Understanding Crypto 7: Nicholas Weaver: A Computer Scientist's Perspective on Cryptocurrencies and Blockchain
Nicholas Weaver received a B.A. in Astrophysics and Computer Science in 1995, and received a Ph.D. in Computer Science in 2003 from the University of California at Berkeley. Although his dissertation was on novel FPGA architectures, he also was highly interested in Computer Security, including postulating the possibility of very fast computer worms in 2001.
In 2003, he joined ICSI, first as a postdoc and then as a staff researcher. His primary research focus is on network security, notably worms, botnets, and other internet-scale attacks, and network measurement. Other areas have included both hardware acceleration and software parallelization of network intrusion detection, defenses for DNS resolvers, and tools for detecting ISP-introduced manipulations of a user's network connection.
Dr. Nicholas Weaver’s well-known lectures on cryptocurrencies explain why he believes it needs to be “burned with fire.” Today, we speak to Dr. Weaver, an expert in computer science and a long-time observer of the cryptocurrency space. He holds a BA in Astrophysics and Computer Science and a Ph.D. in Computer Science from UC Berkeley, where he was also a lecturer until recently. His primary research focus has been network security, among other topics. His interest and search for comedy “godl” have also resulted in published papers on cryptocurrency. In our conversation, Dr. Weaver untangles the complexities of the perceptions of cryptocurrencies with the actual technology. We talk about decentralization, if cryptocurrencies are achieving it, and the underlying concept of blockchain technology, as well as whether or not blockchains are secure and what the potential benefits of cryptocurrencies are to developing countries. We then go into detail about why Dr. Weaver thinks the crypto space is not beneficial, why prestigious academic institutions are teaching about it, and why he thinks it will never work in the log-run. Please tune in for a truly eye-opening, no-holds-barred episode as we learn the harsh truth about cryptocurrencies with expert, Dr. Nicholas Weaver!
Key Points From This Episode:
What public blockchain technology can achieve that was not previously possible. [0:02:51]
How well the original concept, introduced by Satoshi Nakamoto, is living up to its theoretical promise. [0:04:06]
Dr. Weaver explains and outlines the general appeal of decentralization. [0:04:57]
He elaborates on his perspective on trusted and honest authorities. [0:09:39]
An explanation of how Lightning Network attempts to solve throughput limitation. [0:10:36]
We find out if major blockchains like Bitcoin and Ethereum are decentralized. [0:13:08]
Differences between how nodes and miners influence the network. [0:16:23]
How secure public blockchains really are. [0:17:45]
Whether a facility for censorship-resistant transactions, like Bitcoin, is a good thing for society. [0:20:30]
Potential benefits of cryptocurrencies for people in countries with limited access to banking and a good legal system. [0:24:35]
Steps governments can take to regulate and control cryptocurrencies. [0:26:47]
Dr. Weaver’s opinion on why the regulation of cryptocurrencies has been so slow. [0:29:17]
Outline of how Dr. Weaver sees cryptocurrencies developing in future. [0:31:38]
How to incentivize miners not to attack the system to retain the value of Bitcoin. [0:38:24]
Dr. Weaver on how stablecoins fit into the crypto ecosystem and if they’re decentralized. [0:39:18]
A rundown of the new properties the blockchain data structure provides. [0:45:59]
Find out if Walmart using a private blockchain is just marketing hype. [0:51:20]
Why the popularity of cryptocurrencies amongst venture capitalists is rising. [0:53:12]
Why prominent schools like MIT and Cornell are emphasizing blockchain programs. [0:55:23]
Dr. Weaver explains what the smartest way to invest is, in his opinion. [0:58:34]
Who the technologists in the crypto-space are that he respects professionally. [0:59:51]
He tells us whether he has heard any compelling arguments for cryptocurrencies. [1:00:31]
A discussion about Dunning-Kruger economics in relation to cryptocurrencies. [1:04:38]
What the general opinion of other technologists is on crypto and blockchain. [1:06:23]
We end the show with a final takeaway from Dr. Weaver. [1:07:31]
Read the Transcript:
Nick, theoretically, theoretically, what does public blockchain technology accomplish that was not previously possible?
It makes it easier to avoid money laundering loss. So we've had notions of digital money for decades now. Your bank is digital money. We've had people doing peer-to-peer digital money for two decades plus. PayPal is peer-to-peer digital money. The thing is, all those systems have central intermediaries that act to reverse or block undesired transactions. Undesired transactions can be fraud. They can be other issues. They can be using money for drug dealing, etcetera. The idea behind the cryptocurrencies is, let's build a payment system where the record keeping is done in a decentralized fashion that means theoretically nobody's actually responsible for upholding the laws that you have to as a money transmitter.
Okay, wow. That's a very concise explanation. Now, that was theoretical. So it's been more than a decade since this concept was introduced by Satoshi Nakamoto. How well do you think it's living up to its theoretical promise?
Quite well. It's quite good for buying drugs online and paying off multimillion dollar ransoms, and it's quite good for conducting securities fraud at scale, building Ponzi schemes, doing all sorts of activity that basically has been banned by the real financial sector over half a millennia of experience, because they don't work out on a societal basis, but they work out for the boosters. So we've got this system that is basically gleefully and gloriously corrupt, and that's working as intended.
The facilitating crime part makes sense, but there are people who are not criminals who are using this and are compelled by it. What do you think the more general appeal of decentralization is?
I don't really understand decentralization versus distributed systems. Now, in a distributed system, you have multiple actors, multiple entities, but you have some way of identifying the players, trusted third party, hierarchical distribution of trust, et cetera. The idea with the decentralization is let's get rid of the identifiability of the participants and then try to cobble together a system that prevents what are known as Sybils. That is somebody just spinning up a gazillion amount of fake votes.
Now, in a distributed system, Sybils are easy to deal with. They don't exist, because you have some strong notion of identity in the players. In the cryptocurrency space, the problem is, you have to do proof of work, which is, anybody who wants to create Sybils has to waste a lot of effort, or proof of stake, where anybody wants lots of Sybils has a lot of the underlying cryptocurrency. And yes, we are formally enshrining the notion of he who has the gold makes the rules. The irony is, though, when you actually look at these decentralized systems, they aren't actually meaningfully decentralized. They're basically decentralized enough to evade the regulations central authorities would have to do in the same position.
So Bitcoin mining, it's this decentralized structure. Yet really, three to four mining pools control so much of the Bitcoin mining as to effectively have a veto on anything. Ethereum, Ethereum is supposedly decentralized, but again, three to four mining pools effectively have a veto on the thing and conduct activity that they calmly describe as MEV, that is front running, back running, and other illegal market manipulations in a centralized system. So they're not even really decentralized, except for the point of, they don't actually follow the laws that central authorities would have to. At the same time, they don't actually work in a meaningful manner.
So how many transactions globally can Bitcoin support? Do you know?
I think it's six per second or something like that.
Yeah. It's about three to seven transactions per second on a global basis. This can never be a scalable system. The Bitcoin proponents go, "But oh, wait, we have this layer two solution called lightning," which, first of all, really wants centralized actors. So much for distributed and decentralized systems. And you can only on a global basis create fund or close three to seven channels per second, which means it can't scale.
The Ethereum vaunted world computer is one-5000th the compute process of a Raspberry Pi, a $50 computer. But it costs, okay, the price has dropped. It now costs only about a hundred dollars a second to run. We could build a system equivalent to Bitcoin or a system equivalent to Ethereum using 10 identified actors, six of which have to be honest in the end, and the power cost would be 500 watts. So a couple of old light bulbs versus a major country's, because we've identified the actors. Once you've identified the actors, well, it becomes much simpler.
So why aren't the cryptocurrencies doing this? Well, first of all, a lot of them actually are under the hood. A lot of supposedly decentralized cryptocurrencies are really running in this manner. But the other reason they don't do it, and the reason why I'm not offering the Ethereum substitute as a service is, as named entities I would actually have to follow the restrictions on money transmitters and running stock exchanges and all these other financial rules that have built up over half a millennia of painful lessons.
In the example you gave of 10 trusted authorities, where six of them have to be honest, can you just elaborate on that?
So we basically have 10 entities. Each entity has their own $50 computer. They get together, they maintain a log of all transactions and publicize it, and whenever there's a dispute, six of them have to vote to agree. And so if six of the 10 prove to be honest, we now have an honest public transaction lock.
Okay. But in that case, they have to be known named entities, which means they would have to be abiding by the law.
Yes.
Interesting. So much more efficient from a power perspective, but you don't get the ability to bypass the law, I guess.
Yeah, basically that's it. You end up saving nine orders of magnitude on power, but you can't do the crimes.
You mentioned the lightning network. Can you talk a little bit more about how it tries to solve the throughput limitation, and it sounded like you don't think it actually does solve it very well.
So the idea with lightning network is we build a layer two network where you create a channel into the network and pre-fund it with some amount of Bitcoin. In order to do this, you do a Bitcoin transaction. Once this channel is established, you can flow balances through the network to other people, and that just gets recorded in a local way, and only when you need to close out the channel or add more money to the channel, you have to do another Bitcoin transaction.
Now, the problem is, in order to do this, you basically don't have this network of spanning channels. Instead, you have a central authority that creates a channel to everybody and moves data around that, so you lose your decentralization, and you're still limited on this channel creation, channel refunding, channel closing, to the overall Bitcoin transaction rate. So if you look at, say, El Salvador, which has the Bitcoin experiment, because the loony dictator has laser eyes, most of the Bitcoin transactions even never actually occurred even on the lightning network but were through the Chivo app, and so they were literally changing numbers in the database, like the electronic money we've had for decades.
Wow, interesting. So on the lightning network, it sounds like it's, maybe you get more throughput, but it's still ultimately limited by the layer one blockchain?
Yeah. And the lightning network serves one purpose. It allows the cryptocurrency community to claim that a problem will be solved in the future.
Huh.
One of the things the cryptocurrency community is remarkably good at is creating problems and then promising the solution will be six months out. Ethereum has been supposedly going to proof of stake, that is, he who has the gold makes the rules, rather than burning a huge amount of coal and heating the earth, basically since day one.
Right. The other point that you made, though, about lightning network is that it's centralized, so you kind of give up one of the primary objectives of Bitcoin.
Yep.
To follow onto that, how decentralized are major blockchains like Bitcoin and Ethereum?
Very little when you look at who has control. So the mining is very centralized. So there was a paper funded by DARPA recently on how decentralized the systems are, and they're not. The other thing is, decentralized is often conflated with this notion of, quote, unquote, trustlessness, where the idea is you trust the system as a whole but nothing individually. This is a delusion basically, because you trust that enough of the miners are honest, and there's remarkably few dominating to the point where they can be dishonest. You have to trust the code, because the code can have bugs, the code can change. You often hear the cryptocurrency proponents talk about code is law, right? That's a lie.
First of all, let's just discount all the things of, if I go up to a smart contract, say, "Give me all your money," and it does, is that theft? But they don't even uphold that. So in the early days of Ethereum, 10% of all Ethereum got invested in a decentralized autonomous organization, the Dow. The idea was, "We'll be able to basically take this Ethereum, do a voting mutual fund, vote on what to invest in." There was nothing really to invest in. There still isn't. So truth be told, it was a self-organized unintended Ponzi scheme. But 10% of all Ethereum flew into it until somebody figured out, "Hey, wait, there's a bug in the thing."
If you do a withdrawal, it does your withdrawal and then decrements the balance. But in doing your withdrawal, you can have it call another function that does another withdrawal. So basically you do a small deposit and then do a withdrawal that in it does a withdrawal, that does a withdrawal, that does a withdrawal, that does a withdrawal, and basically sweep out all the resources. What was the Ethereum community's reaction to this? Code is law? Ha-ha, because it was a lot of Ethereum developers whose money got rightfully liberated under the terms of the smart contract. And so they go, "This whole code is law business, the terms of the contract saying that you can steal this money? No, we're going to change the underlying code of Ethereum to undo that theft." Well, it wasn't a theft. It was in the contract, and instead steal the money back.
That's a crazy story. You mentioned the miners, so we had a guest a while ago that talked about this, that exact thing about mining concentration and why that's a problem. When we posted that video on YouTube, somebody made a comment about how we're totally misunderstanding how the nodes and the miners influence the network and that the network can still be decentralized even if miners are concentrated. I don't know if I really understand what the point they were making was, but I wanted to ask you, what's the difference between how nodes and miners influence the network?
Nodes just pass messages and have no votes. Miners have the votes. So if you have a world where, say, nine people have the vote and five are religious zealots, the religious zealots win the vote even if the overall population is much more, "God, these guys are crazy." And the mining system is, the miners are the ones with the votes. Nodes don't get a vote.
Is that a common misconception? This person had a ton of conviction in their comment about why we have to think about the nodes.
Yes. And it's also as much a common distraction that one of the things that you have to do in the cryptocurrency space is, because there are no fundamentals, because the technology is, put bluntly, tier tech, because the finance is a self delusion Ponzi scheme, the only way you can make money in cryptocurrency is get more suckers in. People don't hype their apple stock. "You got to get in or have fun staying poor." They do that in cryptocurrency, either consciously or unconsciously, because that's the only way you can make money at it.
How secure are public blockchains like Bitcoin with lots of work securing them?
It depends what you mean by secured. So in terms of being un-malleable, currently Bitcoin is. You can't really undo history, but you can do short-term undos if you're one of the miners. So if you're talking a large transaction in Bitcoin, you're going to want to let it sit for a while before you consider it yours because of this malleability. The other cryptocurrencies that are proof of work that burn less power are provably insecure, because the security of a proof of work system works by wasting X dollars per hour, and if a attack would take Y hours and can earn more than X times Y, the attack will happen and the security is lost. This means that the proof of work systems, in order to be secure, have to be wasting an insane amount of resources every 24/7/365, whether or not they're under attack. So in that sense, Bitcoin is secure.
In another really important sense, it's too secure. So one of the foundational principles is irreversibility, that there's no undo button, which means if you lose your keys, your money's gone. If somebody breaks into your computer, your money's gone. If somebody corrupts the source code, your money's gone. If there's a bug in the random number generator on your Android phone, your money's gone, and this is a very secure system. The problem is, it's secure in the wrong way, that the modern financial system runs on reversible fabrics. There's an undo button, and this makes the system much more resilient in practice, because if some bad guy breaks into my bank account or steals my credit card, there's fraud mitigation, and I'm not looking at all my apes are gone because somebody broke into my computer.
I've heard you say in other interviews that you're a bit of a Silicon Valley libertarian. We talked about the ability to facilitate crime, but more generally, do you think a facility for censorship resistant transactions like Bitcoin is a good thing for society?
No, that's the thing, that I like to say that Bitcoin committed a crime against me. It made me, a fairly honestly libertarian leaning Silicon Valley type, believe in the jackbooted justice of money laundering loss, that the ransomware epidemic can only exist because of cryptocurrency. This is doing tens to hundreds of billions of dollars annually in damage to the global economy. This is a multi-billion dollar criminal enterprise run out of Russia. And the key pain point for the bad guys is getting paid, because if they want a $5 million payment from Colonial Pipeline, there are three options. The banking system, which will flat out refuse. The banking system won't process $500 transactions for counterfeit Viagra. That's how we took out the Viagra spammers a decade ago, cash five million bucks in dollars is 50 kilograms of junk that you have to physically pick up in person, and if you're some Russian ransomware purveyor trying to pick up $5 million from Colonial Pipeline, you're going to also worry about also picking up a 308 caliber gift from a Marine sniper a hill over, or Bitcoin.
If the price of Bitcoin collapses and the interest in it collapses, the bad guys won't be able to hide their ransom deals in the cryptocurrency space. And in many ways, ransom ware's why I've changed my attitude about cryptocurrency. Back in 2013, I thought it was useless and amusing and full of comedy GODL that I could turn into papers. In 2017, I thought it was useless, starting to do significant damage due to the mining and the like, but I could still extract comedy GODL for papers and writing. Now I think it needs to be yeeted into the sun because of the externalities, that having a censorship resistant payment system is such that the damages are just too big to tolerate.
Can you make any arguments in favor of Bitcoin technology?
It produces a lot of comedy GODL. Cryptocurrency joke, they always go, "HODL, HODL, HODL." But really the tech is, that three to seven transactions per second worldwide is a joke. Then when you consider the power consumption, the finance is a self-assembled Ponzi scheme, and the only thing that seems to run on top of it are stock frauds, affinity frauds, and Ponzi schemes. My favorite being what we're now seeing collapsing, a lot of recursive Ponzi schemes. So Celsius isn't a Ponzi scheme. It just invests in DeFi protocols that are Ponzi schemes. So it invests in Three Arrows Capital that's a Ponzi scheme, that invests in somebody else, that invests back in BlockFi, that invests back in Celsius. So along this loop of Ponzis, of self-referential investment, and if people weren't losing big amounts of money, it would be funny.
We've been trying to cover this topic now. We're four episodes in at the time that we're recording this with you on crypto, and one of the things that somebody said to me is that, "Ben, you don't get it because you're coming from the perspective of a North American with a good legal system and a good financial system, but people in countries with less developed financial systems and maybe less developed property rights or precarious property rights, they can benefit from Bitcoin." What do you think about that argument, that Bitcoin is helpful to people in countries with limited access to banking and good legal systems?
First of all, if you say the word M-Pesa, they'll look like you're speaking Swahili. We have good, robust electronic money systems in third world countries that built up organically internally, M-Pesa being the classic example in Kenya. You look at El Salvador, and El Salvador is a disaster, that the same crowd cheering on, "Hey, this could liberate the third world," seems incredibly happy about a lunatic dictator forcing Bitcoin on the population and gambling with the country's own money.
But more importantly, one of the things the cryptocurrency space does is willfully ignore history. So in my wallet, I've got a piece of paper. It's got printed on it a dead president. If I was in Europe, I'd have these pieces of paper that have generic neoclassical architecture. Well, actually plastic. These are what gets used in the cases where the local country, the local currency, is unstable. So you end up using physical cash from other countries, and this is very effective. The excuse of, "Let's do cryptocurrency in the third world," that's basically a distraction. That's basically cryptocurrency backers go, "Ooh, shiny, look over here" without, once again, understanding the actual situation and looking at what actually happens. So they don't even understand what M-Pesa is, how it developed. It was basically started out as a way of swapping phone credits and became a currency, or it became a payment rails, and this is in a country where 25% of the people don't even have electricity.
Incredible. So since cryptocurrencies are censorship resistant, what can governments do or regulators do to stop them?
Control the on-ramps and off-ramps. So within the cryptocurrency space, the cryptocurrency space is basically a lawless nightmare. But what they can do is control the on-ramps and off-ramps where cryptocurrency goes to the real world. The problem is, a lot of these on-ramps and off-ramps are, let's just say, willfully blind.
So there's a case I'm familiar with where BlockFi, back March last year, received three million dollars in Bitcoin from somebody. That someone transmitted and took out a 1.2 million dollar lump, basically a way of cashing out Bitcoin. A few days later, BlockFi noticed under Chainalysis that 15 of the 60 odd Bitcoins were from a mixing service. What did BlockFi do? They made the loan anyway going, when by their own tool, it was 25% dirty money. Then you actually look at the transaction itself that Chainalysis flagged, and you see that it wasn't 25% from a Bitcoin mixer but 100% from a Bitcoin mixer, because it was ChipMixer, which has a very distinct signature. So we have the best, quote unquote, best actors in the field, deliberately or willfully ignorantly, not doing their job.
Can you talk more about what those ramps are? Is that where it interacts with the banking system where money's changing hands?
Yes. So on-ramps and off-ramps are like Coinbase, BlockFi, Celsius, Robinhood, et cetera, et cetera. Basically, any place where you can take Bitcoin in, or any cryptocurrency, and sell it for actual money, and vice versa. The on-ramps and off-ramps to the system are actually very limited, and that's where you control the crime.
Why do you think regulation's been so slow? Like Gary Gensler at the SEC, he gets crypto. I did his course.
I think it was because of a particular problem that regulators have. One of the greatest things they're afraid of is being accused of stifling innovation. During the 2017-2018 bubble, they basically took a hands-off approach, especially the SEC. Basically, every cryptocurrency and every crypto token upon release, except arguably Bitcoin, checks the tally test. It's a security. You're selling unregistered securities. Congratulations. Don't do that. Instead, the SEC did nothing. They basically only ever went after a couple of high profile ICOs after the ICOs failed catastrophically, basically going, "Buy horses, buy horses. Now we're going to close the barn door. See, we can close barn doors." That was, I think, the fear then. I think the fear now is the understanding by regulators that there is no "there" there, that it is a turducken of Ponzi schemes, and they see the lesson of Albania.
So Albania, after the collapse of the Soviet Union, had these huge Ponzi schemes develop, and they ended up collapsing and devastating the country. But the regulators came in just before the collapse happened, and so the regulators were blamed with collapsing the scheme. I think at this point, the regulators understand that the space is garbage. They don't want to touch it, however, because if you regulate before it collapses, you'll be the one blamed for collapsing the Ponzi scheme, rather than the one who gets blamed for not doing anything about the Ponzi scheme. Since they're already getting blamed for not having done anything about it on the way up, well, you're going to get blamed for that anyway. Why also take the blame for crashing it when it goes down?
So the obvious question, Nick, how do you see this all playing out?
Slow, then fast. So the problem is, the previous bubbles in the cryptocurrency space have been very isolated but without debt. So back in 2013, the cryptocurrency bubble was about the size of the Beanie Baby craze, and critically it was fueled by fake transactions on Mt. Gox, fake money on Mt. Gox, and why it collapsed, it went away. The 2017-2018 bubble was driven largely by Tether. It was an order of magnitude bigger but still relatively small, and there wasn't a lot of debt. This time around, there's a lot of debt, but not just a lot of debt, but a lot of debt that takes real money. So over the past year and change, the cryptocurrency miners have seen the price go way up and the mining go way up, and yet they were never selling Bitcoin. They were holding onto it and borrowing against it to pay power bills.
So now there are billions of dollars worth of Bitcoin that have been borrowed against to pay power bills, and power bills don't take funny money. They don't take Tether. They don't take Circle. They don't take Stablecoin. They want actual cold, hard cash or a electronic transfer from a real bank, because let's face it, our money's been digital for a generation now. So this creates mechanisms of positive down feed back loops. We saw a couple happen when these initial round of recursive Ponzis dropped off a couple weeks ago. So a couple of automated liquidations kicked in, sold reportedly about 600 bucks worth of Bitcoin, and the price dropped by a thousand bucks in the space of five seconds... or five minutes. Things have stabilized out for the moment as a lot of these recursive Ponzi schemes have fallen, but there's a lot of pressure that will continue to grind things down.
The other factor is, PT Barnum may have said, "There's a sucker born every minute." That still means there's a limited number of suckers available. How many suckers will be fooled by a Matt Damon commercial now that weren't fooled six months ago, and well, fortune favors those who grab their pitch forks and torches and go to Matt Damon's house to get their money back. So what I think is going to happen is, the price is going to bounce along and grind down, but when it collapses, it's going to collapse very suddenly and will get these feedback loops where it might take a week, a month, a year to grind down a couple more thousand bucks. Then all of a sudden it just, boom, collapses, as you've got these recursive dominoes and houses of cards all stacked up on each other, and it just goes away. Hopefully, it will go away, period, because there is no "there" there. There is no financial innovation in the space. There is no technical innovation in the space, and when it collapses, it's going to have burned so many normal people that there'll be no suckers left.
Do you really think it'll go away? Like, Bitcoin has utility for drug dealers or people who want child pornography and all that kind of stuff.
I don't think it will stick around, because it actually doesn't have very much utility for drug dealers. So the original Silk Road did maybe a quarter of a million in day gross sales. The peak, as far as I know, on the dark markets has been maybe a million dollars a day in gross sales, which for the drug business is pretty much in the noise. So it's fun for the cops to go knock down the black markets. And also the other thing is, a lot of states are finally legalizing pot. At the time of Silk Road, almost all of the business was pot anyway, that it will not provide much utility for them.
For the CSAM pipes, the child sexual abuse material types, Bitcoin was a side chip. There was one site welcome to video that did, allowed you to buy CSAM with Bitcoin. Most of the other CSAM sites, on tour, their currency is child sexual abuse material, which is why they are so vile and deserve to get shut down when they happen, is because the currency for videos of child abuse is child abuse.
Wow.
But that doesn't really rely on Bitcoin. So the only real successful Bitcoin payments rail that is doing multimillion dollar transactions has been the ransomware, and the ransomware only works as long as the casino games of the rest of the cryptocurrency space stay up. So if it crashes some, it crashes lots. There's very little utility left. And the other thing that happens is, you know what I said about that proof of work security? We're already pretty much at the stage where there's a lot of Bitcoin mining equipment that is only profitable if it's turned on for a limited period of time to attack the system. If the price drops even further, the further the price drops, the more this happens. So I think there are feedback loops in the system where, below a given amount, the value just ceases to be.
Wow. In the mining argument there, would miners be incentivized to not attack the system so that the Bitcoin still maintains its value, especially if they haven't been selling?
Well, it depends on the miner. Most rational miners in the past have always sold, because mining and holding are very different businesses. You don't actually need to do one for the other. So anybody who has mining equipment that, say, they were in a mining pool and they took out all their money, that mining equipment is only going to be valuable for attacking.
Wow.
And there are commercial services, NiceHash, which is literally, you can rent out your mining capability for others who want to attack cryptocurrencies. That service already exists.
That does sound pretty precarious. Earlier you mentioned Tether and a few other stable coins. Can you talk more about how stable coins fit into the crypto ecosystem?
So the cryptocurrency space overall is gambling. It's a casino. And in a casino you need chips, and a chip is basically a static store of value within the casino space. And Tether and all these other stable coins are the casino chips for those gambling. I'm going to gamble that Bitcoin's going to go up, or Dogecoin or Shiba Inu, or whatever, over the short period of time. I move into it. Now I want to take my money out. I can't take it out as real money. The actual amount of real dollars in the cryptocurrency system is very low. I take it out as Tether or Circle or Maker or any other stable coin, per se. Then that's my casino chip for when I want to bet on the next one.
Now, the problem is, these casino chips have problems. So there's really three different classes. You have the algorithmic stable coins, you have the over collateralized stable coins, and you have the asset backed stable coins. The algorithmic stable coins are like Iron, Tether and Luna. Waves has one. There's all sorts of... Waves, Neutrino. The idea is, you have some unstable cryptocurrency as well and some arbitrage mechanism where you can go back and forth. This works as fine as long as the pyramid scheme keeps going. These algorithmic stable coins, basically their only actual use is pyramid schemes.
Why did so much get put into Terra Luna? It's because you could take your Terra, put it in the Anchor protocol, and earn 20% return on investment, AKA a Ponzi scheme. The Neutrino, Waves, Vires Finance was a half billion dollar looting. You could take your Circle or Tether dollar denominated, supposedly, backed stable coins, put them in there, get 20% interest. But what that 20% interest was, was people taking insiders in Waves taking the Neutrino stable coin, borrowing against that, taking the Tether and Circle and running and looting stuff for half a billion dollars. So the algorithmic stable coins are broken. They just can't work. Over collateralized stable coins are the same thing, but just the collateral is bigger. To get $1 worth of Maker, you have to have $1.50 worth of Ethereum, and if the Ethereum ever goes below, say, 1.3, you get liquidated. Fifty percent over-collateralization doesn't really work when the market drops 50%, so those don't.
The final one is the backed stable coins. So the idea is, you deposit a dollar, you get a crypto dollar. Now, first of all, we've actually had this in the 19th Century. This is called 19th Century banking, because these are digital bank notes. Now, the problem is, the banks are almost certainly not backed. So in order to believe that Tether is legitimate, and they got up to $70 billion, is you would have to have $70 billion. That is what? Thirty micro strategies worth of investors that wanted to get into the cryptocurrency space but did not want to go to Coinbase or Gemini or Kraken, but instead wanted to gamble on offshore exchanges like Binance or FTX, et cetera, which all are corporate shells that move around the globe to avoid the law, that are known to trade against their customers, that are known to do all sorts of shenanigans, to gamble on third tier coins, or what has undoubtedly actually happened is Tether's customer, say, Alameda Research, borrows a billion dollars worth of Tether, buys a billion dollars worth of Bitcoin, and now that's collateral. So it's basically created a self-referential recursive Ponzi scheme.
Even Circle I don't trust. So right now, Circle, over the past year, has gone from 42 to 55 billion dollars. Over the past year, it's gone from... So starting in July or June, it was 25 billion, and now it's up to 55 billion. So either you had $30 billion of new money going into cryptocurrency, when the market has been grinding down, and critically, the miners have been unable to sell their cryptocurrency for fear of distorting the market, or there's some shenanigans going on. I believe it's probably door number two. Even though Circle is supposedly listed and audited, et cetera, so was Emma.
Are stable coins decentralized?
No. Well, it depends. So the algorithmic and the over collateralized ones are, but Circle and Tether, the backed stable coins, are not, because they are centrally issued, centrally redeemed, and critically actually centrally controlled, that Circle and Tether both have the ability to freeze accounts on the smart contract for intermediary accounts. In fact, I will argue, as a consequence, they are money transmitters, not just for the purposes of turning dollars into Circle and Circle into dollars, but Circle to Circle transactions, because they could, if they wanted to, block every intermediate wallet that wouldn't do know your customer. They just don't.
If we move past the cryptocurrency component that we've been talking about, what new properties does the blockchain data structure provide?
Well, let's split out the public and the private. So the public blockchains are only used for the cryptocurrencies. A private blockchain is a limited writer append only data structure. Translation, there's a few number of people allowed to actually add to the data store, and we can only add to the data store, never delete. A very useful idea that we've had for decades now. So a blockchain is nothing more than a signed hash chain. We've had signed hash chain services in 1999. They just aren't all that useful really. The only thing that blockchain adds in that context is it has a superpower. The superpower is, you say the word blockchain and you can spew whatever bull you want and nobody calls you on it. I've got a few examples of that.
I was expecting a real superpower.
Oh no, that is a real superpower. The other one, and this is the one that is behind every successful private blockchain project, is you go up to the VP in charge of the IT budget and say, "I need 10 million bucks to improve my backend infrastructure. I need a new database, new data store, new portal for doing data, because I've got all these vendors and all that type of stuff," and what are they going to do? They're going to laugh, because God forbid. But you say, "I want 10 million bucks to do a blockchain to do all that stuff," and you get the money. This is so remarkable.
Now, once you get the money, what you do is upgrade your database, improve your data store, all that. And then the other thing you do is you take that database, you have to turn on the database log, and every five minutes you sign a block of the log with a hash going to the previous one, and you say, "Woo-hoo, I have a blockchain, and it's now blockchain powered." That allowed me to get the 10 million bucks to do the IT infrastructure I needed to do, and I spent maybe an afternoon cobbling up, a cobbled together ugly ass shell script that's using open PGP for the signatures to say that I have a blockchain.
So it is a real superpower. People can get funding to improve their databases? Okay.
Yes, it is a real superpower, not for solving any problem other than extracting money, but that can be very useful. So there's basically two audiences for a blockchain pitch. There's the internal audience, where the superpower is, you get the funding to do the IT infrastructure upgrades you needed to do. Walmart Canada's a great example of that. That's what some brilliant engineer over there did, is they got the upper brass to sign off on IT improvements by calling it blockchain.
The other one is the bull. So this actually happened to me. I was sitting in on a blockchain class, mostly to give a rebut. This was back in 2017, 2018, and there were two proponents going, "Blockchain can solve everything" kind of thing. This was a concrete example. Supply chain in India for vaccines. Now, we're familiar now with the notion that vaccines need to be shipped cold, and if they get too hot, this could be a problem. This is especially a problem in a third world environment. This clown was up on stage saying confidently that you could solve this in India with blockchain. No. You solve this problem with shock watch temperature sensitive labels. You stick a sticker on the box, and if that box ever gets out of temperature, the sticker changes color, bada bing bada boom. Anybody with five seconds of familiarity in cold shake, like anybody who's bought shrimp from Florida frozen solid, knows this. So this is why I created my iron law blockchain. Anyone who says publicly blockchain can solve X doesn't understand X, and you can safely ignore them.
That Walmart example is one that I heard on a podcast last week, and it comes across so convincing and compelling. So it just shows the power of a narrative, I guess.
Yeah, because the problem with the Walmart example is they were dealing with how do you integrate multiple data streams from multiple providers and get them all to talk together.
Right.
This is not a recording problem. This is a data format problem, a data exchange problem, a access control problem, who has access to what data. This is a data management problem. How do you find stuff in the database? The actual way of recording the data is, it could be on stone tablets for all you care. It is how you manage the data. And blockchain is bogus for data management, because blockchain only refers to a storage mechanism that we've known how to build for a generation.
And does Walmart using a private blockchain have anything to do with cryptocurrency?
Of course not. It's really good for hyping the cryptocurrency, but no.
It kind of sounds like what you're saying is Walmart maybe did a good thing, like, maybe they really did some good data architecture back there, but calling it a blockchain is irrelevant, but they may have done some really good work.
Yeah. And what they did is in fact used the miraculous power of blockchain to shake loose the internal corporate funding necessary to do the real hard work.
Yeah, that's super interesting.
Yep. Because they talked about how they shrank the invoice disputes that they had dramatically, which was a huge cost savings for them, so they did something that was of benefit, but it became wrapped up in this blockchain story board.
Yeah. And basically, to handle the billing disputes, what you want to do is make sure that people have much clearer understanding of what's going on and can access their data.
Yes.
And that's a data access issue, because I know they did not go with a global view, that they would not want their suppliers to see what the invoices to the other suppliers were, and blockchain does nothing for access control.
Hmm. That is crazy to think about. So I guess the headline, because you can read about this everywhere, about Walmart Canada doing that. Who was that good for? Is it good for Walmart to promote that?
Yeah, and also the guys responsible internally, because it was an IT success. That's always good for your internal politics. Also, there's a lot of crypto boosters who take advantage of people's confusion, and anytime something goes blockchain, you hype it up.
If there's not a lot of "there" there, like you said, why do you think some pretty meaningful venture capital dollars seem to be flowing into the crypto space?
Oh, that's because they've discovered a new way of making money, securities fraud by inducement and securities fraud by proxy. So in the old days, a venture firm like Andreessen Horowitz invests in 10 companies. Now, seven of these companies fail. Three reasonably succeed. Two of those get sold off to bigger companies, and Andreessen Horowitz gets paid. One goes IPO. Andreessen Horowitz does a lot of paperwork. There's a lot of accounting and honesty involved, and then it gets listed. It's a good business. But what Andreessen Horowitz does now is they invest in the company. The company's pitch is privacy preserving machine learning on the blockchain. The company doesn't have a product, doesn't even have an idea really, but does create its own blockchain and token, and this token gets sold to the venture capital. The venture capitalists get a huge amount of the token at a super cheap rate.
Then the goal is, get that token listed on Coinbase, which is coincidentally a non-trivial ownership fraction by Andreessen Horowitz, or worst cases, DeFi exchange, and then you just go dump those tokens on retail for Dentacoin, or whatever it is, which is a promise on an IOU, on a product that doesn't exist that the punters don't actually want. They just think they'll sell it to somebody else, and this is so an unlicensed security. But the SEC has not been policing the ICO space in the unregulated securities. Even if they do, well, they'll go after the company, while Andreessen Horowitz, they just recommended these advisors. They did this. They didn't actually commit securities fraud. They got other people to do it on their behalf.
And to follow on to Ben's question, why do you think prominent schools like MIT and Cornell are emphasizing their blockchain programs?
There's two things on the schools. There's classes designed for internal and classes designed for external consumption. So classes designed for external consumption are a profit center for the university and the departments. So these are going to hype chase. So if there's blockchain hype, cool, get people in the seats, executive MBA program type things, makes you lots of money.
Then there's the one for the internal students existing. Berkeley has had a few of these. These tend to be taught by faculty members who, A, have a particular interest in the space, because, for example, their startup is privacy preserving machine learning on the blockchain, and they do it as a way of engineering things so they don't actually have to teach. So instead, they get a whole bunch of guest speakers, and it's basically an easy gig of getting out of your teaching duty. That's what those blockchain and DeFi classes are. They're either looting the external suckers, or they're faculty members playing games to avoid actually having to teach real classes.
Wow.
So you're a lecturer at Berkeley, or I think previously. You're almost done.
Ex-lecturer. I rage quit this semester.
Right, right.
Something about being in charge either as instructor or co-instructor for 1% of all education at Berkeley over the year and having to fight each semester to get accounted is just an 80% time position.
Makes sense. Makes sense to rage quit.
Yeah.
In that situation. Okay. So what I wanted to ask, though, is you're in that position as a lecturer. You just explained why other people might be lecturing on blockchain. What do you think that you are seeing that they don't see, or are they just blinded by their incentives?
I think they're blinded by their incentives, that what happens in the cryptocurrency space on the tech front is, they either get incentives or go, "It's bull" and ignore it. There's very few people like me on the tech side who've actually been able to follow this space for a decade, and that's because I do have a business model. My business model is extract comedy GODL and turn it into academic papers as well as general purpose writing.
So for example, a couple of students, Julian Piette, and another one he worked with, and I, we just produced a paper on the MEV and how it can destabilize Ethereum. Within the CS department, those who believe in cryptocurrency have their economic investment in cryptocurrency. Everybody else looks at, goes, "Ugh, it's bogus, and I don't even want to pay attention."
Real quick, because he talked about investing, and I heard you say this on another podcast, what is the smartest way to invest, because we agree with you?
Mutual funds, index funds, and ignore it for a few decades. Because on a long-term horizon, an investment is positive sum, that you should have not just what you sell it for later on, but dividends and interest and stock buybacks all feeding into the asset stream. That's the stock market. That's the bond market, on a long-term basis. Short term, day trading, Robinhood, that's gambling, and if you want to gamble, go to Vegas, the food's better. But on a long-term basis, you invest in stocks and hold onto it. Then there's a lot of really advanced math that I forgot basically 10 minutes after I took the class, that showed that you can't actually beat the market. Your goal is to tie the market, and the way to tie the market is with a index fund. My favorite part from that entry in the textbook was the proof ending up with equality if, and only if, the market is a horse race.
Yep. In our normal course of podcasting, we talk mostly about index funds and all sorts of different ways to think about that. When we're not talking about crypto, that's what we're doing.
So I'm curious, Nick. Are there technologists that you respect who are proponents of crypto and blockchain?
There were a couple, but not anymore, because they've basically gone and drunk the Kool-Aid. I don't want to name names, but there are a couple who I had respected who did good work and now have basically just so drunk the cryptocurrency Kool-Aid that, sorry, if you can't see the externalities involved in your systems that it's still orders of magnitude worse than what we've been able to do for 500 years, I'm sorry.
When somebody drinks the Kool-Aid like that that you previously respected, does it at any point make you question your perspective or beliefs?
I find it useful, because being able to understand where they're coming from, I just think for the most part they don't understand how the real systems work, how the real regulated systems work, why the regulated systems are built this way. They also tend to have blind spots on finance, that I'm not a formal economist, but I'm over the Dunning-Kruger threshold, and now in that bottom of, I know just what I know. So the confidence threshold, I'm over here. I understood the financial crisis of 2008. In fact, the key to understanding the financial crisis is the ability to create zero sum instruments. So a option is a zero sum gig, one win, one loss.
So Benjamin, I want to create an option. We do an option exchange. We've basically, no net money has been created or destroyed, but I can tell you on my balance sheet for my bonus, I look positive. On your balance sheet for your bonus, you look positive, and that explains AIG. I called the dot.com bubble. I called the housing bubble. I didn't call the AIG collapse of wackiness. I just knew the housing was crazy. I think the stock market's overvalued and bubbleicious, but that seems to be correcting right now, so who knows? I might be right. So I've got a moderately decent enough understanding of finance to tell real from bull. And that I think is the problem with the technologists in the cryptocurrency space, is they haven't had enough economics inadvertently or deliberately to be able to sort the real from the bull.
But what's the true motivation going on there? When you said you tried to understand from where they're coming, what is that where?
Two of the things. One is there is a belief, rightly, that our current financial system sucks. There are a lot of things that could be made a lot better. Blockchain doesn't help any of those, but it's true, you could make things a lot better than they are.
Okay.
The other one is a belief on financial privacy. So some of the real backers of cryptocurrency truly do believe in anonymous transactions. I'm afraid I don't because of the externalities, that even if we just had anonymous money that was a hundred dollar transaction limit, that would still not work, because we could just do bots and have 10,000 hundred dollar transactions all going at once. So that's more of a fundamental belief, that I no longer believe in financial privacy, that I believe, as I said, in the jack booted authoritarian money laundering laws, because they've proven necessary in the electronic space. So this is really a difference in principles. If you believe that financial privacy is a fundamental right, you're going to come to different conclusions.
That's one of the observations that I've made in trying to learn about this space is that it boils down to, the arguments boil down to ideology. It's not about economics, it's not about technology, it's about ideology. And based on the ideology that you're coming from, you're going to come to very different conclusions about the technology.
And the thing is, there's those who don't understand the technology and those who do, and those who do, for the most part, don't tend to understand the finance side, because the technology... Basically, the technology is garbage, but the finance looks interesting, and then talk to an economist of the finance is garbage, but the technology looks interesting. You really need to have enough of both. You don't need to be an expert. You just have to get past the Dunning-Kruger threshold in either the economics or the technology to not fall into that trap. But if you aren't past the Dunning-Kruger threshold on both, you're going to fall into that trap.
You know what? I think it's even worse than that, because I think there's two different Dunning-Kruger curves for economics, one in actual economics and one in Austrian economics. I think in crypto, you end up with people who have maybe gotten over the hump in Austrian economics, but any-
The problem is, Austrian economics is done in Kruger economics. It's literally smoking weed. Come on. There's a reason we ditched the gold standard.
Right. Yes, I agree with that, and most respectable economists will also agree with that, but because there's an ideology there that people can latch onto and feel like they're learning, I think that makes this whole discussion much more difficult to have.
True.
Okay. We asked about if anybody that you respect likes crypto. What about people who share your views? Do many other technologists view crypto and blockchain the same way that you do?
Yes. So look at the concerned tech letter. Basically, so many technologists who do understand what's going on looked at the space, went, "This is bogus" and ignored it for a decade. Now the externalities have been getting out of hand. The concerned tech letter came together in the space of basically a week in response to a Jorge Stolfi Twitter thread of, "Hey technologists, you know this is garbage. See it as such," that just spiraled out of hand. We got a dozen plus signatories on the main letter. We opened it up for public signatures, and it got 1,500 signatures in what, two weeks? Basically, when technologists look at the space and aren't blinded by dollar signs in their eyes, they come to a very similar conclusion.
All right, I've got one last question before we end here. So when you formally taught your class, now that you're done teaching, when you used to teach your class, your computer science class, about blockchain, what was the main takeaway that you would want to leave them with?
The space is garbage. Either avoid or burn with fire and why. So technology that is useful, Merkel treats, past chains, but then again, they're older than my students. Why do the systems not work? Why are there problems? Why is the trust in the wrong place? So in computer security trust does not reduce to mathematics, but trust can be delegated, and so understanding where trust is delegated in a system, what the trusted entities actually are. And one of the things that I want the students to take away is that so many of these, quote unquote, trustless systems, well, basically all of them, they have trust running through their veins. They're built on trust. They're just good at hiding what you have to trust.
I've got a follow up on that. Since you've taught this class, two generations of students who came up when crypto was going through bubble phases, did you get any pushback from the students when you would teach them this?
Very little. Most of the students tend to like that lecture, and the lecture itself for the blockchain and cryptocurrencies, actually now, I went through the effort this year of making sure it was manually subtitled, so it's up on YouTube publicly.
Yeah, I've seen it. It's a great lecture. We'll put it in the show notes for this episode.
All right, Cameron, do you have any other questions?
I do not, but this has been incredible hour. So Nick, great to meet you, and thanks for your insights.
Thanks for having me.
Participate in our Community Discussion about this Episode:
Links From Today’s Episode:
Rational Reminder on iTunes — https://itunes.apple.com/ca/podcast/the-rational-reminder-podcast/id1426530582.
Rational Reminder Website — https://rationalreminder.ca/
Shop Merch — https://shop.rationalreminder.ca/
Join the Community — https://community.rationalreminder.ca/
Follow us on Twitter — https://twitter.com/RationalRemind
Follow us on Instagram — @rationalreminder
Benjamin on Twitter — https://twitter.com/benjaminwfelix
Cameron on Twitter — https://twitter.com/CameronPassmore
Dr. Nicholas Weaver on Twitter — https://twitter.com/ncweaver
Dr. Nicholas Weaver’s lecture on blockchain and cryptocurrencies — https://www.youtube.com/channel/UC6tTeyrPeR5XLlVGglZowuw