Bitcoin is the Sewer Rat of Currencies
When computer scientist Andreas Antonopoulos first heard about bitcoin in 2011 he dismissed it as “nerd money.” Six months later he happened on bitcoin creator Satoshi Nakamoto’s now-legendary white paper from November 2008. This nine-page, innocent-looking document was a blueprint for replacing large swaths of the financial services industry with a globally-distributed encryption-based transaction network that wasn’t owned by anyone.
Nakamoto’s white paper blew Antonopoulos’ mind. “This isn’t money,” he realized, “it’s a decentralized trust network,” with applications extending far beyond digital currency.
Antonopoulos says he became “obsessed and enthralled” with bitcoin, “spending 12 or more hours each day glued to a screen, reading, writing, coding, and learning” as much as he could. He wrote, “I emerged from the state of fugue, more than 20 pounds lighter from a lack of consistent meals, determined to dedicate myself to working on bitcoin.”
Five years later, Antonopoulos’ work has paid off. The 43-year-old entrepreneur is one of the most respected experts in bitcoin and blockchain technology, and he regularly shares his expertise with businesses and organizations around the world. His 2014 book, Mastering Bitcoin, was called the “best technical reference available on bitcoin today,” by Balaji Srinivasan, the CEO of 21.co, and received high praise from Gavin Andresen, Chief Scientist of the Bitcoin Foundation.
In January 2016, Institute for the Future launched Blockchain Futures Lab, a research initiative and a community for “identifying the opportunities and limits of blockchain technologies and their social, economic, and political impacts on individuals, organizations, and communities over the coming decades.”
I interviewed Antonopoulos to get his thoughts on the current state of blockchain technology and where it’s headed. What he had to say was surprising and enlightening.
(The following transcript was lightly edited for clarity. Listen to the full interview.)
Mark Frauenfelder: How are enterprise/private blockchains different from bitcoin’s blockchain?
Andreas Antonopoulos: First of all, I think we’ve seen this before. It’s very similar to the earlier stages of the consumer Internet, when the Internet was breaking out of the educational environment after the 1993 National Science Foundation decision to open it up to commercial use. At the time, you had people doing things like writing manifestos for the Internet — John Perry Barlow writing “A Declaration of the Independence of Cyberspace.”
The corporations were really excited about this technology. They could see this was going to be disruptive and they certainly wanted to at least defensively prepare themselves for it. The same narrative is playing out with bitcoin.
The banks and the corporations say, “Oh, bitcoin’s awesome. We want that. Only without the open, decentralized, peer-to-peer, borderless, permissionless part. Could we instead have a closed, controlled, tame, identity-laden permission version of that please?”
What they did then was try to create other networks with TCP/IP or intranets or various things like MSN and Compuserve — closed, curated wall ed garden versions of the Internet, where everything would be tame and PG-13 and have all the promise of the Internet, with none of the smut. None of the offensive anything, just a clean Disneyfied version of it.
Of course, those places failed because they were boring and because the real premise of the Internet wasn’t TCP/IP, it wasn’t interconnectivity. It was openness, it was permissionless innovation, it was content creators being individuals rather than top down. It was peer-to-peer. All of those things exploded with innovation until eventually everything else just seemed too boring and became obsolete.
We started saying, “Pay attention to the blockchain, not just bitcoin.” The banks took this all the way to the end and now they’re saying “blockchain without bitcoin.”
The same thing is happening now with bitcoin. Bitcoin is finance without borders, finance without identity, finance for everyone, finance without banks and governments, finance without authority, without central points of control, etc. We basically present bitcoin to the world as open, decentralized, permissionless innovation, peer-to-peer finance without borders. The banks and the corporations say, “Oh, that’s awesome. We want that. Only without the open, decentralized, peer-to-peer, borderless, permissionless part. Could we instead have a closed, controlled, tame, identity-laden permission version of that please?”
We promoted the blockchain as, “Hey, listen. It’s not just about bitcoin.” This was two years ago. A lot of people who are in bitcoins started saying, “Yeah, it’s not about bitcoin alone.” You have to understand because people thought bitcoin was currency. The argument was it’s not about the currency. It’s about this trust network as a platform, as a global platform and what it can do.
We started saying, “Pay attention to the blockchain, not just bitcoin.” The banks took this all the way to the end and now they’re saying “blockchain without bitcoin.” Through 2015 you had this fad transformation where it became very cool and popular to say, “You know, after all, it’s not about bitcoin. It’s about the blockchain instead of bitcoin.”
“Bitcoin is going to fail but blockchain is going to change the world.” That’s like the great whitewashing. It’s hilarious to watch because, for one thing, it exposes their fundamental misunderstanding of what this is, where the value is, and how it works.
Will enterprise blockchains be successful?
They’re going to be successful in replacing even more insidious forms of centralization like centralized clearinghouses, like Swift, Depository Trust & Clearing Corporation, and other quasi-private, monopoly companies that control and clear equities and bonds and commodities.
They’re going to revolutionize Wall Street trading floors. Revolutionize, meaning they’ll drag it out of the 70s and maybe bring it in into the late 90s. That may be exciting to bankers. It gives me a major yawn.
Let’s start with the technical. The really cool thing is the open, decentralized blockchain, that is, a blockchain that has no central authority and is open for anyone to participate without asking permission because the trust mechanism doesn’t depend on access control like everything that came before. It depends on proof-of-work consensus, which is a decentralized approach to achieving consensus and trust over the state of the system.
You can only have that if you have a consensus algorithm that is decentralized and anonymous, such as proof-of-work mining and bitcoin. We haven’t found another good enough one yet. That only works if you have a reward associated with mining to create this game-theoretical competition and balance between the incentives for private profit and trust in the system. We know that works. It works very well and it’s scaled so far to double-digit billion dollars, which had never happened before and that in itself is astonishing.
The idea that you can make a blockchain without bitcoin or without a currency — you can really, but the only way to do that is to replace the decentralized consensus mechanism with a committee, which is what they call “signing instead of mining.” You get ten banks together and, instead of mining blocks competitively in an open way, they sign the blocks, they take turns.
There’s an algorithm for that called Byzantine Paxos, which is a distributed consensus algorithm. It’s basically what we used to call three-phase commit. It’s like a round-robin signing party where you elect a leader each round and they sign for all of the transactions. As long as you trust that nine banks are better than one central clearinghouse and won’t cheat as much, then okay.
Imagine what happens when Lockheed Martin’s blockchain leaks. You’re going to have a time stamp for every financial transaction they’ve ever done. I can’t wait.
But that system is not immutable. The banks can get together and roll back the blockchain and write something else. With bitcoin’s proof-of-work, you can’t do that. No one can do that because you can’t get enough miners to put enormous amounts of hashing power behind that kind of effort. You lose immutability. You lose decentralization.
Because the model is based on this round-robin of trusted partners, you lose the permissionless open access so it becomes something where you have to have permission to join and almost certainly isn’t open in public. It will almost certainly have to have ID added back into it, which brings all of the disadvantages of identity theft if you’re a consumer and will give us an absolute treasure trove if you’re Wikileaks. Imagine what happens when Lockheed Martin’s blockchain leaks. You’re going to have a time stamp for every financial transaction they’ve ever done. I can’t wait.
They’re adding identity back to this, which is absolutely idiotic. You can have a blockchain without the currency. You just can’t have an open, decentralized blockchain where the trust lies in this game-theoretical competition that is neutral and based on rules. Instead, you have to put trust in third parties, counterparties who are the signers.
That then becomes business as usual, which may be revolutionary for banks. But, immediately it cannot be borderless because in order to maintain control over a system like that, you have to control identity. That means you have the old Bank Secrecy Act, Know Your Customer rules, and Anti-Money Laundering rules which, in themselves, are responsible for this problem of global economic exclusion, the reason people can’t participate in the banking system. So you re-erect the borders.
Essentially what it is is an intranet and an intranet is boring. It’s stale content. It’s the less secure version of everything out there. You can’t run the apps you want because corporate IT didn’t allow them. It’s stagnant innovation. In the end, it also lags behind in terms of security because it’s not exposed to the kind of robust peer review that an open Internet-based system has to have to survive. While bitcoin is getting stronger and stronger with security, these things will actually wither.
I did a presentation called Bubble Boy and the Sewer Rat, about that very topic.
The idea being that intranets end up being these really insecure places where you’re running Outlook and FrontPage and old versions of Apache that haven’t been patched. Whereas on the Internet, if you’re Facebook, if you’re Google, if you’re Apple and you’re operating Internet applications, you have to be robust. You have to respond to vulnerability. You have to make systems that are antifragile, resilient to attack and they’re constantly evolving and they’ve become very robust.
Whereas, the things that you put in a bubble are like Bubble Boy. If you remember from the 70s, a kid who lived in a bubble because he had no immune system is like raising a child in a bath of Purell. You’re going to turn them into an allergic weakling.
In meantime, bitcoin isn’t living in a bubble. Bitcoin is a sewer rat. It’s missing a leg. Its snout was badly mangled in an accident in last year. It’s not allergic to anything. In fact, it’s probably got a couple of strains of bubonic plague on it which it treats like a common cold. You have a system that is antifragile and dynamic and robust.
This is the big argument of 2015. It’s the “let’s take bitcoin, cut off its beard, take away its piercings, put it in a suit, call it blockchain, and present it to the board.” It’s safe. It’s got borders. We can apply the same regulations. We can put barriers to entry and create anti-competitive environment to control who has access. It will be more efficient than our existing banking system.
If what they want to do is the Outlook of currencies, go ahead, by all means. That’s what intranets turned into. It’s stale wiki content hosted on FrontPage and Outlook. People hate that. Eventually, successful, vibrant, innovative companies are the ones that turn their IT infrastructure inside out. In the future of finance, successful, vibrant, and innovative banks are the ones that turn their infrastructure inside out and make it part of an open, borderless financial system that serves the other 6 billion, that serves all the people who have been excluded from finance.
Because here is the thing: we’ve had the Internet now for 25 years. If you look at the statistics, financial inclusion is getting worse. That doesn’t make any sense unless you consider that, in the meantime, the traditional currencies and digital finance systems have been getting more and more prone to surveillance and tied to identity where everything is tracked. They’re closing in on themselves in order to maintain this ironclad control over who sends money to who, all in the name of terrorist prevention and anti-money laundering, which is bullshit because HSBC could money-launder billions. Always have since their birth in the Opium Wars and nobody ever goes to jail. God forbid, a kid buys a joint with bitcoin and everybody’s up in arms.
What is this doing in the meantime? The real question is how many billions of people are you going to sacrifice on the altar of money laundering, because the current answer is just a bit over four billion people. Four billion people who are in the deepest parts of poverty are excluded from banking and financial systems and it is because of Know Your Customer and Anti-Money Laundering and borders and control of networks.
You cannot create global finance and economic inclusion on the back of a carefully controlled show-us-your-papers identity-based system where everything is tracked. What you create is a global surveillance dystopia. Our entire financial system is heading into this thing where everything is surveilled. Bitcoin is heading in exact opposite direction. No identity by default, from weak pseudonymity to a stronger and stronger anonymity as time goes by.
As a result, it doesn’t do borders. It doesn’t care about borders. It doesn’t do Know Your Customer. It doesn’t do Anti-Money Laundering. It doesn’t do those things because those things are bourgeois concepts of the privileged financial elite. Those bourgeois concepts have a four-billion-people-in-poverty price tag.
Are banks and financial institutions trying to centralize the blockchain?
Absolutely. It’s not so much centralize the blockchain. It’s that they’re starting with a premise that finance must be controlled in order to be secure. In order to be controlled and secure, it has to have identity. In order to have identity, you have to control all the endpoints and it has to be a closed system. In order for it to be a closed system, you have to control access and centralize everything as much as possible.
You want end-to-end control. You can’t have leaks because leaks are the places where money laundering happens (money laundering for brown people, not money laundering for rich white people. “We don’t care about that,” [say the powers that be]). The end result is that they’re making the system more and more centralized in this desperate attempt to fulfill a dream that started in the 70s that you could create this perfect Disneyfied world economy where everybody’s finances are known.
When they talk about Know Your Customer and Anti-Money Laundering rules, they assume the information all flows in one direction — to them. They don’t assume that the CIA’s money laundering for weapons purchases are seen by the Chinese. They assume what that means is that the Syrian dissidents’ money is known by the CIA. It’s only one-way flow.
Anyway, not to get too political, but the point is that money has been heading away from cash and towards digital and, with that, digital came an attempt to turn it all into a giant mainframe, highly centralized. But the problem is that within the centralized systems, of course, corruptible centralization of power corrupts absolutely. It creates fragility, and so in this march to centralization for reasons of control, which is taken for granted among these circles, the end results have been fragile markets, fragile economies, fragile currencies, and increasingly, corruption.
The companies that have control of a clearing of equities front-run the transactions of their friends and delay the transactions of the ones they’re competing with. You have all of these rigged markets where privileged access is the name of the game.
What the banks want is they want it a hell of a lot more centralized and controlled than bitcoins and they certainly don’t want this ragtag army of misfits to suddenly create competition where they had this nicely parceled-up monopolistic market. But they don’t want it as controlled as the one clearinghouse. For them, a blockchain is this great liberalization. It’s decentralizing their clearinghouses. That’s because their definition of the scale of centralization is very narrow.
For me, bitcoin, a true global currency with liquidity that is completely decentralized, is not even at the edge of the scale. There are far more decentralized, stealthy, and anonymous currencies further off to the side of bitcoin. But this one has promise. It has momentum. It has brand recognition and it has a potential to scale to reach all of those people who are left behind. That’s the whole narrative of 2015.
Why is bitcoin adoption slow among the unbanked billions?
They don’t have access. That’s the whole thing. We’re in year seven. Bitcoin just had its seventh birthday on January 3rd. It spent the first three years in obscurity so you had really four years since it’s been known among the tech elite, maybe two years since it’s become a news item.
We’re very much where the Internet was in 1992. The question is if you asked me in 1992: “Why isn’t social media happening? Why isn’t China picking up the Internet? Why aren’t we seeing popular revolution starting based on this wonderful communication platform? You said it would bring freedom to the world. Where is it?” If you said that in 1992, you’d be right. It’s not there.
That’s where we are. The technology is still immature. We’ve barely had decent smartphone wallets for a year and a half now. Liquidity on the ground? We’re talking about $6 or $7 billion total liquidity in the market, which is not enough to support even the remittances flow, which is a huge market.
We are seeing little pockets of adoption. There’s some really exciting things happening in the Philippines, for example, with rebit.ph. There’s certainly a lot of attention in China and especially now with the yuan devaluing where bitcoin is being used as a safe haven / safety valve / flight from capital.
But it’s naive to assume that a technology that is so new is suddenly going to take spark in and around the world where data plans are barely there, smartphones are barely there. We are seeing some efforts to bring it to feature phones to use SMS, intermediaries. But that brings its own risks because it makes it more centralized and possible to defraud and steal bitcoin from people so that’s not a good idea either.
It’s not going to have a big impact in the Third World for a decade. That’s the truth. You’re going to see pockets. For example, we’re seeing among digiterati of Argentina who are in that perfect set of circumstances where they’ve got the technology infrastructure, the ability to understand and speak English, literacy, numeracy, and tech savvy. They’re young entrepreneurs. They’ve got a massive currency crisis, currency controls, hyperinflation, and no ability to do any import-export business.
There is a small islands of bitcoin activity going on that was grown out of that inefficiency. You have these areas where the circumstance of finance become high enough friction that you see these little islands of bitcoin activity emerge. It’s still really early days. To ask why there’s no bitcoin in the Third World today is to ask why there was no Netflix on the Internet in 1994 or 95.
How will the blockchain enable things such as records of ownership that are easily transferable, permanent identity records , transparent voting systems, and self-executing contracts?
I think all of those things and many more applications that are not simply extensions of what we have today, just decentralized, are in our future and are possible by technologies related to this kind of decentralized model of trust. If you think of it simply as a decentralized trust platform that can securely record it in a mutual ledger, certain facts, whether those are about ownership of a currency or an asset or something else, then all of these things are in our future.
Reputation and identity are really proxies to trust and default risk. They’re very poor proxies as such. There a leftover of old world thinking.
The question is when do they happen and in what sequence do they happen? Which of these are simply skeuomorphic designs of what we think we’re going to need based on what we had before and which ones are really innovative?
For example, I see all of this emphasis on reputation among people who are interested in bitcoin, and to me reputation and identity are really proxies to trust and default risk. They’re very poor proxies as such. There a leftover of old world thinking. I think we can do a lot more things with anonymity, pseudonymity, and diversification of risk without identity, which is far more interesting to me than trying to create some kind of straitjacket identity system on bitcoin.
In terms of voting, the real question is what problem are you trying to solve, which is often the case when geeks come into politics? What problem are you trying to solve with voting: verifying the integrity of the vote? We don’t really have a voting fraud problem, at least not to the United States. There’s some faction among some parties that there is a voting fraud. There’s some vote registration fraud that doesn’t actually translate into voting fraud.
What we have is enormous apathy and complete capture through campaign finance by the 1% and you don’t solve that by doing digital voting. In fact, probably pencil and paper is the most secure system you can yet come up with, with sufficient checks and balances.
Start thinking of bitcoin not as a currency but as a trust platform, one that provides you with a scriptable environment where you can combine conditions that get evaluated neutrally by a network-centric system of trust.
Again, a lot of the thinking around blockchain, I think, is still influenced by what were the problems we’re trying to solve last decade — “Can we stick blockchain or bitcoin on top of that and solve them” — not really thinking about what can we do completely differently with these technologies.
Start thinking of bitcoin not as a currency but as a trust platform, one that provides you with a scriptable environment where you can combine conditions that get evaluated neutrally by a network-centric system of trust. I mean, it’s an enormously powerful idea and you can do all kinds of decentralized things with it that we haven’t yet imagined beyond currency.
Is it possible to have decentralized trust platforms doing these things without mining or some kind of proof-of-work?
Not really. We don’t know yet. Maybe there’s a better consensus algorithm. In terms of what does mining buy you, it doesn’t buy you efficiency. What it buys you is freedom. In fact, you get freedom at the cost of efficiency. The most efficient system you can build for a database is a completely centralized system. You can have a very efficient system that can process billions of transactions per second, as long as you give all of the authority and trust to a single party.
What bitcoin does is it says, “We diversify that control and trust among a very large number of participants or systems, and then we allow them to exert that authority in a very decentralized way and without identifying themselves.” That prevents corruption and it means that you can’t coerce anyone because you don’t even know who they are. It decentralizes the trust mechanism. We don’t know of a better way of doing that.
Maybe there is a better way of doing that. Consensus algorithms are a new area of science that is arguably about four years old, practically. In universities, we’re seeing it about two years now, beginning to develop as an area of steady with research. It’s not easy to do without a currency. It’s not easy to do without mining.
The mining pool is continuing to become more centralized, with fewer players. Is that something to be concerned about? Is there anything that can be done about it?
A lot of things are changing in that way. First of all, there is the concept of a mining pool, which is a specific technical construct, which is a way of coordinating activity between a group of miners in order to smooth out the allocation of rewards. It’s like a lottery pool where you join with many others so that you get smaller but more frequent wins that give you a more steady stream of income rather than saying, “I’m gonna win once every six months but it’s gonna be big enough.”
You don’t want to do that on a poisson distribution because what if you win every nine months? Who’s going to pay the electricity bill? That’s the problem mining pools attempt to solve.
There are really about a dozen mining pools of which three are very powerful. They have behind them probably 100 companies that run mining servers, mining farms. Of those, there are probably only half a dozen that have an enormous amount of hashrate. That is a problem.
At the same time, that is an artifact of a very specific adoption curve, which is that when bitcoin started, mining was happening on CPUs. It migrated from CPU to GPU to FPGA and to ASIC over the first five years. In the last two years, it’s been exclusively application specific integrated circuits (ASIC), which means that people are putting the mining algorithm in parallel on silicon directly, in silicon fabrication. They were doing it at first at 28 nanometers then at 24, then at 22. The latest one that came out is at 16 nanometers.
What does that mean?
16 nanometers of silicon fabrication.
That’s the density of …
Right. Sixteen nanometers is what commercial CPU and GPU is at. Basically, bitcoin mining has gone from something that was a hobbyist thing to increasingly specialized silicon chips being manufactured by the hundreds of thousands. If you have $10 million to get into the game, you can go out to buy these in bulk, have a semiconductor fab manufacture them on demand, create your designs.
That arms race lasted for about 2 years. But at the end of this arms race is a Moore’s Law. At the end of this arms race is a wall that everyone has now hit. There simply isn’t anything beyond 16 nanometers. You’ve got a at least a hundred X improvements in mining capacity in power every year, sometimes a thousand X a year. Now we’re back to Moore’s Law which means two X every 18 months. That’s a wall. And what happens when you hit that wall is it no longer matters how many transistors you put on a chip because you’re already maximizing it, or even how much performance you get out of each Watt of electricity. What matters is how many chips you have and how densely you can pack them, which actually shifts the game.
I have a little computer here, which is built by a company called 21 Inc, which was a big start up in this space. They’re trying to consumerize mining by putting the latest cutting-edge chip in a little device that I run at a small loss but I use it to do other things.
If you distribute 100,000 of these, mining as a centralized occupation is over. People don’t realize that, because the miners don’t have any economy of scale advantage over consumers. If there is reason to run this — and there’s even more reason in places in the world where a tiny income from a solar panel might actually make a difference, a tiny income in an internationally recognized currency from a solar panel you put on your hut — essentially this whole equation could flip on its head and suddenly mining could become immensely decentralized as it always was intended to be.
So this game has not played out yet. A lot of people who are looking at this curve and extrapolating it up into the right don’t realize that we’re at the end of the curve already because we hit 16 nanometers and there ain’t more capacity in the silicon chip to do anything more. Now it’s about volume and volume is done by companies like Samsung and Qualcomm. They do things for consumers. It’s a matter of distribution. The game has changed dramatically. You can’t corner the market in mining anymore.
The market itself, of course, is probably a year behind. It hits that curve because that’s how long it takes to source and deploy, etc. Maybe two before you start seeing consumers getting big. It’s happening.
One of the things that I’ve not been able to understand is what happens when the newly minted bitcoins rewards from mining go away. What’s the incentive then?
The short answer is fees, transaction fees. At the moment, with a thousand transactions per block, just ballpark, 25 bitcoins in reward from seigniorage, from new coinage, and about maybe 0.1 bitcoin in total fees. If you have 1,000 transactions, you’re going to make 0.1 bitcoin.
If you look at blocks using Block Explorer or one of these sites and you look at the first transaction, you’re going to see is a reward to the miner. You’re going to notice it’s an amount that looks like 25.11357. At that, 25 is the new bitcoin and the .11357 is fees. Think of the reward as a declining curve that halves every four years and the transactions per block increasing with adoption, the value of each bitcoin increasing with adoption, and the amount of fees remaining more less steady or in a dollar perspective, if today it’s 250-to-1 (250 times new coin reward versus 1 fees) at the end of the curve, you might have a relationship where it’s 1-to-10,000. So you get 1 Satoshi from reward to mine a block with 10,000 transactions, all of which have a minimum fee of 1 Satoshi. That’s 10,001 Satoshis reward: 1 Satoshi from new coins, 10,000 from fees. You’ve gone from 250-to-1 to 1-to-10,000. Now all of the reward almost is from fees. Those two curves cross over at some point long before you reach the end.
This is intentional. It’s designed to basically, gradually wean miners off rewards when they’re subsidizing the network until there’s enough transaction volume that even with a low fee-per-transaction the job of securing the network through mining is rewarded by the transaction fee itself. Mining never stops.
Mining never stops. However, is there is a fixed cost for electricity. What you’re depending on then if those mining fees to be an incentive, then you have to assume that the value of bitcoin is going to increase a lot, right?
Not necessarily. Yes, it will increase a lot. Think of 21 million bitcoins as your primary monetary supply of what’s in economic terms would be called Mo (pronounced “m-zero”). In the US economy, for example, M0 is about 5 to 8%. So 5 to 8% of our currency circulates as actual coin. Then you have various derivative instruments that increase the velocity of money, such as fractional reserve lending and various other things, so that each unit of currency travels through multiple hands and ends up creating a much bigger economy. Even if you consider 21 million as M0, if bitcoin’s economy was $1 trillion, yes, you’re looking at a significant increase in value.
But here’s the other thing: mining is not very profitable in the long run. In fact, there’s a dynamically adjusted equation within mining that ensures that the more people try, the harder it gets. But that equation also works backwards, which means that if it’s no longer profitable at a current level of effort, miners drop out. As fewer people are trying, the difficulty decreases to recalibrate the system.
If the price doesn’t justify the current level of mining, the current level of mining will decline. The first ones to switch off are going to be the locations that have the lowest efficiency in converting watts to bitcoin. The highest cost of electricity, which naturally pushes mining to places where you have low cost electricity, renewables, free electricity like solar, and it pushes to the most efficient operations in chips possible. Naturally, everybody else is unprofitable. Everybody else is unprofitable most of the time in this space.
There’s currently a limitation on bitcoin transactions of around seven transactions per second. However, that can be changed by doing things like increasing the size of a block. What are your thoughts on that? Are we ever going to see bitcoin have the same number of transactions per second as something like Visa, which I’ve heard can process anywhere between 4,000 to 10,000 transactions per second?
People have already simulated bitcoin with the constraints bitcoin has, but with various things scaled up, as able to substantially approach Visa’s Christmas Eve capacity, which is their busiest day of the year. It’s unlikely that for many reasons that you would want to make bitcoin as efficient as Visa and part of the reason is that that comes at a cost of centralization. One of the things we’re trying to make sure in bitcoin is that we maintain this balance — we’re paying an efficiency price in order to maintain the neutrality of the network, that decentralization of trust.
That compromise is always going to lead to something less efficient than Visa. All you have to do with Visa is give up all your political freedoms and give them complete control of the economy and they’ll deliver as many transactions per second as you want. Just make sure those transactions aren’t going to WikiLeaks because they got pressured by the US government to cut them off. (By the way, that incident is one of the things that motivated a lot of people to look into bitcoin and is why WikiLeaks has its primary funding through bitcoin, but I digress.)
There’s no reason why you can’t scale bitcoin substantially. Several orders of magnitude and, in fact, we’re already seeing multiple proposals in different ways of doing so. The seven transaction per second scalability limit is primarily an anti-denial-of-service requirement at the moment. It’s in place because there isn’t the demand to put more transactions unless you’re doing free transactions, in which case, a lot of them are really spam.
Zero fee is not necessarily a good thing because it encourages network antisocial behavior. Anyway, bitcoin can scale. What is most likely to happen is that, in some ways, bitcoin is more likely to be the reserve currency and clearing system for an ecosystem, a jungle of other species of currencies that have different velocities and are more in tune for things like retail banking and/or retail de-banking, buying a cup of coffee, etc. Bitcoin will serve that purpose, too, but it may end up you have more specialized coins.
The idea that you have to have one coin that wins is not necessarily true anymore. I think what bitcoin should be and I think what we’re trying to do is make it the globally super robust, super trusted, absolutely immutable global record of trust that acts as a clearing and settlement layer. It will be able to do some payments efficiently and for a really tiny micro payments you’re going to use all the technologies in parallel. Think of it as a very thick trunk of the digital currency tree. Some things you want to do on the tiny little wispy branch at the end because the trunk isn’t flexible enough but the trunk is what holds it all together.
What is preventing someone from creating millions of bots that just send one Satoshi to each other that would flood the network and slow it down to a standstill?
There’s a very simple answer. What’s stopping that from happening is that it happened and then it was stopped. Then some other variant happened and it was stopped, again and again, through seven years of grueling denial-of-service, robustness training of this infant currency’s immune system. This is not a bubble boy currency. This is a sewer rat. It’s been fighting in the shit for seven years and it has had everything thrown at it. That happens and in fact that happened in year 2, so some denial-of-service measures were taken and then those were attacked and then they got more robust and more robust.
It’s akin to the question of why in 2016 does Google not get taken down with denial-of-service when in 1999 Google was taken down for a day by denial-of-service? The answer is not because denial-of-service is not happening but because it has been happening non-stop since the beginning. The system has become anti-fragile and immune to it. Bitcoin will go down from time to time and minor service disruptions but it keeps getting stronger because it’s constantly under attack.
At the moment, there are various limits. The seven transaction per second limit on total processing, the minimum transaction amount limits the minimum transaction that various nodes will relay, which is a choice of the node operator, the minimum fee per kilobyte that will be processed, et cetera, all of which make it difficult to do a nuisance attack without spending money and keep most bitcoin transactions in the realm of financially viable transactions as judged by the willingness of the transmitter to pay the fee per kilobyte. It’s a nice market-based system and it works pretty well.
What is your take on Ethereum?
I have some Ethereum. Just in case you need a disclosure, I’m involved in a lot of that scene. I’ve studied Ethereum since the early days. In fact, the creator of Ethereum, Vitalik Buterin, sent me the white paper before he published it for review among a small group of people he sent it to. I was fascinated from the very beginning.
Ethereum is using the same trusted platform dynamics and proof-of-work consensus algorithm to build decentralized contracts. These contracts are essentially much more robust programmable scripts than the money scripts that bitcoin has. Bitcoin has a purposefully limited scripting language inside it, which is not Turing complete. That was done deliberately in order to make it more secure in the early days. It’s gotten more capable since, a lot more capable. So now, a lot of smart contracts you can execute directly on bitcoin.
Ethereum is taking a very different approach. What it’s looking for is complete programmability of contracts. Essentially it’s a blockchain for trusted cloud computing with a decentralized security model. And it is fascinating. It’s a completely different niche. It has a lot of very interesting applications going for it. One important thing to understand is it’s not a competitor to bitcoin. In fact, it’s very synergistic with bitcoin. The best way it works is with bitcoin. I bought my first Ethereum with bitcoin. It will use contracts with bitcoin. Like many of the currencies, it’s a branch off the tree that uses bitcoin as its trunk.
Learn more about IFTF’s Blockchain Futures Lab.