S.1 Ep.9 Futarchy, Prediction Markets & the Future of DAOs — Vaughn McKenzie-Landell, CEO of Butter
In this episode, we sit down with Vaughn McKenzie-Landell to explore how market-driven decision-making can transform DAO governance. Vaughn is CEO & co-founder of Butter, who are bringing Information Finance to Ethereum — starting with Futarchy. He started his career at Dow Jones but exited TradFi after a google search in 2009 delivered him into the BitcoinTalk forum. Together, we break down Butter’s evolution from simple delegate-payments to more complex schemes like proposal auctions and peer prediction, with each iteration teaching valuable lessons. We dig into their latest futarchy experiments – for example, running market-based grant allocations for Optimism’s ecosystem and for the Uniswap Foundation – where clear objectives (like growing total value locked) were set and participants bet on which projects would best achieve them. Along the way we cover practical challenges: choosing objective metrics (often ROI), incentivizing honest participation, and even regulatory hurdles. Finally, we consider what this all means for the broader DAO space: futarchy seems well-suited to big funding decisions with measurable outcomes, while simpler proposals may still be handled by more traditional votes. The episode wraps up by highlighting how continual experimentation – a bit of a “kaizen” (改善 - a Japanese term meaning continuous improvement (“kai” = change, “zen” = good) mindset – is shaping a more innovative, resilient future for decentralized governance and capital allocation. Timestamps 00:00 Introduction to Governance Futures Podcast 04:13 Exploring Futarchy and Its Applications 11:00 Governance Challenges and Information Problems 12:59 Understanding Futarchy in Simple Terms 18:55 The Evolution of Butter and Initial Experiments 26:52 Misconceptions About Futarchy and Prediction Markets 33:40 Concrete Experiments and Their Outcomes 55:30 Future of Governance and Futarchy Models 01:00:22 Navigating Regulatory Challenges in Crypto 01:09:53 Quiz and Closing Remarks
Top Keywords
- futarchy 0.017
- markets 0.013
- market 0.008
- crypto 0.007
- governance 0.007
- prediction 0.007
- decision 0.007
- butter 0.005
- much 0.005
- prediction markets 0.005
- protocol 0.005
- guess 0.004
Transcript
Crypto is just so good that the the scammers are just very, very good at using it. And you can really see crypto as a hyper computer where everything is tokenized. Everything is is given a price, and and therefore, everybody can kind of contribute their own information. But crypto also shows us the kind of dark side of that where, yes, we build protocols that allow people to speculate on everything. People often speculate on speculation itself rather than on things that are productive. TeaTarky was indeed the 2,000 feet and New York Times buzzword, which really is kinda wild, that it was the buzzword the year the Bitcoin paper came out. I think the hype was right. I think markets are incredibly are incredibly used at all. And so, yeah, that's what we're doing at Buzzer. I'd say we use prediction markets to predict the impact of our decisions. We use markets to aggregate aggregate information, and we use the information to inform decision making rules. The reason why some of these things worked is precisely because the founders built things outside of the rules, and it allowed us to explore more of the design space.
Hello, everyone. My name is Jamila. And I'm Eugene. And this is governance futures podcast where we explore the past, present, and future of decentralized governance. We spoke with CEO and cofounder of Better, Von, and we would like to share some of the reflections that we had. Yi Jin, how did you leave the conversation with Von feeling? What was your, perhaps just, yeah, overall thoughts?
Yeah. It's been really cool to see the evolution of butter from the first Ave experiment that they talked about to kind of the current state where, you know, they ran these markets with Optimism and Uniswap. And just being one of the first groups to really start bringing Futarchy into the space, it's just really cool to see this kind of area of experimentation. I also really appreciate that Vonn kinda contextualized that a lot of the focus of their initial work is specifically in the capital allocation, you know, especially towards grant type, programs and that those kind of decisions. But, you know, we got to also zoom out and just talk about Futarchy theory a little bit more and the way that Robert Hanssen thought about it and more setting policies and, you know, just thinking about that kind of future landscape of experimentation around Futarchy. So, yeah, I am both highly interested and still, you know, as someone who is not a market's first person of solving every problem through a market, I am inherently a tad bit skeptical. But I can see how, especially in the domains where, you know, butter is focused that I think that kind of application can make a lot of sense, and we are already seeing some positive results. So I'm really excited, you know, to witness this evolution of which type of mechanisms and governance models lead to the most effective decision making for different types of decisions. So I really appreciate the work that Vaughn and crew are doing to kind of help us clarify the edges of that landscape.
And what about you, Jamila? How are you feeling? I was also entering this conversation with, a bit of skepticism over this whole, like, production market driven approach. But, you know, the perks about being a host of a podcast that you can ask questions and frame them. One might say that and I feel like, yeah, Von definitely, allowed me to see it in a different way and debunked some of the misconceptions. So definitely sparked my interest to look more into that as opposed to just kind of close off myself from that and just not getting in there because from just the initial idea. It's not something that I would be, maybe very much attracted to to explore. So I really love those conversations where you get to change your mind a little bit and open up, about certain preconditioned ideas in your head even. One thing I also liked with our conversation, it's not a direct quote, but he said something like, in order to build something new, you have to break things first. And, yeah, I also like this approach of experimentation, always trying something new and in hopes that that would bring us to a place where we've tried so many things and we know a little bit better what work, what doesn't. And I do agree with him without actually experimentation and sometimes failure, you can't really figure that yet. So that was very interesting, insightful conversation for me and hopefully to our listeners as well. Absolutely. And on that note, I also appreciate that they're very thoughtful with their experimentation,
and they're not sprinting to just put all the decisions onto the system, but they're being very intentional about where to put that in, which I, yeah, just, kudos to them for being mindful and not just, sprinting blindly. So, yeah, with that, also just a reminder for listeners, if you're a fan of what we're doing, please like or subscribe, share with a friend. Feel free to reach out to us if you wanna chat about anything. And with that, here's the conversation with Von. Hello, Von.
Very happy to have you here. I wanted to kick off, today's conversation, with, something that we discussed before. You mentioned that you read Haik and The Road to Serpdom is an all time classic. I have it in my hands right now. And it's one of those books that you often see quoted on crypto Twitter. And, my personal favorite essays by Hayek is Why the Worst Get on Top. And another one, I don't think is part of The Road to Serve Them is, one, not a conservative. And I just realized today that in both of those essays, he quotes Lord Acton, and, obviously, this very famous quote, all power corrupts, absolute power corrupts absolutely. And because you've read it, I'm curious, when did it cross your path, and how did it resonate with where you were at that time?
Wow. That's a great question. I think, I think I just went on a bit of a so I think it was after I shut down my previous startup and I had some, some free time. I was thinking about what I really wanted to learn having spent a bunch of time in crypto, and I really felt like the two big places I wanted to spend time was game theory and I think smart just smart contract design. So just trying to combine those two things. And I felt like that was that was, you know, two skills that became, amazingly it was it was it was incredibly obvious that without a fairly good understanding of those two skills, it would be just harder to navigate crypto. And so I was kind of trying to prep skills for my next foray into into entrepreneurship and crypto. And, and I think as as I was going through the game theory literature, you you realize that your, you know, your obviously, you're standing on the shoulders of of of some of the greats, and so you you want to go up, just keep reading as high up as possible. And so as I was climbing up the ladder from, some of the, I guess, like, simpler papers just on, like, very well known, game theory and mechanism design. So for example, just looking at the standard prisoner's dilemma, and then looking at equilibrium, and and then you suddenly realize that some of these ideas were just explored in a in a much more abstract sense in in much older essays. And so I think eventually I ended up, wanting to go through some of the core economics, writers, and Hayek was obviously high up on that list. And so that's where I ended up with, with Hayek. So sometime in the, like, you know, COVID period, sometime around that, I was reading I was reading that book. Yeah. So fairly recent.
And then just following up on that, do you see protocols as a way of testing hikes, thesis, and practice? Perhaps something that would relate to distributed systems, outperforming centralized authorities and organizing communities and societies. Is there anything that you've taken and applied to your work right now at butter?
Yeah. And that's a great question. I think it's maybe a little bit leading. But, yes, I completely agree. I think there's there's an it's an interesting tension there. Right? I think that parts of crypto are hyper capitalist. Parts of crypto are, I guess, have this sense of, what what do we call it now? What is it? Crypto communism or something or post capitalist? I don't know. It's all all the terms are getting strange. But I think at the core of of crypto has always been this, like, like, ideological, standpoint around, the kind of use of markets and everything. And you can trace that back to the essay by Hayek, which is I think the use of information and knowledge in society, I think it's called. It's, and the basic idea is that you can think about markets as kind of like computers, that they do an incredible job of aggregating information if it's dispersed. A market basically always creates a price for that information. And so if you if you can see a price that is incorrect based on your understanding, you have some incentive to go in and correct it and thereby you contribute your information to the overall price, and therefore it becomes something that's known in society. And you can really see crypto as as being kind of a a hyper, computer in that regard where, everything is tokenized, everything is is is given a price, and and therefore everybody can can kind of contribute their own information, without and I guess this is the one of the core tenants of crypto in a very permissionless way. Whereas lots of markets over time, the world has has has kind of evolved to protect these markets, mostly because it's trying to protect us from from doing things that could cause problems like, you know, there's a really good example of, onion futures, which is still banned in The US because, two guys cornered the onion futures market and created a huge stockpile of, onions outside of Chicago. And in the end that led to people dumping onions into the river because they just had too many onions. And so what you could imagine is that people kind of manipulate markets for profit and therefore lead to adverse, outcomes in real life. And I think that that's why markets ended up with lots of protection and crypto undoes that. And on the one hand, you can argue that's incredible. And I think Hayek would argue that this is a great way of of kind of assuming that, well, if you assume that people don't know best, but in aggregate, we're very good at combining our information to, you know, therefore know best, like the collective wisdom and combining them via using markets is a incredible mechanism of of kind of scaling information. But crypto also shows us the kind of dark side of that where, yes, we build pro schools that allow people to speculate on everything, but speculation on everything isn't necessarily always useful. And, actually, if left to their own devices, people often speculate on speculation itself, rather than on things that are productive. So, yeah, I think kind of a a tale of, like, I guess, like, two halves with, with respect to Hayek and and how it relates to what we're doing in crypto today. But, yeah, I have hope. I think that Hayek was right. I think markets are incredibly are incredibly useful tool, and, and so, yeah, that's what we're doing at Butter. We use, we use markets to, aggregate information, and we use that information to inform decision making. And that's what Butter has been doing for the past couple of years now. So before we go into,
deep dive into Butter and some of the most recent updates and what you guys up to, I'm curious, what, governance challenges interest you more or perhaps, present you as more challenging information as in getting the right signals or incentives getting people to act on them? Or maybe you think they're both equally challenging and important.
So there's I I think there are lots of interesting challenges that aren't they don't seem very useful to solve. And I think that, you know, I was saying before, about, like, how crypto has evolved over time. I think that humans have the people in crypto have kind of clustered around the ones that are, like, not necessarily useful to solve or just, like, more fun to solve or things that people know that someone's gonna turn up and use. Right? So I think with with, with governance, I think when we first started out, we were much more interested in in the problems that we thought were, I guess, really useful to solve, and I think those mostly clustered in, in information problems. But I think we were we kind of, I guess, had picked a path and the path was incentives. So it was really we kind of set the goal as being how do we address governance problems using incentive design? How do we use cryptoeconomics and kind of extend it from kind of low level, like, managing state on a blockchain and having people validate state all the way up to having people make decisions about what things to do in like the governance of a protocol. And that was kind of our stated goal and that was really where we ended up working on incentive problems, not just information problems, but ultimately these lead to information problems. They are two ends of a spectrum, because ultimately in governance, kind of the core problem is that, you you're trying to get clean signals from the people who are who participate about what to do as an information problem. And, ultimately, the problem is always that nobody wants to give you or an on is incentivized to give you that information, and so you end up having to solve an an incentive problem. Thank you so much. There's definitely a lot of food for thought.
I'll be curious before we dive deeper into Futarchy, if you had to explain it to someone who is completely unfamiliar with the concept in one sentence, how would you explain Futarchy
to them? I'd say that and it's only one sentence. Right? Mhmm. And if I need more sentences. Could be two. I'd say I'd say we use prediction markets to predict the impact of our decisions, and then we use the prices from those prediction markets to actually make the decision. And I then I would say, sentence two, is, we basically run this exact same algorithm in our heads. We think, should I grab the very hot cup of tea? I'm gonna say cup of tea because it's the first thing that came to mind. And, we forecast whether we're going to burn our hands or not, based on what we can see, that there's steam rising off that cup. Thank you, Eugene, with the with the mug. Yes. You too. And then and then based on that, we make a decision. I'm gonna grab that mug. Oh, wow. It was hot. That's the penalty. We just lost money. So we're incentivized not to grab the hot cup of tea. So, yeah, that's, like, that's how I normally explain it. I try and make, map it back to something that someone probably very recently did.
Yeah. And I guess before we get into the specifics of butter, I'd actually wanna double click on that kind of framing of Futarchy and use that as a lead in to potentially talk about kind of the evolution of the steps of butter. Because I know the the current iteration of the experiments that have been happening with Optimism and Uniswap Foundation, these are by no means the starting point of your journey on butter. So I also wanna kinda touch on some of the evolution there. But I guess when, you know, when framing that element of, you know, there is the understanding the potential or desired impact and trying the price for that, it does feel like you gave an example that is very tangible. Right? Like, I don't see steam coming from this cup, so I I know I can grab it. But when it's, you know, like, will project a or project b lead to more of a certain outcome for an ecosystem, whether it's, like, long term sustained growth? I guess, how much is our general human challenge with accurately modeling and predicting these things in the first place? How does that either present unique challenges for Futarchy or that's kind of the name of the game is, like, we're all gonna be flawed, but once you aggregate enough of collective intelligence around it, it should tend on many topics towards, you know, like, a realistic average of what the outcome might look like. So, yeah, I guess, how do you, consider that tension?
I think everything's about cost. So in, a collect any collective decision making system, there is a cost to finding the right answer. And also there's a subjective model of what right means. And I think that as we try to make decisions about, increasingly complex or esoteric, subjects, it gets harder and harder and harder and more expensive to figure out what's right, what's correct, what information actually helps you to make the best decision. And you can think about information being in kind of like, one or two different modes. The information you need could be broadly distributed across a large set of people. Right? It could be across the whole world, or it can kind of be, like, in the heads of just a few people, right, just some subject matter experts. And so you could engage in in one of two things. One, you could say, well, I'm going to go and survey every single person until I get my answer. And you can kind of assume that a vote does something like this. A vote's trying to aggregate it's gonna give every everybody a single vote, aggregates all of their preferences. They tell you what they want, and then you use that to make a decision. And there's a reason why we don't have people vote on everything because not everybody has good information. Right? So we've kind of solved this by having a representative governance as, like, a a model of governance. The second model is that you have some subject matter experts and you need to find them. Right? So you have this there's this like really good breakdown of like, time triangulation and trust as being the three T's of transaction costs. And so you need to go out and triangulate the person who's got the information you need. There's some time component, and then you need to trust that they're going to give you a good answer. And so the transaction cost for finding the information you need in lots of these governance decisions is actually just pretty difficult. If you have the experts in the room, then great, but then you have to make sure you trust their opinions and they're not going to be biased to try and get things that they want rather than the things that you want. And so what it looks like is that prediction markets are a pretty good way of solving all of those problems all at once, whether you're in either of these two environments. It doesn't matter if the information is distributed around a lot. Also, people and prediction markers are going to incentivize all of those people to produce that information to you, and it's going to give them good incentives to give you accurately. So it solves the triangulation problem, and it solves the trust problem, and it solves the time problem, so you don't have to go and find those people. And similarly, if you have subject matter experts, you can actually they fall within the same paradigm. They can again, you it doesn't matter whether they're like all in one place or they're distributed. Prediction markets should incentivise them equally to come to market and give you that information. Now, if you, again, I would say if you could find the person who give you the information you need and you could properly incentivise them, you have no need of prediction markets, you already solved the problem. But that's always the problem we're trying to solve in governance, right? Who should we ask? What information do we want? And how do we make sure that they are incentivized to give us that information accurately? And I think this is obviously in voting systems, one, we we run into this problem all the time around like strategic voting because a vote itself doesn't always combine all of our preferences perfectly, and it largely just gives you the preferences, not necessarily the right answer. So that's what we're that's really what we're trying to do when we think about,
like, why why we would use Futarchy to solve some of these problems. So to transition now to more thinking about how this has, led to the specific design decisions that you're making with butter,
where did the journey with butter begin, and what did the initial kind of experimentation look like? Yeah. So we we started with this phrase, like, treasury governance. We were thinking largely about how you govern a treasury. And within, like, very quick order, we, we realized that, DAOs at the time weren't really ready to govern the treasuries alone, and they were kind of just had all these general governance problems. And so we started out actually in a really fairly, I think, quite, a humble place, which which is we just deployed a very simple experiment to use payments to delegates. This was before really there were many protocols paying delegates for governance. Thing really was only MakerDAO. And what we we had at Aave was a simple, a simple payment which would go out once a month to a delegate who won an election and the election was, was run on Snapshot. So again, very, very simple. Just what happens if you introduce payments? Can you increase the number of people who will actually delegate? And I think that that was overall a successful experiment, but not because of, I think, our thesis. Our thesis was that you could actually aggregate more independent smaller holders of Aave to delegate to a delegate and to get away from this idea that, whales just control delegates. And I don't think that was actually very successful, but I think what we did learn is that there's a huge problem, again, with triangulation. It's just incredibly hard to get lots of small Aave holders to turn up and delegate. And why is that? Well, first of all, they're just random addresses on on chain. And second of all, they have very diminished incentives to participate. Actually, whales have strong incentives to participate. So that was that was where we started is we kind of just really dove straight in, deployed something immediately with a very simple thesis, very quickly disproved it, but also proved some other, things like incentives work. Right? Like, there was an incredible, output from that delegate, then they got, a bunch of new delegations. And now I think they're one of the biggest delegates in, in Aave. But at the end, I think it told us that we needed to to kind of explore more deeply. And so we we then moved on from that to, designing what we called proposal markets, which was the idea that you would auction off proposals. That was something that delegates did not like. Sorry. We decided not to put that one. Yeah. They said, no. That's a terrible idea. Don't do that. And then, we moved on to a slightly different idea, which was to use peer prediction, say, well, let's have, rather than it's having, people vote on the outcome they want, let's instead have people rank different contributions. So imagine that there was a proposal, you would actually just have people rank the contributions or score the contributions. And what you'd end up with is a set of scores and at random, you would pick one of the the raters and you would make them the reference and you would score everybody else according to that and you would assign them payments based on how close they were. And that was really interesting, but one of the problems with with that and one of the well known problems with peer prediction is it very quickly devolves into being a, a set of predictors who do all the predicting and a set of people who do contributions. And so you end up with specialization. And, and again, you really just run into like a communication, a triangulation problem. It's hard to get everybody to continually, contribute to that over time unless their incentives are well designed. So that was our our arc. We were trying to kind of iterate closer and closer to what we call this kind of standard model for thinking about governance, which we call the, what do we call it again? The, the objective, it was like this, I forget what it's called. It was really, we have a a a a post of it. Oh, that's so funny. We've moved away from it so much that I've forgotten what it's called, but there's a standard model basically. And it starts with contributions. So everything is a contribution, and then you have some way that you're trying to establish what is the objective truth about that contribution. And you can kind of map all governance systems to that model. And so we came up with that, we we wrote about it and then very quickly found that one of the best implementations of that was, was Futarchy. And so when we embarked on our on our Futarchy journey. So, yeah, that was our that was our arc to Futarchy.
Yeah. That's interesting. I was trying to do some quick googling to find the title of that, and all I got was Nutter Butter and butter related things. So that was not the best use of a distracted time. Before advancing, though, could I I did wanna double click on these two ideas that you brought up with kind of the auction, proposal, and this kind of, like, the peer reviewing element of it. I know the auction proposal, I know we had talked about it, way back when when y'all were, initially coming up with it. And I feel like now it's kind of having its own weird resurgence in the space. And, you know, with groups like LobbyFi and others, though, I know some of these groups are very particular around the terminology used, but effectively, people can choose to, like, bid on amplifying their vote ability in some kind of way. Do you think that, you know, if in a alternate universe, the butter project is just getting kicked off, with where you were doing the AVE work and what followed at this point in time and not where you did,
do you see a potential different way that the proposal auction might have been interpreted these days? So funny enough, I think that, we were that was the original part. So if we if we stopped when we were done with the first kind of election on AVE and we looked forward, the path we we mapped out looks actually a lot like event horizon, which looks like political parties. So people are are basically staking tokens or trying to build a set of of, like, of delegates and then put a lot of, of, like, of of token weight behind them, and do it, like, by sharing the returns on on, like, on that stake weight. And, and, ultimately, we just couldn't get comfortable with that being a positive development for, protocol governance. And I think it's also fundamentally limited because one of the issues is you you you can never become the the there's a fairly good, like natural immune system around that party becoming too dominant. And, you can kind of see it with, like, Lido's dominance, like Lido, you know, kind of having to actively self limit in some ways. And we saw this similarly with, I haven't seen that with Event Horizon, but definitely with Lobbyfy. Right? They're they realize they have to self limit, in order to remain, I guess, legitimate. Like, their power actually is based on them, the kind of exercising some constraint. And so we decided not to go down that path. I think proposal auctions, I think they're awesome. I just wish we could have done them. And I don't know if they would be more acceptable today, but I think that governance is, seems simultaneously to be bifurcating into more radical governance ideas, and also much more, conservative ideas at the same time, just because of the, things like do you know and governance becoming a bit more, like, I guess, like hard coded, I guess, or at least more tradified. And I think that's that's that's a shame, but I do think the proposal auctions has a place because ultimately, we are seeing people start to price these proposals. These things are effectively up for up for I mean, you could basically price every proposal based on the cost of renting the tokens used to pass it. Right? So all these proposals, they do have a price. But I think if you, again, I think if that system became dominant, it might really undermine the legitimacy of any kind of DeFi protocol if every proposal was just passed with a payment. So, yeah, so I think I think there's some unique I think I would temper my exuberance there for proposal markets by saying I don't necessarily think that they could survive if they became dominant, though I do think they could be a nice fringe for certain types of proposals, kind of like easy governance.
You mentioned that decision markets are prediction markets that simulate each future where we pick one of our decisions, and you talk often about some of the common misconceptions when it comes to futarchy. You mentioned that there is several misconceptions about prediction markets, and one of them is that decision market is equivalent to vote by. Could you please elaborate a little bit about this particular misconception or any other common misconceptions or myth that you hear that you want to make sure
to debunk this and put this word out? We're gonna put the word out. Let them know. We've we've we've got I guess, like, people are very quick to engage with Futarchy and as an idea. I think it's been around for a long time. I think in 2008, it was New York Times buzzword of the year, which is absolutely wild. It's twenty twenty fives now. I'm sure it was 2008. That seems insanely long, long time ago. It might be 2018. But if you, if you just hear the word Futarchy, it doesn't really tell you much. So when someone tells you something like decision markets and you map, you have a mapping to prediction markets, then, you know, you have a better idea. But still, I think your model is based on what you can kind of into it from those two words. So a prediction market is obviously I'm going to bet on on predictions. Right? And, actually, you're betting on events. And so those the bets themselves form predictions. But when people hear the word decision market, they just assume that means, oh, I'm going to buy a decision in the same way that I'm going to buy a prediction. So I'm they think you're betting on the decision. So, ultimately, for them, that's equivalent of vote buying. And I think that's what we we mostly end up in. I think, to be honest, I don't have as many of those conversations now, but I do think that we had lots of them before. And I think especially when you're in a public setting and there's lots of people who have to apply and, you know, you're giving a presentation, it does tend to be the the the the level of discourse is always that someone believes they or they that that's how they judge the the, experiment mechanism, whatever.
It seems like you are a little bit annoyed that you have to reexplain certain things. You go on stage and you're like, oh, here we go again. Some questions about, whether it is bolt buying or not. But I'm glad that you are still reintroducing it, and I think it's one of those, like, pinned threads in your ex that, guys, please, let's move on from this.
At least that's that's the type I'm getting. Getting. That's the vibe we got. Yeah. That's funny because I think I just pinned it because I was like, ah, that's actually a pretty good thread. I'm gonna pin that one. But, yeah, I I do think it's something that you again, maybe the fact that I pinned it is why I no longer have those conversations because AM feeds moved on. They've checked out. But I think that was actually a response to to, Anatoly from from Solana saying something like that, and then I just responded. So that is that is a common misconception. I think it's one of those misconceptions that as soon as people understand it, they're fine. They're like, oh, okay. I see. It's not we're not we're not buying decisions. I get it. But I do think that there's an issue there, which is that you still have a you still have this you still haven't really explained what Futaki is. You've just explained what it isn't. And there's still you still have the problem that people can't immediately, into it what that means when you say, okay. You're going to bet on the impact because impact of a decision can literally be anything. And so you you really for people to really understand Futarchy, they have to really take a couple of steps down the, like, the levels of abstraction from Futarchy to decision markets, to prediction markets, to implement the decision, to prediction markets that are working, to predict an impact of a decision. And then what that might look like where you have multiple worlds where you could have different impacts based on your decision, and you're trying to figure out which of those gives you more of the thing you want. And I think that's that's kind of where Futaki ends up is just, like, mired in that conversation about, well, how do I perfectly quantify what I want? And what happens if it's a qualitative answer? And who would who would ascertain truth? And how would we settle the markets? And and so it does end up being something that's, like, much, much more difficult to explain than just a prediction market, for example, or just a standard market or indeed a vote, which has the benefit of being immediately understandable to pretty much everybody in the world. And I think that's why, often, even when people are have are using our our markets, they will still refer to the actions they take as voting. They'll still think in that in those models. It's just such a strong paradigm and and for good reason. I think it's it's simple and people get it. And simplicity sometimes is what you need,
to explain, the full concept. So yeah. So that's I just wanna ask you, how do you define feature to one sentence? But I definitely get, the same maybe feeling when people refer to crypto as just a scam and you're trying to convince them that it's not. But then they ask then what it is, and you sit down for a conversation. But, honestly, I would love to say that I don't have to explain that crypto is not a scam very often. But, I do because I try to also, you know, socialize outside of my little crypto bubble. And I think it's very important that we do and also show other sides of crypto. Crypto can be
a scam, but it's not all that there is to it. Well, it can all be a scam. Every every everything could be a scam. Depends who's who's who's wielding the, the technology, and I think that's the that's true. Crypto is just so good that the the scammers are just very, very good at using it.
Yeah. Yeah. That is very true. And I I did double check. It was, Futarchy was indeed the 2008, New York Times buzzword, which really is kinda wild, to think that it was the buzzword the year the Bitcoin paper came out. That's a very interesting little, little trivia tidbit. And one thing that, because, you know, obviously, Google is now spitting out a whole AI to answer to, what year was it a buzzword. So, you know, the way it briefly describes Futarchy is like a form of governance, obviously proposed by Robert Hanssen, where step one, elected officials define measures of national well-being or organization goals. Step two, prediction markets are used to to determine which policies are most likely to achieve set goals. So as we're actually transitioning to talking about the concrete experiments that have been run this year, which has been really cool to see, it would be interesting to hear what are the kind of parameters or boundaries that are initially being set for some of these, topics versus how much is it just like governance, prediction markets, yay, have fun market. So
what has that design really been like to start? So I think we can't credit, we can't take the credit for really bringing Futarchy back into the fore. A project a project in Solana, called Mesdau, basically just spun up Futarchy, fairly early on, actually. I think it was, like, maybe two years ago now, so 2023. And that, I think, really created, renewed interest in Futarchy. And but one of the things that they did is they they they kind of took, I would say, like, a a fairly, like, a, I don't know, purist approach to Futarchy, which was they just took took the idea, and they just deployed it as is and said, we're gonna use Futarchy to make decisions. And I think there's an interesting kind of wrinkle in in the way that we think about Futarchy. Futarchy is is designed, at least in the original Robin Hanson paper, specifically for policy. It's not it's not governance. It's government. It's through to as a, like, a a reasonable replacement for for kind of, I guess, like, delegated voting effect of your representative governance. And I think that, we are mapping that to what we do in protocol governance, which I guess is really important. It's actually very, very important that they're not exactly the same. And so that's why at the beginning, I was trying to say we're doing decision markets. We're not really doing Futarchy, but, yes, we are using Futarchy to govern these these protocols. Now what we did is we decided that, kind of taking the idea that Futarchy, well, any system, tends to be better at some jobs than others, especially in its embryonic state. And so we decided to try and map Futarchy. Like we said, well, if we were going to do all governance decisions, which decisions do we think Futarchy will be best at immediately, especially when it's not in its prime, right, it's going to take a while to mature. And so we subdivided governance into these four buckets. Eugene, you've probably heard me speak about these before, which was protocol upgrades. So these are decisions around how we actually update the core protocol. Sometimes these things are things are decided by security councils, by delegates. Sometimes they're just decided by the actual team. In Ethereum, they decided kind of by this this meritocratic process, technocratic process. You had all core devs, and that's how things get packaged into into releases and then into the hard forks. And then, sometimes they're just not decided at all by by anybody except the core team, like in the case of like Uniswap Labs, for example. And then, you have the second is like parameter updates, which are really which are really different. They are things that you're live updating. They're variables that you're changing in the protocol that make that change its utility for different people, who are who are all parties to the protocol. So for example, this could be like rates in, Aave or supply factors, like which assets we actually allow onto the platform and so on. And then there's a a third bucket, which is policy. And actually, sorry, a quick clarification
on the second bucket because I feel like in the DeFi DAOs where it's like change in interest rate up or down or chain it feels clearer. What are parameter updates and say l twos or more broad, more general purpose, protocols?
Yeah. So, again, I think the l twos is is quite, funny because I don't know if there are lots of, parameters that end up being exposed to governance. But one of them that you can always imagine is fees or revenue, is typically something you can close. Where does the sequence fee goes? Yeah, where the sequence of fees, yeah, and who the sequence of fee goes to, those are quite important. I think in some other DAOs, like social DAOs, for example, it's much harder because we don't tend to have direct control by the token holders over parameters that are called to the protocol, mostly because there is no protocol to manage, directly. But there may be parameters of the products. And I think that's where the third bucket comes in, which is policy updates, which is sometimes the community is the product. And so the policy around how that community engages, interacts, works together, collaborates, those are the things that you actually want to change. And so policy updates becomes a whole bucket. And then the fourth one, which is where we decided to spend our time was capital allocation, resource allocation, treasury allocation. So some funds underneath the control of the token holders, how do we actually allocate it to benefit the protocol itself? And which is funny because it's hard to disentangle the protocol from the token holders. The token holders effectively are the stakeholders of the protocol, but there are more stakeholders than the token holders. There are the users of the protocol, the suppliers of the protocol, who may not actually be token holders. And so there is this entire ecosystem, including the developers who have some relationships with the protocol and and the question is how does that capital benefit that entire group? And so, and so that's where we decided to focus Futarchy. So rather than being a broad set of any decision, we've kind of focused on these funding decisions.
Yeah. Let's start with what did that look like in terms of the setup and getting to actually go go live? Yeah. So I think in your your previous question, you were asking, like,
rather than did you just pick everything or did you did you kind of, hone down? I think, to be honest, that's probably one of the the problem I think it happens in crypto a lot. We've we're very good at developing the tech, and I'm like, okay. Where do we point it out? Like, what's where does it work best? And I think it took a while, to get really comfortable with exactly what we would focus on. And the way that we we we tried to, I guess, like, again, subdivide the space or at least try and reduce the set of decisions to be made was we tried to think about just what are three main things we need to be clear about. And so the first thing was the objective. So, what is the goal for this decision making system? What are we trying to improve upon? And this is the and I remember the name of the of the project, it's OAE, so it's object of the model. So it's OAE, so the objective alignment engine. So we have some objective and we are now trying to align, the resources. We're trying to make decisions that align our resources, to fulfill that objective. And actually that's the really the goal of every proposal in the DAO is to try and better fulfill the objective of that protocol, right, the whatever the DAO is trying to manage. And so the first one is establishing an objective and the sub goal of that is a metric by which you measure your progress towards that objective. Now with, Futarchy, we want that to be super clear. We want it to be, kind of unequivocal. Like this definitely is a proof that we're making progress. And as you can probably tell, objective, like setting and metric, setting is are both they're both like flawed in in in so many ways. Objective setting has has problems around like, you know, like, agenda setting. Right? If you set the objective, you set the agenda, is a form of political control. And then you also have a problem with, like, the, like, the measurement problem if you have a metric and Goodhart's Law, I nearly swore. Good heart's law, comes down to to kind of nip your nip your wings or clip your wings, a little bit, but that's the first, the first one. The second one is, the, the amount of money. So given given some objective, what's the budget? How much would you want to spend to achieve, to achieve the goal? And the third is is, who are the participants? So what's the kind of locus of your of your intervention? Is it is it projects, like different projects, so you're trying to kind of grow your ecosystem? Is it like strategies? Is it just any proposal that could, lie to achieve your objective? And so with these three things that helped us to kind of, reduce the the the decision space down to these kind of key things. And then after we had those, we were able to, I guess, backfill with all the extra things we needed to figure out. And so for Optimism, the the selections were, ecosystem growth, so growing the ecosystem and actually for optimism, it made more sense to think about that on Superchain, so projects across Superchain rather than just optimism. And I guess one of the things that we also wanted to bear in mind is that there are very good, very successful mechanisms already in play, doing things really, and doing a great job at, say, the optimism level. And so we wanted to look at where there were gaps and so there was maybe a gap at the super chain level. And so the mechanism was deployed there to distribute resources across Superchain. And then the metric was sorry, the goal was increasing growth and then the metric was TVL, so total value locked. And that was picked for a couple of reasons. Obviously, TVL is a fairly well known marker of of, like, success and progress in DeFi. It's also, like everything, fundamentally flawed as a marker of, of of progress and and many people, push back on its use. But, we needed a metric that was harder to manipulate. So the larger the metric is, if you give, if you ask someone to increase a metric and you'll give them money and you do it and it's something that's entirely under their control and it costs them almost nothing to move the metric, For example, like users, it's like, okay, how many accounts do you have on your protocol? It's like, okay, I'm just going to spin up a million accounts. Right? So if it's cheap, then obviously it's very subject to manipulation and so you're guessing no benefit from deploying the capital. So there is this, like, anti manipulation point to selecting the metric. Then we picked projects, obviously, so projects across the ecosystem. I think there were I don't know how many applied in the end, but we ended up with 22 projects we went live with, for the first market. And then the budget was set at, 1,000,000 p, and half of that went to the grants council as a control, and half of it went to the to the Futarchy. And just to clarify that last point, when you're saying half went to the grants council, half went,
to to the platform. So there it's the the grants council was doing their own how they would divvy up 500 k across those 20 some odd projects. And concurrent to that, the markets are sort of being able to figure out their own. That way, you kinda have this sort of AB testing comparison.
Exactly that. Though it was really difficult because ultimately to really compare them, you would have to let them run both in their best setting. But if they both ran in their best setting, probably would not be as comparable. So we tried to normalize a bit. So the grants council had to only pick five projects. The Futarchy picked five projects. There there was, they were both like, they both had to kind of think about TDL. So I think that the grants council's own rules were not necessarily the standard rules. Right. And so it it felt like both were kind of operating slightly, slightly in Hamburg, we should say, but it made made it easier to do comparisons. So that's the yeah. So we got some good comparisons out of that and I think that Futaki did fairly well, in retrospect, but Futaki is designed for those kinds of, like, predict the maximum number, and, obviously, people will will participate trying to do that. And so just to clarify there too, when it came to the top five projects
on the well, I guess, on both sides, was it, oh, if you were the most voted project, then you get x amount of funding. If you were the second, then you get y amount. Or were the actual amounts per project a separate variable that was being kind of independently decided on in in both instances?
In Fuseaki, one of the problems with the prediction market is if you do need to predict something, you need to make sure that the thing you're predicting doesn't change every day. So if, for example, I'm predicting, well, how much of some impact will this investment generate, I need to know what the investment is going to be upfront, right? So you can't have a situation where the prediction itself will change the the investment because then, you know, you're you have this, like, you know, when you're in Excel and you, like, have a self reference and it just breaks and you can't get anything out, it's kind of like that. Right? What we did here is we just had a standard amount, 100,000 OP, which every project would receive, and we ran market saying as I was saying before, we're trying to get people to trade and to basically, for every project, tell us what they think the impact will be. So assuming we gave this project a $100,000, what TVL number do you think they will achieve in three months' time? So that's what every single project was that's basically what every single market was. There were 22 markets each asking us for each project. And so what we ended up at the end, we ended up with a list basically, like a ranking of which projects that Fusarki thought would actually produce the most TVR, from top to bottom. And then that was what we used. We, the top five projects received funding. And that was the that was the decision. And that we made that we made the decision after ten days of running the the Futarchy, and then we ran the evaluation component for three months to see what happened.
And timing wise, where are we at with those three months? Have they concluded or still ongoing?
They concluded. So we ran the we started the markets March 10. Mhmm. And then we we we we shut them off to ten days on the twentieth. And then, we I think June 12, we closed the markets.
And so were what were the markets accurate? Did they get, accurate? At least amongst the five projects chosen, were they able to kinda guess which one had higher probability output of increasing TVL?
Yeah. So this part is really, really interesting. So so I think that in launching them, we learned a couple of things. One of the one of the things that we learned very early on, which nearly caused us to to not run the experiment, was that, we couldn't run the markets in The US, with real money. And so to make sure we got as many people as possible to participate, we made the decision, I guess, like Optimism made the decision to use Playmoney. So we had this really interesting process by which we had to distribute funds to people for them to trade. And because the funds the the the tokens that they held didn't have any value, for some people, and, you know, we you you actually just use the word vote. For some people, it was like a vote. So they would turn up, and we'd literally watch their people turn up, hit the hit the application, and then just dump all of their tokens directly into one project, like all of them, and then leave. And so what we saw was that people didn't necessarily trade as much as they just voted in a lot of regards. And one of the other problems, which is a a rolling problem in Futarchy, is this is this problem called decision selection bias. So unlike in prediction markets, when you are trading in the prediction market, you your incentives are almost directly aligned with the truth. So if you predict that something's going to happen 50% of the time, then the maximum you can make is basically 50¢, right? If you buy in at 50¢ and or you let's say you buy two 50¢, you're holding at 50¢ and eventually this thing does happen, you're going to earn 50¢. If you let's say you find a project and you think that they the TVL is wrong, the forecast number is wrong, and you trade that number down, then as you trade the number down, you might move that project out of the top five, which means that it's not going to win, which means it's not going to pay out, which means that your incentive to move the price or change the TVL is actually not it's just zero because you make no money by by moving up that that market. And so this problem affects a few stockies. There are there are ways to mitigate this, but that that basically was was, kind of at play with our markets. And so what you saw is that and remember that people have no cost for these tokens. So they would turn up. They would try and push the project they they wanted to win into the top five. And so what you saw is that the top five projects, the prices just started to inflate, and all of the prices were the TVL forecast were just much, much higher than we expected them to be and many, many times higher than what they ended up being. So, yeah, I mean, also we had a a huge drawdown in in prices during that the three month period, but it was, it was definitely like over exaggerated, I think, once the prices were over inflated. So that was another learning, which is that when you don't have any downside risks, then your the core tenants of Futarchy as a mechanism fall apart. The incentive compatibility property isn't actually obeyed. And so you end up having to reconstruct the incentives. So you have to kind of tell people there is a downside risk of you like just throwing your money at this, which is you're going to lose some reputation in this leaderboard that you may or may not care about. And maybe if you're successful, there'll be like a prize pool, but we can't tell you how much that prize pool is. And so you kind of reconstruct the incentives that like using dollars just gives you for free. Yeah. Now that doesn't mean that you have to always use dollars. There's a very well known, very functional prediction market platform called Manifold where they just use Play Money and it works really well because everybody in, like everybody in there has a model for like rankings and they're all one group and they very much respond very well to social incentives. It's having a big balance of being aware on there is, like, great. Like, this guy knows what he's talking about. Like, he's got, like, a he's spent a long time, in forecasting and he he's got good predictions. And so there is like something that's doing this reinforcement. Whereas I feel like for our, you know, very early system, we we weren't able to obviously emulate that. So that that was one of the downsides. However, the performance was really interesting. Both, there were some projects that just didn't score. They didn't produce any TVR because they were actually, they'd just been miscategorized. They hadn't they weren't able to increase the the TVR. And so we did have a few projects that just had zero, like at the end. So the forecast was huge and then the outcome was zero. And that's maybe because they were just not deployed yet or that they were trying to they were basically increasing something else that they thought was matched to TVL, but isn't included in our definition of TVL. And so these were some of the issues just like being setting the the the models up front and making sure people understand them. And there's another one another issue, which was just the projects received the funding and they signed off on it and they were happy, but they had no world in which they were being monitored by this market. And so even though there were these targets, the targets didn't really apply to them. They said, okay, well, thank you for the funding. We'll maybe deploy that in like a few months' time. And then they just went back to what they were doing. And so that was another part well was that the projects also needed to be bought into the market. They understood what was going on and they felt like they had some kind of like, urgency to deploy and and produce this, this forecast by or produce this target by some particular time. So we learned lots in running that first one, I think, and it was actually very good that it was a play money because it just meant that there was a lot less risk associated with that first one. And so how did that end up informing
the second one with, with Uniswap?
So the first thing is we made sure that we, would deploy the real money market, which is what we did. The second thing we did is we increased, level of comms, onboarding projects, informing them how the whole thing would work, making sure that what they were doing was attributable to the metric, so, you know, making sure these are aligned. And then also make making sure that that each project understood that there was a benefit to, to actually delivering TVL within a certain time period. So they didn't feel like they would, they could just relax. They got the money, they were like, okay, great. That's the grant done. And now we have this money that we can use when we need to. And they're saying there is an entire program of work here which you're included in. And this is this is funny because I think in in I don't think this is a very obvious idea, but that that people don't the recipient of the funding doesn't necessarily have the incentives to actually do anything with it. They're happy to go the funding. Right? But your market mechanism includes them and their actions, and if they have no urgency, they won't actually partake. So that was really important for us. And I think we we we did a much better job of giving making sure everybody understood the parameters of the mechanism.
Interesting. And so I guess with so with the specifics of the uni market, is that one, planned, live, completed? Can you just remind, at this point in kind of mid to late August where we're at with the experiment?
Mid to late August, we, we ran the market, I guess it was, yeah, it was a while ago now. So we ran the market. It ran for it was a one month market. So we had a a five day period of of of trading. We made the decision at the end and then we, we ran the market after the it was early early, July, and then we ran the market after. So the market's now closed. So we have results. We have a dashboard which we will make available to the community so we can see how these things have progressed. Accuracy was a lot better. So we saw trade, we saw prices that reflected reality, they were much, much better. There was some, there was some inflation, but not much, like much, much, much less. And, and obviously one of the main things we did is we had a lot less projects, so from '22 to '4. The other thing we did as well, which helped make the prices more accurate is we used counterfactual markets. So we we didn't just ask what do you think the TVL will be if we give the project this money? We said, also tell us what you think the TVL will be if we don't give the project any money. So what we could observe is the delta between the two, the counterfactual, right, here's what happens if we make the decision and then we could, we could make the decision using that information, which I guess is the original idea of Futarchy. It's supposed to allow you to compare these two worlds where you make a decision and you don't to inform the decision you should make. If, for example, you just gave all of your funds to one person, but they would have produced the thing you wanted regardless, then your funds are completely useless and the intervention is a pointless one. So so yeah. So that's the so Uniswap the Uniswap market is now complete. We now have some results. We'll be publishing those results to the community and, and then kind of explaining where we're gonna go next.
And so, I guess, touching on, you know, how you're now thinking about the future of governance, and especially how you see both butter specifically and Futarchy style models evolve. I guess, what is the landscape of experimentation that you're most excited for in, say, the next six to twelve months to really drive these ideas further, whether purely in that capital allocation bucket of the those kind of four quadrants, or if thinking of, you know, potentially stepping out of capital allocation into other types of governance
point around, like so we do KPI free talking, which means that we take a metric, and we're we're basically forecasting that metric, and we're using that to make a decision. Now, the other type of Futarchy, which is the one that, the Solana team do is, price feeds hockey. So you just take a price and you say, what's the impact on price? Now you can probably see what the benefits of each are. They're probably quite obvious, but one is that when you have something fairly small price, there's lots of things you can do that might impact your price. As you get much larger, then there's lots of things you might want to make a decision about, but none of them will have an impact on price. And so you actually can't make decisions on them. So you end up with this kind of frontier of decision making where you can literally only take decisions at a certain level and anything below it, as you get larger, becomes things you can never decide with your fees hockey, which is kind of fine, but it just means that your fees hockey stays pretty much at the level of, like, board member. Right? It's like a board that you're already, like you're just scaling over time or maybe it's just your your CEO. So what we wanted is to be something that wanted something that was, could better explore the entire decision space, which is why we used, different kind of KPIs or metrics. Now, what's really exciting to us is, what one of the things that you lose out on in having a, TVM market or, like, a metric market is that there's not lots and lots and lots of people who are in those markets already. There's not a big active market of heterogeneous traders who are all trying to like keep that price accurate. And if they see inconsistency, they're trying to correct it. And so what's what we've been, again, excited about is like just increasing the number of information markets in metrics and TV and, yeah, metrics and KPIs because the more of these that we have and the more of these that we trade, the more that people care about making sure these numbers are accurate, the absolute, like, benefit we end up with in our markets because that just means that there's all of these traders who've already done the work to build models around how to improve metrics that matter to protocols, that matter to DAOs. And once you have that, you kind of have something that looks like price. You have something that lots of people will trade, lots of people are using those markets for risk transfer, speculation, whatever it is. And, the fact that you have that means that you can just have much more robust decision markets. So that's the thing that we're really excited about in the short term. And I think if you go to our website now, you can see some of the markets we've launched. We have some TVR markets, which are which are really cool. They're just TVR for different chains, which we're gonna be launching more of, and you'll see some more, KPI markets over the next few, few months that we'll be launching. So that's that's what we're really excited about. We've, we're also gonna be running out more more funding markets. We think funding markets are still, they still have so much promise and I think we're still seeing the, like, we're kind of nearing the kind of maturity arc where they start to become like things that anybody could just press a couple of buttons and roll out, which actually becomes really exciting. Like, for example, let's say there was like a scroll market for, like, some scroll KPIs. You should just be able to click a button and launch a, like a funding market just on any one of those KPIs, and you enjoy the fact that there's already these traders trading this this, this KPI already. So that's what we're trying to, to move towards. And then, the, and then I think in general, like, Futarchy is going through a bit of a, like, of a growth spurt. So we have more Futarchy protocols launching. I think when we launched our last market, we had a really nice Twitter space where there was like three or four few target protocols on it. But what's coming around that is a ton of information markets, a ton of prediction market, projects. Honestly, way too many. Like, it's like hundreds of them launching. So I think that the prediction market space is is going to end up with way more traders, way more liquidity, way more, like, like coverage of the output. So that's gonna be really exciting for us. And I I don't think that we're we're we're I don't think we're ready to move away from just doing funding markets. I think we still love funding markets. We think they hold the most promise. But I do think that there's there are other people doing, like, other types of of, decision markets. So
Yeah. And I guess as we're we're nearing the end, I wanna touch on you know, as you mentioned with the optimism example, there was this, kind of, you know, regulatory adjustment that needed to get made in terms of what can realistically be deployed, and in which jurisdictions. And, yeah, just wanted to kinda get your sense on how you view this kind of tension between, you know, let's call it, say, compliant protocols and sovereign protocols evolving. You know, how do you see these kind of different projects? Are there gonna be some that choose to focus in specific jurisdictions or, others that focus on, you know, finding ways to, maintain regulatory compliance in different ways, you know, as the landscape of projects goes from four to, you know, I'm sure dozens, if not hundreds. Like, what's your own guess in how you see that evolving?
Yeah. I think, crypto has always been in this weird, in between space, as it, on the one hand, fights for legitimacy and on the other hand as it tries to stick to its ideals that it shouldn't be governed by any particular jurisdiction, and it doesn't exist in any jurisdiction. I think in the last, like, year or so, we've seen that that, like, wall that was up has fallen. And I think now we see I mean, I mean, from my perspective, I see a ton of projects where you speak to them and you're speaking to them a lot, and they're they're saying all the same things that you are, and then realize they're not even in crypto. They're just building a very traditional Web two stack thing, but it's doing your crypto thing. It's doing the same thing, but they are, they are firmly within the jurisdiction and they are using that to their advantage. This is the Kalshi versus Polymarket, kind of debate. So Polymarket, obviously, they ended up offshore. They ended up outside of The US. They're boxing lots of territories and jurisdictions and then Cauchy just spent like a ton of time getting a license and licensed in The US and now they can serve all these US customers and PolyMarkets had to go and acquire exchange to get back into The US. And the reality is that The US is still a huge, huge market, and, it's the biggest from a financial markets perspective. So it just pays to be here. Like, if you don't have that market, your is going to really suck. And I but I I just feel like the I I keep saying I think there's an ideological path for it though that seems to be on the back foot at the moment. I think there's a pragmatism to founders, which comes from a pragmatism that's come from investors, which comes from a horrible slap on the on the wrist from markets when, you know, lots of projects launched that didn't actually end up producing returns. Yep. And I think that's carrying through to what we see ourselves doing now. So I think we're getting more pragmatic, and I think we know that we need to be in these markets to capture some of this demand. And so I do think there's this this feed through system and, you know, The US is making some great strides in in in trying to legitimize some of this stuff and give us more avenues to to build all these things in, you know, inside The US. So it's a tension I don't think is resolved. And I think if if we just end up building things that are perfectly licensed, they fit within a regime, we sit sit ourselves inside countries and we do things according to exactly what the rules are in that country. I think we're going to lose some of the, like, innovative frontier of crypto because we are working within rules. The reason why some of these things worked is precisely because the founders built things outside of the rules and it allowed us to explore more of the design space, regardless of whether that's like an arb on regulation or something. It just allowed us to explore new things that we just couldn't explore within this set rules. And that's that is how systems change and evolve. Right? People have to break the rules and flap the rules to find out what's possible that the rules didn't even know they were constraining. So yeah. So, like, I think that's where we are today and I think that's a great place to be. I think it's I think it's it's great that we have some legitimacy, and I think that first wave you know, like, when you, like, settle, like, a block of transactions? So we're settling our first block, and now we need to go out and build the next block of transactions. But, yeah, these ones are stitched in. I think, like, we're going to have more on, like, onshore regulated, trading venues. We're going to have, you know, like, institutions deploying tons of captions, crypto. I think that's great. I think we've done that first layer. Trading venues are locked in. But let's see what else we can build. I think there's more stuff to build that maybe doesn't fit into this paradigm. So
Mhmm. Applied to DAO governance, which decisions do you think are best made through Futarchy and which are really not? Maybe you have any concrete examples.
So if you just think about the mechanism itself, then there are there are lots of decisions that don't require it shouldn't require a Futarchy because a Futarchy just has a cost. And if it and if the impact of that decision is below that cost, there's just no sense in in in in running this. I think that's the first thing is just cost, and I think that's there are lots of decisions that are just cheap, right, they're cheap to make. So an example could be, let's say, we take a a governance protocol. We wouldn't sorry. A governance protocol of a DeFi sorry. Governance of a DeFi protocol, you couldn't imagine that, you might want to just decide whether to add one one asset to the protocol. And, running a market might be so expensive that it's just not worth doing. So maybe you should just add all the markets and then run a market to decide which to remove. Right? You can probably just have a much simpler mechanism. And I think there is a there is a sense that people believe Futaki can solve all problems, and I'm just not convinced that that's true. I think that it solves some problems really well, and we obviously believe that that's funding problems. And the reason being is that, and, actually, it's just really simple. One of the one of the, problems you have with decision making is, as I said before, is what decision is best is subjective in lots of regards and intersubjective at best. And I think that ROI is almost almost completely objective. It's very hard to argue that ROI an increased ROI is bad. Mhmm. And I guess that does make funding, much easier to evaluate with a market. It's like unequivocally better certain decision is made, whereas it's much, it's a bit harder with other decisions. So that's why we think funded decisions work really well, because it's just outcomes are always easy to stack rank. If I think about Ethereum and I think about if it would make sense for Ethereum to say have a vote, or a Futarchy on, like, EIPs, I think that's quite a, it's quite a seductive idea that you would just have markets for all of these. But, some of these decisions are so, so important, and some of them are not easily reducible to whether it increases Ethereum's token price that I think it would be hard to I don't think you would want to have it for all decisions. Or would that would that because I think that you just can't you can't evaluate all of those decisions by the same, logic. And in the same way that when we were running the optimism market and we compared our results to the to the grants council, the grants council, and I think very valid point, they're not really trying to increase TVL in the next three months. That's not really why they're deploying the capsule. But then the team is good. The team should be in the ecosystem. They think they need the support. It will help them derisk the the the the grant or, sorry, derisk the protocol so they can actually do something in the future. Yet there are loads of things that grants council can consider. Futarchy is blind. It's just like, I'm looking at this one thing. You tell me if it increases that thing, and I will I will give you a good answer. That's I I guess that's the difference. So sometimes we want this, like, this, like, one note decision, and that's great. Super, something we can repeat over and over again. Scales really well. But sometimes we do need to have, like, a much more, deliberative decision making system, and I do think that that's that's like there's other mechanisms to do that, but that's not Futaki's job.
This is very refreshing to hear from Peter's CEO that not everything can be solved by Futarchy.
But I take it back. Everything can be solved with Futarchy.
Well done. I suggest we move to our quiz. So we usually ask you questions, and we ask you to answer with just one word. Yep. Take your time. No rush. So the first question is, what is one word you would describe your journey so far?
It's that's very tough. Probably, like, haizen. Does that make sense? It's the it's the Japanese word for, like, CI, for, like, continuous improvement.
Thank you for that. Thank you for, my good little. Yeah. You know, I know it, absolutely does. It doesn't have to make sense to anyone but you.
But also I snuck in two words.
You tend to do that, Mon. You tend to answer Very crafty. Two sentences. Ask for one. But we love Gotta
find a hack.
What do you think would be the buzzword of 2026 in crypto?
'26?
Oh, we I think we're approaching, like, second half of the year in 2025. So let's look a little bit in the future.
L two is not one word, is it?
It could be.
It could be in this context.
Yeah. The hyphenate.
Good hyphenate. Layer two. Yeah. I'd say I'd say everyone's gonna launch layer twos. Layer two two is like that seems to be the move. It but App chain. I'll just call it App chain. Let's go to App chain. App chain? Yeah. Yeah. I mean, that's easier.
Still too odd, isn't it? I don't know. Who do you think listens to our podcast? Come on now. Are we I mean, we would assume people have some familiarity with l twos. I would hope so. But, no, we do update, additional context. Okay. If butter wasn't called butter, what's the one word that you'd want it to be known by? Does it have to be a proper noun? Could be a proper adjective too. It could be anything.
It's a little, twee, but I would probably pick, like, autonomous because it's a word that I used internally a lot. The idea of, like, a like, a cellular automata. Mhmm. It's like the idea of having this, like, very simple thing that you've designed, and then just by some simple rules, it kind of being able to build something almost looks lifelike. And so autonomous is like a play on that, that includes this idea. This is not just one word, obviously. I'm saying now loads of words. But, the idea that you can kind of design something simple, some simple rule set, and it can give us something that approaches like a truly autonomous governance system, which I think actually the governor contract kind of does. Like, it already is kind of part of the way there.
And so, yeah, that's that's why I pick Autonoma. One book that you would recommend everyone to read who thinks about current problems.
With one word titles. No. Just kidding. Demon.
Demon?
Yeah.
By whom?
I forget his his first name. Suarez. But his his name is normally normally backwards. Daniel Suarez. Yeah.
Our last but not least question that we ask every guest, what is the future of governance
in one word? I would probably say an an acronym, and it's not a nice one, but it would be USAF.
USAF as in United States as fuck? No.
No. It's just the United States Air Force.
Right.
Which the the point there being, I feel like, like, governance of protocols would eventually just kinda be backstopped by the US government. Again, it just feels like that's the like, the direction of travel is that ultimately we would just, like, be undergirded by these systems. I do feel like there's, like, a you know, we're kinda giving way to some of these systems that are just, like, incredibly powerful already, these these systems of governance. You know, we're wrapping our DAOs in, like, things that are then, you know, backed up by the government. Right? And I guess that's the that's my current thinking about about governance the future of governance. It feels very much like, okay. We'll eventually back into these systems. They're very secure. Right? It's like like, like, hog tying ourselves to the l one.
Now I want to talk to you for another hour about the politics, economics, policy, government intervention, and its effect on governance and innovation. But we only have that one hour that we spent talking, and I really enjoyed it. And I hope you enjoyed talking to us as much as we did.
No. I really I absolutely loved it. So, yeah, it was a great chat. I think we got to cover some good ground, and I like the high end beginning.
Good stuff. Thank you so much, guys, for inviting me. Thanks for tuning in. The Governance Futures podcast is sponsored by the Skrull Foundation and produced by the governance team at the foundation, Jamila Kamalova and Eugene Leventhal. Any music and photos are attested in the episode description. Feel free to subscribe, leave a review, or share with a friend. Until next time.