MEV for Public Goods Funding with Nikete Della Penna

Transcript

foreign [Music] ation on the Pod today I've got the cat Della henna who is a formerly trained mechanism designer with a PhD on the intersections of economics in machine learning the cat I was really excited to have on the podcast because he's the most advanced thinker in the space exploring how Mev could be made of public good so basically on this part we we cover what is Mev we talk about why it's a powerful mechanism to support the public good and how can we use Mev in suede away from financialized and Monopoly or red seeking interests in the ecosystem and redirecting some of that

to the public so an interesting pod to learn about Mev and how it could be used to support public goods and um I just like the cat's just such a Galaxy brain and so I think this is a really fun episode because of of that I think we go deep in this episode on how Meb could be used to fund public goods indicate is one of the most advanced thinkers in that Mev I think is a multi-million if not billion dollar opportunity to support the public good in the future as we talk about the trade-offs of what would it look like if there was an Mev burn at the flare one or what would it look like if there was Mev supporting the public goods with on layer twos and layer ones and the

trade-offs inherit within that so I think it's a an interesting conversation and like I said naked is just really an advanced thinker and so you're going to enjoy this episode coordination ballet Dao is an open Global Collective working to finance and democratize the governance of synthetic biology Technologies to protect the future of our planet research and progress into synthetic biology is critically constrained by the world of nation-state regulations and bureaucracy from middlemen but the potential that synthetic biology has for the world is massive it can produce a world in total harmony with nature we can grow stuff using cells like software it's awesome and the best part is we no longer have to pull carbon out of the ground to grow and maintain civilization don't you think we should be funding this Valley

Dow does and that's what they're here to do to try and make the solar Punk World of Tomorrow a reality today so join Valley Dao right now and save the planet at valleydow.bio what's up Nick how you doing hello Kevin thank you for having me yeah really excited to get into your research so maybe we could just see you off of the softball what is what is Meb Mev uh was originally defined uh in a paper by Diane at all as the minor extractable value so when you have a distributed uh system that relies on miners for uh creating and propagating blocks those miners are in a privileged position in that they can see and they can censor and they can insert based on

what they saw and what they censored uh transactions and they can reorder transactions around with other people um in in deciding how they build a block so there's a certain amount of value that is extractable by the nature of being in this privileged position uh so this is the original definition of of Mev and by and large this will be the definition of Mev appropriately generalized to to validators um that I will be using in this stock some people then took this in a very maximalist Direction where they Define Mev as the maximal extractable value that definition is a little bit trickier to work on on because this kind of maximal Mev can depend on your

privileged access to information outside the domain that you are ordering or reordering um and so it is more difficult in in protocol design to make guarantees about this more maximalist view of Mev so in in terms of informing policy or informing the decisions that the networks take that one is is there's maybe less positive results or or there's less Clarity still on these more maximal decisions of what they mean it's a much more active research area the other one we we understand a lot better ways my example that I always like to give of Mev is a sandwich attack where user x uh wants to buy an asset at at Price Y and uh user Z wants to uh wants

to sell it at uh y minus two and so there's a sandwich attack inserted by the the miners who can see the M Pool sorry the mempool um or anyone who can see that member what's a transaction in between them then they bundle the transactions together and then they take the diff the difference between the the two prices there so uh that's a sandwich attack and that's sort of an example of the value that's created by having access to the mempool and uh I guess an example of Meb although maybe a truly trivially uh simple example of Meb so how much value is there right now on the ethereum layer one associated with Meb this is a very

hard number uh to to precisely articulate but so so a very naive view would be what's the slippage tolerance of all uniswap orders that pass through the mempool so that's like the bare minimum amount of Mev so basically when that because of the any amount of slippers that you allow is is creating pure Mev for someone that front and backs running but backgrounds you so my slippage is just the difference between Y and Y minus two in my example yeah yeah so if you when when an interface to UNI subcontracts you're allowed to say okay I'm trying to get this price plus minus something um and so the miner can always guarantee that that you execute right at the

boundary of what that tolerance is and so they're able to extract all the difference between the the current price of the market maker and the tolerance that you had well now that we've established a baseline understanding of Mev I wonder if you can block us down to the depth of Mev research and uh you know what are the frontiers of deciding how to how to design a network in order to either maximize uh minimize Mev uh or maximize the amount of funding they could go to public goods or maximize the amount of value that can be captured back to a network by minimizing Med uh it feels like this is a really Advanced Frontier and I'm and I'm curious if you can if we can Splunk down to to some of

these Advanced ideas and what you spend your day-to-day on now the first instinct is to try to minimize this Mev because it looks like it's bad right it's value that's extractable by intermediaries the subtlety that needs to be taken into account is that you you can trivially minimize Meb by having no trade um right so you can make the transaction fees Infinity no one trades there's no value to extract so so there's there's this question of you want to limit how much value the intermediaries can extract while still providing them with enough incentives um to carry out their their intermediation in the network um and so mevs is is one of the ways in in fact at the moment the the main way

um that that many of these intermediaries for right Searchers that that bind liquidity between different domains um into each other so so there are many things that that increase the welfare or increase the utility of users of the network that also generate some Mev um and and that Mev is in in a deeper sense um a part of of any permission on the system so any permission the system must allow the people that make the effort to to update the state of the system to extract some value because otherwise there's no motivation uh to update the state so the the frontier of Mev research is about

um trying to understand what kind of gains do we want people to to be playing in terms of this of this value extraction um that that allow us to create a lot of welfare uh and a lot of that has to do with with having some amount of competition between providers of services to the network now efficiency wants us not to have any competition right we like for efficiency we wouldn't want to compute anything more than once um for robustness and incentive compatibility and decentralization we would like to have maximum um duplication so we would like everyone to compute everything um and and a lot of Mev research is in in high speed systems

um or and the latest systems this is it's very difficult to have everyone compute everything you you pay huge performance penalties uh in systems with with a very centralized actor it's very hard to check that they're not extracting too much it's very hard to constraint constraint uh the amount of extraction that they're able to carry out um and so a lot of a lot of current research is trying to understand what are the problems where we can constrain the actors that are extracting value as intermediaries um in ways that still allow uh for for welfare um enhancing measures um so for example um so in in different networks people

use different non-meclature um so in ethereum it used to be Miners and now it's either block Builders or block proposers or validators depending at what level obstruction you're thinking of um some of the some of the more emergent networks use the notion of solvers so in the sense that you know users of the network have some some intention and and the solvers are helping solve the problem of matching the intentions on on both sides to each other um so that's their intermediation role that they're that they're solving foreign so we're trying to create incentives uh for them to solve well a lot of the boundary right now which is where where the stars you get fairly spicy fairly fast is which

applications should be privileged um so if you if you're privileging applications that look a lot like equities trading um maybe like a very unregulated equities Market um you're going to you're going and end those same assets can also be traded in centralized exchanges you're a very difficult problem you've set up a very difficult problem for yourself because you're an inherent disadvantage in terms of the amount of computational effort that it's going to be required and and the communication complexity of that effort that's going to be required in the decentralized versus the centralized setting right so you you're never going to be able to run

an order book with the same latency as like a very large centralized exchange that shall remain nameless unless you have your your your your cluster of compute as centralized as they do right because a lot of that is is just going to be you know if if if it's mostly microprocessor through L2 cache of the microprocessor on the same die latency that's determining how quickly um you can run those order books which is in fact the case um because the order book is a relatively small object and you care a lot about it so you can basically fit it entirely in the in the cache of the of the processor it's going to be very hard for the centralized system to compete directly there um so a lot of it has to do with what

kind of specifics of the application like require that the centralization or the decidualization provides real leverage and um and leveraging those um for for for finding ways of constraining the the mediation oh wow I I feel like uh I touched too many things it's it yeah no it's it's actually kind of amazing like it I'll be out like with my wife my wife's friends and they'll like be talking about white and all the different flavors and like the ways they relate to each other and um and I'm such like a simpleton that I I just hear like oh yeah it's like it's red wine it tastes like a little bit sour that's like that's how I feel listening to you talk about all the

different dynamics of Mev and you really spelunked us down there and I'm gonna try to bring us back up to to a more relatable version of that and let me try to summarize what you said um so there's there's money to be made by by having a privileged spot in the system where you can execute transactions um insert transactions in between other transactions and like the old Wall Street way of of doing this would just be to be close to the data center to be close to the compute and that's created this whole market around Wall Street where people will try to rent high-speed fiber optic cable and and like high fast compute close to Wall Street so they can inject trades into the order books

um and and that's sort of analogous to what you can do with a centralized exchange but now because we have a decentralized ecosystem of Builders and Searchers and solvers then the protocol kind of dictates the rules of who can be advantaged in this game and so a lot of what I heard in in sort of like the beautiful um description of of the of the of the line that we're cooking up here is how do we decide what is even the values of the ethereum network or different ethereum networks to uh and who do we want to have these privilege position and privileged positions and you know should the privileged positions uh but should the money the value that's

created there go back to the protocol or should it go to public goods there's kind of differing values about how we should uh build the protocol so let me I just tried to play that back in in more relatable terms for the audience so I don't know if I got close at all yeah yeah I think I think you go close I think there's there's a couple of things that are worth emphasizing one of them is if you want to design a permission on the system you're going to have some amount of value that is extracted by by the mediators so so this is this is like an inescapable property so it this is this is not the case necessarily in in a very centralized system you know you could you could use kind of in in Mev they're usually called out of the main

um so you could use you know legal contractual constraints um to try to reduce it much more aggressively in the centralized domain so the centralized domains they because they have these other ways of enforcing things onto their mediators they're they're playing a more expressive game so so you're already at a disadvantage in terms of in the decentralized permission on the setting you can only ask people to do what you can verify verify while staying decentralized in the centralized setting they can verify you know I mean admittedly that the law yearly cost of doing So eventually get get a bit expensive but but you can you can verify much more right because you can go out of the domain um to verify what's inside the domain so

by having recourse to courts and and and so forth so I think that like this is the first level of difficulty the second level of difficulty is that like the physical separation of those CPUs and that memory or also puts you at a disadvantage so so these are these are true different sources of this advantage that you have when you're trying to design um for the decentralized setting these are these are two sets of constraints that in in the centralized like legally compellable setting you can you can also use contracts you can also people so for example on Wall Street people don't even compete just with having the fast lines the the marketplaces themselves can have an

auction where they sell the right to be collocated in the same rack as the machine that's actually solving the market problem um oh wow so you know like those data centers like can charge much more rent and so the the exchange can can choose to to split the revenue the extra revenue from from from collocation there um with its participants these are all things that you can do via contracts um that that reduce the amount of rents that those mediators can extract um right because it forces them to compete with each other for the rent and so you're always trying to it's much better to have them compete with each other for the rent and then the rent money can go to the market that can figure out how to distribute it and

having compete with each other in like laying slightly more straight um Fiber Optic Cables so that they can beat the other guy by like a tenth of a millisecond um which is very expensive and and you spend the money on bulldozers and holes in the ground and it kind of doesn't really doesn't really do much for welfare there's no like further step that you can do it so so these are the two areas of difficulty yeah and if I recall uh I remember Carl Flores back in like 2009 talking about Mev auction uh uh auctioning off the rights to order transactions at a block and then taking those funds and distributing them to public goods yeah there's sort of an analogous mechanism there in India right now we are we are about to to have this

debate in in ethereum so so this will be extremely relevant once again how timing yes perfectly timely but what are the sides of this debate that you see unfolding and and you know what's at play here so there's there's a desire to start capturing in the protocol um the value that can be extracted from from a block or from a series of blocks um so exactly in what way you capture you capture that value is is quite subtle and and there's many ways of doing it but let's say let's let's black box that part of the conversation for now that gets very technical very fast um and and let's say okay you have some some kind of option some some ethereum version of the of the analogous option

of the of the of the data centers in the in the solver rack um and so now people that can extract Mev have to compete with each other for for having this slot and the way that they compete is by by by paying for the right to do so um and now there's a question of what you do with the source of Revenue um so in one five five nine there there have been uh some proposals or at least like the idea have been floated that um it could be used to pay for for public goods and there was a counter argument that this would lead to Too Much governance in some sense that that how to divide that tie could then uh be a source of corruption in the of the system or or a

difficult you know it would add a lot of governance uh surface uh to the system and there's this desire to to minimize governance uh to make the system more Economist and more resilient you can definitely I mean Bitcoin is kind of the the living example that you can take this too far um so you can definitely fossilize your protocol to the point where it's it's not actually viable for for any of its of its proposed uses um and and were in the long term you're you're really like committing yourself to to a strategy that doesn't doesn't have a lot of good exits um and and leaves a lot of circles on the table right so so for example it yeah it's almost Unthinkable how to change proof of work even through its very you know once you've like secured

like hundreds of billions of dollars of other value without burning the electricity you're like okay but you really like it's like almost like they become like ossified to the level of like the religious beliefs like they become like a part of what the object itself is and and they cannot be changed anymore and and that can also be very bad in the long term for the object there is a reason why there is not a thriving angle system of smart contracts in Bitcoin um and why Bitcoin holders do not occur the the fees that instead go to burning allergic on the other hand I think that that now we're about to have this discussion again with Mev that that originated in 1559 I think with time the the arguments against some kind of

redistribution and again some kind of uh way of measuring of guiding those flows to to things that generate uh positive externalities for the network itself of helping internalize some of the some of the benefits that that people bring to the whole network uh become weaker and weaker in in time because what happens if they don't become weaker and recurring time is is that the network becomes more and more about serving current ethereum holders right so because burning is equivalent to proportionally Distributing two ethereum holders yeah deflationary yeah yeah so so the more so this this both both against the centralization because

it's an it it it concentrates wealth more and more um and it also um goes against the future health of the network because it means that the network becomes more and more a value extraction engine for for ethereum holders instead of an engine or you know a a commitment device that that that that provides welfare um by by acting in this credible permissionless manner um to to people that want our or computers that want to enter contracts with each other so there is you know how how well for maximizing do you make the permissionless Contracting platform versus how much do you reward the people that provided the initial capital for the creation of of that platform and so if we if if every time that there's a

choice to allocate value in the network it goes to becoming the ball you know then then this becomes itself its own anchored expectation right we're next time this happens it's like oh yeah you have to burn because that's what you do in ethereum you always burn and so it becomes harder and harder in time to have this more positive vision of what the network is trying to achieve let me try to play that back so there's this ability to uh take the Mev and to allow the protocol to capture some of that value instead of just people who have Searchers and Builders and bundlers and the question is what do we do what do we the colloquial the the The Wider we do with that what does the protocol

do with it does it burn it and therefore create a deflationary effect for all Eep holders um or does it go to somewhere else that uh maybe funds public goods or some other positive externalities and there's two camps um well so I'll say that now that ethereum has this L2 based architecture that every L2 is going to make its own decision about what to do with this um and and I think that there's and there's also the the question of the of the L1 and it feels like there's there's one camp that says we want to reward each holders and we want to create a deflationary deflationary kind of effect for for each holders and we also don't

want to make the protocol credibly neutral or we break the protocols credible neutrality by asking which public goods does this go to ethereum itself is a public good and therefore it we should just burn it and that's a reward for all the holders and then there's maybe another Camp that's more welfare maximizing and wants to use it for other public goods Beyond just the ethereum one and um and and it isn't as concerned about the credible neutrality issues there no I I think I think that there's there's there's a more sophisticated and I think there's there's a more sophisticated view of of the second Camp I'm sure there's a more sophisticated view of the first one but I'm probably not the right guest for it

uh so okay the the the public goods that we found can be entirely about ethereum right so so it's perfectly fine just say look we're gonna fund like people that bring new forms of transactions to the network or going to fund people that bring more volume to the network we're going to fund people that bring you know more different people to the network more more different sets of validators people that improve the efficiency of the software for the validators for the for the users all of these are ethereum public goods they all also benefit each holders indirectly by raising the demand for for for yeah or block space and therefore like raising the value of it it's just a

different channel for which so one of them is this very direct channel of like you burn and you're basically Distributing it to take holder directly and the other one is like okay you invested in the ecosystem and they told us like are already aligned with the ecosystem by by virtual being the token um that is used for validation staking so one of the things I think is really important here is how verifiable is the value back to the ethereum network because you know the public good the the the uh that miners used to provide and people who run saving box provide is cryptographically verifiable at securing the network and it's verifiable cryptographically whereas if you were to fund like open source software development well I don't know if this feature that I developed is going to be valuable and even when I pushed to production I don't even know if anyone's

going to use it and so there's this spectrum of verifiability and then there's another spectrum of is it indisputably a public good for ethereum or or is it not and I'm just remembering the most recent Bitcoin round controversy where there were certain people who said oh we really care about diversity Equity inclusion and there was an opposite camp that said no we don't want this Americanized version of plurality uh baked into our public goods protocol so it feels like the public goods that you were advocating for are a indisputably good for the ethereum network and B kind of cryptographically verifiable is is that a fairer summary uh I don't know if they're all cryptographic verifiability is a very high standard

um so so I would say Okay I I would be I would Advocate at this point for for some kind of like exposed evaluation that has good properties but but maybe it's not quite a it's not quite like a cryptographic property um so you know it it it you have to be careful about you know because in honest majority assumption um is it is it really cryptographic or is it just an honest majority assumption and then you're having some kind of consensus on on where the value was in the network um the most obvious place is the gas fees um so right now people that deploy contracts to the network that provides huge value um receive nothing but legal liability

and the cost of deploying the the contract so so we we do the opposite of incentivizing and to some extent you know contracts that people paid a lot of gas for using its there's the problem of people kind of like washing you know wash trading their own contracts um but as long as you're giving less than the amount of gas that was spent on the contract um that's much less of a concern right because it's it's so so there's various ideas for for gas rebates which which work is Maybe let's say let's call them the the gentlest form of public goods funding where there it becomes really evident that at the moment these must be outrageously

under provisioned relatively efficient Frontier right like there's the if I don't pay for something then I I have to expect that and and there's no way for the person building it to to benefit from it because it's a pure probably good so so any contract that doesn't charge fees like right now this creates wealth for for validators because it increases the demand for Block space it creates wealth for for each holders it creates wealth for users agrees well for everybody except the guy that deploys the contract and and perversely the guy could applies the contract is the legally liable party for the behavior of the contract so not only is there and and many of many of the many such contracts have right for them to be efficient they kind of have to be

like Fearless so there's there's a real desire to have contracts that that mediate and don't take a chunk right so for example the grab the rat is contract you like you know everybody wants to coordinate on one rep contract it doesn't really matter which wrap this implementation it is as long as it's correct but it can't possibly be one that's charging fees so so right now this very basic public good of the network yeah there is no good way of funding it in protocol so you have to rely on being able to to play some like social social economic game to like bundle it with something else or to ask for grants in the appropriate way or so so finding like for the protocol to find ways of permissions funding

these protocol Goods I think is is a worthwhile experiment to do you know not dedicating any of the extractable value of the network to this yeah totally so so this is the question you know the question is right now we're at this like total extreme where like everything gets burned and something new comes along and everything wants to get burned again and it's like maybe you know my my first paper that that mentions ethereum like I started it in 2014 so it's about to have its like decade anniversary so you know maybe after a decade we can afford like some amount of experimentation how we pay for the stuff there's two directions where this um makes a lot of sense um going forward to to experiment and

and one of them is this like pure public good type contract that doesn't charge any fees and promotes a lot of use and the other one is is trying to find ways of of protocolizing um incentives for for client developers so so anyone who's who's taken an economics of the Soviet Union class and and and and sits through like an all developer call will find it strikingly familiar um so it like current ethereum development like client development looks a lot like a Five-Year Plan um and you know there's like the sub goals of the plan and there's how many tons of steel I mean how much block space we need uh and you know is it one giant nail you know oh it's just blobs and don't

worry about what's in the blobs and it's like well there's probably like some pretty severe limits to to not worrying what's in the blobs right so we probably need so I think that it's it's a good time for the network to start experimenting with with with with being um I guess more permissionless in this other aspects of its of its development that that currently have been the exact opposite of permissionless right there's been like extremely centered on very specific individuals well there's one piece of information that I want to unveil to listeners and to you and that's that uh I was the author of something called vip1890 about three or four years ago at this point and it was uh block rewards funding for public goods and um I think

it was called a sustainable commitment to ecosystem funding and it and just started to think about what if we took some of the block rewards and directed them to public goods that even specified which public goods and the blowback to that was pretty hard because people didn't want to sacrifice the credible neutrality uh of ethereum and you know I'm the public goods guy at this point like this is before any like regen stuff and um you know I'm wondering how much of that debate we're going to revisit and how much did the 1559 debate we're going to revisit with Meb stuff and one of the areas where it feels like there's wiggle room to me in this is what l2s do with their 1559 piece and their Mev and and I wonder if there's sort of like a a model where the l2s are where we

experiment we around and we find out um and then the L1 is is really where there's the most fees and the most ossified protocol and and you know I I'm curious what we can learn from um in the 1890 debates and then also the 1559 debates but also I wonder how if you think the Bell 2's might be a little bit of wiggle room here and and which which exact Camp you find yourself in so we're gonna plug down another Rabbit Hole so I I'm not sure that in proof of stake there are really L choose um so I mean you can anchor your trust assumptions to a specific Network and that Network can can anchor their trust assumptions to you

um so this is something that is is much more obvious in the cosmo Cycle System um in the way that they do Bridges um ethereum has had this kind of like almost Microsoft esque desire not to not to provide interoperability in the places where it's currently dominant um so you know there's there's people had to really um I mean people eventually pulled it off I I know so sync Labs um has a two-sided Bridge um I think there's there's a couple of others um coming up online but um but the amount of of technical wizard that had to be deployed to do these

things in in the ZK way without like any updates to ethereum was really rather unnecessary you know ethereum could have chosen to provide the facility you know to to put more effort into providing the facilities to make verification of the of the of this of the state of the chain from from a different chain um more more straightforward and many other chains did it um and I think that you know it's hopeful in in that eventually it gets resolved through it's interesting that it doesn't get resolved through updates to the client it gets resolved through cryptographic wizardly um and and the ZK people like going through so so it might be the case that I have a hard I have a hard time calling current ethereum incredibly neutral

right so so the bus number of the all developers call is probably not even double digit if if tomorrow there was like an anti-protocol Guild so the protocol Guild is all about providing funding to improve the lives of protocol Guild members and the portable build members are Anonymous which address corresponds to which um you could imagine an anti-particle will Guild contract getting deployed to ethereum and I would be in favor of of of censoring such a contract I believe that that the Dow hack provides a good a good roadmap as to why that contract should in fact be censored um you know and it will be a contract that would provide money for for harming the members of the protocol Guild and and providing negative incentives for

for for the ethereum network you know every every competitor that is not sufficiently complementary to ethereum of ethereum would be interested in funding such a contract right so you you can I it's it's hard for me to believe that that people in the protocol Guild that have these like total credible neutrality we will never censor absolutely any transaction I'm not so sure they've like fully thought through um what what that game tree uh looks like what like inverting the signs of some of the contracts that we've already deployed would imply about like the the kind of behaviors that you'd be incentivizing so I think it's people that go early and people that contribute a lot to the public good of a community

we want to have social norms that that reward and protect those people so I think it's it's always I always go back to the dial hack because it's like okay then why aren't you deploying ethereum classic right so in in India in the standard go client there's on under the misc there's something called Dao dot go and that's not like doing for making a dial that's like a state transition that is different from all the other state transitions that we observe that says like Oh Let's do an basically a retractive airdrop um that cancels out exactly the Dao right because the people developing the Dao were providing a huge public good to every to every dial that comes after

right because they're they're paying the cost of innovation and it doesn't make any sense for them to like be the only ones that suffer that cost and most of us choose to to deploy and develop on on what we call the main Network which has already said that it's not incredibly unusual right like there is ethereum classic like you don't have to participate on the on that chain right yeah I mean I think in some ways the the Dow hack was interesting because it was early in a I I think in some ways it wasn't a precedented because ip9 999 funds recovery of the parody wallet where there was uh I think it was tens of millions of dollars in the parody multisig that got accidentally killed it's a much smaller share of the

ecosystem yeah and so you know I I think interesting it's debatable whether the Dow hack is is uh is precedent setting or not but um you know I think you you bring up some interesting questions I agree that it's it's it's arguable whether whether it whether it is or not my argument is a normative one that it should be but like we shouldn't forget that we're already like choosing to participate in the not totally credibly neutral substrate so so you know tens of millions of dollars when the network was 300 million dollars wasn't enough but like a few million dollars when it was a third of the value on the network was enough or sorry I think it was 100 million dollars when the value of the network was 300. um so when it was a third of the value of the network it was

enough I I think you know if it was if it was a contract that like you know incredibly threatened the well-being of of of like most people in the protocol Guild it would it would be totally correct to censor it and and and to fight it in in every way um it you know and and it would make sense for for the network not to want to lose its developers and in some sense to to step in to to protect them um so it's not so clear to me why like if you know if you know so some amount of of none of of non-neutrality is required like you have to favor it's like a cell right like there has to be some inside outside separation like like the idea that you're going to be credibly neutral to the Insiders and The

Outsiders is itself not very belief right is itself not credible it's just not believable that yeah it's an interesting argument um I'll be curious to see how things unfold over the next um over the next several months and I do not see myself entering The Fray Over one side or another and maybe that's just my my burns from the 1890 debates back in the day but I just want to create a channel for for people to have the debate and hey both options are great uh Mev going back to youth holders in the form of deflationary burn and also going to public goods is another is another great direction for it to go and and across all the l2s and I want to I'll be curious to see where it goes and I'm wondering if if we could uh rabbit hole and collective intelligence for for

a few minutes here and basically markets as a form of collective intelligence so could you tell me about your ideas ideas there oh the ideas aren't mine many other people have have very similar ideas um there's one view that sees markets as as basically 500 years ago like double entry bookkeeping and Banks and so forth um that were connected to each other through networks of debts and and repayments um establish some kind of global computational substrate um over which like Global goods allocations were suddenly computed um there's much more evidence once people start putting like Telegraph lines across the ocean and using them almost only to transmit prices in those

markets um so this is equivalent to having like like a neural network that tells people what they can buy and what they can't buy um and and that sounds very dystopian in a way um but but current money is just limiting yourself to neural networks that can only keep one weight per person which is basically their their net worth and so if if what you want up if your net worth minus the cost of what you want to buy is is above zero you're allowed to have it and and if not you're not and and that's how we allocate Goods globally and and there's some adjustment of the prices that goes on yeah in in different markets in different ways that kind of uh work off each other these Global networks that allow for richer things that than than pen and paper like

double entry bookkeeping did they allow for much richer uh mechanisms so potentially mechanisms that don't rely on on having a single number per people but that we can have kind of richer lives if we have like more multi-dimensional uh representations of people's in in these mechanisms um that are able to to match them based not only on on how much money they have but on on other properties that that might make them better matches or Worse matches or generate more so um so in in the long term I find this is part of my modulation for for studying the systems comes because they they present a very a very interesting potential such such substrate they they they give us a way of making this kind

of thing real and and I guess a part of my a part of my desire to I'm not saying that they give everything to public goods you know maybe at the start we only give one percent of public goods and 99 we burn and then we let the people that have to burn decide if they want to donate a little bit more ah so you know it it doesn't have to be an all or nothing kind of thing but I think that it would it would be a Pity um and we would give up a lot of potential um Surplus generating applications that that are also good for the people that care about the financial applications because they help legitimize the system so it's going to be very hard to legitimize the system um if if all the system is useful for his regulatory Arbitrage of existing

institutions like that leaves you in a place you know it's like a fiscal Paradise like okay yeah it's hard to make good treaties or it's hard to have a lot of allies in your Coalition it's the only value that you're doing is this like very unregulated kind of extraction so I think that being the ecosystem that has these other um Surplus producing um activities also makes it sort of politically or more robust let's call it it provides uh there are like practical reasons so I was trained as an economist which probably shows um and that I'm always trying to appeal to people's green you know there's a couple threads that I'm just really excited about and what you said this idea of Market size collective intelligence and

um you know going from telegraphs to computational markets to ethereum means that we have a richer substrate for building a more plural dimension of of of what value is you know as opposed to everything collapsing down to the US dollar maybe we now have local currencies that um you know our bioregional regeneration currencies or for some people diversity Equity inclusive for some people it's open source software and now we have different ways of expressing that and I like the idea of web3 exporting positive externalities to the world um either just by having a better richer substrate that's one public good but also this idea of exporting public goods uh monetarily in some way to to all these

things I think it you're you're right about it being a good in itself but it's also it's a great look it's a great look for the web3 ecosystem to be exporting public goods to the world and that's what excites me about what you just said if the argument is convincing to the people that only care about greed it's like that much more convincing to everybody else um so so there is so I guess this is why this focus on like look you you have to do like you are at the stage of the building of the movement where you start having to do more Coalition building um in the wider World um or so you know that both involves like in the future maybe being a bit more careful about like how how closely we associate ourselves with with people that that claim to obtain returns that

are not really feasible um because we know how that's gonna end and you know every every time that one of those shocks comes along um so for example there there there's a conference that that studies uh computational economics so in a way like the natural place for for very sophisticated uh analysis of the theorem ecosystem and and at this point they're not accepting like any paper on on with a blockchain or on defy at all because you know the blowback like oh no well I mean I I have some empathy I mean I I don't think it's the right decision but but I can see how we but I can see how how it ended up happening so so I think you know in the future we need

this more like robust political coalitions um and I and I think that those large will depend on on being able to bring something um that that money I think that what you said was interesting and that you were thinking of like local currencies and and different forms of equity and these are still like very like uni-dimensional in some sense I mean there are different dimensions but they're still very separable so we might even think of of much richer mechanisms that really think about like every interaction as like a like a an opportunity for for co-consumption and co-creation um and and many of the other values of the people involved in in those transactions are coming through so kind of the opposite of financialization

um bizarrely enough because of having a richer substrate for transactions so so you make more things for those actions but they become less money like uh by by nature of having more expressive substrates it's really beautiful uh venkatesh Rao gave this talk at Devcon this year about how Humanity advances by uh by creating a richer substrate uh for which that we can coordinate and create positive some games without having to think about it so like all the way from on the Savannah uh a handshake me reaching out to you and shaking your hand is a way of showing that I don't have a weapon and therefore starting a positive some engagement to you all the way up to uh democracy to quadratic funding and and the idea is that if there's this thing

called Whitehead Whitehead advances where basically the more computational substrate you have and protocols that can take advantage of that computational subject that's how Humanity advances and in a way I think that that jives very well with what you're saying about uh markets in ethereum as a substrate for collective intelligence and that being an important place to advance Humanity yeah so I'm wondering if you have any comments on that oh yeah so I can I can plug it there's a there's a great talk by uh by Jason Martin a few days ago on the on the summer of protocols um talks which uh venkat is organizing on money um where it was it was it was it was lovely for me to hear it because he expressed a lot of these ideas of like

okay look you already have the like one parametal neural network um like it sounds bad but that's already like the current state of the art like it's not like you're gonna go into some dystopian something thing like you live in the dystopian sounding thing um you've just become accustomed to it um I think that this this sequence of engagements um I I like to place them at the at the very beginning in in the first time that that a human shares fire with someone else um so the domestication of fire like allows us to have much more refined uh vocal cords and to communicate with with more subtle sounds and and sort of expand their vocabulary um to to its richness and also it allows us to to be like this like long distance like Mass Hunters that like you know

create fire and reshape the environment over like Continental scales um so I I think it goes It goes back through to to what being human is at the deepest possible sense and and the Very technology that we start with which is which is fire and and yeah being endurance Hunters that walk on true leg which is a very energetically inefficient way of moving um already shins that that we must be doing some kind of crazy coordination what a beautiful place to end I mean it makes you so optimistic about the future of this technology and and and what it can do for Humanity if we just choose to coordinate on it well I think that we're about over on time but I'm wondering uh one last question is there anything that I didn't ask you want to say no I think I think we're good I think we

we covered a lot of grounds well uh I'm gonna have to have you back as the debates about Mev and what to do with it advancing the ecosystem and I think that you're just a really rich thinker as a mechanism designer and your background in economics and machine learning and um I'm I'm just really excited to continue the conversation and thank you so much for appearing on on the Greek podcast thank you so much for for having me Kevin peace and move foreign [Music]