The Biology of Token Networks with Thomas Rush

if it doesn't I can trade with you thank you hi everybody thank you for having me really psyched to be here talking about biology of token networks which is something I've been uh focused on for some time um and also just wanted to start right off the bat with an invitation to collaborate if you're deeply interested in the space and mechanism design token design would love to connect with you after afterwards um so just real quick I mean she gave me an intro so I'll preserve this I'm a partnering head of crypto economics at consensus mesh uh focused on all of these things you know design treasury management a number of things uh all related to liquid tokens and then of course making new Investments on behalf of our firm as well as managing our existing portfolio uh which is very sizable uh we have around 175 positions currently and a number of those are from historical around about seven years of investing in the space um if you're not familiar with us we've incubated and invested in in a lot of the most influential uh blockchain projects in the space just tons of phenomenal teams building phenomenal products uh so these are just a smattering of them um and one thing we've noticed over the years is all of those teams often struggle with how to design their token do mechanism design deal with you know monetary policy distributions there's a lot of factors to account for and um it's it's just common across every single project but generally speaking they all fall follow a broad system right you set out a goal uh you understand what you're trying to achieve you consider your constraints uh some of them are internal some of them are external and then of course you design mechanisms around those things to understand how can you achieve that goal so just if you're not familiar with this kind of like very broad framework which which was uh from Eddie lazarin over at a16c um I'll give you the example of of Bitcoin so it's they want to maximize network security uh there's a couple constraints they chose one of you know Satoshi chose one of 21 million exogenous uh constraint of energy costs and obviously this is not comprehensive um and then the main main design being just proof of work uh sorry mechanism being proof of work and you know this is a really elegant uh solution to the problem of trying to secure a network and it's worked obviously incredibly well over the past uh 15 years and so what happens and what do we do when we want to solve more complex tokens you get into a huge number of solutions so here's a ton of Frameworks uh this is just a couple of them that exist for thinking through token design so you may look at your token Network that you're considering developing and you view this landscape uh these use cases or capabilities and you say all right I'm an internal currency but I also want to have a reputational element and you can kind of understand where you are in the landscape of things understand uh what other projects are doing similar things and compare and contrast that's great you can also run simulations use CAD CAD use token spice all of these different modeling tools look at Analytics you can read read papers of course there's just like a huge a body of work that you can pull on um but again the problem is you have to absorb all that information and you have to condense it and digest it and then you just start making decisions and it's frankly not easy um and so here's a slide with there we go uh that brooch is flawed and we know that because we're still iterating on this approach and there's a lot of tools out there but none of them are perfect yet um but I want to highlight today one key element that exists within every single token project that I think is really the core element of how we can move along much more quickly and move the space more is this cutting in and out I don't think okay cool I don't think they're paying attention so um cool so that element that's common to every single project is that they evolve or die and that's the most critical element that I want to like drive home is the fact that every token project does this right and so we can look at an example of liquidity mining uh most people consider compound as kind of like the father of um liquidity mining they kicked off Deep by summer in 2020 there was actually a project called f coin that did it uh early in 2018 and they didn't design it properly and they died so that project you know had watch trading and all of it you know they're I think their trading volume shot up to like 17 billion dollars uh but all of it was just just back and forth like across stable coins and things like that um and ultimately people are just you know scraping the rewards out of the system selling them Etc and so that project died and then you see this continued iteration and evolution of liquidity mining over time to where we are today roughly speaking with you know ve tokens bribes Etc and and there's just ongoing elements and I think the the main thing to call out here in regards to those those Evolutions is that we still need to my point still need more work on this and we need a universal language to document all these experiments that are being run more visibility into the very specific mechanisms the specific traits that are uh that are being passed along from generation to generation and then just doing a better job of isolating the variables across all of these experiments um and that's sounds ambitious and I think it is but it's why it needs to be that's why I'm on the public good stage talking about it because I think it needs to be an open project where many people are contributing uh and luckily we have some lessons and some models that we can lean on so biology is also a very complex system we've we've you know more or less kind of quote-unquote tackled this field on a very in a very rigorous manner in a scientific manner of course so what they did with Biology is this guy Carl Linnaeus came up with a system he uh he you know named all of the all of the species uh Kings play chest on funny green squares in case you remember it from elementary school and now everything is much more structured and and scientists have a number of of different benefits as a result of this system right everything from Charles Darwin to your eighth grade science class uh is effectively built off of this and so there's a universal terminology there's ability to study the interaction between species and ecosystems and all this and then obviously you can track survival of the fittest all of which is very similar to what we're trying to do with token uh with token models and so I just want to highlight the fact that you know really I think one of the biggest elements of this is that we can uh we can better understand the evolutionary lineage of projects over time and doing that if you go back to the liquidity mining example right the biggest opportunity for learning from Project to project is not when a project launches it's not watching a project live with its analytics or whatever it might be in my view it's the it's what happens between those projects and understanding what's changed from one generation to the next right it's it's why scientists choose fruit flies to study uh for a number of reasons but because they reproduce very quickly and you can understand a number of iterations and monitor those very closely you know within a few days time um and I think that's kind of what we need to aim for as a space um so how do we do it uh it's going to be complicated to some degree but you know taking that compound example we have this is quote unquote you know illustrative but the original compound protocol uh everything's baked into a smart contract and what I'm suggesting is that we extract certain mechanisms out into distinct smart contracts uh I love the nodding um and so being able to do that allows for a number of things right so suddenly now when you see just that's very specific liquidity mining contract that's out there on its own that's independent and you're seeing people Fork it you're seeing different versions pop up right kind of the family tree of of what liquidity mining could look like versus a family tree of the compound contract which bakes in so many other factors that it's hard to parse out what's driving the success of uh of Any Given project suddenly you're able to see again that evolutionary tree start to evolve uh and so now you know this is again a rough example but you would see that of version 11 being deployed into convex or convex are deployed into another project um and then we can see that what is the reason that you know liquidity mining 7 through 10 failed but 11 was the winner like what is it that they tweaked and made more successful as a result um and again it's kind of like great great Theory how do we do this like what's the reality I don't want to just propose a bunch of theoretical Solutions uh and so we've started to work on this and collected the data around this and this is kind of where the invitation rate comes from uh currently we're indexing existing mechanisms so got a database of the top 100 projects other notable projects and we're tying the specific token token design mechanisms to those networks so we can understand very distinctly here is the mechanism that's being used here's how it's being used across different protocols at least at least we're aligned on the nomenclature uh of how these these networks are being designed and then the second thing uh beginning to tie different correlations or different success metrics to each mechanism so using again the the is again illustrative but the liquidity mining example you have uh inflationary rewards you know compound has a tvl of just under 2 billion and then Ave does a fee share they have a tvl of four and a half billion so you can now you can compare two mechanisms side by side seeing what one is obviously achieving what you would imagine the goal of these two protocols to be which is TBL and locking up value and you know serving a larger basically market share um and then you do this and you do it not just for two projects but you do it for a hundreds right over time and ideally statistically you're able to see the correlations with which mechanisms are more successful um and which versions are more successful going back to that that evolutionary tree and there's some side benefits as well as a result you make mechanisms composable so by pulling them out into distinct contracts you now have the ability to uh I think I have it in here to write your saving development time because you're just working a specific mechanism you can improve security very similar to the open Zeppelin uh ERC ERC standards that they published there were many eyes on a single implementation type of thing uh it has the possibility of increasing resiliency there's there's obviously a trade-off with upgradability and some security um you know security considerations there but being able to swap out mechanisms right you can imagine a dow voting to upgrade their protocol it's not a ton of potentially coding and development time you could potentially just plug and play a new mechanism that serves a similar purpose but has different attributes or different traits uh and then the biggest one I think honestly is interoperability and understanding how how multiple mechanisms can interact with each other um and so this is just an illustration of of you know how scientists view organisms and I think this is kind of a good North Star for us to get to in terms of tokens right it's a it's a very distinct uh view of the species across the globe you see rates of the species birth you can pack all this data in there um and we don't have this for token networks despite the fact that all that data is available and on chain we don't have a great view from the bird's eye level of What mechanisms are being used which ones are being most are most successful how often are they used across different protocols how often are they being replicated or forked um all of this will be in theory possible but we need to start somewhere which is again why I was proposing those first few steps um and luckily just to kind of summarize there is some momentum building around this right so you have an ERC standard that uh that was proposed I don't think it's been accepted yet but this is very similar process uh that's that Nick Mudge proposed in terms of like isolating different elements of a smart contract and pulling it out so that multi-facet proxy is what he calls it it's similar in concept uh similar in execution just a different kind of origin behind the thinking block science labs is doing some phenomenal work around their modeling uh six-man Ventures just put out a study that's really interesting around mechanism design and really they were working to try to isolate very specifically which elements of of different token models you know Drive different outcomes so there's there's a lot of work being driven around it and I I think what I want to propose and kind of drive home here is building a framework where we can collect all of that formation and ultimately generate the insights out of it so that it's not just a medium post or a mirror post um you know that you need to then absorb yourself and and rinse and repeat you know just back from the beginning of the presentation um so that's it I mean what's next in terms of how we can put this into action uh would love contributions to that that database if I do it myself it's not going to be worth anything uh because it's just gonna be somebody's opinion but if everyone contributes it's going to be I think massively valuable uh and of course please feel free to reach out I'm on Twitter telegram all of the platforms uh no one has an extra underscore and um would love to hear from you all and and thank you so so much for listening really appreciate it