. Links to the series available in audio on Bitcoin Audible, will also be available in the show Notes Generally, though, the properties of Bitcoin's proof of work have not been fully explored. It has suffered a few reorgs in its history, but as far as we know, no deliberate adversarial reorganizations where money was stolen. And we know that miners allocate a staggering amount of real world resources into mining transactions. This means that recipients of a Bitcoin transaction can have extremely high confidence that once buried under a few blocks, a transaction is unlikely to be reversed. However, this isn't the case for many competing cryptocurrencies. While they look cosmetically similar to Bitcoin, in many cases none have the same settlement assurances. This isn't necessarily because of any design flaw, but simply because Bitcoin's block space has more accumulated costliness and hence cost to attack per unit of time. And because Bitcoin is a near monopolist on its hash function and has dedicated hardware, somewhat surprisingly many weaker chains haven't been exploited, even if the cost to do so has been low. This is likely due to the fact that monetizing a 51% attack requires exploiting an exchange, which introduces additional complexities and quite frankly, most smaller coins aren't worth much in the first place and don't have any liquidity on the short side, capping the yield from an attack. To get an idea of just how vulnerable many cryptocurrencies are, take a cursory look at crypto51 app. The methodology, somewhat unrealistically assumes an attacker can rent sufficient hardware on Nicehash, but it still nicely depicts a lower bound of the cost to attack these systems. So what are the key variables for evaluating settlement in a public blockchain system? Let's divide them into the easily quantifiable ones and the harder to quantify variables. Before we jump in, let's pause for a tiny literature review to credit some. [00:12:42] Speaker B: Prior work in the space. [00:12:44] Speaker A: For a much more succinct take on the matter, read Anthony Lissardi's Understanding and Mitigating Reorgs. For a comprehensive investigation into the qualities. [00:12:54] Speaker B: Of Bitcoin's proof of work, see beyond. [00:12:57] Speaker A: The Doomsday Economics of Proof of Work in Cryptocurrencies by Rafael Ayur of the bank for International Settlements. For a fascinating implementation of what a model incorporating some of these variables might look like, see A Lower Bound on Minor Rewards by Kevin Liu of bkcm. Quantifiable Settlement Variables Ledger Costliness Ledger costliness is the most profound and direct variable available to us to evaluate a blockchain's settlement guarantees. Put simply, it is the equivalent to the amount paid to validators or transaction selectors per unit of time. In Bitcoin. Miners receive a per block subsidy and transaction fees as an incentive to stay honest and play by the rules. In proof of work, miners attach an unforgeable proof that they have burned some energy and hence incurred a cost to each block proposed. At the time. Of winning a block, the miner necessarily has to have burned resources roughly equivalent to the value of the block, typically with a small margin, unless they are extraordinarily lucky. Because of this, miners are incentivized to create valid and rule following blocks. Think of it as a bit like a school project where you had to read a book and produce a book report. You need to prove to your teacher that you read the book, so you produce a book report, a valid block hash with a sufficient number of leading zeros, which you could only have created if you actually read the book, computed sufficient hashes. Because your teacher is a stickler for style, you also have to format your book report correctly, produce a well formed and valid block. It would be a tragedy to read the whole book only to present a digest which is malformed and ends with you getting an F. Proof of work is the same. The work is upfront, with the payoff only coming later. You've incurred a real cost and your business depends on you carrying out the final bureaucratic steps to collect your reward, so you do your best not to screw that part up. Recently a miner did all the requisite work to be eligible for a block, but fell at the last hurdle by creating an invalid block. For a more complete description of how the proof of work incentive works, read Hugo Huynh's piece the Anatomy of Proof of Work, also available in Bitcoin. Audible in audio. The link will be in the show Notes so why does more ledger costliness per unit of time mean more security for transactors? Because a greater salary to miners who are presumed honest means you need a larger army of mercenaries to defeat them. These resources have to come from somewhere. You need to marshal resources and hardware capable of producing hashes, electricity, and so on. There's an argument out there that since attackers collect the subsidy when 51%, attacking only the fees provide security and proof. [00:16:10] Speaker B: Of work, I don't have the space. [00:16:11] Speaker A: Here to engage with this fully. For now, I'll just maintain that this subsidy, especially with dedicated hardware, is itself an enormous cliff which must be scaled before 51% scenarios can be theorized. To sum up, outbidding the set of honest miners dutifully producing blocks on bitcoin is very expensive. They collectively take a salary of $6.9 billion per year right now, and many of them have presumably invested in their businesses in anticipation of future cash flows, meaning that the hardware active on the network might even be higher than current miner revenue would imply. So bitcoin is protected not only by the daily salary that the protocol pays its miners, but by the discounted rewards these miners expect to earn in the future. This means bitcoin isn't just protected by the reality on the ground today, but miner expectations about rewards in the future. We don't have an easy way to model expectations, so the easiest thing to do is to simply take the miner salary per unit of time and compare blockchains on that basis. If you stop reading this article now and just retained that one sentence, you would already have a better understanding of security than most people. Very few entities, even those for whom the stakes are very high, like exchanges, bother benchmarking blockchains like this Usefully, Anthony Lissardi has already done some great expository work on the topic. He introduces the bitconf, demonstrating how many confirmations are required for one bitcoin confirmation's worth of security on other blockchains like litecoin. Suffice to say, most people do not use bitconfs or try to index settlement to work done. Quite the contrary, the folk theory of settlement holds that settlement is a linear function of the number of confirmations. This is, sadly, a very common view. Even the Litecoin foundation website implicitly makes this litecoin transactions are confirmed faster than other cryptocurrencies like Bitcoin. Because it generates a block every 2.5 minutes, as opposed to Bitcoin's 10 minutes. This means your money gets to its destination quicker. The initial moment when a transaction is plucked out of the mempool and included in the chain is indeed reliably faster in litecoin. But in cryptocurrency, probabilistic settlement must be contemplated. In other words, if you only care about the first confirmation, then litecoin is faster. But the moment you start to care about longer term settlement over multiple confirmations, it becomes clear that it is much slower. If you believe that litecoin and Bitcoin confirmations confer the same amount of settlement guarantees, then you might depict settlement as follows. With Bitcoin apparently slower. [00:19:10] Speaker B: Here he just lines up the expected. [00:19:12] Speaker A: Time of blocks and just shows that there's many more blocks of litecoin for every block that passes in bitcoin. But this is mistaken. Litecoin has more blocks per unit of time, but it accumulates ledger costliness much more slowly. In reality, Bitcoin pays its private army of miners far better, and as a consequence they produce far more security per minute in the form of hashes. Bitcoin blocks are heavier with accumulated cost than litecoin blocks are. Even if litecoin had a 10 minute block time, a bitcoin block would still be worth 14.5 times more than its litecoin equivalent. Confirmations don't really matter. The opportunity cost incurred by miners per unit of time does. You could alternatively visualize ledger costliness as blocks getting piled on top of their predecessors, with transactions getting more and more final as they are buried deeper and deeper in the pile of blocks. Here he actually has stacks of blocks comparing Bitcoin, Ethereum and Litecoin and showing essentially the quote unquote weight through the size of the blocks. [00:20:31] Speaker B: And you can just clearly see and. [00:20:34] Speaker A: Visualize that the bitcoin blocks are far, far larger then both the very thin and small blocks that come in extremely. [00:20:44] Speaker B: Frequently on Ethereum and then the slightly. [00:20:47] Speaker A: In the middle of Litecoin. But both are tiny in comparison to the amount of weight, the amount of. [00:20:52] Speaker B: Security that a single bitcoin block or. [00:20:55] Speaker A: Confirmation can provide, and you can just visually see this. [00:20:58] Speaker B: So I highly encourage you go to. [00:21:00] Speaker A: The link and see that graphic. As more and more blocks get added to the heap, it becomes more and more implausible that they would be reverted and transactions become more final. In this graphic, I've scaled the width of the blocks to the relative ledger cost incurred and depicted the granularity of blocks. The point here is that settlement in a blockchain system is a flow. Block time is largely irrelevant. Ethereum has many more blocks per hour than Bitcoin does, but settlement should be compared between the two based on ledger cost rather than number of confirmations. Yield from Reversal Transaction Size Ledger costliness isn't the only thing that matters in settlement. Also important is the incentive someone might have to try to reverse a transaction. The purest codification of this incentive is simply the size of the transaction. If you are a recipient of a 50,000 Bitcoin transaction, you might wait more than the six block rule of thumb out of an abundance of caution. If you are receiving 1000 SATs, one confirmation is likely sufficient. In short, transactions have more or less perceived settledness based on the stakes at hand. Elaine Ou formalized this concept in a fantastic Bloomberg article, arguing that recipients should wait until the transaction's value and ledger costliness match to consider a transaction settled. Elaine's formulation handily conjoins two of the most important quantitative variables in blockchain settlement ledger cost and yield from reversal. If you wanted to settle a $10 million inbound transaction in BTC according to this rule, you'd wait 60 blocks or 10 hours. It's a neat coincidence that at a price of $13,330, Bitcoin accumulates ledger costliness at a rate of exactly $1 million per hour. Henceforth, I'll refer to this simple formula as the OO rule. Now that we have the two most critical settlement variables enumerated, let's put down some numbers and compare the major proof of work networks. At the time of this writing, Bitcoin has a daily miner revenue of $23,972 million. Revenue per 10 minutes is 166,000 days to settle $1 million is 0.04. So the BTC finality multiplier is 1. This is our point of reference. Ethereum below it gets 4.1 million in daily miner revenue, 28,000 per 10 minutes, and a 24 settlement for $1 million, which is a multiplier of Bitcoin by 5.8 times longer. [00:24:01] Speaker B: And without going through all of these. [00:24:02] Speaker A: Elements on all of these charts, we'll just go down the multiplier. Bitcoin is 1 Ethereum 5.8 Litecoin 14.5 Zcash 33.4 Bitcoin Cash 33.7 Bitcoin SV 69 Dash 91 Ethereum Classic 127 Monero 133 Bitcoin Gold 488 Dogecoin 495 Verge. [00:24:28] Speaker B: 1737 Vertcoin 3686 needless to say, Bitcoin. [00:24:36] Speaker A: Is by far the fastest settling blockchain, just including these two variables and none of the other salient ones. Settling even a $1 million inbound transaction can be extremely slow on many blockchains. Aside from Bitcoin, Ethereum and Litecoin, it takes over a day for every other decentralized ledger. I'm not including Ripple and Stellar in these examples because they don't have meaningfully decentralized validation. Smaller chains simply do not have enough miner reward to make settlement suitably quick. Luke Childs how many comps offers a dynamically updated version of parts of this? It's also worth calling attention to the fact that Bitcoin cash and Bitcoin SV settle transactions 33 and 69 times more slowly than Bitcoin, respectively. While they are functionally identical to Bitcoin in most respects, because they offer miners less of a bounty, they are vastly slower. This directly contrasts with their common positioning as faster blockchains. This is also an interesting case study in how Bitcoin resists duplication. You can create something which looks cosmetically similar to bitcoin, but you cannot replicate these settlement assurances which derive from the costliness of the ledger. Miners obey economic reality and cannot be cajoled to lend their support to a protocol which doesn't pay them well enough. In fact, as we will learn, Bitcoin cash and bitcoinsv are even worse off than this table suggests because of a third variable monopolist on its own hash function. So far I haven't mentioned a third critical variable which directly affects the settlement guarantees of a given blockchain, whether or not it holds an effective monopoly over the hardware which is addressable to its hash function. As I implied above, Bitcoin cash and Bitcoin SV are at a massive disadvantage relative to Bitcoin because they have a minute fraction of all the SHA256 ASICs. What this means is that even a mid sized or small pool mining bitcoin could temporarily redirect its hash power to one of bitcoin's smaller forks and 51% attack it at will. The fact that these blockchains have not been attacked yet is not evidence of their security. It may well be the case that there are no miners on bitcoin willing to maliciously interfere with either minority fork today, but depending on the goodwill of miners, makes for an extremely tenuous security model. Since this risk is ever present, it could be posited that neither blockchain ever reaches effective finality, regardless of the number of confirmations. This is because there are ample mining pools on Bitcoin, which could create a 100 plus deep reorganization in BSV, for instance, without too much difficulty. This variable introduces more complexity into the analysis. It is not the case that more hashrate means that a blockchain is more secure. It must also occupy a large fraction of the addressable hardware. In this example, I'd characterize blockchain A as less secure than blockchain B even though it has more ledger costliness in absolute terms. Because it is theoretically easier to marshal enough hardware to attack A, he has a graphic here where there is a large pool of the hardware, and then. [00:28:19] Speaker B: Blockchain A in the larger pool is. [00:28:22] Speaker A: A small fraction of the entire space of hardware where there is a smaller blockchain B that has almost all of its addressable hardware market and therefore is actually more secure against the hardware that is available than blockchain A, even though blockchain A has far more overall hashrate or costliness of the ledger. So consider this variable to be a Boolean. If the blockchain is a monopolist on its own hardware, the analysis is straightforward. If it is in the unfortunate position of splitting hardware with one or many other blockchains and retains a minority share of that hash function specific hardware, it is likely fundamentally unsafe, but it's hard to determine just how unsafe it is. The risk of an attack is a function of the attacker's ability to amass sufficient electricity and hardware. Less Quantifiable Settlement Variables the three variables mentioned above aren't exhaustive, but simply the easiest to quantify. With those, you could probably build a plausible model which is superior than those used by many exchanges today, but there are many more factors to consider. Yield from Goldfinger attacks Goldfinger attacks take their name from the Bond film in which the villain plans to irradiate all the gold in Fort Knox, making all of his gold more valuable. The term describes a class of attacks where the attacker is motivated by some extra protocol financial interest. Joseph Bonnu, more scientifically describes them as attacks where the attackers have an extrinsic motivation to disrupt the consensus process. The risk of these attacks is virtually impossible to quantify, since attackers have a variety of different motivations and they tend not to disclose them a priori before an attack. Here I'll give two further examples where the yield from reversal dramatically increases, rendering settlement guarantees less certain top heaviness this refers to the condition in which a large number of financially significant assets are created as tokens on top of some base layer protocol, for instance Omni assets on Bitcoin or ERC 20s on Ethereum. As these tokens inherit their security from and are wholly dependent on the base layer, they are vulnerable to attacks on the underlying chain. As the asymmetry develops between the value of the instruments on top and the cost to attack the base layer, the top heaviness problem starts to manifest. If the asymmetry becomes large enough, an attacker might seek to take out a short on some instrument on the top layer and simultaneously attack the base layer, either by mining empty blocks and dossing the tokens in question or creating reorgs and confusion. We have real world examples of the consequences of top heavy systems. Attackers have recently made a habit of attacking the underlying index, which sets the price for derivatives on Bitmex. Since there's a big asymmetry between the collateral present on Bitmex and the underlying reference market, the bottom it's lucrative to burn funds market selling on bitstamp because the attacker can monetize by causing an outsized move on Bitmex as margin positions are liquidated. I don't believe any blockchain faces this problem today, but as more instruments are tokenized and inserted on top of blockchains, the returns from attacking the base layer will increase liquid derivatives markets this is rather straightforward. Derivatives options in particular give financial market participants the ability to obtain leverage and magnify their returns, even relative to a small move in the underlying as with the top heaviness condition, the risk to the blockchain comes when a significant asymmetry exists between the cost to mount an attack and the returns from an attack. The creation of liquid derivatives markets enables attackers to magnify their returns from predicting price action, and if they can induce a drop in the price of the asset by mounting an attack, the settlement guarantees of the chain are potentially at risk. As the return from an attack grows, so does the amount of resources that an attacker is willing to contribute to an attack. So the creation of leverage on the short side potentially impairs a blockchain's settlement assurances. But due to the heterogeneity of actors and uncertainty about the ability to monetize such an attack, it's impossible to quantify this risk and add an appropriate security discount. Of course, one counterbalancing factor here is the potential unwillingness of an exchange to pay out on a successful bet if they suspect that the trader in question was coordinating with an attacker to interfere with the blockchain. Additional Hardware Considerations Implicit in the earlier point on hash function specific hardware is the well documented notion that GPU mined coins cannot ever be monopolists on their hardware because there are so many GPUs in the world thanks to gaming and other non cryptocurrency applications. I won't belabor this point. David Vorick has cleanly laid out the case for why GPU mined chains are fundamentally at risk and why long term incentive alignment in the form of asics is so critical. Thus, GPU mined coins should always be assessed. [00:34:12] Speaker B: Additional Confirmations it's hard to know exactly. [00:34:15] Speaker A: What the ratio should be for one GPU mined unit of ledger costliness to. [00:34:20] Speaker B: An ASIC mined unit, but there absolutely. [00:34:23] Speaker A: Should be a discount for GPU produced security. It's simply too easy to acquire hardware to mine a GPU mined chain Case Study Kraken's Confirmation Requirements Startlingly, from my conversations with exchanges who have a lot to lose from miscalibrated rules around settlement, it appears to me that they tend to give little thought to confirmation rules. I couldn't find much detail on how many inbound confirmations exchanges reserve until a transaction is considered settled. Helpfully, Kraken have made their criteria freely available. I decided to benchmark Kraken's confirmation requirements against what a naive implementation of Lizardi's bitconf would look like, simply requiring that all chains provide the equivalent of six confirmations on Bitcoin. The results are startling. Depending on how you put it, Kraken makes either extremely stringent demands of Bitcoin transactions or extremely loose demands of non Bitcoin chains. While Kraken asks for six Bitcoin confirmations to consider deposits settled, they ask a mere 12 of Litecoin, where the equivalent in Bitcoin security terms would be 174. [00:35:37] Speaker B: 30 for Ethereum Bitcoin equivalent 173 and. [00:35:42] Speaker A: 15 for Monero, where Bitcoin indexed security would demand 2000. My guess is that 6 confirmations is massive overkill for Bitcoin, making Kraken's lesser settlement demands of other chains more reasonable. Still, when the ledger costliness variable is consistently applied, the results are occasionally comical. Qtum or Quantum for instance, if held to the same standard in bitcoin, would need 67,000 confirmations equivalent to a wait of 115 days. Quantum may well have some alternative settlement mode. [00:36:19] Speaker B: I'm not familiar with. I computed the numbers simply based on. [00:36:22] Speaker A: The payouts it makes to validators. Of course, this is a very naive implementation of the model. A more sophisticated version would include higher security demands for non monopolist chains, GPU mined coins, large inbound transactions, and so on. I would encourage exchanges like Kraken to consider a systematic rule set for inbound transactions if they don't already. Whatever the particular formula chosen, it would likely suggest fewer confirmations for Bitcoin and more for smaller chains. Some takeaways what's the practical significance of all of this? Well, as we continue to await the formalization of these variables into a model that makes sense and is directly applicable to everyday usage of cryptocurrency, here are a few takeaways. [00:37:11] Speaker B: 1. Block time is arbitrary and changes little. The only thing that a lower block. [00:37:20] Speaker A: Time alters is reducing variance in the time to the initial confirmation. If you are impatient, you probably prefer a blockchain with a 2.5 minute block time, but this doesn't mean that settlement is any faster. Ledger costliness still accrues at the same rate, being a function of issuance and unit value per coin. Indeed, Bitcoin could reduce its block size by 25% and switch to a 2.5 minute block time and virtually no one would notice the difference. The system would be functionally identical. The six block rule of thumb would become a 24 block rule of thumb. Satoshi opted for 10 minute blocks because he did not know how well the system would be able to come to convergence. Latency and large blocks interfere with validation and make convergence among nodes more difficult. A healthy 10 minute block time gives the system plenty of breathing room and also gives us an indication of what kind of a system Satoshi was envisioning. Hint not suited for in person petty cash payments. It's true that the first confirmation matters some small amount since your transaction cannot start to be buried under the weight of subsequent blocks until it is included in a mined block. Additionally, a lower block time reduces variance in variables like daily issuance. However, aside from that block time is completely arbitrary. The security spend per unit of time in addition to the quality of that ledger costliness is what matters for settlement. A lower block time just means that you're chopping up that security flow into smaller bits. It doesn't make final settlement any faster. 2. Bitcoin is either providing massive security overkill or other blockchains are critically at risk. This is the clearest takeaway from the various benchmarking exercises I did for this article. If you measure blockchains purely based on the salary paid to transactions selectors, miners and validators per unit of time. For the most part, they look devastatingly weak compared to Bitcoin. Just have a look at this chart. Aside from Bitcoin, Ethereum and Litecoin, virtually nothing is visible on the chart. Because their security spend is so minimal, this isn't necessarily fatal. [00:39:48] Speaker B: It could be the case that Bitcoin. [00:39:49] Speaker A: Is way overpaying for security, for instance. [00:39:52] Speaker B: And that proof of work is better than we think. This is actually my current view that. [00:39:58] Speaker A: Due to the current subsidy conjoined with the high unit value of Bitcoin, Bitcoin is probably spending too much on security. But it does wrap the protocol in a warm blanket, which gives it a good degree of protection as it enters its teenage years. So this data is not necessarily apocalyptic for smaller blockchains. After all, even though Satoshi ordained the six block rule of thumb, it could be the case that for most transactions, one or two blocks are sufficient. This would lessen the heavy load placed on other blockchains trying to match Bitcoin's security spend. 3. Settlement is always probabilistic I will admit that I chafe a little bit when. [00:40:42] Speaker B: New blockchains tout their absolute finality. [00:40:46] Speaker A: The only way to truly have finality. [00:40:49] Speaker B: Is to have an organization vouch for transactions, effectively endorsing them. [00:40:54] Speaker A: But when this happens, authorities that might. [00:40:56] Speaker B: Have an interest in reversing transactions, say. [00:40:59] Speaker A: If they suspect they are related to. [00:41:00] Speaker B: Criminal activity, will typically ask that entity. [00:41:03] Speaker A: To facilitate the rollback, poking a hole in the perceived finality. Take the example of eos. EOS has a concept called the last irreversible block, which according to EOS Canada, means that you can trust with 100% confidence that that transaction is final, fully confirmed, and immutable. If the block number of a block is lower than the last irreversible block, that means it is considered final. According to EOS Network Monitor, the current last irreversible block is trailing the chain tip by 330 blocks, equivalent to about. [00:41:37] Speaker B: 2 minutes and 40 seconds altogether. [00:41:40] Speaker A: This makes EOS claimed time to finality very short. Except there's a catch. EOS has had a bureaucratic process through. [00:41:50] Speaker B: Which individuals could appeal to the EOS. [00:41:52] Speaker A: Core Arbitration Forum and ask for funds from suspended thefts to be frozen and returned to the victims, effectively reversing long settled transactions. One batch of these reversals took place in June 2018. This was possible because there were only 21 entities, the block producers tasked with processing transactions and all were known to the leadership and hence accountable, while many onlookers cheered the return of stolen funds. From a settlement perspective, this undoes the qualities that transactors seek when they use a blockchain. In practice, any mechanism which can reverse settlement can be abused. The reason credit cards embed a fee into transactions is because chargeback fraud is rampant. Imagine a sophisticated scam where someone sold eos for fiat in a peer to. [00:42:49] Speaker B: Peer exchange and then appealed the transaction. [00:42:51] Speaker A: To the ECAF and managed to get the EOs in the transaction returned to him under the guise of having been scammed. These are the kinds of schemes that result from administrative exceptions to finality. There are any number of examples I could give on this topic, but I'll stick with one for now. In practice, many of the blockchains that claim to have full and effective finality also insert the capacity to create discretionary rollbacks and account freezes into their system. You still have to consider the probability of a reversal, even if it's not explicitly codified. 4 By being open about its security model, Bitcoin's proof of work is usefully transparent Echoing Elaine Oo once again, one of the most useful features of Bitcoin's security model is how transparent and easily apprehensible it is. The precise guarantees are not easy to determine how many confirmations to settle a billion dollars, but the resources being spent to backstop the system are at any point, an onlooker can trivially determine how many hashes and, by rough extension, how much energy it would take to overpower the system. For years now, it has been clear that no entity outside the most potent state actors could muster sufficient resources to outweigh the honest majority. By contrast, other blockchains seek security through obscurity, security through complexity, or through untransparent institutional modes of finality. Verge, for instance, conjoined five different hash functions in its exotic proof of work. [00:44:34] Speaker B: Model, and that was ultimately its downfall. [00:44:38] Speaker A: An attacker realized they could perform a time warp attack by targeting just one of the hash functions and lowering difficulty to 1. Far from providing extra security, the insertion of more complexity into the system introduced new attack vectors. Summing up if there's anything I could have you take away from this piece, it's the Instead of viewing settlement as a function of some preconceived number of confirmations, think of settling a transaction in a proof of work system as the process of wood petrifying slowly. It happens at a given rate and can't be accelerated. The rate is determined by the variables enumerated above, chiefly ledger costliness, transaction size, and the availability of addressable hardware. Once completed, the wood has been replaced by minerals and is rock solid. No longer soft and malleable, the features of the wood are forever frozen in time. Similarly, as Nick Szabo has said, blockchains are computational amber. Amber starts life as tree SAP, only later becoming hardened in the process of storing bits of information, insect DNA, and so on within it. The essential process of burying past changes to the ledger under unforgeable cost provided by proof of cost incurred, provides the same slow moving settlement assurances. As more blocks accumulate, the gravity of the blockchain exerts itself and makes distant rewrites colossally expensive and unwieldy. The bounty available to miners and hence the cost incurred is a function of issuance, unit price and fees. None of these, aside from issuance, can be directly programmed, and a high issuance. [00:46:33] Speaker B: Alone cannot guarantee security as investors have. [00:46:36] Speaker A: To buy into the chain's prospects and backstop its value. In this sense, strong settlement assurances in a proof of work system cannot be planned for they can only emerge. Whether you find this to be a dismal conclusion or not is up to you. In this article I tried to enumerate the variables which I believe are most critical for evaluating the settlement assurances of blockchains, especially those built on proof of work. But you'll notice I provide no formal model nor a recommended solution to the problem. Many of these variables cannot be easily quantified and there are likely some which I am leaving out. A more comprehensive or implementation focused model I will leave to subsequent authors. If we ignore these questions, they will be forced upon us through necessity. As short side liquidity emerges for a larger share of the market, whole new classes of attacks will open up and exchanges will find themselves targeted more and more equally as major custodians and clearinghouses start to take cryptocurrency deposits totaling hundreds of millions or billions, they will need to devise formal rules for what constitutes settlement. They would do well to think deeply about the security of the blockchains that they are reliant on. All right, quick shout out to fold. They have a bunch of gift cards. [00:48:01] Speaker B: Right now with a chance to win. [00:48:02] Speaker A: 250,000 SATs that you enter when you get gift cards. [00:48:06] Speaker B: And if you are buying a bunch of holiday season stuff, using their gift cards is literally the best. I mean this is something that I do every single year and I always get like a very, very solid stacking of sats. Right now I've got, let's see, $7,193 worth of sats stacked and that's in two years. This cost me $10 a month for fold premium. I have $7,200 out of it. Or excuse me, these are not dollars. [00:48:38] Speaker A: It's 7.6 million SATs. I'm telling you, if you were trying. [00:48:41] Speaker B: To be on a bitcoin standard, out of all of my tools, I think. [00:48:46] Speaker A: Fold is the most indispensable. [00:48:49] Speaker B: You can buy bitcoin directly in it without any fees. [00:48:51] Speaker A: You can deposit bitcoin and sell it and push it straight to your card without any fees. You get sats back on purchases. You get even more sats back on gift cards. [00:49:01] Speaker B: You can do roundups, you can do recurring buys. It is honestly one of the most complete packages of integrating bitcoin and banking that I know of. [00:49:12] Speaker A: At least that I've ever used. [00:49:14] Speaker B: I have been a big proponent of. [00:49:15] Speaker A: These guys for a long time. [00:49:17] Speaker B: I highly recommend it. [00:49:18] Speaker A: There is a link right in the. [00:49:19] Speaker B: Show Notes to check it out. All right, so this article, I think one of the best things about this framing is to understand what it is that you get out of a bitcoin. [00:49:32] Speaker A: Transaction is to understand what it is that proof of work does. [00:49:37] Speaker B: It is a way to have a measurable means of finality that does not. [00:49:45] Speaker A: Have trust in a person involved. [00:49:49] Speaker B: I want you to think about that. [00:49:52] Speaker A: This is a digital ledger. [00:49:55] Speaker B: This is just a network where somebody. [00:49:58] Speaker A: Is writing a note that says I used to own these 100 units and now I send them to you and. [00:50:05] Speaker B: You own these 100 units and they. [00:50:08] Speaker A: Are signed cryptographically by a set of keys. [00:50:12] Speaker B: And when that note is published, there is a strictly measurable, there is a. [00:50:19] Speaker A: Quantifiable and completely non trust dependent degree of this is never going to be changed. Proof of work is like an independent force field. [00:50:36] Speaker B: It is a force field around a digital record keeping system. And literally the only way to edit. [00:50:45] Speaker A: That record keeping system to alter who owns what in the past is to use the energy created to make the force field in exactly the same degree. [00:50:58] Speaker B: Reverse it back in time in order to reach through it, change something and then continue forward. [00:51:07] Speaker A: Proof of work provides a measurable this can't be changed metric. [00:51:14] Speaker B: And it does not have a person. [00:51:15] Speaker A: Involved, it does not have an authority. [00:51:18] Speaker B: Involved, it does not have some bureaucratic. [00:51:22] Speaker A: Voting system and forms to fill out and all of the systems that think. [00:51:28] Speaker B: That they can get around it. The Ethereum consensus proof of stake, permissioned validators nonsense, the EOS fill out my forms to counter a transaction board members. [00:51:44] Speaker A: Crap like all of it. It's either independent or there is a mechanism of trust. [00:51:52] Speaker B: And the whole idea is to replace. [00:51:55] Speaker A: Trust is to create a digital form and economic incentive structure that creates trust as a metric that you can simply see and is not dependent upon anybody, any authority, any vote, any board, any foundation or anything. That is the value of Bitcoin. That is what makes it a profound innovation. That is what makes it a thousand year leap in the context of monetary guarantees. It is a digital network with a degree of physical world security that you can just see. And this gets us back. [00:52:37] Speaker B: I had a conversation with Milian from Primal this morning and we talked about, we talked about Nostr. Obviously he's the CEO and founder and everybody who's doing Primal and they have. [00:52:51] Speaker A: A really awesome project over there. [00:52:53] Speaker B: I've been using the Primal web app. [00:52:54] Speaker A: For a really long time. [00:52:55] Speaker B: But it was funny. One of the things we talked about. [00:52:57] Speaker A: Is just why he thought Nostr was it and it didn't matter. Everything that happens in the short to. [00:53:03] Speaker B: Medium term because Nostr works. [00:53:05] Speaker A: And one of the reasons Nostr works is because it is simple is because. [00:53:11] Speaker B: The degree of how to look at and it's just signed notes sent by relay. That's it. It's just signed notes sent over a web server or a node and what. [00:53:27] Speaker A: Kind of relay and how those connections are established, totally agnostic, not relevant. It is just signed notes in a simple format and it just works. It's very intuitive, it's easy to build on because of that. [00:53:40] Speaker B: And in the famous words of in. [00:53:42] Speaker A: Fact, this is actually in anil sets. [00:53:43] Speaker B: Those list is why dumb networks are better by Andreas Antonopoulos. That is also one that I highly recommend. And then there's also. [00:53:54] Speaker A: Worse is better. [00:53:55] Speaker B: By oh my God, Grisha. Grisha, I think that's right. I can't remember the name right now, but it's Bitcoin is worse is better. And it's talking about the Linux philosophy of worse is better better and just the general protocol philosophy is worse is better. [00:54:15] Speaker A: And it's very, very much aligned with. [00:54:17] Speaker B: The whole dumb networks idea is that dumb networks are reliable. [00:54:21] Speaker A: The beauty of simple things is that they just work and all of these other systems in some effort to get. [00:54:28] Speaker B: Something that's quote unquote better, faster or whatever it is, whatever arbitrary thing that they're trying to say, their token or. [00:54:37] Speaker A: Crypto or whatever is more accomplished at, inevitably they make something that's vastly more. [00:54:43] Speaker B: Complex, less reliable, has moves out of the realm of quantifiable statistics and moves entirely into the realm of how do. [00:54:53] Speaker A: We even measure or understand how secure this is. [00:54:57] Speaker B: That is what their whole proof of stake permissioned voting system that Ethereum has now if you notice all the links. So this was written back in 2019, or at least it was updated in 2019. I guess it was probably early 2019 that it was written. But if you actually go to the links, a lot of the things that they actually have are still there, like how many comp. How many comps.com, crypto5, one app is. [00:55:23] Speaker A: Still there and these are the ones. [00:55:24] Speaker B: That he linked to that compare Bitcoin to a bunch of other coins. [00:55:28] Speaker A: But you'll notice something is that Ethereum has had their proof of stake switch since then. [00:55:33] Speaker B: And because of that it's not in any of these things because you can't compare them. [00:55:37] Speaker A: Because they have explicitly moved to a. [00:55:39] Speaker B: System that is now so complex and. [00:55:42] Speaker A: Has so many other random variables that. [00:55:45] Speaker B: We can't even test and can't even quantify. [00:55:47] Speaker A: But there's no way to compare them. [00:55:49] Speaker B: It's like asking how secure is my bank transfer in comparison to Bitcoin? [00:55:55] Speaker A: There's no comparison. [00:55:58] Speaker B: There's no how much energy is put into the Bitcoin bank transfer. [00:56:02] Speaker A: It's like nothing. It doesn't even exist. It's not there. [00:56:05] Speaker B: Their security model has nothing to do with how Bitcoin works. [00:56:09] Speaker A: And that's the same for Ethereum. [00:56:11] Speaker B: The only one you'll still find on. [00:56:12] Speaker A: All of those lists is now Ethereum Classic, which is still, which still has. [00:56:16] Speaker B: The proof of work. [00:56:18] Speaker A: But the beauty of proof of work. [00:56:20] Speaker B: The reason Bitcoin uses proof of work and the reason why, you know, you know, I don't think a proof of stake, I don't think Ethereum ever, ever. [00:56:29] Speaker A: Would have even gotten off the ground as their Casper proof of stake system. [00:56:34] Speaker B: Had it not been for bitcoin coming first. [00:56:37] Speaker A: If the cypherpunks were trying to launch. [00:56:39] Speaker B: What Ethereum is Today, back in 2006, 2007, they would have just gotten laughed out of the room. [00:56:48] Speaker A: Nobody ever would have used it. [00:56:49] Speaker B: It would have died on the vine. [00:56:51] Speaker A: The only reason I think Ethereum has. [00:56:53] Speaker B: Survived as long as it has is. [00:56:56] Speaker A: Because it was originally just a copy of Bitcoin with a giant pre mine. [00:57:01] Speaker B: And you know, insiders that they just. [00:57:03] Speaker A: Hand out a bunch of coins to. [00:57:04] Speaker B: And that they own 70% of the supply. But they got to position themselves as if they are decentralized because they had proof of work, kind of like fiat. [00:57:13] Speaker A: In the gold standard. Is that fiat by itself, like out. [00:57:17] Speaker B: The gate never really worked. You can't just like, be like, oh, well, here's my token, buy it. I can print as much as I want, but I'm trustworthy, so don't worry. [00:57:27] Speaker A: You'Re going to use it as money. Like that never works. [00:57:30] Speaker B: You can't sell people on that. That's ridiculous. But if you get them into a gold system where it says these tokens are worth exactly this amount of gold. [00:57:39] Speaker A: And then for a hundred years you actually redeem that amount of gold and. [00:57:43] Speaker B: People just come to build trust, they. [00:57:47] Speaker A: Associate the trust with the token itself. [00:57:49] Speaker B: Rather than the gold, and people just. [00:57:51] Speaker A: Start using the token and they forget about the gold. Well, then you can switch from that system to a completely fiat system. But it is only because you have. You've basically stolen the trust. You've defrauded the fundamental nature of the system. And it takes a really long time. [00:58:09] Speaker B: For that trust to be bred out of a population. [00:58:12] Speaker A: Essentially. They just get used to money being a certain way. They get used to it looking a certain way. And it takes a very long time. [00:58:19] Speaker B: For the price to correct and for. [00:58:21] Speaker A: The trust to completely break down. [00:58:23] Speaker B: But that's kind of where we are right now. [00:58:25] Speaker A: That's where I. This is the transition that I feel like we are in. And in a similar sense, the buildup of trust in a completely different system is also very slow, very difficult. But that's exactly what we're seeing. And the novel system is proof of work. Consensus on top of Bitcoin. Bitcoin is the innovation. And the beauty of it is that. [00:58:48] Speaker B: It'S simple, is that the security doesn't. [00:58:52] Speaker A: Have that many assumptions. [00:58:53] Speaker B: I mean, bitcoin, the network itself is. [00:58:56] Speaker A: Complex and you do have to make economic assumptions. It is a probabilistic settlement, just like. [00:59:02] Speaker B: Nick Carter talks about. [00:59:03] Speaker A: But ultimately it requires as few assumptions. [00:59:07] Speaker B: As we could potentially give it in. [00:59:11] Speaker A: Order to provide the value that bitcoin provides, which is settlement without trust in third parties. The idea that that transaction is done, period, how done is it? [00:59:25] Speaker B: Well, in the last this many hours, $30 million worth of ledger depth, ledger. [00:59:34] Speaker A: Weight has been added on top of it. And the only possible way to undo. [00:59:39] Speaker B: That is to redo all of that. [00:59:42] Speaker A: Power to fight back through the force. [00:59:45] Speaker B: Field at that $30 million weight. [00:59:48] Speaker A: But here's another interesting thing, is that. [00:59:50] Speaker B: If you're specifically attacking the network externally, like from outside, and you actually have. [00:59:56] Speaker A: All of this hash power, it's important to remember. [00:59:58] Speaker B: And of course, if you've been listening to the show, you know this. But it's important also to remember that. [01:00:03] Speaker A: This is also while the rest of the network is moving forward. [01:00:06] Speaker B: So every 10 minutes, like let's say we're talking about a 10 hour attack and somebody has exactly the hash power. [01:00:13] Speaker A: Of the entire network and they are. [01:00:14] Speaker B: Trying to go back 10 hours. Well, it takes them 10 hours to go back through it because it takes. [01:00:19] Speaker A: 10 minutes to produce a hash if. [01:00:21] Speaker B: They have the same amount of hash power, which means that every 10 minutes they go backward one block, they add another block that they have to go backward through. [01:00:29] Speaker A: This is why it's referred to as a 51% attack, because with 50% of. [01:00:33] Speaker B: The network, nobody makes any progress. [01:00:36] Speaker A: You're basically in stalemate. And at a 51% attack, you only have a 1% advantage, which means that. [01:00:43] Speaker B: If you're moving back 100 blocks, it's. [01:00:46] Speaker A: Going to take you 10,000 blocks worth of time at a 1% advantage. [01:00:52] Speaker B: If everything continues forward with 51% hashing against and 49% hashing forward, in reality, there's mostly just going to be a. [01:01:01] Speaker A: Battle at the chain tip of who's. [01:01:03] Speaker B: Got the next block or who's reversing one block. But the likelihood of going back 100 blocks, like going 10 hours deep, is almost non existent with specifically that amount of hash power. If we're talking about a 51% attack specifically. Now, Nick Carter also brought something up. [01:01:19] Speaker A: During the early parts of this that. [01:01:21] Speaker B: Says, quote, there's an argument out there. [01:01:23] Speaker A: That since attackers collect the subsidy when. [01:01:25] Speaker B: 51% attacking, only fees provide security and proof of work, I don't have the. [01:01:30] Speaker A: Space here to engage with this fully. For now, I'll just maintain that the subsidy, especially with dedicated hardware, is itself an enormous cliff which must be scaled before 51% scenarios can be theorized. [01:01:42] Speaker B: Now, since then I think Nick Carter has written about this and probably expanded and made his views clear on it. But I also want to make a clarification. People talk about the. When a 51% attack, the subsidy is. [01:01:58] Speaker A: In fact a wall that has to be broken. [01:02:01] Speaker B: Broken through. [01:02:02] Speaker A: It does actually protect. [01:02:04] Speaker B: That's exactly the thing that does protect. [01:02:07] Speaker A: Against that level of attack. Because they are specifically giving up their ability to go forward, they have to get, they have to use the subsidy. Either way, they would get the subsidy. [01:02:18] Speaker B: Going forward or they get to fight. [01:02:20] Speaker A: Through the subsidy and do the work twice. [01:02:23] Speaker B: So no matter what the subsidy that is contributing to the amount of work has to Be beaten back, beaten back through. Now what I think this might be referring to or why some people make this misconception or suggested that there is this element is because fees only provide. [01:02:41] Speaker A: Security to censorship resistance. So this is specifically about getting certain transactions in a block and basically having. [01:02:48] Speaker B: Transactions be entirely neutral. So an example is, you know, let's say every block is paying out $100,000 or $500,000 in value to every single miner. So every single block is whatever the price is. [01:03:06] Speaker A: Well, that means that whether you put. [01:03:09] Speaker B: Any transactions in a block at all. [01:03:11] Speaker A: Or whether you decide to explicitly exclude certain transactions because they're from bad people. [01:03:17] Speaker B: Or whatever you want to say, if. [01:03:20] Speaker A: You were trying to censor those blocks in some way, you still get paid $500,000 the value of the subsidy of those blocks. [01:03:29] Speaker B: Now if you want to reverse those. [01:03:31] Speaker A: Blocks, it's still going to cost you. [01:03:32] Speaker B: The $500,000 to reverse it. [01:03:35] Speaker A: And only if you manage to succeed and completely outpace the rest of the. [01:03:39] Speaker B: Market that's trying to get that $500,000 and you double the amount. You basically outpace them to a degree that you can rely on it. [01:03:48] Speaker A: Yes, you still get the subsidy, but you still have to do the proof of work. [01:03:52] Speaker B: And you do the proof of work. [01:03:53] Speaker A: Under the assumption or under the risk that you might get nothing at all. So it remains the proof of work doesn't change one way or the other. And the deeper you go, the more proof of work you have to do and the hard it is to fight. [01:04:04] Speaker B: Against the honest miners moving forward. [01:04:06] Speaker A: But in the censorship scenario, the fees, if the fees are only $10,000 of. [01:04:12] Speaker B: That block, well then, you know, what's the big deal in not adding transactions in if you're going to get $500,000 per block. [01:04:21] Speaker A: Now if the fees are the same as the block subsidy, now you have a massive cost. Now your decision to censor a transaction. [01:04:30] Speaker B: Has a massive economic trade off. So this is why some people have said and have explained in the past that the subsidy is a is protection, is 100% protection from a 51% attack and from all attacks to attempt to reverse transactions. The subsidy protects against that. [01:04:51] Speaker A: But the subsidy does not protect against censorship resistance. Protect the censorship resistance of the Bitcoin. [01:04:59] Speaker B: Network because you can still censor and. [01:05:01] Speaker A: You will still get paid the subsidies. [01:05:03] Speaker B: It is transaction fees that ensure people do not censor Bitcoin and that the censorship of the network or the transactions allowed on the network are in fact considered. [01:05:16] Speaker A: All are held equal, so to speak. And the more and more we progress. [01:05:20] Speaker B: Toward a lower subsidy and higher transaction fees, the more secure and the higher that settlement assurance will be and the higher the censorship resistance will be at. [01:05:30] Speaker A: The exact same time. [01:05:32] Speaker B: So that's just an important caveat right there. And again. And we'll have a another roundtable very soon. [01:05:38] Speaker A: It'll be this week. [01:05:39] Speaker B: We'll be recording on Wednesday. Not sure when we will publish it. [01:05:42] Speaker A: Hopefully sooner rather than later. [01:05:44] Speaker B: It'll probably be this week, so stay tuned. But Steve, Steve uses words or simple Steve I think this is. I can't remember. Got his old ones. His old handle stuck in my head so bad. [01:05:56] Speaker A: But he runs UTXO Live and one. [01:05:58] Speaker B: Of my favorite things about it is that everybody talks about like fees going crazy and how all over the place everything is. And one of the things that he has shown that I just think is so cool is take all the variants out and the Bitcoin the SATs payout of fees, they've obviously gone up a. [01:06:18] Speaker A: Ton in dollar value. But the SATs amount of fees on. [01:06:21] Speaker B: The bitcoin network have actually been very consistent over very long spans of time. It's like back to 2012 and 2013. The amount that people were paying in fees was almost nothing, but it was. [01:06:32] Speaker A: The it was the same sad amount. [01:06:34] Speaker B: That people are paying today on average. And this is obviously stretched out. Obviously there are times where fees spiked to 55 and $100 per transaction, all this crap. But those are short term bursts that don't really have anything to do with the overall trend. [01:06:48] Speaker A: So anyway, that's that was just an. [01:06:50] Speaker B: Interesting tidbit that I that stays stuck in my head. But I want to repeat the important line for understanding what a confirmation is, what a blockchain, what blockchain security gives you, because this is the most important. And he says specifically, if you just remember this one thing, you'll be way better off than the overwhelming majority of. [01:07:14] Speaker A: People says, quote the easiest thing to. [01:07:18] Speaker B: Do is simply take the miner salary per unit of time and compare blockchains on that basis. [01:07:26] Speaker A: If you stopped reading this article now and just retained that one sentence, you'd. [01:07:31] Speaker B: Already have a better understanding of security than most people. [01:07:33] Speaker A: Very few entities, even those for whom the stakes are very high, like exchanges, bother benchmarking blockchains like this. [01:07:42] Speaker B: And it's really crazy that they don't do this for things like bitcoin cash and Bitcoin sv, things that have basically ASIC secure excuse me, not asic but hash rate security in which they hold a tiny, tiny portion of the market of available hardware. And this is one of those things we've read about this. I think I've read the David Vorick. [01:08:06] Speaker A: Piece he was linking to. [01:08:08] Speaker B: That one might be one of the dead links in this article, but I'll try to remember have all the links. [01:08:14] Speaker A: To the ones mentioned. [01:08:15] Speaker B: But if you, if I miss one or you can't find it, it is literally in this article he puts posts all the links right there. So just something to keep in mind, these are all these can be found. [01:08:25] Speaker A: Except for the ones that are dead. [01:08:28] Speaker B: There's a couple of blogs that he was linking to that just aren't available anymore. But this is the thing about GPU coins and about why ASICS like ultimately and you know, we thought this for a long time that you know, ASICS seemed to be a net negative and there was a lot of discussion, a lot of articles that we talked about on this show like kind of during the transition back in 2014, 15, 16 and all of this stuff where it. [01:08:55] Speaker A: Was this huge question as to, you. [01:08:57] Speaker B: Know, what, how do we think about security when it comes to hardware? Because everybody kind of had that one cpu, one vote thing. In Satoshi's explanation, his mental model for how it was that miners would be. [01:09:14] Speaker A: Outpaced or excuse me, honest miners would outpace dishonest ones. But one of the interesting things about. [01:09:22] Speaker B: The GPU mined coins is and which Nick Carter specifically says that they should. [01:09:28] Speaker A: Be assessed additional confirmations. [01:09:31] Speaker B: And it's really, really hard to know. [01:09:33] Speaker A: What ratio this should be because GPUs are so easy to get a hold of. [01:09:39] Speaker B: And especially after AI, there's going to. [01:09:41] Speaker A: Be all these data centers that could. [01:09:43] Speaker B: Just completely demolish a bunch of these other crypto tokens. And these specific cryptos say they argue. [01:09:53] Speaker A: They are more secure and more decentralized because of this hardware being out there. [01:09:57] Speaker B: When it's actually not true. And their security is actually threatened because somebody who has made an investment in ChatGPT or some LLM running service now has all this hardware that they can. [01:10:10] Speaker A: Just devote, they could just completely demolish. [01:10:13] Speaker B: You know, 10,000 blocks of Monero or something. And there's no real way to secure against that because essentially there is an. [01:10:23] Speaker A: Alternative investment, an alternative way to make. [01:10:26] Speaker B: Money on the hardware hardware, which means. [01:10:28] Speaker A: The hardware is not an investment in. [01:10:29] Speaker B: The token, it's not an investment in the network itself. Whereas Bitcoin being ASIC mind and being the only thing that you can really. [01:10:37] Speaker A: Do with SHA256 ASICS, which has any. [01:10:41] Speaker B: Market to speak of, actually it actually makes it a security benefit that someone has to like, if Bitcoin becomes worthless, so is the asic. [01:10:52] Speaker A: So the investment in the infrastructure itself, in the hardware, in the machines that are plugged into the wall and the electricity that it is eating all goes to Bitcoin. [01:11:03] Speaker B: It actually becomes a really interesting security benefit because of the lack of versatility. [01:11:10] Speaker A: Of the hardware used for Bitcoin mining, because of how unbelievably specific it is. [01:11:15] Speaker B: And what's funny is that there's no way, like you can always make more. [01:11:20] Speaker A: Specific hardware that's better than less specific. [01:11:22] Speaker B: Hardware that's just axiomatic to the nature of things. [01:11:26] Speaker A: If you build generalization, it is less compute effic. [01:11:30] Speaker B: And this is why I've talked about this with AI on AI Unchained and in previous episodes, which you can all. [01:11:35] Speaker A: Just now find in this feed now. [01:11:38] Speaker B: But I've talked about why it is that I think the race to general AI is such a lost cause or. [01:11:46] Speaker A: Is so misguided, is because general intelligence is just so unbelievably inefficient. And if you're trying to complete a task, if you're trying to get intelligence. [01:11:55] Speaker B: On one thing, it makes sense that you would want to build one, build. [01:12:00] Speaker A: Something that's extremely specific toward that kind of intelligence. [01:12:04] Speaker B: Like, if you wanted something that had like great, like really fantastic network intelligence, you wouldn't want to bother using a massive amount of compute on getting it. [01:12:13] Speaker A: To understand the English language at the. [01:12:16] Speaker B: Same time so that you could talk to it. [01:12:18] Speaker A: It's network design. [01:12:20] Speaker B: It's a. [01:12:21] Speaker A: It's a model for network connections. [01:12:23] Speaker B: It should only do network stuff. And it's very similar to the hardware problem of Bitcoin mining in general. [01:12:31] Speaker A: And the idea of a GPU versus. [01:12:33] Speaker B: An ASIC is that there will always be an ASIC that can do it. [01:12:37] Speaker A: Better than a gpu. [01:12:38] Speaker B: And what's funny is that Monero and a bunch of other coins have actually hard forked a bunch of times in. [01:12:44] Speaker A: Order to try to make sure that. [01:12:45] Speaker B: People can't build ASICs for them, which is such a problem, which is, I think, a huge security loss rather than a gain. And some of them have changed their. [01:12:56] Speaker A: Algorithms, which is a huge loss to the trust and the settlement assurances. Because not only are you restarting the. [01:13:02] Speaker B: Network effect, not only are you restarting the cumulative buildup of security in the system, but you've also just let it shift it over to something that nobody has any trust basis for now. People have to just assume that, okay, well, maybe this new model works, is. [01:13:19] Speaker A: Just as good as the other, the old one and Verge being the example. [01:13:23] Speaker B: That Nick Carter I thought brought up, which was so cool. We've, I think we dug into that on this show probably like five years ago or something like that. But it was a really great test case in why, oh my God, it would be so funny if it was in the commentary of unpacking Bitcoin's assurances, the other article by Nick Carter from like literally half a decade ago. Because if that was the case, it's such a perfect example of why you can't just like split up the hash rate. Going back to the idea of simplicity as the ultimate form of being trustworthy and being reliable is that they were like, oh well, we'll get it super. [01:13:59] Speaker A: Decentralized, super mega decentralized because we're going. [01:14:03] Speaker B: To have five different hashing algorithms and five different types of hardware and we're. [01:14:08] Speaker A: Going to be the best. [01:14:09] Speaker B: And what's funny is that they actually made the cost of attack so much cheaper because the attacker specifically did a time warp attack on one difficulty, one hash rate or one kind of hashing and hardware on the network and basically. [01:14:30] Speaker A: Tricked its mode of interacting or measuring them against each other so that he. [01:14:35] Speaker B: Could have an unbelievable. [01:14:39] Speaker A: Rather than being one fifth of the. [01:14:41] Speaker B: Network, he managed to trick it into thinking it was a thousand times the size of its potential impact and in doing so literally just pwned the whole network, just destroyed it. [01:14:55] Speaker A: So their attempt to be more decentralized. [01:14:57] Speaker B: Through more complexity ended up literally being its downfall. And why the entire thing just blew up. Again, dumb networks are better, simple metrics. [01:15:08] Speaker A: Simple, easy to understand, easy to see, easy to quantify. [01:15:13] Speaker B: That is a critically important part of security. And almost, almost every single network, every single alternative, every single fork of Bitcoin is making multiple massive trade offs that they fail to recognize and they certainly aren't going to talk about openly when. [01:15:32] Speaker A: They are trying to sell it to a bunch of people that are being duped. [01:15:35] Speaker B: Luckily that it literally seems like most of that is just done. I don't think we'll have another cycle. I don't think this cycle will have the secondary crypto blow up or like hype cycle that it has had in the past. I think it will be muted, the last one was muted from the 2017 one and I think this one will be even more muted because more and more I think it's just nothing even compares. If you, if you go back to all the things, all the comparisons that Nick Carter Talked about in this one. All of them are worse. Like everything in relation to Bitcoin has just gone down. Bitcoin is more secure in every metric. It has not changed, it has not been hard forked. [01:16:21] Speaker A: It is still the same assumptions. It is more robust, the market is. [01:16:25] Speaker B: Larger, the liquidity is deeper, the hardware is more advanced and more widespread and more invested in the ecosystem built on top of it is vastly larger and more diverse. There are numerous layers and protocols and. [01:16:40] Speaker A: Still things even being tested and implemented. [01:16:44] Speaker B: And innovated on Lightning has been growing significantly. Like constantly growing as form, as a metric in payments and in the amount of traffic has been incredible. I think the last report we saw was. Wait, that might have actually been all the way back in 2022. [01:17:05] Speaker A: What was the latest arc? [01:17:07] Speaker B: I think it was Ark Invest. You had like really good like growth graphics. But the, the growth metric then was. I think it's like 1,200%, 1,300% over the year in amount of traffic through existing nodes, the amount of payments and honestly, I think with Nostr, lightning is crazy ubiquitous. 22 years ago I couldn't use Lightning almost anywhere. Now it's the default when I'm making a transaction. I think people really don't appreciate that. Like I interact with. I pay everybody and interact with everybody with Lightning. And if I'm doing something, it's like Lightning is expected to be there. You just get a Lightning invoice these. [01:17:55] Speaker A: Days or a Lightning address. And that's like the beauty of Nostr is that I just like zap people. [01:17:59] Speaker B: Like crazy and there's not even any friction. Like I don't even think about it. Like there's just. [01:18:04] Speaker A: You just pay people to their addresses. [01:18:07] Speaker B: But I think all of my, like practically all of my services can you. That's one thing I haven't checked on Fold. I usually do. I do on Chain for that one. I know river has Cash app has Lightning and I know Fold does Lightning because you can get gift cards with Fold. There's gotta be a. It doesn't look like you can deposit Lightning, even though I can just like buy stuff with Lightning. Anyway, Lightning is all over the place to the point that I'm not even sure when it's not available because it's mostly just expected and. Albie, Dude. Alby Hub. I know I've talked about this a couple of different times, but if you're. If you're running your own node and you want to be able to connect. [01:18:54] Speaker A: To stuff and use Nostr with your. [01:18:56] Speaker B: Own node, Albie Hub is such a fantastic tool. If you have not tried it yet. [01:19:03] Speaker A: Please take the time. [01:19:04] Speaker B: It took me like three months of seeing people post about it and having it my list of like yes, definitely go to do this, you got to check this out. I know I'm going to be using Albi Hub. I can tell that this is going to be a great tool. And I still procrastinated, procrastinated, procrastinated. And I just could not find the time to just set aside from it. It literally only took me like 15 or 20 minutes and it has been. [01:19:25] Speaker A: A godsend ever since. [01:19:26] Speaker B: Like just being able to like I just open up the Albigo wallet and I'll tell you, every other mode that I have connected to my main. My start 9 wallet and like back in my node too has been slow, clunky and frustrating. And it's basically because it all runs. It all has to run over Tor and Tor sucks. Albie Hub is far and away the fastest. The one that just kind of acts like I would think it would. That loads fast enough that it doesn't even register that like I just, I just open up my wallet and it works. And then because of its use of Noster Wallet Connect, I can zap people on Damis and Primal and I don't have like, I don't wait for it, I don't do anything, I don't confirm. I literally just, I tap and then it, it zaps them. And Prime Milian from Primal I think has and I highly recommend this episode one when it comes out. It'll be probably this week maybe, I'm not sure. But I highly recommend listening to it because he thinks, and I completely agree with him, that in the coming years the number one onboarding into bitcoin, into the bitcoin ecosystem, into the bitcoin technology and into actually using it day to. [01:20:51] Speaker A: Day will be Nostr. [01:20:52] Speaker B: Like he thinks it will just trump. It will basically destroy everything else in terms of metrics of people who come to Nostr and then find and discover and go down the bitcoin rabbit hole and start using Bitcoin. That will be the number one funnel of no coiners to Bitcoiners. But anyway, back to bitcoin itself and just talking about how all of this is expanded. Bitcoin is literally, and not even just in price, but literally it is 100 times the Bitcoin it was three to four years ago. Everything has grown massively. And I honestly think the people who are still thinking that some side crypto or smart contract platform is the future. [01:21:45] Speaker A: Still just do not understand the monetary. [01:21:47] Speaker B: Economics, do not understand the value that this thing is actually providing. What it is like what the utility. [01:21:53] Speaker A: Of money is, where the value of money comes from. [01:21:57] Speaker B: And don't realize that they are. [01:21:59] Speaker A: They've just completely misunderstood. [01:22:02] Speaker B: They've just completely misunderstood. And I think this is one of the simplest framings, one of the easiest ways to begin to see the difference in framing that this is a tank, a bitcoin transaction, an on chain transaction is like taking a tank to the grocery store, putting your groceries in the back and then driving home. [01:22:24] Speaker A: Yes, it is unbelievably secure, but no. [01:22:28] Speaker B: It is not a system designed for retail payments. It is optimized for the best, most reliable, highest trusted, most decentralized, global neutral settlement of value. Satoshi created a brilliant, an absolutely brilliant. [01:22:51] Speaker A: Means of coming to global consensus without. [01:22:55] Speaker B: Any trusted entities, without any authority whatsoever. [01:23:00] Speaker A: And he had to make explicitly a. [01:23:02] Speaker B: Ton of trade offs that make it suck for retail. A 10 minute block time for in person retail. The fact that it's on a global bidding war and you don't even know. [01:23:14] Speaker A: If it's going to have a confirmation that it's trying to get into the block that the fees aren't reliable, they can change. [01:23:22] Speaker B: I had great example actually I tried. [01:23:23] Speaker A: To send an on chain transaction the. [01:23:25] Speaker B: Other day and I we kind of needed it in like a quick amount of time and I just didn't really worry about or wasn't thinking about the fee. And just after I sent it a handful of higher fee transactions, enough higher fee transactions kind of flooded in that I ended up being two blocks behind. And so what I thought was going to be in the next block ended up not being in the next block. And luckily I just had rbf. Nunchuck just has those things. Just bump the fee, you resign and republish. So it was really easy to fix. And I got into the next block. [01:24:00] Speaker A: After I realized that a block had. [01:24:01] Speaker B: Gone by and I didn't get in. But it's a perfect example of something that require you have to have a system like this in order for global consensus to actually be sorted out. And but what you end up, what. [01:24:15] Speaker A: You end up with is an auction. It's an auction like you don't know. [01:24:19] Speaker B: If your transaction is going through. It's a global broadcast network and that. [01:24:25] Speaker A: Auction is specifically its defense against DDoS. [01:24:29] Speaker B: Its defense against being just flooded with. [01:24:32] Speaker A: A bunch of arbitrary or nonsense data. [01:24:34] Speaker B: Those, those fees are another critical security mechanisms. It's not even Just the fact that. [01:24:40] Speaker A: You have settlement assurances. [01:24:41] Speaker B: It matters how you provide those settlement assurances, how do you require and how. [01:24:46] Speaker A: Many people are requiring that that proof of work is even attached to those blocks. How do you define the network? How do you create decentralization? There are so many different factors here. [01:24:56] Speaker B: And people who think that because somebody said that, oh, it would be great. [01:25:02] Speaker A: If you could use this for retail. [01:25:03] Speaker B: Payments and then it's not being used. [01:25:05] Speaker A: For retail payments without looking at it and seeing what it is and what. [01:25:09] Speaker B: Value it actually provides. I mean, of course they're going to. [01:25:12] Speaker A: Be confused, of course they're going to. [01:25:14] Speaker B: Get it wrong and they're going to redesign and they're not going to give it the value or understand what its actual purpose is. Because, you know, it's a little bit like you don't know, you know, nobody knows what this should look like to begin with. [01:25:29] Speaker A: So when you give it the label. [01:25:30] Speaker B: When you categorize it as, oh, this. [01:25:33] Speaker A: Should be used for retail payments. [01:25:35] Speaker B: And when Bitcoin is small, it's actually. [01:25:37] Speaker A: Really useful for that. [01:25:38] Speaker B: And it's not because this design is optimized for that. [01:25:41] Speaker A: It's because there are no attackers. [01:25:43] Speaker B: It's because you don't have to worry about settlement. It's because it's still a closed. It's very much like, you know, IRC and the early Usenet and all of these things is that before they had to work, worry about spam, it didn't matter. You could just go up and anybody. [01:25:57] Speaker A: Could just share and post, whatever. But as soon as it starts to scale, as soon as it starts to. [01:26:01] Speaker B: Grow, that model falls apart. This is one of the reasons why. [01:26:05] Speaker A: So many past decentralized networks ended up in centralized silos. [01:26:09] Speaker B: It's because they were fundamental spam. And I mean this is where proof of work came from. This was the point of email and how Finney and RPOW reusable proofs of work like this was not. This was not a new problem. This is the age old problem of the Internet. [01:26:26] Speaker A: Bitcoin solved it. [01:26:27] Speaker B: And while Bitcoin was still small, yeah. [01:26:30] Speaker A: It kind of looked like you could. [01:26:31] Speaker B: Just go up and buy stuff online with on chain transactions, but it was never designed for that. [01:26:38] Speaker A: It's not a retail designed system and. [01:26:41] Speaker B: It doesn't make sense at that layer. [01:26:44] Speaker A: It's a monetary network. It's a monetary structure and consensus system. [01:26:50] Speaker B: And not only is that far more difficult, like extremely difficult to maintain and has far different and more nuanced security. [01:27:02] Speaker A: Assumptions and economic incentives. [01:27:04] Speaker B: Not Only is all that the case, but it's also orders and orders of magnitude more important and more valuable than. [01:27:15] Speaker A: A retail payment system. We have tons of retail payment systems. We have exactly zero trustless digital monies up until Bitcoin existed. Just the nature of saying that was. [01:27:31] Speaker B: An absurdity in 2008. So if you still think this battle is going to be won by some crypto token that has slightly better privacy, even though you could divert a third of ChatGPT and completely ruin weeks and. [01:27:51] Speaker A: Weeks and weeks of the history of. [01:27:52] Speaker B: The network, if you think it's going. [01:27:54] Speaker A: To be because of some smart contract. [01:27:56] Speaker B: When all you have to do is tie software, the hash of some piece of software to a Bitcoin transaction and you can basically have a non interactive. [01:28:07] Speaker A: Oracle system for any network, any trade, any, anything the world. [01:28:11] Speaker B: It's completely agnostic to any kind of code. You do not have to have smart contracts built right into the system itself. And if you still believe that Bitcoin. [01:28:21] Speaker A: Isn'T a smart contract system, when that's literally the only thing that it is, and Lightning Network is literally a smart contract payment network that has instant settlement and proof of reserves and is global. [01:28:33] Speaker B: And can make thousands of payments, tens of thousands, millions of payments in no. [01:28:38] Speaker A: Time, at very, very small, small amounts. If you are still confused by these. [01:28:42] Speaker B: Realities or think that EOS or Solana. I don't even know what the token. [01:28:49] Speaker A: Of the month is these days, but. [01:28:51] Speaker B: One of those things is going to replace it or is better and has. [01:28:56] Speaker A: Cooler features and is going to, you. [01:28:58] Speaker B: Know, Bitcoin is going to be the. [01:29:00] Speaker A: MySpace and this is going to be the Facebook. [01:29:02] Speaker B: You've wholly and fundamentally misunderstood everything. You have completely miscategorized what this is. [01:29:09] Speaker A: Why it is valuable and what it. [01:29:11] Speaker B: Is doing for the world. And I highly suggest you go through Anil Sedso's entire list, go through that list, listen to those episodes or read the pieces. You'll get a lot more out of it if you actually just read it. But I know a lot of people don't have time. That's why Bitcoin audible exists. If you don't pull anything meaningful from it, that's fine, I will just say good luck. But if you think Bitcoiners are worried about your crypto token, nobody cares, man. I don't even know what any of them are doing anymore. I could, I could not care less. [01:29:44] Speaker A: And I hope for your sake you. [01:29:46] Speaker B: Realize it and you move to spend your time, your investment, your building and put your idea that you're hoping to. [01:29:53] Speaker A: Accomplish or the feature thing that you. [01:29:55] Speaker B: Want into a bitcoin network, into lightning, into something that actually has a foundation and is going to last for the long haul because. Because simply, it has better settlement assurances. It's the settlement assurances, stupid. And nothing competes with it. As Michael Saylor says, there is no second best. And with that, I will catch you on the next episode. Don't forget to check out Fold and the mini affiliate. We have a shout out to everybody who boost and zaps. I am Guy Swan. And until next time, everybody. Take it easy, guys. [01:30:46] Speaker A: Any fool can know the point is to understand Albert Einstein.