Fireside Chat I Kain Warwick, Synthetix, and Kartik Talwar, ETHGlobal | Pragma New York 2023

foreign [Music] so for our next interview want to bring on Kane who you may all know as one of the co-founders of synthetics and if you were early enough as coventors of Haven we're going to talk about sort of how the default landscape has evolved so far can I what are the things that we're seeing or things we should be excited about or things that kind of are not going so well and kind of everything in between for how this space is evolving so without further Ado please join me in welcoming Kane from synthetics thanks for making it thanks for having me since this was prepping for what we want to talk about here I ended up having five different things that end up sort of changing topics halfway so we're going to cover whatever we can we'll try to see if they all fit together at the end of the day but uh mostly I want to start by talking about how d5's change I mean it's not what I thought it was five years ago and in a good and a bad way that's not kind of what we were pitched when everything started that's not really what we see is consistently persisting either we're seeing so much experimentation for for better some of them are just small some of them are impressively uh amazing engineering changes so I guess I'm curious about what you think in retrospect what the last three years have been and what do you think D5 looks like now or where it's going yeah I mean it's a big topic I suppose right um so I think the probably the primary change in D5 over the last say 12 to 18 months is the fact that we now have uh layer twos um so we've got you know multiple production layer twos that you can build on that allow us to do things that were just not possible on L1 you know even back in the day um when you know we were like sub one-way transactions there was just stuff that was not possible Right like you know too much complexity in transactions Etc so I think that that is like fundamentally unlocking um that additional computation has really changed things and allow people to do a bunch of stuff that we couldn't do um for me the primary thing I guess you know in synthetics is allowing for a bunch of new perps protocols to emerge um it's just something that we couldn't do on L1 uh and you know we tried and it didn't work out very well um and we learned that lesson so you know that that's something that uh I think in the bear Market looking at the potential for product Market fit is one of the biggest opportunities um that you know we know that there are still a ton of users who are living on centralized exchanges binance by bit Etc and the notional volume is in the tens of billions per day and if you look at dexes it's like sub a billion dollars so there's like a 50x opportunity kind of sitting there for whoever is going to be able to put that together and I think that we've got you know probably three or four uh solid contenders um for you know figuring that out and right now in the bear Market the name of the game I guess in Defy is trying to capture the existing users are already here because there aren't too many new users showing up um it's awesome to see all of these faces here you guys are obviously still hanging around for some reason um but uh you know we um we don't really have a large influx of new users right so um you know tapping into um C5 and you know lending is a great example right all of the C5 lenders blew themselves up over the last 18 months which is awesome um and so now you know there's an opportunity in D5 lending I think to capture some of those users and figure that out and you know so we're seeing the emergence of like D5 combined with real world assets some kind of weird cd5 solution but I think that that's a big opportunity as well um and so there's you know I think there's a ton of things that uh have either knob impossible or have taken a lot of work to get to a point where they're now viable and so when I look around at kind of the defy infrastructure today versus say you know 2018 2019 or even into D5 summer in 2020 I see that we now actually have feature parity with a lot of the C5 or tradify Solutions and so then the real question is okay cool if we've got feature parity across a number of different um you know protocol types what is stopping us from getting users right and I think that's the the kind of adoption question of like how do we actually take it to the next step and focus on like user acquisition user experience things like that because the infrastructure is there now uh that's a that's a that's an amazing kind of perspective I think it ends up touching on a couple things I want to talk about maybe maybe one follow-up question to that is kind of from a personal perspective for you is the defy that we have now the same defy that the definition that you thought we had back in D5 summer or four years ago yeah I think it is I mean you know let's all forget a lot of what D5 summer was was like crazy yield farming Ponzi schemes right and so you know there wasn't a lot of sustainability there and uh you know the transition from crazy Ponzi schemes in you know D5 1.0 that were really driven by like kind of weird growth hacks um and and you know some utility some utility but combined with a lot of like growth hacking and D5 growth hacks um I think what we're seeing now is the transition to uh sort of real yield if you will right or you know Revenue generating protocols so there's a ton of protocols out there that were making no money back in 2020 and now they're actually you know making significant Revenue potentially they're profitable whatever that means in the you know in D5 right but they make more money than they burn or something like that and you know some of them even pay token holders um so you know there's I think there's a significant shift there in terms of like the sustainability of what's happening it's not just maturity altogether yeah exactly yeah would it be fair to say that effectively what we think about D5 now is that look this is feature parody because of l2s and we're kind of able to do a lot of things at traditional Finance uh does from like the set of products and our goal is to just get that same audience to switch so that we have better guarantees but at the end of the day it's the the same one-to-one mapping just on a better Network or do you actually think there's still some opportunity to kind of do things or offer products that are just not possible in in personal finance I think there's absolutely the the possibility of kind of making novel things right I think uni swap is a novel thing that tradify can't absolutely right the ability for a bunch of people to kind of come together who don't know each other and provide liquidity into an amm and facilitate trades is just not something that tradify really can contemplate or do so we do have Noble Solutions out there um I think that there are some things that we can't do yet right like we're not able to issue mortgages on chain right which is something that I think would be pretty but that falls into the technically possible operationally not yes yeah I mean there's also there's a component I guess of like social consensus right like you want to yeah yeah like you know we're we're kind of uh if we're going to do mortgages houses unfortunately live in the real world right we haven't been able to tokenize them in a way that is sensible right so you know houses are a thing that is a physical object and ultimately you're going to have to resolve ownership of a house via courts or violence right like at the absolute base level right um you know guns right someone it's not like you lose your private key and suddenly you can get into your door anymore exactly yeah thankfully otherwise we'd all probably be sleeping on the street um but you know there's there's kind of this idea of like okay how do we tokenize this thing or fractionalize it or you know create some weird derivative or something like that that allows us to bring the real world on chain um and you know I was very bearish on rwas for a really long time but I do think that we're now getting to a point where there's like enough social consensus to start to uh you know see these things being brought Unchained right and you know one of the other shifts I think is the um you know the interest rate environment as well right um in a zero interest rate environment there's no yield in the real world all of a sudden now you have assets that have genuine yield um some of which are Bare Assets and so you know those kinds of things are easier to tokenize or make more sense or whatever so I do think that we're seeing like some macro shifts but also like a social consensus that real world assets can actually be brought on chain and and we can find ways to manage them and wrap them up in a sensible way and uh and that's a multi-trillion dollar market so I hope it works out because that that has a lot of opportunity um one thing you kind of said which uh which I definitely agree with is just that l2s in the last 15 to 18 months have made a lot of this easier and possible um I'm kind of curious about from your perspective do you think we're just going to keep seeing different D5 apps spin out to their own new chains because that's kind of what it Trends to if you have a notion of having a yield in Revenue then it kind of gives you a good argument for being separate so you can track that versus kind of sharing the the pie um how are you going to think about that in like a multi-chain world or just what what comes to mind when you see these types of incentives here yeah I wrote about this recently um this idea of you know let's say there's 10 l2s or something like that right and so if you have a protocol that um is going to be deployed to all 10 of these chains which may or may not make sense right depending on what you're doing um but you know if you believe that they're kind of distinct users on each of these chains and you want to be able to generate revenue or generate you know yield from all the different chains and you kind of need to deploy to all of them and then there's this question of like okay having deployed all of them how do you collect the revenue where does it Flow To You know which chain is going to be the core chain or whatever and so I think that's where potentially we do start seeing um you know dap chains emerge right the problem that I I see with adapt chain that is like its own independent chain that's maybe collecting all the revenue or Distributing it or something like that is you do lose composability so I think you need to be on the chains where there's liquidity but may maybe you have your own like canonical chain that you sort of collect all of the revenue onto and then distributor because it does get a little bit weird one of the challenges is you know who earns the revenue on each chain is it you know just LPS if you're sharing with token holders what does that look like what's the split of creating something just governance altogether yeah chaotic exactly yeah it's very much an unsolved problem so you know I do think that there is maybe a role to be played with an app chain to say okay here's the credibly neutral chain where the governance token lives and you know where we do some kind of Revenue distribution or token Burns or whatever the method is and that sits over here and it's not you know maybe there's not even transactions happening on there in the sense of the protocol isn't transacting or there isn't even liquidity there the liquidity is actually spread across all the other you know generalized l2s but I've got my specialized app chain that is doing my stuff whatever that is are you sort of answering this in the perspective of this is how we can make it work or are you saying that it probably shouldn't happen I I well I definitely don't want to go is there is there a value in trying to make it work maybe let's answer this way because fragmentation is is a big deal especially when you have money locked into these systems where it is a zero-sum uh compared to I guess traditional Finance where you can kind of think of fractional liquidity but uh it it's not obvious right in in the case of defy it's not are the arbitrim guys still here they left us left um all right good I'm sure they're monitoring everything that I say anyway um but you know I tried for a long time right when this original L2 war was kind of Brewing before anything had been deployed to get everyone to agree to just deploy onto one chain which was optimism obviously that didn't work out very well for me um but you know the reality is that uh we're just past that point there's no way to kind of get everyone to coordinate to deploy to 1l2 now um I think we're going to see a you know proliferation of different l2s and again maybe app chains and whatever um and you know it might be that we have five or ten that have like meaningful liquidity um you know base launching shifts that calculus even further openness coinbase has their own users and there's just we'll we're gonna see like a general power law distribution of it these five things are doing this and then then you have like a long tail of weird Behavior yeah and God help us if it works right because then you have crack and chain and buy a bit change I mean bye but already has mantle I guess my name's already has one too yeah they're on L2 so you know every centralized exchange has an L2 and you know whatever right so um you know I don't think that it's possible to uh put the genie back in the bottle and get everyone to coordinate onto a single chain which means we need to think about this and you know for each D5 project if we're talking just D5 um there's going to be sort of idiosyncratic challenges right if you're uni swap well you don't pay fees to anyone so that's not a problem for you and you can just deploy sort of immutable code to any chain and then once they have and a lot of changes it's super easy right like that you know uni swap has done something in in the way that they've approached the world that makes their lives much easier right um if you are synthetics or some other fairly you know complicated protocol that requires a lot of governance and upgrading it just is much harder um it's not obvious how you share liquidity across Chains It's not obvious how you ensure fungibility of assets um there's a whole bunch of challenges that are really unsolved hence why we're still camped out on an optimism at the moment I'm gonna I'm gonna do another tricky question here um does evm compatibility matter in this new world of how we see D5 evolve like do you have any thoughts on like what happens here with or what's happening here with dydx moving away from from ethereum all twos is that a good thing do you have any comments you want to make do you do you think that works out or people don't care because they get the the service figured out at the end of the day and that's what they care about I mean I'm pretty much bearish on everything that dydx does and then proven wrong so I don't know if I would necessarily listen to me on that um you know maybe the cosmos ecosystem is going to blow up and uh and you know take over uh ethereum but you know so I'm pretty skeptical of my own takes when it comes to dydx but I do think that we are in a position now where you know l1s are migrating to become ethereum l2s we're seeing that narrative right um and it just makes it much easier I think if you're if you're a ghost chain that's been you know operating for the last two or three years with minimal activity pivoting to be in ethereum L2 is a pretty good narrative play right it's sort of tapping into that um now maybe that makes it easier if you're say Cosmos or polkador or whatever to kind of hold strong and you know there's only gonna be two or three Ulta ones left it might be like Solana and a couple of others um but you know it certainly feels like the evm compatibility and being part of some kind of ethereum L2 narrative is going to be pretty powerful this cycle you know it feels like that's kind of where the momentum is for sure that's great I I won't uh dig more in here um where does synthetics fit in all of this uh and what do you see the kind of the big opportunity and how or the direction defense evolving and kind of from a product standpoint what's the the best correlation that you have to uh traditional Finance so I think for synthetics what I realized and you know I kind of hinted to this earlier is that we actually have feature parity now with centralized exchanges the liquidity's there um you know we've now got I think like 50 or 60 different markets um obviously you know uh binance and a few others have like 200 but a lot of them are um you know denominated in different quote assets or whatever so you know having the top 100 assets or maybe the top 200 assets is pretty close to Future parity um and so then the question is okay we have this now this is working the liquidity is there why are we still doing a 50th or a hundredth of the volume of Finance when like binance is maybe on the point of you know implosion or something right if you believe crypto Twitter like what is going on here why are people still doing tens of billions of dollars on on binance and not on dydx or synthetics or GMX or whoever um and the conclusion that I came to after kind of looking at from every angle is that it is genuinely a user experience issue right it is onboarding it's bridging it's you know so we the part of the problem I think is right at the time when we solved perps dexes we solved it by going to l2s and the vast majority of liquidity was still sitting on L1 so it's like hey guys we've solved this we're over here now and everyone's like what are you talking about like you know I'm I'm still on on mainnet right and that made it really problematic to get users over there and we ended up kind of fighting over the scraps of people that were willing to bridge Del two so bridging is a huge problem there's a bunch of other things but essentially what I what I realized is that we need to build a new front end for um specifically synthetics but you know probably the same goes for GMX or dydx that abstracts away all of that complexity removes the need to bridge removes the need to get weird assets as margin removes the need to even sign transactions right that you have an embedded browser key um also you know handles custody and account recovery and all of that stuff allows you to sign up with a username and password there's a whole bunch of user experience things that we can do that will make the experience much closer to a centralized exchange and that was sort of a pointless exercise when the liquidity was terrible like when the core product didn't work there's no point building like this nice amazing layer on top of it right but it's only in the last kind of you know maybe six months that we've gone to the point where it's like okay we're there now let's actually make sure people can use it I guess one of the interesting pieces that I just picked up from what you said is looks like you're not betting on people converting or getting their traditional Fiat into Engineers is that the market is big enough that there's already enough liquidity on chain that they can just trade without needing a centralized exchange to do this thing is that is that like the the core bet here too I think that's the next six to 12 months right like there's no new users walking in this room right um with you know a pile of especially not this year yeah exactly right so um you know the Fiat conversion is is really not the issue at the moment right I think that there are a ton of people with a lot of assets on centralized exchanges right now that are doing whatever they can do on centralized exchanges and maybe have never done an on-chain transaction other than you know moving assets between different centralized exchanges right like I think there's a lot of users out there and you know they're still transacting a lot like we're talking you know tens of billions of notional volume every single day um and so that I think in the short term is the group of users that we should be targeting and trying to convert them off of centralized exchanges onto dexes The Next Step once the market Cycles sort of shifts and you know maybe we're getting close to that but I think it's still a ways away is to ensure that we have like Fiat on ramps and that's the next layer of ux in my mind but absolutely that's important for for somebody to take on and not not really have those two camps so hey you you figure out how to get on the centralized side and then once you have the money or the stable coins come back and then you can do stuff here yeah so I I think just to be really clear like you know if we can convert the majority of users that are currently transacting only on centralized exchanges to dexes which I think is now technically possible like from a functionality perspective we can actually bring them over and it will be as good if not better for them to be on a decks if we can do that then when a bunch of new users stop falling out of the sky which at some point they will right we need to make sure that they don't fall into someone's database we need to make sure that they actually go into a decks and that is going to require some additional functionality that we don't have right and at the end of the day if it's one to one experience we would want them to use the thing that has all these better guarantees is this I assume what infinix is and uh I guess what I'm curious on is what's kind of new about synthetics V3 that sort of helps you get there faster or just better so I think there are there are a number of things that so infinix is that UI layer right it's it's making this new user experience that's the the approach that we've taken um in terms of synthetics V3 what we're enabling in my mind that probably the biggest difference is permissionless markets so in the past synthetics has required significant governance overhead to add new markets which means you know World coin launches and we want to have a world current coin perp and it takes maybe two or three weeks by the time you're going through the governance process um and if you want to build some novel Financial instrument um you basically can't right there's there's almost no Pathway to getting that done and so synthetics V3 will enable anyone to turn up and tap into synthetics liquidity of which there's a lot and build some novel derivative so you know as it stands now if you want to build squeeze for example right you need to build the entire protocol maybe for audience could you also describe what sweet is it's basically a weird here to eat derivative right if you want to build something that tracks the price of eth or you know even if you want like a leveraged eth token or something as simple as that right you need to build the entire stack you need to go through audits you need to you know maybe have your own token to be able to pay people to add liquidity to it there's a absolute you know a huge amount of overhead to be able to do Financial engineering still right because in order to do that Financial engineering you need to have all the base stuff right and so what we're saying with synthetics V3 is don't do any of that you can literally turn up tap into liquidity and you just need to do the financial engineering side and the rest of it synthetics will manage and so I think that that's something that's going to be very interesting for people to play around with it'll make experimentation much faster I mean that sounds extremely powerful this is like a a leg up on composability but on top of exactly engineering different derivatives and verbs um I mean you kind of talked about this little governance process that kind of V3 in a way skips uh synthetics also is a really interesting governance um way of doing governance with a council um could you talk about sort of how you've set up governance and sort of how it differs from what other protocols have done and sort of why you made those trade-offs yeah so we started from the standpoint of direct on-chain voting I like make her style boating um is insane right um which I think it had been fairly well proven even before maker did it but you know maker does some weird stuff sometimes so um that was not really a viable path right as far as we were concerned um but we also didn't want to have kind of some centralized decision-making system especially once we got rid of all of our entities so you know we deprecated the synthetics Foundation there are no legal entities at all that participate in synthetics governance only token holders that's it so there's no Foundation there's no Labs there's nothing like that it's literally pure token holders control what happens and so we didn't want to have a centralized entity and we didn't want to have something that was like pure chaotic on-chain governance we wanted to find a middle brand and the way that we did that was to essentially set up delegated democracy so we have delegates did the difference between some other delegated democracy systems in crypto is that rather than having like an unbounded number of delegates so you know you could have a thousand delegates with like long-tailed distribution of governance power and the vast majority of them can't do anything and so you end up with you know maybe five or just the top ten yeah and the top ten that kind of make decisions and the rest of them are kind of pointless is we basically said okay we're going to have a council that has seven seats or eight seats or five seats or whatever and token holders vote for that Council and that Council sits for an Epoch which might be you know three months four months six months a year and those people are the representatives of their constituent token holders and they make decisions how do you sort of measure if this is going well in this system you just replace them if it's not because there's a fixed term or something else yeah I mean you know they get replaced every three or four months I think the e-book's now four months um and you know in terms of how you measure whether it's going well um I guess it's difficult hopefully the protocol rather it'll be obvious what's slow and and not well but you don't get to Benchmark this with any other model because no one no one is following the same thing yeah I mean you know we can certainly look at other protocols and and you know what their governance Cadence is I guess you know like how how kind of rapidly um and there's a few uh projects out there that track this you know like the number of proposals that are passed or whatever um and I think synthetics is up to you know 300 plus for different proposals that are possible that's a lot yeah so you know that's over the last like four years or whatever um so we you know are good at making bad decisions that need to be fixed and we're pretty fast at kind of deploying those new decisions and then you know replacing them to something else over time so um would you recommend this as the path for someone else right now and trying to think about decentralized governance or their own protocols or do you have any better advice for how they can set up governance or take any lessons that you've learned since the council is introduced yeah so you know we started out with a single Council and now there are five different councils that I think we're about to introduce a new one um so I think as a sort of new protocol having most of the governance uh controlled in a single Council makes a lot of sense and letting them manage the treasury and deployments and you know even potentially contributor salaries and you know contributor payments hiring firing Etc that kind of makes sense because you know as a sort of new protocol um you have an expectation that specialization is going to be difficult like most people have most of the context over time though I do think that it makes sense to introduce specialization that's what we've done in synthetics so with infinix we're going to have a single Council that's going to be the starting point again and it will govern contributors it will govern the treasury so this is just a council to govern infinix not synthetics related exactly yeah so it's a separate Council um and we're doing some things in terms of how we uh control governance power so synthetics token holders will get some governance power but we're also going to do like a governance mining program similar to like yams for example um where there you go we're bringing back the ldp 100 exactly D5 summer three now or whatever we're up to um and so you know the the idea being that there are multiple different groups that need to have some say in governance and Traders is one of them right and so we're allowing traders to connect uh read-only API and basically prove that they are trading on buy bit or binance or whatever and they'll be able to earn governance power and then they will have you know seats on the council to be able to hopefully have their voice heard I mean that's a really cool incentive because you're literally proving that they're an active participant um that's awesome uh another interesting point that you made which I want to just maybe get a quick answer on I don't want to go too deep into it but you said they get rid of all legal entities to kind of govern the feature of synthetics any advice to Founders who want to think about doing the same thing which is going from centralize to fully decentralized yeah I mean this has become I guess a little more difficult question to answer um I think there's some precedent out there for the risks of not having any entities um I don't necessarily you know I'm talking about ukidao right I don't think that those precedents are great precedents necessarily like I think a lot of people over index on that and say like oh ukidao you know if you don't have an entity and you're pure Dao then you know someone's going to come off to you and say you're an unincorporated Corporation or you know partnership or whatever um the reality is that ukidao could not really have up as many things as they like they kind of maxed out how many things you could up simultaneously and so I think that that probably was more indicative of why uh they had legal issues than anything else and so I think that you know when you don't have entities the primary risk in my mind is like investor protection that investors could come after you if someone you know buys a token so that means your personal liability yeah like personal liability there right like you don't really have that kind of investor coverage the the limited liability um and and you know potentially someone could come after you in that respect right I think when you don't have entities and you are pure Dao and it is genuinely down and you're not like just doing decentralization theater or LARPing is decentralized right um and you know it's great to see uh people like arbitrum that have pushed it even further right where there's like direct on-chain deployments and and all those sorts of things like I think you can't just be like oh yeah we've got a dow and you know there's actually four people in the room exactly yeah um you know it has to be really decentralized if you do that um I do think that over time um you sort of accrue uh like a regulatory resistance to some extent right um on the basis that a regulator you know we already saw the SEC um a long time ago say that the theorem is sufficiently decentralized now that same organization is on the month 100 right you know the same the same uh organization is looking to kind of roll that uh that comment back I think they would love to be able to um you know delete that from the internet somehow um but the reality is that uh there is a power in being decentralized for long enough right and being genuinely decentralized and so I think you know the more decentralized a project can get over time and that includes not having entities um I think the the better off you'll be um but you know it's not as clear-cut for me as it was a year it's not everyone says fits all and it'll have to be unique um I want to end on it could I get that uh disclaimer that uh was was up here I just added there that not Financial advice um I want to end on two uh questions but that kind of concern different audiences or applicants and audiences one is obviously this here has been interesting because as most of us like to say the tourists are gone which is a good thing but at the same time kind of effects where the things were finally ready for Consumer adoption for my tech standpoint but then nobody's here as a consumer wanted to come in but at the same time the people who really are passionate about this space are building and they're working on things that will Define what the next few years look like uh do you have any advice for Founders to think about what it means like to build in in this type of environment yeah I mean you know if you have a tourism economy and all the tourists leave that's not amazing um so you know there's definitely downsides um but I would say that uh having no tourists means that you don't need to be operating um and and kind of maintaining um systems in the same way um and you know there isn't the same level of pressure in terms of like onboarding new people and you know all of the overheads that go along with having just a ton of new users pouring into a protocol um and so the opportunity really is to like kind of take stock of what you built or what you are building um and really go back to First principles and kind of question all the assumptions and you can get away with doing really crazy stuff in a bear Market um you know much more so than you can in a bull market in a bull market the optimal play is really just to kind of find whatever the narrative is and try and ride that wave as closely as possible and don't fall off um in a bear Market you can do weird style you get to define those narratives yeah you get to define the narratives but you also get to do strange things so you know a really good example of that was synthetics um back in March of 2019 deciding that we were going to go from 100 million fixed token Supply to an inflatory token that now has 330 million tokens um trying to do that at a bull market yeah yeah exactly you know it's three three and a half times the the number of tokens that we started with right the original fixed Supply token that we issued and when we went around and asked large token holders and other people in the ecosystem whether they thought that was a good idea a bad idea no one cared like literally there was like sure do whatever we don't care right and so you know when there's minimal attention on your project and you know when um you're kind of thinking like okay what are some crazy experiments that we can run um and and you know see whether that helps us get tractional see whether that improves the that flywheel or you know the incentive mechanism within the protocol this is the time to do it and test and test and test and iterate and then hopefully with the you know small number of kind of uh I guess you know hardcore users you have you can get an updated to find something that is workable so that when you users do start to arrive um you know you've really got the incentive structure as uh you know tight as possible yeah you you have a way bigger surface area to experiment with so that one of these things work and you have more shots on goal versus the musical chair strategy uh which uh you definitely don't want to risk on uh live um my last question to you is especially in this room there's uh there's a lot of people who are coming from traditional Finance uh they were curious about how things are moving in this space especially on D5 I'm sorry for your loss so my question would be uh what comments or advice or sort of just general things you want to say to them uh to people who are sort of still hesitant about playing with D5 because all they hear about in the news is somebody something going wrong or somebody getting screwed uh what can we do to kind of help them be more comfortable or at least experiment and and how they should think about this space altogether so I think the the challenge is that a lot of the obvious things have already been done right um and so you know if you're coming into the space now and you're like okay I've got some domain expertise in some particular area of Trad Phi and I think that that would be a useful thing for me to try and do it can be difficult sometimes to find the Gap in the market for like you know we already have compound and Ave and you know whatever right um you know we already have a bunch of dexes we already have you know a lot of things um and so the the marginal Improvement that I think you can make on some of these core protocols is um is maybe pretty small but at the same time we have absolutely not solved all the problems there are definitely a ton of unsolved problems and so finding that unsolved problem and and you know working through uh maybe a more optimal solution is really the key thing in my mind and that takes a kind of a deep dive you really need to get into the space and and use every thing and see there's nothing like kind of practical experience in that world Nuance matters a lot more yeah absolutely right you know um if you kind of look at something like Ave and say okay you know the advantage you have is yes Ave is there right um and yes Ave has you know a ton of market share a ton of mine share really good brand Etc but it is also a slow moving incumbent right um you know in the same way that you have in Trad five there's you know big banks that it's really hard for them to do stuff it's really hard for Ave or you know even synthetics to do something right and so if you turn up and you're like okay I see what you know this protocol is doing or what that protocol is doing um and I think that there is a potential Improvement that could be made identifying that potential Improvement and then finding a way to test that really rapidly um will give you I guess the the best opportunity of kind of differentiation and if you do find something that is genuinely novel like a genuinely novel solution to even in a category that has a dominant incumbent you can out iterate them and you can you know kind of improve things uh you know much faster than they can and again in a bear Market the noise floor is much lower and so you know there's only a few hundred people is attention you need to get to be able to you know really start to get some mind share probably less yeah maybe they're probably all in this room um and so you know I think that um it's it's a fine line right it's really hard because it's like you look at a dominant player like Abe and think uh lending let me do something else right somebody has one so yeah what's the point lending right um you know and like unusual would love you to believe that amms are solved right but don't believe the psyops they're not solved there's definitely some tweaks that could be made if we believed you knew swap then you know there'd be no curve first just ask curve yeah exactly right so you know there's there's no question there unsolved um things and you know they might be more Niche but sometimes the niche use cases like curve turn out to actually be huge markets so um that's what I would say if you're coming from Trad find you've got some domain expertise probably very few people in crypto have that domain expertise and and trying to find the way the place to iterate that that Gap to drive into is is the right approach awesome well thank you so much for an amazing conversation that was super insightful and uh hopefully everybody else also enjoyed it too my pleasure thank you [Applause]