More Than Swaps: What DEXs Must Build to Flippen CEXs l KBW2024

[Music] all right good afternoon everyone uh great to see you all here um so today we're going to talk about the uh decentralized exchanges and how they compare to centralized exchanges uh but before we do that uh I would like to introduce our panel uh myself uh Vladimir um I'm the GP at New Hong Kong VC in a studio Venture studio called uh meas Venture Partners uh we have U Joseph here uh founder of uh warp uh which is a novel uh Dex with a nonchain uh order book uh we have JP founder of uh Tor chain uh which is a cosmos based um uh L1 which enables cross chain swaps um real yield and lowest fees in in the space and finally we have almir founder of um wowmax uh which is a uh Dex aggregation protocol that aims to um find the best exchange routing for the for the users right um we don't really have like anyone representing centralized exchanges so guys like please uh don't be like too harsh on it um so I'll start with a question uh for uh JP um so the last time I checked the uh Dex trading volume um was at 12% of the uh centralized exchange trading volume and historically it has never exceeded more than 20% what are the reasons for that I would say at the moment it's it's actually a user experience problem uh interfaces wallets uh there needs to be an improvement of all of that and that is coming there's probably another four years of grinding ahead of us to massively improve the US experience but there is a bit of uh there's two sided the use experience will bring the users on chain but then there needs to be enough onchain liquidity so unfortunately a lot of that liquidity over the last 10 years has being kept in the central exchanges I mean coinbase has a million over a million Bitcoin binance have you can check me out at the think of that 4 500,000 Bitcoin and that's I mean binance has been around since 2017 so it's a relatively new exchange and over those years they've accumulated over half a million Bitcoin through an amazing use experience an amazing brand that gets out and appeals to users so what we need to think about is how can we bring a million Bitcoin back on chain again not not locked up in Ledges and cold wallets but actually in onchain liquidity pools uh you know W Bitcoin wasn't a unfortunate rap Bitcoin is navigating a bit of chaos at the moment uh but there's a lot of at least this cycle we can see with the LST and liquid staking and liquid reaking there is now an Avenue to bring a million Bitcoin out of cold volts out of the exchanges and bring them in on chain on chain defi at least for Bitcoin and then once we have that liquidity on chain and we can generate a sustainable source of yield to bring users to staking and that's going to be like through the up chain thesis uh for right now base make 40 million a year so it's about how can we get that $40 million of fees from base L2 and package it up and give it to bitcoin stakers that that there is a giant source of yield so user experience more on chain liquidity okay perfect and um I guess from the uh the vast choice that users have um alir right so there's thousands of dexes deployed on various layer ones layer twos they have different approaches to uh their own design and architecture um in your opinion what makes users choose certain dexes over the others well yeah it's true uh what we see here at at wmax we aggregate a lot of dexes and we see that the new dexes emerge like every uh two three days uh currently there was a lot of innovation in terms of like providing concentrated liquidity and many uh projects experimenting with also onchain offchain kind of setup where the match orders um uh offchain and and they executed onchain um but to answer your question is is basically the centralized exchanges uh they also have like also established Brands everyone knows Unis swap so and uh other projects also like innovating a lot when preparing for this panel I spoke to many of our partners like isumi Maverick n exchange they all like innovate a lot but Unis swap has a very strong brand so actually you know maybe some execution efficiency is lagging there but they still the the users stick to Unis Swope to its brand because they trust it so uh I guess any D5 protocol needs to establish trust this is the first thing you need to do because uh you we might have a smaller number of users but our users like our whales like some of our users connect their wallets to our exchange and they they have like $50 million on the wallet and if something goes wrong the user can lose that that money like like in a second right so it's important that you educate your users and you assure your users that your protocol is safe so for defi safety is number one priority and of course deep liquidity is also an issue but like most of uh dexes now um deploy concentrated liquidity models which actually could be very efficient uh compared to even centralized exchanges in fact I was starting doing like Define 2017 and that time the gas cost were very high so I remember we were like even sometimes paying a few hundred dollars to execute one trade especially like nft minting nowadays l2s brought those cost for like fraction of cents so that's also stimulates the onchain activity and helps dexus to overtake Sexes so you you you touched a little bit on liquidity um and I guess the different ways to uh Attract it um Joseph from your point of view what what are the key differences between the liquidity provisioning on dexes versus centralized exchanges and and how dexes can attract more liquidity yeah so one thing to fully understand is that what you see today in defi is a function of Technology right like when you people often say the medium is the message but technology often times shapes the medium itself and the medium that we've been dealing with for the past seven years is the amm design and obviously we've done some tweaks with it with like concentrated liquidity but fundamentally you're dealing with this type of design in Defi and understand that real Financial system don't do this this is not what real Financial systems do real Financial systems contemporary ones do an order book and historically was open outcry but everyone switched to a central limit order book why haven't we done that since 2017 right the only ones that we see are just centralized systems some type of on offchain matching which just amounts to being a server right we we deal with that because we need to deal with adversarial constraints when everyone else is an Anon and a jerk you're not dealing with the same problems that NASDAQ or the Korean Stock Exchange or even coinbase deals with and this type of algorithm historically has not existed um this is what we're working on at warp um we'll be doing a release soon it'll be open source and it's essentially a way to do onchain matching where all of the data is onchain the matching engine itself is onchain we can shut down every single computer we run and it still runs there's no validator set it just runs it's run onchain it's executed on chain even the depth chart is a view function right like that's what you need to build and the reason why it hasn't been built is just it requires a lot of development it requires a lot of research and you know I believe one year out everyone will be using an algorithm like this and that'll reshape defi the problem with the amm model is like if you did it on the NASDAQ or whatever it was right like if I want to put in you know Market make one Apple share I also need to put in you know a thousand bucks or whatever you know like it doesn't it just doesn't make sense and then after I'm done Market making I don't know what ratio I'm going to get that's a little bit you know like I think that's why a lot of defi stuff you know like people were talking 7 years ago okay we're everything is going to move on chain why hasn't that transpired like we should be able to be out executing and out pricing centralized exchanges y see I disagree I I think a centralized limit order book is a FSE errand to pursue ever since I started there was ether Delta and before ether Delta there were attempts to put limit order books on chain but fundamentally a limit order books are maker versus taker for every making order you need a taking order sure whereas the am design with what uniso with V1 V2 V3 and our V4 with HS your it's essentially taker versus a pool and that pool could be passive liquidity I think Contra liquidity was a diversion and an optimization in a specific Direction which has made liquidity providers compete against each other when it used to be makers competing with takers and now it's LPS competing with each other so it's chased out a lot of the retail that's why most of the meme coin amm have gone back to V2 Styles not V3 Conant liqu it's just too hard to manage uh contract liquidity when it when you come to volatile assets so good luck but I just I think it's a false error so to follow up on that fundamentally I disagree because if you look at let's go yeah let's go um because if you look at what this automated Market maker role is playing it's ultimately a lagging function right like you cannot dispute that is functionally a lagging function I am basing this on History 300 400 years of History what we've had in terms of the intense type design which is what everyone's obsessed with right now that historically has been called open outcry open outcry does not exist yeah but we in defy we need to rebuild the history destroy the Legacy threat five systems for the inefficiency right no one does that anymore and there's a reason why since 1992 let's call the anymore um JP so I I actually want to sort of the higher level like ask ask you a question like the following question so does the um I guess decentralization itself and the transparent nature of the blockchain opens up the path to certain inefficiency or unfair Advantage versus the centralized private more opaque systems or entities uh proper decentralization uh should enable anyone to come to well in the case of defi like a liquidity pool and interact with it the the rules of that liquidity pool should be open and transparent and everyone knows unfortunately because of the consensus design of ethereum for so long that splits consensus with the application layer a lot of the transactions have monetary value and that monetary value is not shared with the validators which results in maximum extraction value me so there now becomes uh an off-channel bribing which is what me is and you then you you set up the conditions for sandwich attacks and back running front running and Etc uh so I think bar chain which is yet to launch uh which is going to recombine liquidity with consensus is the correct design and it's one that we built thought chain with back in 2018 with an incentive pendulum so the nodes actually participate in the liquidity yields of the network itself so I think that will stand the test of time and that demonstrates if you properly couple a system security liquidity and consensus you can actually design a cohesive system that works what we've seen today I mean in another example ethereum by not providing for aked eth it was also another mistake that they've made and so they've resulted in Lio having $25 billion of eth and a large centralizing effect through achieving economies of scale so bad bad L1 design can result in bad ux and erode the Integrity of the entire system so unfortunately eth has a reckoning I'm I'm I've been a massive subscriber since 2015 of eth and building on it but eth is is been challenged by the monolithic versus modul Theory but also is going to get challenged by bar chain which couples liquidity back with consensus so I think the E theorum Community needed a lot of work to do to actually solve some they've generated problems and they've solved some problems so let's say thanks um and just U on a like a slightly different topic but then again to align with the sort of the thesis uh that we need to bring uh liquidity on chain um alir uh what is cdy and why has it been uh so popular in this market cycle well uh CD stands for centralized and decentralized finance basically those kind of projects they try to get the best from centralized and decentralized worlds uh but to be honest like you know I know many F D5 Founders here right but like in 99% we all not D5 protocols we are cdy protocols because like in Even in our case we we kind of have like onchain settlement protocol but we have a massive infrastructure that do doing like offchain uh simulation and calculations and then when the user wants to find the best way to swap coins we actually make a millions of calculations at the back end and actually the most of the protocols using a lot of centralized infrastructures that's how technology works right only very simple block smart contracts can be only like live be live onchain and then actually can be called without any uh offchain preparation but um uh cdy is developing now because also the regulation pressure you know like for example uh last week uh when maker da announced the new token uh basically it's called Sky instead of die you know it has a lot of uh abilities to for the team to uh um regulate the token to regulate the IM mission to prevent some people swapping it so it's basically maybe a natural response uh of dii protocols to fit the current regulation landscape and JP on a regulation front um so I guess you would be an expert by now um is it coming into Defi and um can like will the need to kyc users kill dexes so when we first started thought chain we were worried about getting shut down by n nation state actors so we built the network of high churn anonymous validators no delegation didn't run any front ends paid the front ends to run themselves just run ran this kind of like faces protocol about 12 months ago a large amounts of illicit flows uh started being swapped over thorchain and it wasn't long before we started getting the letters and the the the phone calls and we argued at first that uh the protocol was as as decentralized and it's just responding to these onchain transactions and generating outband transactions and the protocol itself um because of its high churn Anonymous nature it's very difficult to regulate and enforce kyc and and it's actually will erode the protocol itself and harm 99.9% of the honest users and that was accepted from from sources and we shifted all the burden of uh address checking to the interface layer so if you if you do a swap on trust wallet and a user store chain then trust wallet will do the address checking but to date there is no oppression no address checking no censorship at the Bas layer which is which is the correct way to do it uh our also thesis is thorchain is a very large cross chain transactional graph essentially a very large Honeypot of information so someone coin joining Bitcoin and swapping over to Ethan going out through a cise exchange can be very easily followed so The Regulators are actually viewing th chain now as this giant honey pot of cross chain information which anyone can query by the way and it's actually helping them to find and track the elicit flows through the ciz exchanges before they get cashed out so for now the relationship is quite productive there is no censorship on Thor chains base layer the address checking happens at the interface layer and the AGG layer and now thought chain is being viewed as I mean to get the information out of binance you need a subpoena then to say this transaction came in where did it go you need to subpoena them un thought you just quering AP it's all there so it's right now being viewed favorably and that's I think the the correct relationship to go is say hey it's there um and yeah super interesting um well I guess one more question and then we'll round it up um Joseph just from your perspective what uh tools do we still need to build to technologically Advance the deck space yeah right now I think the deck space is at a very very weak position even relative to last year if you look at the antisocial participation within the space like look at the complete drop off in Salon activity why is that happening right that timed is very very well with the activity coming into salana like it wasn't just pump fun you know if if you you look at the principal existential threat to this ecosystem it is me and if you think about this everyone is talking about how do we live with me it's not how do we get rid of it and fundamentally if you think about it right A lot of these people are ragging on l2s l2s right now the rollups is the principal Avenue of me protection because it creates a competitive environment between different different rollups where if one sequencer starts doing me or letting meev in guess what's going to happen everyone is going to leave and that's doesn't completely solve me but it's these types of con considerations these types of designs where we need to be thinking about and we need to be thinking about okay how do we get rid of this problem as opposed to how do we help the antisocial people in this space live in it thank you and um just U as a final question in one two sentences we'll start with alir uh what's your prediction for the deck space in the next uh two years well I think D5 protocols and dexes will be kind of like underlying infrastructure for maybe other more like a centralized protocols because this is the only solution now uh which the world has for actually really cross-chain border exchanges where no one has you know custody of your funds no one can actually stop your transaction no one can exclude uh your transaction so basically this is the nature of crypto we did it not to makebe to just bring the fees down the core idea remember is a self- custody and the ownership of the assets so only dexes can provide that all web2 companies all AI companies all centralized exchanges are going to be forced onchain and we will solve the user experience and we Sol we will solve me which I agree with and the the liquidity rails of the future will control the labor force of the future the Bots will be the future labor force and the Bots will be paying each other at first they'll use tokenized us treasuries stable coins that will get censored and have to move to bitcoin and Bitcoin being a cens resistent currency will be the currency of the Bots and so we should all be trying to build Bitcoin liquidity and this is something where we definitely can end on a note where we both agree upon um I think that you know the future is definitely brighter um I think everyone in this crowd you know definitely build the tools to help us get to that space and I think one you're out um you'll definitely see significantly more activity and one in which we can start to compete with not just centralized exchanges but trfy all right thanks everyone thank you thank you