In this episode, we’re joined by Vitalik Buterin, the co-founder of Ethereum to discuss governance, political economy, and crypto. We also discuss land value taxes, Georgism, optimal voting systems and public goods provisioning. This episode was co-hosted by Lars Doucet.
Will Jarvis 0:05
Hey folks, welcome to narratives. narratives is a podcast exploring the ways in which the world is better than in the past, the ways it is worse in the past, where it’s a better, more definite vision of the future. I’m your host, William Jarvis. And I want to thank you for taking the time out of your day to listen to this episode. I hope you enjoy it. You can find show notes, transcripts and videos at narratives podcast.com.
Additionally, in this episode, my friend Lars du se joins us as a co host, or Vitalik. How are you doing this morning?
Vitalik 0:44
It’s been a good morning in the process of writing an article about where to use and not use blockchains. What other things have a Yebin? up to just a bunch of random things, but generally, it’s been good. And it’s been fun. How about you?
Will Jarvis 1:04
doing really well, we really appreciate you taking the time to come on the show. Do you mind giving us kind of a brief bio and some of the big ideas you’re interested in?
Vitalik 1:12
Sure. So I think people in general mostly know me for crypto stuff. So I joined the Bitcoin ecosystem back in 2011. So I was still in high school, I heard about Bitcoin a couple of times, once on the internet once from my dad, and I thought, hey, you know, this is interesting, this seems like an ecosystem that’s doing something really cool. And so I started, like really jumping in and participating in it. I started off as a writer, so I co founded Bitcoin magazine with this guy named Mihai Alesia from Romania. And I was writing just a bunch of articles on all kinds of Bitcoin topics, including the technology, the economics, the politics, and society related issues. And, you know, culture just about everything. And a couple of years after that, I went into the crypto space full time started exploring more projects, and eventually Aetherium started, and I have been part of the Etherium project ever since. So the stuff that I generally do for Aetherium tends to have to do with research. So basically, they’re coming up with the ways to improve the Etherium protocol, kind of test, testing out some of those approaches, they get coordinating with researchers and developers and Nick, explaining things on those kinds of issues. Though, I also have continued to kind of write anything like think and do big picture stuff on a bunch of blockchain related issues. I guess, the thing that’s fascinating about blockchains, and that I think becomes even more true about systems like Ethereum than it was about Bitcoin is that blockchains are this environments where you can create new organizations, you can create new systems of rules, you can create applications, that Unicode ways in which large numbers of people interact with each other. And so there’s a lot of ideas around things like economics and sociology and like thinking through the different disciplines about, like, how would people actually interact and use a particular application? What some application break? What would some application not break and what different circumstances? So I ended up thinking about these ideas that are very relevant to things that are happening in Aetherium lands, but that also do end up stretching further.
Will Jarvis 3:51
That makes makes a ton of sense. Metallic, I’m curious, a lot of people think about you as a crypto person, but actually think of you kind of like being more interested in governance, political economy, and then you know, crypto is almost like a way to get there. It’s a way to innovate in that space in a way that people have not been able to do that in the in the recent past. Is that a fair assessment? Or is kind of crypto actually the main thing?
Vitalik 4:14
No, I think that’s definitely true. I think like, I mean, I’m definitely you know, not like, into the coins for the coins. I’m like, to me, blockchains definitely are this kind of way to open up this big space for people to, you know, experiments have enough governance, communication, open up the ways that we can coordinate without relying only on governments and corporations, and all of these kinds of things. So, you know, and I think I’m definitely very interested in these broader ends that have always see in the crypto space as being about trying to accomplish
Lars Doucet 4:51
interesting so on that note, you know, our audience is you know, our audience has has some degree of knowledge of crypto but it’s more of a general audience. We talk here mostly about out policy and big ideas and kind of things like that. So well, I guess we’ll just jump right into it. We invited you on the show, when we saw that you had retweeted some of her work. You know, I had written this series of articles for Astro Codex 10, on the topic of Land Policy, and Georgia ism. And I saw you had read it. So I guess we’ll just jump right into that. First of all, you’ve read the series. What do you think of it?
Vitalik 5:25
I thought it was very interesting. I had been a fan of Georgia ism for a really long time, I think it was one of the political philosophies that really appealed to me, even back when I was just starting my learnings about what the different political philosophies are. And when I was still a Bitcoin magazine person, I was definitely, I guess, surprised about the conclusion of just how much revenue a land value tax might be able to capture, like, as I remember, like, it was basically coming close to be able to replace like, a whole bunch of other taxes. And that’s, I think, not the usual take along kind of people out there in the world that are vaguely pro georgeous. I’m right there, you know, like, the more usual viewpoint is like, okay, yeah, I mean, I see the rationale for this sort of stuff. But like, I mean, oh, come on, it’s not Oh, land is just such a tiny part of the economy, it’s not going to get very far. And, again, yeah, reading through that post, it was interesting to see like, actually, you know, just how big a part of the economy land deals. And you know, on some level, it does make sense, right, like, especially given how people are like to talk about housing prices so much. And when you actually look at like, the price of a house, in most places only, like a small portion of it is the actual house, right? Like one way to look at this as if you did home insurance, and you’ll look at like, what is the maximum amount that they pay out in the case that the, you know, you got a fire in the house, and we would break so like, relative to the price of a house, often it’s pretty tiny. And so, and then, you know, comparing, like, empty, lots to full lots, and you notice that the empty a lot is like three quarters, and you have the full load. And if you’re looking at buying a house, from the point of view of buying a house, you’re like, WTF made, and there’s, and then, you know, even things like there was that study that said, like, if you will let people build, like, even to a medium extent, in places like San Francisco, the US GDP could have been 36%, higher, and like all these different, you know, things that stack up on top of each other and you realize that like, Wait, did I actually get going on land is pretty significant that it is big? And yeah, that’s I thought that was just generally a really interesting of view. Picture.
Lars Doucet 7:50
Cool. Did Did I leave anything out? Did you disagree with anything?
Vitalik 7:55
I don’t think I disagreed with anything. And I do think I have this kind of module in my mind that says, Don’t let your mind get changed too much by like any one person’s a single opinion. And there’s always a chance that like, you know, some unknown unknowns are being left out. But I definitely would HIGHLY welcome and encourage more people, and of researching and double checking all this and trying to figure it out. I think, in practice the biggest obstacles to like actually trying to land value Taxify, or they’re either America or other countries is that land value, instituting a large land value tax is basically a one time transfer away from existing landowners. And, you know, existing landowners are gonna get upset by that. And in a lot of cases, existing landowners are this fairly powerful political lobby. And, you know, there are these kind of emotionally appealing arguments of like, you know, oh, you know, are all I have? Is this home? Are you going to tax away the home? And, you know, in reality, like, with land value schemes, you can always fix them and say, Okay, fine, a third of the revenue goes into US citizens dividend, that’s actually totally enough to be able to afford an affordable home and the people complaining are actually in the kind of top fifth percentile of rich this but you know, even still, like it’s an emotionally appealing argument, and politics works off of that sort of stuff. So it’s, like, it does feel like those kinds of political issues are probably more likely to be the NF source of problems, almost done, even the math itself. And just to kind of dial that back to a previous topic. I think one of the things that interests me about like doing crypto in particular, instead of doing getting into politics and the traditional way instead, whenever you chain you wants to change an existing political system like there’s always going to be a lot of status quo pressure, you have to change huge numbers of people. There’s like this real really long lead time between developing an idea and any hope of having that idea implemented. And when when it does get implemented, it’s always going to be a kludge. Whereas, you know, with a blockchain, it’s like, you create a new system. And you know, you build a new system, and it runs. And sometimes the thing breaks, as we saw happen with Tara a couple of months ago, but sometimes the thing doesn’t break. And eventually, we learn over time. And I’m hopeful that, you know, at some points, these insights that we get from the crypto space are going to start making a bigger and bigger impact in the wider economy and society.
Lars Doucet 10:38
So that actually brings up a really interesting point that ties back into crypto, I think the general term for this phenomenon was coined by a guy called Gordon Tullock. He called it the transitional games trap. And he used the example of taxi medallions were the first generation to get some kind of particular benefit. Mix out like bandits, and then the next generation, like, the next generation, like the value of that thing might get like priced in. So like taxis might not be particularly profitable. But removing the medallion system will create overnight winners and losers. So now it’s very politically fraught, to do it. And so that’s just to name the phenomenon we’re both talking about. And he tried to like remove some existing system that has a spot in interest. So you’ve just brought up it’s like, one of the things that appeals to you about crypto is that it’s like, okay, well, instead of having to, like deal with all these politicians, and something, create this new ecosystem, how long until that new ecosystem you’ve created has its own winners and losers baked into the status quo, like, like Aetherium is trying to move from proof of work to proof of stake right now. Are you dealing with your own transitional gains trap? And you have any insights there?
Vitalik 11:47
Yeah, that’s actually fascinating that you asked the question, because I was literally just about to raise that exact point. I think it definitely is an issue. I do think that it has been much less of an issue than especially people looking on Ethereum from the outside, sometimes tend to assume like I think we get more concerns rolling about miners being expropriated, then, you know, even miners complaining about the the switch to proof of stake. And I think the big reason why this happened is because proof of stake Isn’t this like new thing that’s being foisted on an ecosystem that never expected proof of stake was part of the discussion in Aetherium, since I made that first post on us last year in January 2014. And then there was a definite plan to move to proof of stake since at least November 2014. And maybe a bit earlier, right. So it feels like the the goal of switching to proof of stake their goal of implementing some scalability stuff like sharding, it was kind of part of Ethereum is a living Constitution almost from the beginning of the project. So it’s this kind of interesting experiments, where you have a community with these kinds of constitution, like norms that are not just about preserving a particular thing that’s already there, but also about, you know, trying our best to execute on this specific roadmap that everyone was aligned on. And, you know, there definitely are many people who do not like the roadmap, and a lot of people went somewhere else. When the Dow hack happened in 2016, I think that actually was a moment where Aetherium was just hit with a very unexpected situation. And it did act in a way that ended up violating a lot of people’s expectations. And a lot of those people who were very unhappy about it rebelled and went over to Aetherium classic. But one of the interesting side effects of that is that I think if you’re Iam classic also actually managed to attract people who were unhappy with Aetherium for other reasons. So a lot of proof of work fans went over to Aetherium classic. I’m a lot of fans of Ethereum having a less aggressive and much more conservative roadmap and not bothering with sharding, once to Aetherium classic. And so that kind of moments, even actually kind of purified Ethereum to some extent, and made it more cohesive and more able to execute on its own particular values, which is also something that I found interesting. So I guess, I definitely think that I’m going to avoiding transition audience traps is going to be an important thing for crypto ecosystems that going forward as well, especially as they become more mature. But I also think that crypto is this very large design space. And so there are different both technical tools and social ends up political tools that you can use to deal with those kinds of issues. And I’m gonna think Etherium actually has done a pretty good job of that and I think other blockchains can as well.
Will Jarvis 14:58
Botallack that’s a good A transition there, I want to talk about, you know, kind of your preferences, what constitutes good governance, you know, kind of flows from base values, you see very pragmatic and kind of non ideological, which is an interesting contrast, because I went to an early, you know, 20s, in the early 2010s, I went to these early Bitcoin meetups, and you know, you’d go out to the parking lot afterwards, and people would show you their automatic rifles in the back of the trunk of their car, you know, and all these like, really wild things very libertarian, very committed to these ideological things and unwilling to budge, when you think, much more pragmatic, what are kind of your base values that guide your ideal kind of governance structures?
Vitalik 15:37
Yeah, and I think openness is important. Freedom is important. I think cosmopolitan ethical values are importance to me. So you know, the idea of like, not just caring about people that are either part of your country or that are otherwise going to very close to you, but also thinking about the needs of people who are further away. I think, just like, I would, how would I call it him? Having positive some relations with the outside world is also I think, something that’s definitely important to me. So I guess that stands in contrast with this kind of approach that tries to kind of like get internal cohesion by emphasizing its like extreme differences with everything that’s outside of itself. Like that, I think is closer to something that a lot of people in say the Bitcoin maximalist camp are in favor of, but I think Ethereum has always not really been that ends. And I think even pragmatism itself is kind of an ideological value to like, trying to kind of emphasize that Nick, emphasize middle finding the good in all the different viewpoints. And like, even if you don’t agree with something, if you have, if you can identify some very cheap way to make people who do agree with that viewpoint, happy, you should just go do it. Yeah, so I think, a combination of all of those.
Lars Doucet 17:06
So one thing, just kind of circling back to some things we had said before, you know, you’d mentioned freedoms, very important to you, you’d mentioned that, you know, you kind of have a little different viewpoint than the Bitcoin maxis. But one thing I’ve noticed is that crypto culture in general has a very pretty strong anarchist streak. And I’m defining anarchy very, very broadly, generally, is people wanting to route around government in general and not having to deal with the political system, which is something you self say, bought into, but do you identify with anarchism in general, big capital A or lowercase a? And with the way that it manifests in crypto culture at all, to any degree? You know?
Vitalik 17:48
Yeah, I mean, I think a lot of ideas from that camp are ideals that I align and agree with. I mean, I, even this morning made that tweet where I complained about how people, like keep using the word real name to refer to government names, which is something that I really hate. And they’re, by the way, there’s actually, I guess, kind of mainstream from mainstream Western people, they might be a bit confusing to kind of really try because whatever your name is probably your name, but you know, there are these kinds of other groups of people like foreigners, I’m especially Asian, so I’m, I talk to them a lot, right? You know, names are more fluid thing, transgender people extra wasn’t even thinking about them. But that was something that came up in the replies. But that the, in general, that’s definitely kind of like one small example of that, you know, I do think that the, the crypto space is about trying to, like, create alternatives to things like governance governments in for, for a lot of the kind of functions that people wants to do like, well, both governments and centralized centralized corporations, I do think that there are two different ways in which you can kind of be, you know, quote, anti government. One way is to identify the government and its goals as an enemy, and set yourself up to try to resist them. But the other way is to try to say, okay, the government is a medicine that is trying to solve very particular diseases that are very real. But you know, like chemotherapy in the 1960s, it’s American medicine that has really bad side effects. And so can we create different medicines that have a word, that don’t have those side effects or have side effects that are less bad in some way? So I think in general, like, the second approach, in general appeals to me more than the first one, and I feel like in general, like, if you try to do the first one without doing any of the second one, like I just worry that it just kind of sets you up for conflict with the rest of the world too much. I do think that there are cases where the first one is just, you know what you’re going for, and that’s just kind of the way it is and you’re going Whenever if the world and you have to deal with it, right. So like, for example, I’m gonna, like I personally, you know, do like talks to Russian activists quite a bit recently, and you know, people that are anti war are trying to kind of push Russia into a healthier direction. And they are, you know, obviously, he’s having a very, very hard time because of their own government. And they’re like, you know, it just is a fact of life that, like, they’re in an environment where in a lot of cases, like, they are trying to kind of protect their activities from, you know, being interfered with by the local banking system, and, you know, privacy infringement, and the government seeing things and so forth. But at the same time, like, trying to use that kind of frame as a frame for like, this is the default way that all of society works. Like, I think, if everyone has that kind of conflict orientation, mindset, then things or who can just easily completely break? Right? So my, I guess, yeah, my general approach toward the freedom thing is basically, so also, like, focus even more on this question of like, well, you know, what are the things that people are actually looking for? So it could be protection of certain different kinds, it could be some kind of kind of filtering and curation function, it could be public goods funding, and, like, try to find ways to provide at least a little, a little bit of that, you know, using the crypto tools that we have. I mean, as I think, as I’ve already said, you know, I am a moderate, like, I’m not trying to, like literally replace the entire government. So I definitely admire or align much more with lowercase A and not not really uppercase A but that’s in general where I’m coming from.
Lars Doucet 21:47
So this kind of reminds me of to what degree does governance in crypto ultimately depend on real world governance, we have smart contracts, but we’ve seen instances where there’s disputes, and somehow or other these things get dragged into real world courts and settled by dumb contracts. So and I wonder to what degree this is, like related to like the Oracle problem or things like that. So to what degree does ultimately the blockchain and smart contracts and any any any human agreement you want to put on a blockchain ultimately wind up getting appealed to the real world legal systems? You know, I was wondering like to what what sense does the Oracle problem kind of imply that real world government will always have a say in what goes on in crypto land?
Vitalik 22:36
Hmm. So I do think that there have been a couple of instances of people kind of definitely going to some kind of social recourse in order to solve a problem that happens in their blockchain, but where that social recourse is still local to that community, and they ended up solving the problem without relying on a government court completely. So one of those examples is the steam versus hive situation that I’ve written about a bunch and I keep coming back to it right, basically, and to
Lars Doucet 23:05
be clear steam here is Ste M. Blockchain, not the non blockchain video game store,
Vitalik 23:11
correct? Yeah. So steam is a social network, a blockchain based social network, the goal was that, you know, you have a token, and you can like pay to vote, and people could get paid for making good content and all of these different ideas. So it was originally owned by one group of people. But then Justin sun in this very, not reputable crypto character, I came in and bought the company. And the company originally held, I think it was 20% of all the steam tokens. But there was this community expectation that those coins were kind of held interest, and they’re supposed to be used for the goal of making this ecosystem better. But then the steam the company ended up being transferred to Justin sun. And Justin Sun is this outsider who has completely different values, where, you know, basically, as far as I can tell, Justin signs, ideological values, or just makes Justin sound great again. So he, this immediately started a bunch of tension. And eventually, Justin ended up like, basically getting enough like kind of validators on like, on like voting nodes on the proof of stake network, to walk out of the the people who are opposing him, and basically kind of secure full control of the steam chain. Right. So the steam blockchain like it, it had this version of proof of stake, which is totally not the same as Aetherium. Right, and Aetherium. Like all that the proof of stake validators are doing is they’re just keeping the chain going in accepting transactions in steam. They also have this power to kind of vote and make pretty much arbitrary governance decisions. So they did, they kind of walked everyone else out. And the community was unhappy about this. And so the community made a fork of Steam. And incidentally, they had to call it hive. And here I think the existence of government law and trade actually went against what I thought was the good outcome. Because I steam, the company had the trademark. And so they had to name it something else. And so in some ways, they kind of lost half the battle right there. But they created hive, and there was this massive migration to hive. And on the hive chain, the coins belonging to anyone who has participated in Justin suns attack and got deleted, right? It’s like when you when you make a new coin, you can just kind of make the supply of the coin, do whatever you want, right. And, you know, anyone could make a version of a theory. And that’s like, if you’re young Vitalik is Evil Edition, where they just delete my balance. And if people really decide that I’m evil, so much, they want that they want to switch to it, then like they’ve done it, right. So in now, I think hive has like a roughly comparable valuation to steam, I think, I forget, last time I checked, it was a bit higher, but like, sometimes it’s higher, sometimes it’s lower. But like, that’s an example of off chain resolution, that is also not legal system based, right. So that happens sometimes. Other times, though, there’s definitely cases where the resolution is legal system based, right. So like, for example, a lot of the time contracts and projects within the blockchain space get hacked. And sometimes the hackers end up returning the money because, in part, you know, they’re afraid of legal consequences going to them, or sometimes the hacker is offered to kind of voluntarily return 95% in exchange for the remaining 5% of their money getting legalized. And because it’s so hard to like, actually spend $60 million, if you have illegally obtained money from a hack like that, that actually turns out to be a win win. So you know.
So, you know, a hacker turns into a security bounty. But so, there are lots of cases of that kind of legal recourse happening, too. I think, in general, like my view on this is that if courts and law disappeared, and assuming that in such a world, the Internet was able to keep functioning, which I think is probably not true of the internet today, but probably would be true of an internet that actually was optimized around the world being more dangerous. So one partial example of this might be me know, the war in Ukraine, and how Elon just brought in all of these Starlink nodes, and they actually have been very effective at keeping all the different parts of Ukraine connected, like I think, even you know, kept as of still connected to the world almost up until the very end. So that’s, you know, if you imagine it, like, basically, if the internet did exist, then you know, people would be able to would be able to cooperate, and you know, you have cryptographic keys, they would be able to solve things off chain. But at the same time, the number of problems that you have would just kind of be, you know, one or two levels higher than the number of problems that you have today. Right. So I think it’s not an either or thing. It’s like, the government exists, whether we like it or not, and courts exist, whether we like them or not. And so there are a lot of cases where you can kind of gain some benefit by appealing to them in some situations. And so people do, and, you know, on net, that’s probably reduced the number of the amount of terrible stuff have happening in the crypto space, which is good.
Lars Doucet 28:28
Yeah, to to follow up questions on that. And then I’m gonna want to kick it back more to questions of governance. But I thought it was really interesting that in the in the steam in the hive divorce, it’s interesting that you kind of pointed out that one of the things that didn’t come over was the trademark, which is something that only exists in the off chain legal world, and is something that I mean, it’s an off chain asset, right, a trademark, and it’s a court that adjudicates that. So it was interesting that they could get everything but that. So I was wondering if you had any thoughts on that, as well as when you say the word hack, there’s different kinds of hacks, right? There’s the kind of hack where you steal a bunch of passwords and gain unauthorized access. And then there’s the kind of hack where you do what the smart contract allows, but it’s not what anyone actually meant, and cause all kinds of havoc. And there’s the whole like, code is law question. So I was wondering if you had any thoughts on the tension between those two subjects?
Vitalik 29:26
What sorry. So what are the two subjects here
Lars Doucet 29:28
the two subjects is is that yeah, I probably should have asked him one at a time. But the first one was some of the steam hive thing. They were able to do off chain resolution. But the one thing they weren’t able to resolve was a legal conflict ultimately over the trademark. Right. And so is there anything interesting to say about that, like is that just the Oracle problem is there any way to stuff the trademark rights on chain
Vitalik 29:53
right I mean, theoretically there is right like I’m you know, we have DNS and Aetherium name system, and like I have Vitalik agility if the name, but you know, if you manage to hack that particular wallet, then you know, you can have italic dot eath for yourself. And, you know, I could sue you, but um, you know, you could transfer it to an anon account and it might never be able to find it again. And I might be able to do things like, you know, Tim plane to ether scan and complain to some of the centralized ecosystem providers, and they kind of add the interface away or might blacklist fidelity, right. So like, that might be possible, but what they’re not going to do is I’m never going to get the telethon eath back. So if now, of course, you know, if I want to I can make this our DOD, the my holding of intelligence, I’d still be very secure, right. So in the crypto space, there is this concept that I’ve been a big proponent of for a long time called social recovery wallets, where basically, you it’s a special kind of wallet accounts, which is a smart contract. And the rules basically say that you define a fairly large group of people. So in the case of the wallet that contains like, all of my money, for example, this is public, this is public on chain information, it’s a four of six. And, basically, so there’s some group of people and the majority of those people need to sign off in order to make some particular something happen. So like, you know, it could move with our bodies over to that wallet. No, it could even move it’s our.es into a contract where kind of I could unilaterally do things with that name. But then if it ever gets stolen, or if it ever gets lost, I could kind of call off that group of four of six and force them to transfer back. And so crypto does offer these kinds of things that could and if gift, the risk of things like loss or theft to be reduced to a very low number of the technique is actually very similar to what WeChat to use this for account recovery, like in WeChat. When you recover your account, you also have to, like basically call up the two of your friends and get them to send a six digit code to you. It’s, which is interesting, right? Like, even in those cases, where it’s centralized, the resolution is obviously completely available. Like sometimes these kinds of techniques also have to be used just for efficiency purposes. Right? Yeah. So I think in the trademark justice on Steam case, one way to read that situation from a political economy perspective is where basically, you have a local level of governments, which is the steam blockchain, and then you have a kind of central more distant layer of government, which is the trademark ecosystem. And the higher, the more central layer of governments was just socially too far away from the situation and didn’t understand the specific specifics of the situation, and like local community morals, well enough to be able to give the correct outcome. Right. So that’s kind of more generally, I guess, an arguments for trying to make some of these government functions more kind of local. And I don’t necessarily mean specialty, local, because I think in the 21st century, you know, individual talents are kind of less important as a yellow cross of community than they were historically I mean, like, socially local. And in. So the, I do think that like blockchains, and Seminole, Dallas and things happening in the crypto space and kind of trying to handle most things themselves as a good example of making that happen.
Lars Doucet 33:28
Okay, so well as a transition question to get us back to some topics of governance. One of the chief pitches of crypto is decentralization. But we keep seeing this trend of centralization anyway, and this comes in many forms, you know, whether it’s services to just will stick a private database right in the middle, between the blockchain and its users, or consolidation of mining power, concentrated coin ownership, or Dows, that are owned effectively by three people, how do you fight this tendency towards re centralization or, you know, assuming you think it’s a bad thing? Just do you have any thoughts on that that kind of trend towards centralization versus the tension of the initial pitch of decentralization? Hmm.
Vitalik 34:10
I mean, I definitely think that it’s a concern. And I definitely have been the get into getting more and more open to this viewpoint, that if you want some particular like, layer of an ecosystem to be decentralized, then that decentralization has to be to some degree and intentional and ongoing effort. Right. Like, that’s the philosophy behind the antitrust law, for example, to some extent, right like that, you know, you want to have some kind of, like explicit government’s thing that pushes for competition to continue to exist, because otherwise they get their competitors do end up kind of calling each other and banding together. I think, in the crypto space, obviously, there isn’t a government enforcing antitrust law, but there are these other kinds of community driven levers right. So one of them is Just reputation and people’s desire to be a good person and be seen as a good person. And this happened in both Bitcoin and Aetherium, right? Like, I think in Bitcoin, there was a situation back in 2013, when a single mining pool it was called a G hash.io got to 51% of the total Bitcoin network hash power, which basically meant that they were would be a have been able to unilaterally rewrite the chain to be whatever they want. And everyone noticed this, and there was this large movements to try to get people off of G hash, it actually works, right and the percentage once I kind of down pretty quickly. And I think after that I forget the details. This is a long time ago, but I believe that they even made a public commitments to not let their market share go above 40%, after that, which I thought was a very good kind of lovely and pro social move of them. Right. So there’s those kinds of levers. And other Weaver is, of course, like protocol design, and trying really hard to actually design the protocol to not have economies of scale, and even to have diseconomies of scale to some extent. So one interesting example of this in theory, in Aetherium, proof of stake design, is we have this concept of anti correlation penalties. What an anti correlation penalty means is basically that, when you as a proof of stake validator node, do something clearly bad. So like sign two conflicting blocks at the same time, for example, then the amounts that you get penalized is proportional to the number of other validators that do something wrong at almost the same time. So you get penalized more, if you fail at the same time as other people. And you get penalized, actually very little, like only about 3% of the amounts, if you get penalized at a time when no one else is getting penalized. So the idea here is basically to discourage people from joining the largest oil, to discourage people from even putting their staking on the same cloud service as everyone else, to discourage people from sometimes even running the same software or, you know, relying on being in the same geographic zone as everyone else. So in some ways, it’s like this intentional effect of efforts to create the exact opposite of the nobody got fired for buying IBM effects, right. You know, this is this kind of, you know, conformist effects that exists, you know, in a lot of your old contacts, where if you make a decision that leads to something bad, but you are just following common wisdom that you never actually get get punished, because you know, what, the guy was just holding common wisdom. But, you know, if you do something different from other people, and that leads to something bad, then like, all the blame goes on you. And so the, you know, in the, in the proof of stake case, like it’s intentionally trying to create the exact opposite, right? It’s like, no, if you do something that’s the same as everyone else, and that leads to something bad, that’s when you lose all your money. And there are signs that it’s effective, like there aren’t like we’ve talked to even staking pool providers, some of the big ones. And the some of them have been decentralizing between different providers, and the different software implementations internally, because they were afraid of like one of them failing, leading to a huge amount of money getting lost. So that was actually interesting. You’d like even align selfish incentives with the decentralization goal. So yeah, and I think there’s a lot that you can do with these kind of protocol changes that try to, like, deliberately get middle remove economies of scale, where possible and add these diseconomies of scale where possible. But then also, I think, there is some extent to which there might be economies of scale that can’t be removed. And the only way that we really have to compensate for that is by appealing to this kind of community morals and social governance layer sometimes.
Lars Doucet 38:59
So I’ll kick this section over to you well, because that I think a good is a good bridge to questions about voting systems. Absolutely, absolutely.
Will Jarvis 39:07
So recently, Vitalik actually, like three days ago, we talked to someone who’s working on approval voting. They’ve had a lot of traction. They’ve gotten, you know, a lot of success in implementing of approval voting in South Dakota, they’re working on Seattle, Washington now, they raised a lot of money from open philanthropy, and it’s been going really well. I’m curious, you’re interested in quadratic voting a lot of these other interesting systems. Can you contrast quadratic voting with approval voting? And is there in some sense, like, I guess, a trade off and Lars you might have a better phrasing than I do for this between like a system that’s like more optimal, but it’s harder to explain and kind of illegible to people to one that’s, like, easy to explain, and somewhat easier to implement.
Vitalik 39:47
Right. So I guess the general like theory of why existing voting is broken like starts with the arrows theorem, right where you basically there’s like this list of probably have a fairly simple and logic very intuitive property instead a voting system that st should obviously have. And it turns out that it’s mathematically impossible to create one if there’s more than two candidates. And so you know that like, that’s why we have issues like the whole strategic voting thing where, you know, if every, there might be a lot of people that like one particular candidate, but if there is this kind of self reinforcing impression, that candidate has no chance that everyone votes for just the lesser of two evils out of the two that do have a chance. I guess. So approval voting is one way of getting around arrow CRM, right? i i Do you think arrow CRM is overrated by a lot of people like era era CRM, basically, yeah. It often gets cited as this kind of general theory, why politics sucks, and why it has to suck. And like, the reason why that’s not true is because arrow CRM is like specifically reliant on this concept of ordinal preferences, right that like, like, it assumes that the the thing that you’re inputting is, you know, is a better than B is a better than CSP better than C, without being able to say, you know, by how much is a better than B or No, which ones do I just think are fine, and which ones do I think are not fine? You know, I actually have my own candidate for kind of the mathematical theory, that is the general reason why politics sucks, which is, I talked about this in one of my blog posts on coordination. It’s basically the fact that in the individual choice model, so where games are, like in game theory, where everyone’s making decisions separately, you can have games that have equilibria. But in the cooperative game theory setting, so where you assume that kind of subsets of players can actually talk to each other and have shared collective strategies. There’s actually a wide class of games where there isn’t even an equilibrium like it’s basically gyms to be chaotic. So I thought that was just a bit of a fascinating aside, right. But approval voting does change the issue or solve the aerosteon problem. And basically, because instead of saying, you know, who’s better than who you’re just saying, you know, which candidates Am I fine with? Which candidates? Am I not fine with? Quadratic voting is interesting in that it’s like range voting in that you get to kind of score like in a very fine grained way, well, how much do you like each candidate, right? But range voting has this problem that like, if you think some candidate is good at all, then you know, if you know that, they’re going to do an average. And let’s say if the average is 55, and your actual opinion is 70, votes, 70, it might increase it to 55.01. But if you instead vote 100, then it’s going to, then you’re going to increase the average to 55.03. Right. So it’s just always everyone’s incentive to vote either zero or 100. And it collapses down to approval voting. But But quadratic voting fixes that problem by making the your votes like have a cost, right and have a quadratic cost. So if you want to vote with a score of one that cost you one credit, if you want to vote with a score of two, that costs four credits, if you want to vote with a score of three that has nine credits. And so the more extreme your vote gets, the more expensive in terms of kind of sacrificing your ability to vote on all the other issues doesn’t get to vote even what kind of one unit further right. And it turns out that there’s a bunch of really nice math why quadratic voting is optimal. And it does basically perfectly compensate for this kind of zero and 100 bias, that range voting house. So quadratic voting, basically, it’s great because it allows you to express these, these preferences that have varying levels of strength, where some people might be weakly against an issue. So other people might be strongly in favor of, or against an issue. And it does it in this kind of more inclusive way, where it’s part of the protocol. Because as opposed to like, say, for example, like nimbyism, right, where you do have a small number of people that’s like heavily against, you know, housing being built, but the way that kind of city hall meetings meetings, and the whole system is structured work is that it basically, you know, really selects for like a particular group of people that just have a lot of time on their hands to like, to, you know, go into these meetings and keep talking and shutting everyone down. Right. So, whereas, you know, with quadratic voting, it’s like, there is a way to express the fact that you feel something strongly or feel something weakly, but it’s done in this kind of more, you know, equal way where just everyone gets some number of credits, and you can just go and vote.
Lars Doucet 44:28
You know, speaking more on, you know, nimbyism and your kind of thoughts on governance. You’ve written this article talking about this dynamic between V talk cracy and bulldozer mentality. I was wondering if you could briefly summarize, you know, just in a couple sentences, what that means for our audience and how that drives your thinking and ties into the things we’ve talked about today.
Vitalik 44:51
Yeah, so if you talk cracy versus bulldozer was my attempt to, like show a different conceptualization of political divides that we have, right. So I think people are used to left versus right and authoritarian versus libertarian. But there’s a lot of people like a lot of movements and viewpoints that really cut across both in a way that feels inconsistent, but actually, is very consistent. And so the authority or the autocracy versus bulldozer or kind of access basically says, On the VT zakresie side, you have systems that are biased toward inaction, and toward doing nothing, unless, like, almost like almost everyone agrees on a move in a particular direction, and giving lots of different participants veto power, whereas bulldozer is, is basically, you know, systems where you do give like, particular single individuals or potentially large groups, just the ability to kind of move things and that’s, you know, the don’t ask for permission, beg forgiveness, sort of ethos. So, one of the criticisms of contemporary America that you see recently, I think, it was hard for people to articulate, especially if you think in the left versus right, in the libertarian versus authoritarian sense, because the criticism has aspects to it that appear on all sides of the spectrum, right. And the criticism basically does say that, you know, hey, look, America is at the top per se, right? So some examples of this are that, like, you get people who milk for example, criticize, you know, government for being ineffective and not being able to agree on new laws, or even be able to do local things. Like, there was this fairly famous big article, like, I think one or two years back, just talking about, like, why Penn Station is terrible, and how no one has managed to upgrade it for a really long time.
At the same time, um, you know, you have, like, the criticism that like, oh, you know, private property gets kind of really respected. And it’s really hard to do even very needed things if they go against private property in some senses. And at the same time, um, you know, you get this criticism of nimbyism, and which is weird, right? Because if your viewpoint is like, oh, like, the problem with America is that it doesn’t value the need to do a collective good things because it respects individual freedom too much. But like, California zoning color codes are the exact opposite of respecting individual freedom, right. So like, that’s kind of the totally middle 45 degrees off wrong access by which to even make that case, right. So if instead you say, Well, you know, the problem actually, is that America has veto kradic. And that this is a big problem, then, like, that actually makes sense, right. And I think there is a good kind of way to shape the historical story for why this happens, right? Because like, there is a left photography, there is right photography, and I think you can view a lot of it as being a reaction to authoritarian high modernism in the early 20th century, right. So both Hitler type stuff, but also Robert Moses, and some of that kind of urban planning stuff that just did a lot of bulldozing that ended up really hurting people. And so, you know, there was this viewpoints that like, okay, look, like, you know, if you let great man just like go around and have power like that, then you know, they’re going to do terrible things. And so, you know, you can have the left wing response to that, which is dominated by things by saying, you know, oh, you know, we need community input. And we need to make systems that lead to make sure that something like that can’t happen without community input. And then there is the more right leaning over here and responsive says, We’ll see, this is why personal property really needs to be protected. But you know, really, they’re, in some ways, part of the same trend. And in some ways, the kinds of problems that too much of that mentality causes are also very similar problems. Right? Like, you know, if you have a system where you have total private property, and there just is no eminent domain, then, you know, how are you ever going to build something like a high speed rail, right? Like, at some, once you start buying up the houses, everyone else that’s kind of in the middle of the path, they’re going to realize that this is going on, and that’s their consent is like, basically required for this multibillion dollar project to finish. And so they’re going to build things up by a factor of five. And just, you know, coordination between like the UK kind of different government bodies is often a problem. And all of these said, you know, different reasons why stuff is expensive in, in the US. So, if you make that criticism on the like, if you feed zakresie versus Baudot, bulldozer access, and you basically say, well, oh, the, the problem is that, you know, maybe America needs to be willing to be a little more bold or sorry, then, like, the argument just feels kind of more coherence. Right. So that’s in the US context. In the crypto context. I think one of the things I mentioned in the article is that I think crypto has this interesting duality to it in that crypto often is accomplishing some Pretty bulldozer react, right? It’s basically saying, you know, look at these sclerotic financial systems that are not updating to the modern age, it’s still incredibly hard to send money internationally. So we’re just not going to bother interfacing with this existing system, we’re just gonna, like, go ahead and build our own. And, you know, here we have a currency, you can now transfer it from Augustana, Guatemala as quickly as you can transfer an email and, you know, you can go use it, and it’s great. And, you know, you know, good luck trying to stop us. But at the same time, the mechanisms that cryptocurrency uses to protect itself internally are very veto kradic. Right. And this is especially true in the Bitcoin space. Because, you know, in the Bitcoin space, there’s this very conservative developer ethos, and there basically is a group of like a couple of dozen people where any single one of them can basically Stonewall a particular proposal to the point where it’s either just can’t happen or takes a Yeah, a really long time to happen, right? And there’s this really strong bias that like Bitcoin is not even a majority vote system. It’s not a democracy, you know, you actually do need consensus and consensus basically means no, no significant disagreements that kind of the Bitcoin already, you know, we just decide is technically reasonable. So in theory, I’m I think also has this to some extent, like it is a weaker I think, in part because in theory M also has this like constitution around like moving to proof of stake and moving to sharding. But you know, there are also a pretty strong Fatah cracy norms, so especially around, you know, things like like immutability, and just not going around and changing the state to, you know, like rescue accounts that got hacked, for
Lars Doucet 51:42
example. Yeah, so that kind of brings up like the whole first mover second mover advantage, you know, there’s this common criticism of blockchain is the environmental issue. A lot of modern l ones and l twos that are coming out are now kind of proof of stake is kind of table stakes for environmental reasons. And also just for, you know, transaction time and cost reasons. Bitcoin was the first and so they had the advantage of got all this mindshare, but then they’re also kind of locked into this V ta cracy. As you say, like, I don’t see Bitcoin changing, like in major ways, anytime soon. Ethereum had your second mover advantages, because you could see what happened with Bitcoin. But um, I guess the question I’m kind of asking is, if you could talk about first mover and second mover advantages and disadvantages. And like, if crypto if you had all the knowledge you have now, and crypto revolution hadn’t started yet, and you were now going to found a theory? I’m like, what would you do different basically, with knowledge of these first and second mover advantages and disadvantages.
Vitalik 52:39
I do think that there were a lot of these kind of specific technical things that I would want to, like transmit to myself back in 2014. So things like for example, you know, don’t bother trying to make this really complicated and perfect proof of stake from the beginning instead, start off with a simpler proof of stake and upgraded over time. Things like a lot of particular places in the protocol where the protocol could have been designed to be a lot simpler. So there’s a lot of these different lessons that I think we’ve learned from Ethereum is progress so far, and a lot of those things are fixable. But just because if your aim is a living system, it takes much longer to fix them. And Aetherium itself did ends up learning quite a bit from Bitcoin, right? So like, for example, one of the political problems that Bitcoin has, I think, is because there is this one central Bitcoin clients, the bitcoin core software implementation there that basically creates this elite of core developers where those are the ones that have veto power over a change. Right? And it’s very hard to basically yeah, for an outsider to come in and start participating in Bitcoin governance on their own terms, kind of without, you know, respecting the existing establishment and, and its values and so that were easily lead stead being kind of sclerotic in a lot of list. So the way what Aetherium did is it basically said, we’re not going to have one clients, we’re going to have a bunch of clients, and anyone will start their own clients, the Ethereum foundation is going to work on a client’s and but we’re going to work on a client that’s written in Python, which is a very slow programming language. So it’s basically just like any implementation that’s set up to fail. And in a market, I was just there to kind of show people how it’s done. And just experiment with the ideas, right? So that’s something that, you know, it’s a choice that we made, actually, that we made two iterations of the choice, interestingly enough, so when Aetherium first launched, there was, you know, the Go client, the C++ clients, and But what happens was that we ended up launching, like basically being forced to launch more quickly, fairly quickly, because I’ll actually it was it ended up taking much longer than anyone expected, it ended up taking about, like 20 months, and people were just so tired of a theory, I’m not launching, but they were like, hey, you know, this thing has to come out already. And so because of that only one clients death was really ready at lunchtime. And so the other clients were not ready. And so like, none of the other clients ever really had a chance. So then after that, there was the parody clients, which was started by an outside team. And so that succeeded for a while, but then the parody team went off to do polka dot. And so they stopped, we’re working on that client. And now actually, we have like, nevermind, Aragon, basically, we have a couple of alternatives. And it’s finally becoming healthier. But that took a long time. But in theory comes during our switch to proof of stake, we actually learns from our own kind of like, rounds, one failure. And we get, that’s when we decided on this approach that the foundation is only going to do the Python client. And the foundation will kind of actively subsidize any kind of competent seeming group that’s trying to make a different client. And so now we actually do have this like diverse ecosystem of the four or five pretty high quality of implementations of Ethereum is proof of stake side. So that’s an example. So that there was an example of I guess, a theory morning for Bitcoin and then Aetherium warning from itself. And then, I mean, there is, of course, these newer projects that claim to be learning from Aetherium. And that tried to kind of avoid what they see as being some of Ethereum issues, which is interesting to see, I think it’s interesting, because sometimes they do identify things that like, if Etherium easily could, we would love to adopt, there are some times the identify things that like actually don’t really matter that much. So and then sometimes they identify things that I just disagree with, right? So a lot of those chains tends to be much more comfortable with higher levels of centralization, for example, right. And that’s something that I think both I and the Etherium community are not comfortable with. So that’s a case of just them deciding to, you know, make a kind of fundamentally incompatible different products. And so we have, you know, two ecosystems, and everyone just goes to whichever one appeals more to them personally.
Lars Doucet 57:26
So I wanted to tie this back to get crypto and Georgia ism together, I wanted to ask you a quick question to kind of tie those two together was, there’s been this whole trend of virtual real estate on chain. And I was wondering if you had any thoughts about that phenomenon? I have my own thoughts, and I’ve written extensively about it. But I think it’s kind of this interesting. confluence of land economics and crypto. Yeah,
Vitalik 57:52
I think it’s a very fascinating topic. I think it’s part of this like a broader topic in crypto that I’ve been trying to push for, which is like, if you have a crypto project that can create this, like, limited supply asset that people ended up wanting and being willing to pay huge amounts of money for, then I think we should create systems that try to kind of channel that revenue into funding, like public goods in the ecosystem that would otherwise get underfunded, right. Like I think the way that Glen Weil says it is a tactic ingestible subsidize the increasing returns, right? So it basically it’s like, tax the thing where if more people participate in the the experience becomes worse for each person, and use that revenue to subsidize the thing where if more people were dissipated, becomes better for each person. And so you kind of fix one of the big economic imbalances there. Right? So in terms of like, things that are ingestible, in crypto lands, like obviously, there is the just the the token supply, right, so there’s a fixed number of tokens. And if you issue an unlimited number of tokens, the whole thing breaks. And that the ability to issue that limited number of tokens becomes the sort of key ingestible resource that you can get revenue from allocating. And other one is block space, right? If block space was infinite, and everyone could send as many transactions on Aetherium as they wanted, it would be impossible to run a node. And so there’s a fixed amount of space and so Aetherium automatically auctions that space. And since about a year now, the revenues from those auctions get burned. Right? This was EIP 1559. The main goal of VIP 1559 was originally actually to get rid of first price auction efficiency is it was this kind of technocratic credit economic thing that basically said hey, you know, looking at what game theory says about what how auctions work today, and let’s make a better one, but it had the side effects that also it actually burns those coins. But then also there are other kinds of congestion effects in blockchains where is an excellent one. ENs names are also an excellent one, right? Like there’s only a limited number of three letter and four letter ens names. There’s things like the right to sit in include transactions in a block in a block first, this is the thing that the cool kids in crypto will call them nev extraction. Like basically, there’s a lot of revenue that you can gain from being able to choose just the first transaction, like, you know, you can do arbitrage of on chain things and so forth. There’s a limited issue. NFT is obviously at red. So there’s this long list of things that are in the crypto space that Bennett that have like some kind of congestion effect or benefit from being a limited supply in some way. And I think, you know, the idea of actually yeah, capturing that revenue, and then putting that revenue toward, you know, funding things that would otherwise be incredibly hard to find is somewhat one of those things that I think many other crypto space already is doing to some extent, but potentially could do really well. And it’s also one of those places where I’m hopeful that I’m you know, there might be some innovations there that then get exported to other contexts.
Will Jarvis 1:01:11
That’s great. Well, metallic, one more quick question. You’ve got your whole life ahead of you get the attention to the world. What problems do you want to help solve through the rest of your life? And how do you prioritize those problems?
Vitalik 1:01:22
Hmm, that’s a definitely a good one. I think, like arounds now is an interesting time for myself personally, because it’s getting to the point to where Ethereum research is ossifying. Right? So it’s, the amount of stuff that’s changing in the Etherium roadmap is becoming lower and lower, the Ethereum Foundation’s a level of quality has just shot up massively over the last three or four years, to the point where if it’s gone from me being like, basically indispensable to any kind of research happening to the point where there are these really amazing and smart teams and individuals that are capable of, you know, making these very high quality proposals themselves. So and at the same time, it’s also getting to the point where like, we understand what blockchains are like, and we understand what blockchains will be like. And so a lot of the stuff that hasn’t been done yet is going to happen at the higher levels of the ecosystem. And so, lately, I have been getting more interested in the higher level questions. And, you know, and the, these questions of like, what are some good Ethereum applications that don’t exist yet? How to make Dallas work better, how to make decentralized identity work better, Ian’s mo these kinds of things all start end up kind of quickly, heading into the questions of kind of mental political economy and psychology and how people interact with each other, and even more than blockchains themselves do. So no, no, it feels like I’m going to be whacking my head against that. And, you know, both in the Etherium ecosystem and also in a kind of increasingly in need of some of these efforts outside and I think increasingly over time, those spheres are going to start merging for quite a while.
Will Jarvis 1:03:12
That’s excellent. Well, metallic, thank you so much for taking the time to come on the show. We really appreciate it.
Vitalik 1:03:17
Thank you to you. This is a great one.
Will Jarvis 1:03:25
Thanks for listening. We’ll be back next week with a new episode of narratives.
