Casey Muratori: Hello everyone and welcome to another cryptocurrency interview today I’m talking to Amaury Séchet and I’ve talked to him before actually but we spent so much time talking about just tech stuff because it was a really interesting interview that I didn’t get a chance to actually ask him about some of the more like systemic adoption questions or like why this will lead to any actual different outcomes than our current financial system, which is kind of something that I want to talk about a lot because even though it’s not specifically technical, it implicates the technology because it each thing we do technologically has to be justified by thinking it will achieve something, right? Because if it won’t achieve something then we’re just solving cs problems because we want to have fun, right? which is a different thing, so uh Amaury welcome back to the program.
Amaury Séchet: Hi, hello.
C: And uh so I guess just to get us started today uh yeah could you give me a little bit of an overview on, I think you understand my question generally about why is this technology going to produce something different, but just to set it up I’ll simply say you know “technically” there’s nothing that stops any people in the United States for example from running their own bank. In theory anyone can do it, but in practice only a very select few people can do it because the government has sort of created a system over time whereby they have so much control over the ability to do things like accept customer money and hold it or loan it out, all those things that really even though there’s no technical reason anyone could start a bank, the practical reality is almost no one can start a bank, so what I’m kind of interested in getting at is are these technologies we’re talking about with cryptocurrencies actually going to change that fact on the ground or is the end game for cryptocurrencies just going to be the exact same thing which is technically anyone could run a Bitcoin wallet or validator node but in practice the government has it locked down so hard that nobody can accept the big banks or whatever.
A: Yeah, so that’s an interesting question. I think it’s going to be way harder to lock down than traditional banks, right? Because the traditional banks operates in like real buildings with you know real money and real people working there and stuff like that. Bitcoin and other cryptocurrency are inherently protocols so you can you can harass someone or you can shut down one server but that doesn’t do very much for the network and in fact every time you do that you kind of like create a lot of publicity around that specific network and so it’s likely that that you’re acting you know against your best interest. So I don’t see that as much of a threat at least not the direct attack. I think there are more indirect threat though. One of them is crypto people you know being focused a bit too much on number go up and not enough on the on the bigger goal, even though like I don’t want to say number go up is bad, right, like it’s very good a lot of people are making money I have nothing against that, but at the end of the day there is a promise to change things, right? And that money or that value is created on the promise that this is going to change something in the real world and deliver value there, right?
C: It’s sort of part of the sales pitch you could say, although not everyone is doing it for that reason. The sort of zeitgeist or the public facing statements about it tend to be along those lines.
A: Yeah and you want it to be along those lines, right? Because everything that is not along those lines is hype and hype can work for a while but it cannot sustain itself for very long, and so everything that is hype in like three years we’re not going to even talk about it anymore right? We may talk about other stuff that are also based on hype, but the one that are doing hype right now they’re going to cycle around, right?
C: Yeah, there’s going to be another hype.
A: Yeah you know they always are hype right but those things that are here to stay they need to deliver value in some way or they are not going to stay, right? And there are risks there, right? Because what we’ve seen, a recent trend in crypto that have been worrying me quite a bit and that is like a big motivation on working on the project I work on are that there have been scalability issue or other kind of usability issues with various projects and everybody is using custodial solution so that you know the custodial person can deal with those technical orders, and that defeats the whole point. Like in in many ways this is even worse than banks because at least for all the negatives that banks have, they have insurance and stuff like that. If an exchange has custody of your funds and the exchange goes bankrupt well tough luck right? Like all the money is gone right so in some way those actors they don’t have the guarantees that bank can provide you. They are much more risky and inherently they are custodial as well, right? So that doesn’t strike me as a very good development personally and the way we avoid that is by making the custodial unnecessary. That doesn’t mean that they’re never going to exist. Maybe someone can make a service that is custodial that provides some value to people, but generally you don’t want that to be necessary. And in many way for the person providing the service being a custodian is also a liability, because now they have your money so they need to take a ton of precaution so that they don’t lose it. They also probably need to have some personal information about you, and you know if there is a hack or something all of that is going to spread online so a lot of that business they do because they have to, but they don’t really want to be doing that. It’s not their core business right? And so on that front if we can live in a world where the custodial is not necessary, then I think we are in the much better place. If we cannot make the solution good enough to work that way at scale then I think the whole point is kind of mood.
C: So basically if I understand you correctly you’re sort of saying, and I think I would agree with this intuitively because it seems to make sense from what I’ve heard, you’re basically saying that if we don’t get this working as a true global peer-to-peer system, meaning the people who are doing the transactions are themselves in control of their fundamental base ID, like no one else has the like final signing authority, then really we don’t need to bother. That’s the only actual end goal that we would bother talking about. Is that a fair statement?
A: Yeah maybe that’s a bit extreme, but yeah that’s the simplified idea yeah.
C: What’s the less simplified version, or less extreme version I should say, of what you think would be the minimum necessary accomplishment in order to start considering this system valuable at the limit?
A: Oh, so I mean there are a ton of stuff that those systems do already that are valuable, but none of those stuff are like, none of those stuff are really like about civilization changing, right? Whereas what we described is really civilization changing, so you know all those other stuff for me they are more or less of a running error you know? They’re interesting, and if you look there are many projects that explore many stuff and some of them do useful stuff, but really I think for me that’s where the the big goal is right?
That’s where we have a problem, that there is like no reasonable way today that exists that we’re going to solve it, and that’s a real problem in the real world because now like everybody’s money is controlled by someone else. And I think that’s a pretty big deal when you think about it. It’s actually a pretty huge deal, right? Because the power that is over you is way greater. It’s even worse than having an army or something like that because you can still hide you know? Like if you are a clandestine somewhere, it’s going to be a very tough life, a very risky life, you don’t want to be in that position, but it happened you know during history there have been periods like that and people have made it through, but well if people can catch you completely from the…
C: I mean like Edward Snowden for example is still alive right?
A: Yeah, but the deal is like if you can be cut out of the financial system completely, it’s way worse, right? Because you cannot live anywhere, you cannot buy food, you cannot do anything right? And what you would do so far is use cash, but as more and more things are electronic the solution of cash is you know just evaporating as every day there is less cash around, right?
A: And so they need to be like a replacement for that cash technology that comes online. You know so there is this saying from Linus Torvalds that I think is very interesting where he says you don’t have to trust me. Like you can’t trust me because you don’t have to is the actual quote.
C: Haha okay okay, gotcha.
A: Right? And it means that you have these checks and balances, right? If people know that you can use cash, but they control your money in other aspects, well the actual control that they can exert is fairly small because if they overplay their hand, everybody is going to revert back to using cash.
A: But as less and less options are available with cash that power that exists somewhere else is more and more likely to be abused.
C: Yes, uh that’s a very good point. In a sense having an alternative is in some ways, it’s similar to games like nuclear weapons and things like that. It’s not so much about whether or not they’re used, it’s whether or not there’s a plausible threat that they would be used, and that threat keeps the other system more honest because even though no one really wants to go all the way to the other thing, the fact that there’s that backstop creates incentives in the other domain. Would that be a fair statement?
A: Yes, yes, and this is something that is funny because the threat has to be has to be very credible, right? And the more credible the threat is, the less likely it is to actually happen, which is an interesting paradox, but this is how it works.
C: Uh so let me drill down on a few things you said there, because I think we already mentioned, or I should say you already mentioned, a few things that I’m curious about because I don’t see exactly how they play out. One of them I sort of have a hypothesis about how it would play out, so I’ll ask you that one first because I do have an idea, and the other ones I don’t so much have an idea.
So the first one is with regards to if the end goal has to be that individuals have the ability to run their own cryptocurrency as a node, meaning no one can interfere with that because they’re not forced to go through some centralized exchange, whether or not we actually need this or not, like to your point about it just being a threat, the system at least has to plausibly be such that large numbers of people could do this in order to make other banking get more honest or something like that. So assuming that we have to live in that world, what is the solution to uh, for lack of a better term, off-chain welching, off-chain refusal to honor contracts. So for example in this world in order for this to be a credible threat, I have to be able to use it like cash. I have to able to go to the grocery store and pay for a box of cereal with it. I have to be able to go to the movies and watch a movie with it. Anything that a person might do, buy a car whatever. Suppose the person doesn’t give me the car right? Suppose the person doesn’t give me the cereal. There’s no way to cryptographically insure these things because it’s a two general’s problem. It happens off the chain so it cannot be validated. What happens there?
A: I mean, yeah that’s outside of the crypto system and what happens there is what happen with any other you know exchange method. Like you would have to go to your justice system, and if the justice system is inadequate, then what’s going to happen is society turns violent usually, right? Because people then takes matter into their own hands, which you know is not desirable, but like um I don’t think there is something specific to crypto there, or I am missing a part of your question.
C: Yeah, so I think there is something specific to crypto there, so I’m glad you answered the way you just answered, because that would at least be my hypothesis. The reason I think this is specific to crypto is because doesn’t that provide the obvious place where the government immediately exerts all their control? They just say no lawsuits about any transactions in crypto. Don’t they just disappear overnight basically? Except for illicit transactions at that point?
A: Yes and no, right? Because you don’t have like one government, and so then it becomes like a huge legal arbitrage opportunity.
C: I’m not sure I understand why, because when I buy all the things I buy, 99% of them are just transactions in my country. I mean I have to get cereal here.
A: Yeah, so that that’s not going to fix things in the country per se, but that’s going to cause a lot of wealth to move away from that country, which is an incentive for other countries not to do it, and also like for the people able to move, probably many of them are probably going to do it right? Like that would be economic suicide. And there are many kind of that like you see country doing it all the time like Venezuela what not, if this system becomes big enough, that would effectively amount to doing some kind of economic suicide to do that.
C: Well I mean would it? Because credit cards and all that stuff still exist, so why would like…I mean if you took a country like the US and they you know they said that about Bitcoin. They were just like all transactions on Bitcoin are considered effectively non-justiciable, we won’t enforce them with the police, we won’t enforce them uh through the court system, or anything like that. You’re on your own. I don’t see how that would be economic suicide. People would just pay for the groceries in cash, right? Credit cards I should say.
A: Well, there are a few reasons there, like first, the US has an economic system that works pretty well, right? Overall. And Bitcoin itself, I mean like if the US would want to control Bitcoin it’s even easier to just control the custodian right? Because people are using Bitcoins through custodians, so if the US wanted to block that, right now that would be completely possible, so I’m assuming some world where this crypto space is big enough, so that matters right? Right now it’s not big enough so that it matters very much, so that wouldn’t be that big of a deal. Unless it becomes big enough in another country and then you get back into that arbitrage situation.
C: Can you explain a little bit more about the arbitrage situation? Because I might sort of get that, but I might not. In other words, suppose take the example you were talking about. Let’s assume the most favorable example to the cryptocurrency where we find ourselves in a world, for some reason where people are literally paying for their groceries in Bitcoin in the US, or it doesn’t have to be Bitcoin right, but some cryptocurrency solution takes enough hold that people are routinely paying for their groceries in crypto. For some reason a government decides they don’t like it, you know some president gets elected, or some congress, and they say that’s it, we’re taking this out of here. If you want to do Bitcoin we’re not going to outlaw it, but it’s just it’s wild west now. We’re not going to enforce it, we’re not going to consider them valid transactions so you know you you’re on your own. What do you mean by arbitrage? Like what would the citizens of this country actually do to arbitrage their wealth or whatever it is you imagine would create the counterbalance to that?
A: Well, it’s not so much the citizen, but it’s like the opportunities that involve crypto are going to happen somewhere else and not there anymore, right? So let me take a very concrete example that happened in France where I lived a few years ago. So um almost 10 years ago now, Fancois Hollande was his executive president and one of his one of the promises that he made you know is that his enemy be would be the financial system, and that he would fight it right? You know remarkably enough, this is a promise that he held, which is very uncommon for politicians. But this time he yeah, so he passed a bunch of measure that made it that made it worse, effectively, for financial institutions to operate in France, and as a result the whole business in the area decreased by like 70% since then or something like that. Like it’s been decimated in France, and there have been plenty of negative consequences, right? Because it’s not just the financial system that is impacted. No, it’s like you want to make your business, you need the loan to get started, you know like buy equipment or whatnot. Well there is nobody to make you a loan now, right because the people that were doing that they are not there. So it has like a huge impact and you know like those kind of decisions they are very bad for the country over time. Think about that like you go back 20 years and France had like the best train, the best airplane, the best internet, and a few other stuff, and right now … eh. Right? Like the best train are made in Asia now, and the best internet is … I don’t know. But like none of that in France anymore, right? And if you go in France in the early 2000s like everybody had like TV that comes through the internet like you know Apple TV and or Chromecast or stuff like that. People have right now but well like everybody had that at home 20 years ago, right? But since then, and why all of that stopped? Well because the opportunity are not there to create those businesses in France anymore, and they were there 20 years ago. So this is what’s going to happen right? Like this is not a huge break that’s going to stop everything, but it’s just like the opportunity that would have been created there over the years are going to start to be created elsewhere, and little by little the economic weight of the country that make those kind of decisions are going to become like less and less, and other countries are going to become more relevant.
C: I guess I’m not sure I follow that because while I absolutely agree that if a, and I think we have probably more examples in history than just France, certainly, so I think it’s a very strong case that, in general …
A: Yeah I’m using France because you know this is where I live and I know the situation but many countries did way worse.
C: Exactly, so I totally agree with that, and I’m sure I could think of a number of examples that were you know even more deleterious. My only problem with that example is those are like the kinds of regulations that involve what you can do with money, not which money you can use, so to me it doesn’t sound that plausible only because the kinds of people who have the sort of wealth you’re describing where they’re going to be funding businesses and actually like putting up capital, they have access to banking in basically every form, so if somebody told them that they couldn’t use bitcoin in France, but you could still use any other banking technology from traditional banking, they would simply use that. I don’t see them going, well, because I can’t literally just directly use Bitcoin in France I’m not going to convert those things to just like straight fiat Francs or something and fund the company, right? So I don’t think I get it I don’t think I get why this is relevant?
A: Well because if we assume that economy to be big then effectively like you are adding friction. Everything else being equal you are going to do stuff where there is less friction. You know like why would you do those international money transfer and you know exchange your crypto to the local currency, but in some other country because you cannot do it in the US anymore, and then you know like move the money around internationally to fund those stuff. Where you have an alternative opportunity in some other country that is you know just as good.
C: But what’s the friction? Large players already have trivial ways of moving money, what’s the friction? Like what’s the difference? If I want to move a million dollars into France right now to fund a train company, well, I shouldn’t say France because like you said there are regulations, but let’s say take some place that didn’t do that. I want to move a million dollars into the US to fund it, what’s the friction I’m going to suffer relative to Bitcoin in this case?
A: Well if you have bitcoin you would have to transform it in US dollars somewhere inside of the US and then wire it to the US and the whole process is going to take like at least a week and involve I don’t know how many intermediaries, which I mean is okay, you can do if the opportunity is really good, but everything being equal you know like people are just not going to do it there because it’s just like when you have like many huddle to go through, then you go to the place where there is less hurdles like very simply.
It’s like you know I can start that business in the US and then you know like every time I want to move the money it’s going to take a week and then I’m going to have to change it various time, or I can create my business in like Switzerland or what not, and then everything is done like right away. Like everything is a phone call away right? Everything else being equal you’re just going to not do it in the US in those circumstances.
C: I mean again, sorry to push back so hard on this, but like I said, this is this is the part that I’m much more skeptical about, right, is the part about like will these play out well? So my questions are a lot more sort of like specific because I feel like the devil is definitely in the details for these sorts of things.
I mean one man’s friction is another man’s insurance might be what I would say. So you know okay the way Bitcoin works is I have to make a transfer. I have to make it to basically an anonymous address because I have no way of guaranteeing that the thing on the other side is actually even the correct thing to send it to, or who’s actually controlling it, or any of those things. And it can’t be reversed, right? Like I can’t. Once it goes, it’s never coming back unless I go convince 51% of the mining pools that it should, which if you can do that with any regularity it defeats obviously the purpose of crypto in the first place. So as a person investing capital, why wouldn’t I want it to be something that goes through the court systems and can be reversed by a government? Like don’t I want that extra friction where there’s actually people who can reverse this transaction if it goes bad?
A: Yeah so you can have, so the reverse part no, but otherwise you can have protocols that solve like all the other problems that you mentioned. There is like for instance there is the stamp protocol. With that protocol if you want to send a transaction to someone you are not going to like broadcast it on the network, which is kind of dumb actually, because like imagine if the supermarket would work that way, right? Like there would be a big box, and everybody go to the cash out, right, and throw the money, like throw like a check or whatever in a giant box and there’s like hundreds of them in there and the cashier has to like go in the box and move them all around to find yours, and be like okay this is like this guy paid the right amount. That’s a bit retarded, to put it mildly.
So the way the stamp protocol works is that the person that pays is going to send, effectively, like an encrypted transaction to the person that received the money, and the person that received the money is going to decrypt it and broadcast it to the network. I’m simplifying a bit. There is like a bit of an exchange that is going on, you can like, you can negotiate various stuff when that happens, but effectively like the basic idea is like you know one send to the user. And by the fact of decrypting, you know that the person you are sending the money to has like, he’s in control of the money right? Because otherwise they cannot decrypt the transaction and so you know that you’re not sending that to the wrong person or to a dead address or to whatever.
C: Okay so basically you encrypt it first, effectively with some kind of a key that they give you so that you know that they’re the only person who could have received it kind of a thing or…?
A: Yeah so in practice you do that um.. in practice you don’t do that with a key that they give you. They effectively like have a public key that everybody knows and then you do Diffie-Hellman to generate like a shared secret that only the two of you have, and based on that you can derive various keys and use those key to do the thing, right? But if they can decrypt your stuff, it means that they can do the Diffie-Hellman and derive the key and all of that just like you did. And if they can do that then they are in control of the coins like you know that for a fact. And nobody else other than the two of you can identify those address either because they’re generated you know from a shared secret so this kind of system can solve most of the problem that you mentioned except like the justice system specific one.
C: Right, which is kind of the place where I feel like I’m not as convinced that there’s any real incentives for the country not to do that, but it’s like at least I know you’ve got kind of you’ve gone over your thinking on it which is now I know.
A: Right, yeah yeah so obviously like this did not happen so we can only speculate, and we may evaluate different things differently, but for me personally that would be a huge endorsement actually. If some country did that because that means that this thing has grown big enough so that they take it seriously, right? Which like in the price that may reflect negatively and stuff like that because like the more casual player they may be like, oh you’re like this is not so nice anymore, like this is like this is something else now and maybe I don’t want to get involved, right? So maybe some players are going to exit but this also would be a very good sign that this is actually important now.
C: Okay so let’s talk a little bit about currencies versus investments, because this is sort of a separate issue as far as not so much I don’t think it matters so much for the end game, because basically my previous questions were about like where do we end up, right? Like best case for crypto where do we end up?
A: It’s interesting, like can we stay on that question a bit, because I think actually I think the main issue you did not touch, and I think the main issue is the volatility.
C: Yes, well actually that was kind of my next question, but okay. Yeah so you can feel free to expand on this more than what I’m saying then because it sounds like you have a separate idea too about what you’re thinking, but just to set the stage for it, right now cryptocurrency isn’t really used as a currency as much as it’s used as like an investment. I don’t know if investment’s the right term. I mean some people just outright call it a ponzi scheme at this point right, but no matter how you look at it, it’s obviously the case that there’s a huge fraction of the people who are actually using these systems, aren’t using them primarily as a currency, meaning the value of Bitcoin right now is absolutely not in even for illicit transactions separate from like you know like things that would be legal. So because we’re starting off in a place where everyone’s sort of treating it as something where they want to see, I think you use the phrase number go up, which I’ve heard a lot as the as the phrase for that, where a lot of people are looking at this as more like gold where they’re betting on whether the price of this thing relative to some currency in their home country will go up or down, because then they can exchange it out for more money later if they’re right. If you’re in a situation like that, how does it transition from being like gold, which no one uses as a currency because we switched away from it specifically because gold doesn’t make a very good currency for a number of reasons. Why would it naturally transition over to a currency whose primary value is not, is that it has a predictable slow inflation rate but otherwise very little volatility.
A: So I think there is an interesting tidbit of history that I don’t know if you’re aware of or not, but I think it’s worth going over. The whole gold stuff is fairly recent. Well maybe not that much anymore, but it’s maybe like five years old something like that.
C: What gold stuff?
A: Yeah yeah considering those stuff as like you know a gold like kind of like digital gold kind of thing.
C: Oh you mean for cryptocurrency being considered like gold?
A: Yes, and before that it was used very much as a currency. Much more than it is now. If you go back to 2014 for instance, in 2014 you could buy like Dell computer with Bitcoin, you could buy game on Steam you could um…
C: “Stripe” actually had it as one of their payment options. You could just enable. I remember that because we use stripe on our website and they sent out mails like you can accept Bitcoin and then like a couple like a year later or something, they were like you can’t accept bitcoin anymore.
A: Yeah so the whole stuff started more as a currency than what it is now, and there is something that changed the direction of the space quite dramatically. It’s basically when those systems started running into capacity issue right because then the property that they have as a convenient medium of exchange become horrendously bad right? But historically the value have been going up so this part like did not really change, and people still invested in that for that very reason, and there are some property that are desirable in that front as well. So I think this is not much of a mystery if this can be a currency or not, because I think this already has proved that it can for as long as you don’t screw it up with capacity issue, right? Now that being said, it’s probably going to take some time before people realize that again because this whole space has been destroyed, and it’s always easier to convince people the first time then it is the second time when you screw it up. So I assume that it’s going to take some time you know for that idea to be taken seriously again, but I don’t see it as a blocker necessarily, like this technology has proven that if it’s done correctly, it can be used in that direction and be quite popular in that direction. Though that being said, yeah the volatility issue is more serious because some people can take the volatility, but uh not everybody can, right? Like the more you can’t the more of a problem this is, and so this part is not ideal. But there is one thing that is very interesting and that nobody is talking about in the crypto space. It’s why there is so much volatility. Like there is a reason right, so what’s happening is that you have an economic system you know people buy and sell and stuff like that it determines the price, and you have a bit of randomness that is associated with that, right? Because it depends on the action of like many people, it’s not stable by nature and it moves a bit, right? And most systems of price, for most goods out there, they tend to stabilize, because say if the price of oil go up then some people are going to use less of it, or maybe use alternative, and at the same time people who are drilling oil they may start exploiting some like more difficult to exploit places that are more expensive to exploit as a result, but because now the price is higher they can do so profitably. So on one hand the supply increase and on the other hand the demand decrease such as you know the price tend to stabilize again right? So there is there is some stabilization effect that is going on there, and this is the case for most good and services that are out there. For Bitcoins it is different, and it is different because there is so little actual use outside of being a speculation asset, and the structure of things make it so that it’s actually an amplifier of randomness because…
C: Oh, okay please elaborate.
A: Yeah you have a large portion of actor in the space that make their revenue in Bitcoin, so most of that would be miners for instance, but those people they have expenses that are in fiat, so miners they need to pay the electricity, but even if you’re a business you have like offices to pay that are priced in fiat and stuff like that, so what happens is that when the price goes down now those people need to sell more to pay their expenses in fiat, and so the price goes down even more, and so they need to sell even more and you have a runaway effect until the less profitable business in the space go bankrupt, effectively, right? And suddenly there are less coins that arrive on the market and you can reverse the trend.
C: If I may, if I’m understanding you correctly, the opposite would also be true because if the price of Bitcoin goes up, I have to sell very little of that Bitcoin to buy new GPUs for mining or energy to power them so I’m spending less to mine, so I can mine more if I want to, so in some sense both directions are the opposite of the original example you gave.
A: Yeah yeah so exactly, this acts as an amplifier of like the random noise that is on the input, up to the point where you reach something that is non-linear, and when you go down the non-linearity is like some people they just don’t have Bitcoin to sell anymore, and on the way up it becomes like well the Bitcoins that people have are worth so much more now that you know they kind of want to cash out some of it at least, right? And this is where like you have those behavioral change that can reverse the trend. But otherwise the machine is just like going down and down and down or up and up and up, right, and the stabilizing factor is people actually using it as a currency, or if it’s a smart contract platform using the smart contract, or whatever right? Because people using it as a currency, well when the price is higher they’re richer so they can buy more stuff and create more supply on the market, and when the price is lower, then you know they are poorer so they buy less stuff and…
C: So in some sense it’s just like until the energy in your company, sorry in your country is denominated also optionally in bitcoin, so you can make a direct Bitcoin transaction to pay for energy or something like that, you’re going to be in this nasty situation where because it’s not a system-wide stabilization, until you reach that point, I mean basically everyone has to…
A: Not necessarily no, not necessarily. You would need to have the group of economic actors that act as a stabilizing force become bigger than the one that acts as a destabilizing force. And when that happens you get more stability. Now the problem is the volatility is more of a problem for the second category of actors, so you get a bit of a chicken and egg issue that is going on there, right? So this is why people have been looking into stable coins and stuff like that to try to break that that chicken and egg situation. I can’t say that I have a good solution for that, or like I don’t have a great plan to propose to you, but what I suspect is that like maybe stable coins are way to push through that, maybe the situation is going to be so bad in some country that people you know like move over, and then it creates some mass and the bigger the mass is and the more stable the system is effectively, right, so maybe it happens through that way, maybe it happens through something that you know did not exist whatsoever, right, really if we look at it historically and assume maybe the same kind of things are going to happen, really the thing that made Bitcoin start being used, actually being used is Silkroad, right. because this is the first time that you have something in there that you can buy with Bitcoin that you cannot buy with anything else, and that doesn’t mean that Silkroad was the only use for Bitcoin, or even the biggest use for Bitcoin, there were plenty of legitimate businesses and stuff like that, but for those legitimate businesses you have other options, but the fact that there is a category of users that are interested in something for which that system is fundamentally different is what bootstraps it, for better or worse. It doesn’t mean that it’s the biggest use case or, the use case that everybody use it for, but it’s a bootstrap mechanism so you need some kind of bootstrap mechanism to get the machine going. So now that the Silkroad ship kind of have sailed, so it’s not going to be Silkroad, but who knows, it’s very difficult to predict the future.
C: Yeah so in that sense it’s kind of like, well, you know you’re just kind of at the mercy of the fates. It’s like either something naturally arises that gets enough people on it, or it doesn’t and no one necessarily has a plan for why that will or won’t happen, I guess.
A: Oh many people have planned for that I guess, or like hypothesis on what that could be, but I’m much more focused on working on the protocol and making sure that it has the right properties and stuff like that, so I’m not the best person to describe those plans or even like tell you what are the most likely to pan out or not right, but I can tell you that some people are spending the whole day looking at that.
C: So I guess I feel like we’ve already got a lot of great stuff here. This has been very interesting, but so maybe let’s wind down with the, if you could, we’ve talked about a lot of stuff, and it seems like at least in your vision of the end game for cryptocurrency, the main thing we’re going to get is the fact that governments wouldn’t be able, well not just governments but people in general like strong actors let’s say wouldn’t be able to cut people out of the payment system, which I think we both think would be nice because right now it’s maybe a bit too easy for that to occur from what I’ve seen.
A: Yeah actually I’m fairly convinced that in the coming years we’re going to see much more of that. It kind of happened you know like with Patreon kicking out a lot of people and stuff like that, but as cash disappear, this is going to happen more and more. And you always have this stuff like well it happened to the extreme and those guys they are not very nice anyway. We don’t really like them so we don’t feel that sorry for them, but the thing is like the next guy that was a bit less extreme is now the extreme right, and little by little it erodes, and at some point you’re like, yeah, this is getting pretty close you know.
C: Yes absolutely, so given that, I mean first of all would it be correct to say that that’s the primary thing we would get out of this before I go further? Or do you think there’s other things we’re going to get out of this just so we start with that.
A: I think there are two very big things that we get out of this, and this is the first one, and the second one is like much more of a bigger time scale. So if you look at pretty much the history of every single civilization that exists, you can map it on the history of their monetary system, right, and always like the same thing happened always for every single one of them, right? At the beginning they use some kind of hard currency like precious metal or stones or like something that you cannot make out of thin air, right? And they use that and they have some success in their civilization. The whole thing grows and stuff like that and it becomes not very practical anymore to use gold, so they come up with like a ton of system and they create effectively like fiat currency, and then it works much better as a medium of exchange though it doesn’t have those hard characteristic that you know prevents some kind of like catastrophic scenarios, right? And so the whole civilization like grow much more and become like the Byzantine empire or the Roman empire or whatever, and at some point you get the inflation becomes too big, effectively, and the value of the money goes down and the empire cannot sustain itself anymore and it fizzles out.
And the reason why you see that happening again and again and again, it’s not random and it’s not because the people participating they have an evil plan like, oh I’m doing that and in 300 years the empire is going to collapse, right? Like it’s not what’s happening right, the people they’re just solving the problem that they are having and the problem that they’re having is that the hard money is not convenient to use as a medium of exchange, and the things that are a good medium of exchange, they don’t have good hard money properties. And well we have a technology that I think can can be both, which is a first in human history, and what are the consequences of that are probably going to be like very unpredictable. Because that changes some stuff at like you know the fundamental level in the way we operate and we interact with each other. That’s probably going to be very strange, like I would really be curious to see in a few hundred years what is the result of all of that. I’m probably not going to be there right, but I would really like to see that. Because when you look it on that front ,it’s kind of like on the same plane as say the printing press or the internet itself. You know like those kinds of technologies they’re really changing things in a profound way. So okay, the internet change, especially the internet, but in particular social media, changed something that I think most people have not caught on that is extremely fundamental. For the whole of human history information was scarce. So if you were a scientist or stuff like that, you would want to read like all the paper that you can, even the one that are bad because it’s so difficult to get them all right? Because someone has to like you know copy them on some like dead skin animal and ship them across Europe with some dude that is going to carry them on horses and stuff right, so like this is a complicated process, right, so when this is the process, well you want to have all of it, but about ten years ago I’d say we switched from a situation where information is scarce to a situation where information is like absolutely abundant, and actually we have like so much of it that we cannot possibly process it all. Like you could spend your day scrolling various social media platforms and never run out of content. And actually you would barely scratch the surface on all the content that exists on the platform.
C: And that is true of the aforementioned scientific papers as well. There’s no way for an individual to read all the scientific papers. Potentially even in some fields there are so many, let alone all of them across all fields.
A: Yeah and the whole stuff is going ballistic right, like this is it’s exponential, and I think like a lot of the madness that we see especially over the past year come from that. Because you know like what makes the difference between the madness of the crowd like let’s say the mob, and the wisdom of the crowd. It’s very funny because you have those two phenomena right, and so what makes the difference typically when the crowd is wise is because everybody is making a decision with little coordination. So for instance there is this very famous experiment where you put a box of pasta, and you ask people how many of them are there in the box, right? And obviously like I don’t really know. I can try to guess, but my guess is probably going to be fairly inaccurate. But the thing is if you ask a bunch of people they all make a guess and the guesses like vary widely in all kind of direction, but at the end of the day like you average everything out and you get something that is like remarkably close to what there is in the box. Like to a precision that is really surprising. You know like typically you get like a few percent of error or something like that, which is quite remarkable. And the thing is like some people are going to overestimate, some people are going to underestimate, and if you ask enough people at some point it cancels each other out sufficiently so that you get the very accurate result. But now if you put all the people in the room and ask them okay you’re going to discuss with each other and at the end of the day you’re going to come back to me and tell me how many pasta there is in the box, like the answer that you get is random crap because now they’re going to debate with each other and try to find a way to estimate it and whatnot, and yeah so this is what happens when you have this high information flow between the people. They tend to create mobs, violence or not, but they tend to go crazy generally, and this is what, well I’m going to get a bit political. so I apologize in advance….
C: Cryptocurrency is already pretty political so…
A: Look at trump in the US, like half of the country thinks he’s literally Hitler, and the other half thinks he’s the second coming of Jesus, right? And when you look at it objectively he did remarkably not that many things, like compared to…
C: Not to also be remarkably political but I very much agree on that, which is that if you look at the actual in practice things that occurred during that time period, if you hid that information from someone like what actually happened and you showed them what people were saying about it, like an alien who came down would be like wow there must have been like incredibly crazy stuff happening, that this administration was doing or something, but the but the actual truth is, no, it’s just it’s just what people imagined on various sides of an issue that never really happened.
A: Yeah and well this is that phenomenon right? And that phenomenon happens because we are now in an abundant information world and it probably means that we’re going to develop new cultural norms to deal with that, so all the norms that we have around information are like oh censorship is bad, and the freedom of the press, and all of that. There are norms that are geared toward making sure that information flow in an information scarce world, because we discovered over the years that this is better that way, but those norms are completely inadequate to the problem that we’re facing right now. Like it’s literally impossible to censor someone right now, like there’s so many platforms that if you wanted to ban that person from them all, but it doesn’t mean that that person is going to reach like so many people, right, because that person is going to be drowned in the noise, and this is more of the problem that we have. It’s not making sure that people are not censored or whatnot, it’s not that relevant anymore, what is relevant is making sure that the interesting or valuable information or whatever comes out from the general noise, and you know in the next hundred year or so, we’re probably going to have very different social norms around all of that because the shift is so drastic, and it was a long-winded, you know side quest, but if I come back to crypto I think, like if it catches on and it becomes big, some kind of similar things is going to happen, and that means that we’re going to end up in a space that looks very different from where we are now and in a way that are very unpredictable to us, because you know we don’t really look at reality, we look at like we have models in our minds of what the reality is like and we use those models to understand what’s going on around us, but in those models there are a bunch of heuristics right, and we kind of like need them because you know if we don’t have them then it becomes like way too complicated to process any situation.
C: Yeah our brains aren’t set up for dealing with that much stuff right, they have to simplify to function.
A: Yeah so there is an amount that happened like at the hardware level in our brain, but there is an amount that is learned as well, right? Like when you see the situation plenty of time in your life, and so that part is malleable, but we’re not that conscious of it because it’s basically always true in everything that we see around us. And you know like maybe it’s going to be false you know when you go see a magic show or something like that, because this is effectively what they do, like you have a model in your mind of what this, like the trick happened in your mind actually when you look at magic show, right, because you have a mental model of what the situation is, and suddenly something happened that don’t fit into that model, and this is what a magic show is.
But this is uncommon enough so that you can make a show out of it right? So that means that it is very uncommon. But for those things this is true like always when we think about it. And like as a result it’s very unlikely that you and I can have any kind of current idea of what it’s going to look like. We can speculate about it, but it’s probably going to be completely wrong, and actually I was watching your your presentation about how a CPU works and that kind of reminded me of, yeah I watched that this week, and it kind of reminded me of something that I find is like an amazing piece of computing history.
You know like there was a time where there was a lot of debate between RISC and CISC right? And the RISC people were like well this is simpler to make a processor that way, this is also simpler to write the compiler, and so at the end of the day RISC is like certainly going to be better, right? And the CISC people are like oh but we can do instructions that do better stuff that are more suited to what the program is actually doing, and that argument went on forever. And one thing that is quite remarkable is that it’s pretty obvious now that CISC has one, because like even ARM that used to be a RISC platform now they have like this new coding, well it’s almost 10 years old now, but they have those new encoding and stuff like that, so RISC now is also a CISC platform in many ways.
C: And vice versa to a certain extent. You could also say vice versa. I mean one of the things that’s interesting about it is, I would say in line with your comment as far as CPUs are concerned, very few people back then would have predicted that what we would basically have is a CISC-like instruction set because you want to represent the instructions as concisely as possible but a RISC-like execution engine because you simply can’t build something that complicated under the hood, so it’s translated along the way, which is like you’re both right and wrong.
A: This is where I was getting at. All the argument that the RISC people made at the time, they are basically all correct. And all the argument that the CISC people were making at the time, they are almost all wrong. And yet CISC won. Why? Because what happened is that you effectively make your RISC CPU and then you wrap it around with something that translates the CISC instruction into RISC construction, and you’d be like why is that better right? Like you could have just the RISC CPU. Well it is better because the instruction set is more compact, and so you use less memory bandwidth, and you can store more of it in your cache and you have less cache miss and all of that, and it turns out that it’s a cheap man compression algorithm for your instructions, which is something that nobody saw coming at the time. And you know like it’s one of the problem where people they were mapping the situation where the mental tool that they had at the time, and the people that understood the situation better ended up being wrong even though everything that they were saying was right, and the other group was, I’d say kind of right in their conclusion even though the reasoning was completely wrong because they didn’t have one key piece of information, which is that memory became so slow compared to the processor, and all those debates that are around crypto are on like social media as well as I explained, I think they fall probably in the same category where everybody is wrong.
C: So you might say that like from your perspective, a lot of the debates and what you should do, or will this be good or won’t it be good, are sort of like there are things we’re doing on the way, but if this ends up being important to society we probably won’t really know why or how because it’s going to look different than what we can exactly guess at this point.
A: So we can know what core property it changed, but predicting all the snowball effect from that is like virtually impossible I’d say. We just don’t have the tools, and what I say the tools like not physical but mental tools to think about that very clearly, because it’s just something that never happened.
C: So I guess, you know I don’t think I’ve really thought too hard about that part of it, so I don’t think I have very many questions, so let me just go to what my last question was going to be. I’ll have to like think about that for a little while, because I think I would agree in obviously broadly that if this does turn out to be something very important, I think it probably is likely, I mean anything where you change a financial system it’s really hard to guess what that’s going to do, because they’re very, they have a lot of behavioral aspects to them, a lot of things that are very difficult to quantify even today, so you know thinking that we know what would happen if we change a few things is probably very presumptuous.
A: Yeah it’s like you know think about it, you go back to the Roman empire when they switched to fiat and you ask the people like even the economist, you know people that studied that all their life, what do you think the consequence would be, and none of them is going to tell you, well, there’s going to be banks and they’re going to do fractional reserve and they’re going to do lending and stuff like that, and they’re going to rebalance their stuff and then they’re going to do leverage trading and whatnot. Like none of them is going to be like even the world expert on the matter is not going to be able to tell you anything that is even remotely close to looking like that. And I think it’s very much the same, so we can say like what basic property of the word is going to change, and I think that you know if we’re successful, the basic property of the world that we change is that we have something that is a good medium exchange and store of value, and if it becomes big enough it’s also a good unit of account, though it’s clearly not right now. But that’s in itself it’s not a technological problem as we discussed more of a chicken and egg problem. So we know it changed that fundamental mechanism, and we also know that that fundamental mechanism is at the core of the cycle of civilization growing and collapsing. So it is very important, and that means that the life cycle of civilization is going to be different, right, and the consequences of that are wildly unpredictable I’d say.
C: So let me close by asking the question about that fundamental difference I guess, which is technologically speaking, based on what we talked about last time, I’m now going to give my ballpark guess of what you’re thinking is the technical requirement for this change, meaning what is the actual technology that we need in order to do the thing you’re talking about. If I understand correctly, you’re thinking that there needs to be one system, so a single thing basically, that both has a predictable supply level, meaning it can’t be altered by anyone, so you know the Bitcoin proof of work means that the computing power of the day is the only thing that tells you how many there will be. You can’t just arbitrarily change it.
A: Not even that because difficulty adjusts right? So there are going to be how many there are going to be.
C: That’s true. So you kind of know, like all right there’s a specific bound that unless somebody finds bugs on a particular thing or whatever, like you know most of the time you’re going to have this very predictable supply, and a very efficient censorship resistant transaction model on top of it are the two technical requirements, and at least from your perspective right now, the best we can do here is proof of work for the I guess for lack of a better term, security, like the preventing the denial of service attacks that would come if we didn’t have some hard bound on the number of things you could broadcast per second or whatever, so there’s like a proof of work thing that requires a slowness to an underlying thing, and then Avalanche was your choice for the fast transaction system. Am I close?
A: Yeah I think it’s basically all correct except for the proof-of-work part. So the reason I think proof-of-work is important even though like it’s obviously wasteful. Actually, it’s like the whole point of proof of work is to be wastful.
C: Exactly, that’s what it does.
A: Yeah that’s exactly what it does. And so that’s something that like especially computer scientists and especially the one that have cared about optimization a bit are going to be like ehhh. This is like the opposite of everything I believe in, how is that possible? So here is why I think it’s necessary, and I don’t think we have anything better on the table right now. Maybe you know at some point in the future, but right now we don’t. It’s that it anchors the system in the real world so to speak, right? Because anything that’s going to come from within the system is de facto self-referential right? So if you do like a blockchain just the the same way it worked with proof of work but say you do it with proof of stake, and the different stakers design the block, and the more coins sign the block effectively the more work there is in it, or the more stake there is in it I guess, and the chain that have the more stake is going to be considered you know like the heaviest chain and the one that that everybody is on, right? The problem with that is that I can basically sign all the blocks that come my way, right? And even worse, you know once the system has been going on for a year, let’s say you know like in the beginning I had a lot of those coins, but I managed my money very poorly, I lost a lot of it, and now I’m not so important of an actor, but I can take back the system like it was a year ago and produce a bunch of block with all the money that I have assigned it, and as I do that I get like even more block reward and so I have even more money, right, and so the weight that I can put in each block is effectively like growing up quadratically, because each block adds weight, and plus the weight of each block is linear in the number of blocks, right? Because I make more money with each block so my chain grows quadratically, so if I make it long enough eventually it’s going to be the heaviest chain.
C: Okay so it is so in a sense you’re saying that there will just basically be races of people like trying to produce as many of these things as they can to produce the heaviest weighted chain. I just want to make sure I understand this correctly. I think I do but can you be 100% clear just for everyone listening and me.
A: That has a name, right? It’s the nothing at stake problem. It’s like if I committed on one chain, I can commit it also on the other chain, because on the other chain there is a separate state, and because the whole thing is self-referential then it just goes on. And this is a problem because if I’m a new participant on the network and you’ve been doing that for like quite some time and your change is like humongous now, existing participant they may say oh you know like this chain I’ve not seen it for like a year and now the guy comes, he has this humongous chain, but I’m going to reject it because I don’t consider that the value chain. But it’s called weak subjectivity, which means like you make a subjective decision based on parameters that you assume everybody’s going to agree on
C: Yeah you’re kind of back to, I mean at that point it’s unclear why you even needed it because if all of the nodes are just kind of thinking, do I trust this person or not, well you didn’t really need anything, we could have just had that, like when you receive something you make an arbitrary decision about whether you accept it.
A: So all those systems that use resources from within the system they can be much more efficient, but they have this self-referential problem, and if I come a year later, I have literally no way to know if your chain or the other chain is the real one, except you know like trusting, like asking someone I trust that’s going to tell me, well I was there for the whole time and here is the reality. But at this point like you trust the person, like the whole system, the person is effectively the equivalent of the bank now that’s telling you what is the balance in everybody’s account. So you have a problem here, and the only way we have to solve that problem right now is to take a resource that comes from outside the system and weight the chain with that. So that can be proof of work, there are stuff like SIA that use like proof of disk space. There are like different system that use proof of whatever. Like pretty much any computing resource that you can imagine, someone has made a proof of that since then. But you kind of need one that is tied into some real resources that come from outside the system, because otherwise it’s self-referential and you get this subjectivity problem. So I mean maybe at some point some crazy genius is going to come up with something, but like right now we don’t have anything else.
C: Gotcha gotcha, well Amaury it’s been a pleasure just like last time. I thought that was absolutely fabulous. I guess if you have any closing comments and also if you could tell people who are watching at home where to find you online if they want to follow you or whatever.
A: Yeah so I’m @deadalnix on Twitter and Github and I don’t know like pretty much everywhere there is an account. It’s going to be me. I’ve worked on BCH historically, like Bitcoin Cash, one of the big crypto. I’m not working on that anymore and I’m working on a project that is called eCash that is a fork of Bitcoin on top of which we are doing the Avalanche consensus so that we can have like instant transactions, much higher security levels and stuff like that. I guess people can learn more about that in the other interview.
C: Yeah, that one had a lot of details about that.
A: So yeah that’s what I’m working on these days and yeah that’s pretty much it.
C: Well thank you so much. It’s been a pleasure talking with you and hopefully we’ll do it again sometime.
(Click here for Part 1.)