Bankless - dYdX Abandons Ethereum?! with Founder, Antonio Juliano

Episode Date: June 29, 2022

Last week, one of Ethereum’s biggest L2-integrated dapps — derivatives giant dYdX — announced plans to depart Ethereum upon its upcoming v4 release. The destination? A fully sovereign blockchain on the Cosmos Tendermint consensus protocol, built using the Cosmos software development kit. We brought on dYdX, Founder, Antonio Juliano to discuss this announcement and so much more. ------ 📣 NOTIONAL | Real DeFi Yield https://bankless.cc/Notional  ------ 🚀 SUBSCRIBE TO NEWSLETTER: https://newsletter.banklesshq.com/  🎙️ SUBSCRIBE TO PODCAST: http://podcast.banklesshq.com/  ------ BANKLESS SPONSOR TOOLS: ⚖️ ARBITRUM | SCALED ETHEREUM https://bankless.cc/Arbitrum  ❎ ACROSS | BRIDGE TO LAYER 2 https://bankless.cc/Across  🏦 ROCKET POOL | STAKE YOUR ETH https://bankless.cc/RocketPool  👻 AAVE V3 | LEND & BORROW CRYPTO https://bankless.cc/aave  ⚡️ LIDO | LIQUID ETH STAKING https://bankless.cc/lido  🔐 LEDGER | NANO S PLUS WALLET https://bankless.cc/Ledger  ------ Topics Covered: 0:00 Intro 7:00 The Announcement 14:38 Creating the Best Product 25:10 Decentralization 33:50 The Risks 37:55 Wallets 40:07 Staking Validators 46:05 Monetary Policy & Latency 54:44 Tradeoffs with Ethereum 59:55 Legal Reasons 1:06:00 Real Centralization 1:14:20 Bridging & Composability 1:24:00 Closing ------ Resources: Antonio on Twitter: https://twitter.com/AntonioMJuliano  Announcement: https://twitter.com/dYdX/status/1539607062291877889  Why Would a dApp Leave Ethereum Right Now? https://newsletter.banklesshq.com/p/dydx-cosmos-ethereum?  ----- Not financial or tax advice. This channel is strictly educational and is not investment advice or a solicitation to buy or sell any assets or to make any financial decisions. This video is not tax advice. Talk to your accountant. Do your own research. Disclosure. From time-to-time I may add links in this newsletter to products I use. I may receive commission if you make a purchase through one of these links. Additionally, the Bankless writers hold crypto assets. See our investment disclosures here: https://newsletter.banklesshq.com/p/bankless-disclosures 

Hey, Bankless Nation, welcome to another episode of State of the Nation where we do a deep dive into a topic. I'm fresh back from a week off, David. Feels pretty good, except I got a little bit of a case of COVID. And so I might be muting through some of this and coughing. But I'm really excited to get into this because this was some breaking news when I was out. DYDX. Are they abandoning Ethereum? That's the question we're putting forward to Antonio, who is the founder of DYDX. D-YDX. And of course, D-YD-X is a very popular, very successful derivatives exchange that has

formerly been built entirely on Ethereum, first on Mainnet and then on layer two. We've had Antonio on the podcast a couple of times to discuss this. David, what are we going to dig into today with Antonio? Yeah, this has really taken the app chain world by Storm, not just the app chain, just like smart contracting world, because where will the applications live when this crypto industry finally matures and settles into its long-term future, is it going to be an app chain world or is it going to be a layer two world? And it seems with this move from D-Y-D-X from the Ethereum layer-2 ecosystem to the cosmos app-chain world, that the app chain world has gotten a big victory mark, a big point.

And so we're going to explore this world and explore all of these questions. Is this going to become a trend? Is this alpha? Is D-Y-D-X leading? Is D-YD-X the first of many apps? to leave an Ethereum layer 2 and go on to an app-specific chain? Or, from the Ethereum perspective, is this just short-term thinking and impatience on layer 2 development? We're going to explore these things. You're going to understand why D-YDX decided to make the move to an app-specific chain, why Cosmos was the right place, and all the pros that come out of this, and also explore some of the cons as well.

Yeah, I'm really excited to dig in and get a lot of questions from Antonio. As does the bankless community, I think you guys have been feeding us some questions. We'll get those into the episode as well. Of course, one other thing to discuss is, well, all of the centralized, not all of them, but many of the centralized lending and borrowing exchanges are kind of melting down. There's still some safety and still some solace in Defy because Defy has actually performed very well. And our friends at Notional wanted to remind you that you could still get a fixed rate Defy loan. And let me share some of the rates here on USDC. So there's a fixed APY, 4% annual USDC alone. You can lock that in for up to a year. Also,

ETH as well on Notional. And it's really, I think not, it's definitely exciting to see how well Defi performed, particularly the lending and borrowing protocol side of Defi throughout the bare market so far. It's held up. What I love about Notional is you can actually look at one of their dashboards and see where all of the money is being backed, right, and see where all of the liquidations are taking place. And the nice thing about everything on defy in the collateralized lending and borrowing space is everything is fully transparent on chain. There's no black box, and you don't have to give up custody of your funds to a third party. So if you guys are interested in that, now is the time to make some yield with some of that capital that's sitting idle.

Go check out Notional. There will be a link in the show notes so you can attach. happen to as well. And we know this because it is Defi that they are just not re-hypothicating your funds right back to three hours capital on the other side of things. And it can also never prevent you from withdrawing the assets because it's baked into code. These are the things, these are the properties that have saved some people in the world of crypto from gritting wrecked because Defi has those powers. David, I could ask you the question. We begin every one of these episodes, which is what is the state of the nation today? Oh, Ryan, the state of the nation is tugging. There's a tug of war

of going on between these two communities. The Cosmos people got the dub and so they have tugged over, D-Y-D-X over to their Cosmos ecosystem. Meanwhile, the Ethereum Layer 2's ecosystems are fighting back and saying, no, this is short-term thinking.

There is a tug of war going on and D-Y-D-X is in the middle of it. And so, Ryan, that is the state of the nation we are tugging. It's good to have some competition, some tugging going on. I my hope is, of course, the winners at the end of the day are the users. The winners are the product, and the winners ultimately are those that are going in a more decentralized direction.

So we'll be back with a lot of questions for Antonio about all of these things when we come back. But before we do, we want to thank the sponsors that made this episode possible. Rocket Pool is your friendly, decentralized Ethereum staking protocol. You can stake your eth with Rocket Pool and get our ETH in return, allowing you to stake your ETH and use it in Defi at the same. time. You can get 4% on your eth by staking it with Rocket Pool, but you can get even more by running a node. Rocket Pool is the only staking provider that allows anyone to permissionlessly join their network of validating notes. Running a Rocket Pool node is easier to set up than running a solo node and you only need 16Eth to get started.

Why would you do this? You get an extra 15% staking commission on the pool to ETH, so your API is boosted. So if you're bullish-eathe staking, you can increase your API and get some extra tokens by adding your node to the decentralized Rocket Pool network, which currently has over a thousand independent validators. It's yield farming, but with Ethereum notes. You can get started at rocket pool.net and also join the rocket pool community in their Discord. You can find me hanging out there sometimes in the chat, so I'll see you there. AVE is the leading decentralized liquidity protocol,

and now AVEV3 is here. AVE V3 has powerful new features to enable you to get the most out of DFI, including isolation mode, which allows for many more markets to be launched with more exotic collateral types, and also efficiency mode, which allows for a higher loan-to-value-ratios, and of course, portals, allowing users to port their office, AVE position across all of the networks that AVEA operates on, like Polygon, Phantom, Avalanche,

Arbitrum, optimism, and harmony. The beautiful thing about AVE is that it's completely open source, decentralized, and governed by its community, enabling a truly bankless future for us all. To get your first crypto-collateralized loan, get started at AVE.com. That's A-A-B-E-D-E-D-com. And also check out the Avey Protocol governance forums to see what more than 100,000 Dow members are all ravening about at governance.a-Ave.com. Arbitrum is an Ethereum layer 2 scaling solution that's going to completely change how we use defy and NFTs. Over 300 projects have already deployed to Arbitrum and the defy and NFT ecosystems are growing rapidly. Some of the coolest and newest NFT collections have chosen Arbitrum as their home, all the while DeFi protocols continue to see increased usage and liquidity.

Using Arbitrum has never been easier, especially with the ability to deposit directly into Arbitrum through all the exchanges, including Binance, FTX, Quibi, and Crypto.com. Once inside, you'll notice Arbitrum increases Ethereum speed by order. of magnitude for a fraction of the cost of the average gas fee. If you're a developer who wants low gas fees and instant transactions for your users, visit arbitram.io slash developer to start building your DAAP on Arbitrum. If you're a Dgen, many of your favorite daps on Ethereum are already on Arbitrum, with many moving over every day. Go to bridge.arbitrum.com.com now to start bridging over your ETH and other tokens

in order to experience defy and empties in the way it was always meant to be. Fast, cheap, secure, and friction-free. Hey guys, we are back speaking with Antonio about DYDX's recent decision to abandon Ethereum. That's what it looks like. Moving to a cosmos app chain. Of course, you guys know Antonio. He's the founder of DYDX. He left Coinbase about five years ago because he wanted to make a better exchange.

And he did make a better exchange on Ethereum Maynet first. And then more recently, DYDX moved, migrated to Starkware and Ethereum layer two. and they've been there ever since. Last week, though, Antonio and the team in D-Y-D-X just announced they are moving towards to their own app chain, something built on the Cosmos ecosystem. We're going to talk a little bit more about what that means.

Antonio, it's great to have you back at Bankless. How are you doing? I'm doing well. Thanks so much for having me, and looking forward to the conversation. Yeah, I think we're going to have some fun questions for you, right? Because more than anything, I think I'm just curious about this move. But like, got to say, it could have been worse.

You could have decided to move to Tron or something like this. It was Cosmos, your own app chain. So can you tell us about the reason for this? One thing we've appreciated about talking to you in our conversation with you in the past, Antonio, is you've always pitched yourself as, hey, I'm not a chain maximalist. I'm a product maximalist. I want to make the best possible product experience, exchange experience for my users. And that drives all of my technology choice.

and all of my platform choices. Can you talk a little bit about this starting tweet here? I'm going to actually pull this up. This was your announcement tweet, and you give us some context for this. You said, and this is when you're announcing the move to DYDX's own app chain, for everyone with hot takes on why DYDX is developing its own chain,

we believe it gives us a chance to develop the best possible product. And that's it. It's all we care about. I'm not sure how many times I can say, say product. I have to say product in the past five years. Product means the best possible combination of user experience, features, decentralization, and security. That's the lead in, Antonio. Tell us about it. Well, you heard it from me first. I mean, I feel like I've been one of the most product-focused founders in D-Fi for a while. There are a lot of great products that are built

on D-Fi-Fi, but I'm proud that D-YDX is one of those as well. And, you know, I tweet about it constantly, talk about it on all the different podcasts. Really, what we're aiming to go out and do is build the best possible product. Like if you look at the success of tech companies throughout literally any tech company, success usually means for a tech company the best possible product times the best possible distribution. So you can't really win without having one of the best products. And that's really what we're going out to be able to build. We really don't care too much about any.

sort of chain maximalism. I would say we're sort of intentionally, and this is a tradeoff, it's not necessarily better or worse, but we're not really ingrained deeply in the community of any particular layer one or any other technology for that matter in some of the ways that other DFI products are. And I think that's been an intentional choice. We've just always really been as focused as possible on building the best possible product, adopting the best technologies. And I think that is the fundamental driver that led to this decision. Excited to dive into the details. And I think that's what always matters.

Like, why do we think that we can potentially build a better possible protocol, better possible product on top of a cosmos space chain than Ethereum? We can get into that. But at the end of the day, it comes down to what I would describe as the opportunity to build the best possible product. And that's one of the other things that I'll say about this decision. I think I said this in my very first tweet kind of announcing the move to Cosmos,

but this is a risky decision. Like, I think this is a risky decision. We don't know if we can build the best possible product on top of Cosmos or not. But we really wanted to try something that was fundamentally different than a lot of the different approaches that at least the most popular Defi apps have taken so far.

And we feel like this gives us a good operational. opportunity to potentially continue to push the needle on what's possible with the technologies. And then the last thing I'll say is you really saw DYDX be a super early adopter of new technologies in the past, right? We were one of the very first legit DFI protocols to move to layer two, where by far the biggest protocol on layer two today. And at least we think we want to continue to push the needle on what's possible with technology. I always say this to people, that we interview for DYDX, one of the things that continues to excite me about the blockchain in crypto industry is that the technology is always improving in a lot of different ways.

Like Ethereum is improving, layer 2s are improving, app chains are improving. But the way that it's improving is exponential. And that's what gets me really excited, right? It's like we can process 10x more transactions than we could last year. The latency is 10x lower. The decentralization is, you know, to 10x better. whatever metrics that you care about with the product. And then I'll just bring it back to kind of my second tweet there.

What does product actually mean? I think people have kind of a misconception in Defy about what product means and they think it's only like Ux and features. But when I say product, I mean what I said there, the best possible combination of UX, features, decentralization, and security. Because that's at the end of the day what people care about. And I think those four properties are really going to be a roadmap for this conversation because those are the parameters. that we are trying to optimize for. And I think this move to a Cosmos App Chain is saying that these four parameters are,

they score higher, the aggregate score is higher on an App Chain than they are on Ethereum Layer 2. So that will be sort of this like guiding North Star for this conversation. But also I think whatever happens as a result of this DYDX move to a Cosmos app chain, what we get as an ecosystem is a ton of information, a ton of data.

And something I've always appreciated about you Antonio is that you are a product maximalist. You are here to make DYDX the best product of all time. And you've used decentralization in order to achieve that goal. And this has been the explicit ethos of you building

a DYDX from the get-go, which turns this opportunity that we have as an ecosystem watching DYDX, something that has clear user adoption, clear volume, obvious product market fit. When it migrates from an Ethereum layer two to a cosmos app chain, we're going to see

as an ecosystem the data that comes out of this as a result. As somebody who is going after the product first approach, this is going to be a learning, this is going to be a lesson that we learn from the entire ecosystem. So I'm just really excited just to see the data that comes out of this. And just to go back to those four parameters, UX, quality of features, the level of decentralization and the degree of security, along those four parameters above, can you just like give us a high level pitch about why a D-YD-X app chain produces a better optimization on those four parameters? Yeah, for sure. And the first kind of premise I also want to come to this conversation with, I don't think that Cosmos is 100% better for all cases. DYDX just has a really unique

kind of set of tradeoffs that we care about, just to kind of set the groundwork. Roughly on DYDX right now on V3, on our current system with Starkware, the way the system works is there is two pieces. One of them is decentralized and one of them is centralized. The centralized piece is the order books and the order matching engine that run on our servers, on AWS right now. Of course, it's still non-custodial, but that handles a lot of the heavy lifting for the throughput. And then, of course, there's the decentralized system built on Starkware. That's effectively the smart contracts for our system. So going into V4, so maybe on V3, like how do we rank right now for Defi.

Like very good for Ux on Defi. I would say, you know, still getting to the level where DeFi in general is on the level of centralized exchanges for UX, but I like to think we push that forward a lot. Features, like I think we do pretty well for features as well. I think probably more features than pretty much anybody else in Defi. But again, like a lot less than centralized exchanges. It's just harder to build in Defi and it takes more time to build things. Decentralization, sort of like not that great.

Like it's non-custodial, which is. is really great. And you guys talk about the benefits of that a lot, which I certainly subscribe to. It's not fully decentralized. It's not censorship resistant. So like sort of not super awesome on that one. And then security is pretty high because like obviously we always care about security. So the real thing that we wanted to tackle in V4 is that this third thing here in decentralization going to a system that is, quote, fully decentralized. And we mean that. And one of the pieces that I think is important to understand about decentralization is the decentralization of a system is really equal to the minimum decentralization of any of the

components that make up that system to use the entire product. So you could look at something like a Starknet or an optimism and look at those kind of systems as decentralized in the sense that people can withdraw from them, right? like you can always get your money back, but at least for our current deployments on Starkware and a lot of other roll-ups that rely on a centralized sequencer, they don't necessarily have decentralization of that kind of sequencer layer. You can still get your money back. You can like get out of the system, but there could potentially be censorship of the core

product experience itself. So we wanted to get to a place where the entire system is decentralized. So okay, let's like move to V4 in Cosmos, like how should we rank on that? ideally the ux and features are pretty on par with v3 and that's actually pretty hard to do like to make the the ux and features even stay close to the same level on a fully decentralized system so our goal basically is to get those as close as possible to what we experience on v3 so v3 right now it's super nice if you go on it it feels just like using a centralized exchange like you place a trade confirms immediately um there's no gas keys when you trade all of that good stuff

on V4, I think we can get it to a similar level. Likely, the latency of trading will be slightly higher. I think it will still be a really good product experience. Just there's certain tradeoffs where like this system is decentralized and it like takes the speed of light, you know, 400 milliseconds to go around the globe or whatever. So probably like the system will be slightly higher latency, but still well within the bounds of what you would, you know, not that far off from what you'd experience on a centralized exchange. From a feature perspective, we want it to be.

at least as many features as V3, talked about, we'll just continue on V4 with our focus on perpetuals, which is the core type of derivative product that we offer, and then likely scale that out to other types of trading like spot, like margin, potentially other types of derivatives over time, so probably even better on the features over time. Decentralization, you know, talked about this. We believe that this is the system we can build that gives us the maximum decentralization. If you think about decentralization in that kind of definition I just gave, where it's equal to the minimum decentralization of any component of the system. And also I want to call out what I was saying when I began this, which is that our level of scalability is very

different than the requirements for like the scalability of most other decentralized apps. roughly right now we handle about a thousand order places and cancellations per second on our centralized matching engine, which is pretty similar to what a lot of centralized exchanges experience, probably even on the lower end of that. And then the actual trades or transactions that go on the blockchain is more on the order of 10 trades per second, sometimes flexing up to like 30 or so, but it's not anywhere close to like a thousand. So we approaching this, we kind of came to it with like, okay, first of all, what do we want to build? And we actually took that question to heart as well. We were like,

okay, is the DYDX products like the best possible product that we could build? Or are there other things that are worth building? Like should we pivot to an automated market maker? Should we pivot to like a request for quote system? And these could have significant impacts on the level of scalability that we require. And kind of the conclusion that we came to with this is from a product perspective, it's really important for us to continue to support this orderbook trading experience. The main reason for that is we've actually been extremely successful over the past six months, getting a lot of the institutional traders on crypto. Like pretty much all of the top market makers are trading on DYDX now. A lot of the institutional firms are trading on DYDX. And they trade on

D-YDX before they trade on pretty much anything else in D-Fy. Because we built a product that's really approachable to them and has an order book, has a lot of the advanced trade types. So we're like, okay, we still want to have this order book-based system. We still want to build this products that we have. Yes, we want to build more features into it, but we'll do that eventually. So, okay, how do we do that? And we took a look at the problem.

And we were like, okay, we have 1,000 order places and cancellations per second. So the natural question you would ask is what blockchain can handle a thousand order places and cancellations per second, ideally with minimal to no fees, right? Because on D. YDXV3 right now, again, if we're not trying to take a huge step backwards in terms of UX, we want that to be basically free or as close to it as we can manage. Took a look around and we're like no blockchain, Cosmos included, is even close to like a thousand's transactions per second right now at all, let alone. like it doesn't even matter if you pay like the highest gas fees ever. It's just not within the bounds of the throughput of the system. So then that brought us to this pretty quick realization that we need to build a second what we call off-chain network to run this orderbook and this matching.

It's kind of a similar idea to how do IDXV3 works right now in terms of there's a separate network, if you will, which is just like our centralized servers on B3, running the order book and matching engine, but in V4, this orderbook and matching engine function will be taken on by a decentralized network, but importantly, one that does not have to come to consensus. And what do I mean by that? Like, the network doesn't have to come to consensus because all of the validators will effectively store their own version of the orderbook, and they don't have to come to a consensus on it. So like, let's say the three of us are the only validators on the DYDX same. Obviously, it'll be more, but I could have like orders ABC, like you could have orders like BCD, and then, you know,

somebody else could have different orders. And that's okay. And we realized that that was actually a decent tradeoff to make from a product perspective. And that realization that we didn't have to run the order book on chain allowed us to let the scalability of the order book, be orders of magnitudes higher because the fundamental bottleneck in all blockchains is coming to consensus. So if you don't have to come to consensus, it's just way more scale. level. So we were thinking about it and, you know, there's a lot here. So I'm sure we can dive

it more into any of this that you want to talk more about, but just to finish the story, we were thinking about it. We're like, okay, we need to develop our own network to run the orderbook and matching engine. What's the best possible technology we could use to develop that network. Okay, probably it would be a Cosmos SDK blockchain built on top of tendermint. Potentially that could settle to another blockchain. And then, we were thinking more through it, we came back to that initial definition I gave of what decentralization is. And like, I'm going to keep coming back to this because I think this is the critical, like, insight. Decentralization is the minimum, decentralization of a system is equal to

the minimum decentralization of all of its components. So let's just say for the sake of arguments that we like built this off-chain, order book and matching engine, and then we rolled it up to layer one Ethereum. And let's just suppose that layer one Ethereum could handle the throughput needed for trades and it was like super decentralized on layer one on Ethereum and then the decentralization of our network was the best we could possibly make it. Well like what does that get you? Like go back to the definition. Like actually what you care about is the decentralization of the least decentralized components. And that quickly brought us to the realization that we should just build the entire thing on Cosmos because that's the maximum decentralization

that we can achieve anyways. And you know, what is that level of decentralization? I think that's certainly arguable. I think the level of decentralization is actually really high. And I think we care a lot about decentralization of the system. Like, that's the entire reason that we're building before. But that's the general thought process as to like, you know, why didn't we just build on Starknet? Why didn't we build on Solana? Why didn't we build on, you know, like finance chain or something like that. And instead, that kind of led us to building our own chain based on Cosmos. Okay. So if I'm tracking correct, Stark. Stark your StarkX exchange, DIYDX,

as a centralized sequencer, which is basically like centralized block production. And once we go to a Cosmos app chain, the DYDX chain, the DYDX token turns into a validating token, allowing a more permissionless level of block production for the system, which I think as that cool quirk that you were talking about, where every single validator gets to host their own order book, which doesn't need to be on chain,

the order book kind of turns into like a mempool kind of thing, Every single validator has their own version of the order book, and that adds a bunch of scalability. But then it adds a bunch of decentralization because it decentralizes the block production of the D-Y-D-X exchange. But just to play devil's advocate here, I mean, decentralization is supposed to be a means to an end, right? You're supposed to be able to achieve something with decentralization, not just like going for the heart of decentralization. There's like a decentralization helps you get to a goal. So going from a centralized sequencer, which is what you have currently in your L2 with StarkX, to going a decentralized block production with the DYDX chain, what does that actually get you?

Yeah, it's a good question. And also, so I think basically everything that you just said is pretty much exactly spot on. The only thing that I'll mention is we don't actually have control over the DYDX token. So it's sort of up to the DYDX community, whether the base layer one token of this chain is DYDX. or not, but we look forward to engaging with the community on that. In terms of your question, what do you actually get at a B4 and what do you get at a decentralization? I imagine it's pretty similar to what you guys talk with all the different founders about week over week, right? What do we actually care about? We care about non-custodial trading. We care about no censorship resistance.

We care about a quality of access. And I think those are a lot of the things that you get with a fully decentralized system. And that's just something we never possible. could have achieved by keeping a centralized sort of sequencer, if you will, with us just running our own order book in Matching Engine. Antonio, is the centralized sequencer piece of it as part of, you know, Stark X or Stark Ware or any roll-up right now? Is that just a temporary condition? Isn't the plan for most roll-ups to have, to move to a much more decentralized sequencer?

And if that's the case, why not wait for that? Yeah, great question. So absolutely. that's the plan. And we've been talking with at least Darkware a lot about this. And I know other teams with other roll-ups have plans to decentralize their sequencers as well. But if you think about it, we didn't really feel like there was a great reason to believe that whatever could be done on decentralizing the sequencers is necessarily a lot more decentralized than what we could achieve with our own base level validators on the DYDX chain. You kind of look at the reputation, the brand of

what we've built. We think that we can go out and build a very decentralized system and that at least order of magnitude will be on a similar level of censorship resistance and therefore decentralization as what any of the decentralized sequencers can offer as well. And arguably, it's a harder problem to decentralize a lot of the sequencers as well for Starkware. And they have some of the very smartest people in the world working on this. And I fully believe they will figure it out. But it's just a hard problem, right? Like how do you decentralize the provofer? network in terms of making sure all of the provers that are doing these pretty computationally expensive zero knowledge proofs, how do you make sure that's as decentralized as a network

that's just validating in sort of a simpler way, a lot of transactions like any other layer one blockchain would right now. So I'm sure that can be solved and I'm sure it will be solved. But one other point that I'll make is that we always care about delivering the best possible product right now, or at least in kind of a two to three year time horizon, I would say. We're not happy to just wait around for something that could potentially be done in the future. And again, like, there are really awesome people working on this. We love the Starkware team. And I fully believe they will figure this out in like another year or so.

But we want to build the best possible product right now. And I think the other thing that gave us a lot of confidence to make, again, what I self-admit to as being a risky decision. at least from a technological perspective, is that we're not afraid to move on to new technologies and sort of move past everything that we've built before. And I think that's a really unique mindset, especially in DFI. You hear people talk a lot about in DFI, like, oh, like EVM compatibility. Like that's so important.

We need to get like all the developers to just be able to seamlessly deploy their contracts. You hear certain DFI founders being like, oh, we would value. value EVM compatibility because we've built up this smart contract system. We know it's secure. We just want to deploy that on another EVM compatible chain. And you just see generally a lot of reluctance to build new things. And that's not wrong. It's just a tradeoff. It's like, you know, that's probably a little bit more in the camp of, okay, let's just like absolutely ensure like security. Let's absolutely ensure we don't break what we have. But one of the things that I talked about in the blog post that we came up with is we came up with like company values

or whatever for DYDX and going, you know, it's going into this process. I was like most company values are BS, but at least kind of articulating them. I think it just articulates a little bit better, like the way that I and the company think about things. And one of our values is think 10x bigger. And the way I kind of articulated that is we should focus on what we can achieve rather than protecting what we have. Like, we should play basically like we have nothing to lose because we don't have anything

to lose. And I think that's the mentality that we come into this with. A lot of people in DeFi are like, oh, wow, look at these amazing systems that have been built. And in a lot of ways, I agree. Like we in Uniswap and Compound and everybody else have built some really awesome stuff in DeFi, stuff that just fundamentally wasn't possible before. But how do you think about that?

Are you just sort of like championing it? Like, hey, look at all this awesome stuff that's being built. It's like, yeah, we'll just keep improving it at like a pretty steady rate. And then eventually like the technology will be big enough, will be best enough that, you know, we can build a great product. Or instead, do you, like, how do you measure yourself? And for us at DYDX, I always say this. And I think most people think it's sort of a meme or like we're not serious about it. But like, I just want to build the best possible exchange.

And I want to build one of the biggest exchanges in crypto. So how do we measure ourselves in that world? We're about 1% of all trading volume for perpetuals right now. So it's like, okay, like if you look at it through that lens, it's like we have like, we can lose like 1% of all trading volume. Well, our goal is like 30%. So like who even cares about like 1%. It's just like we should only play for whatever the best possible chance of like building

an awesome product that can get us more to that 30%, you know, competing with like FTX and finance mark. And again, like, I think this gives us the best possible chance to do it. We looked at a lot of different potential technologies and pretty much everything except this had been done before by somebody that was pretty good. So we probably thought we could probably build it better than other people. I have a lot of confidence in our engineering team. But this is kind of a new approach.

Like nobody's built like a decentralized off-chain order book network before that handles on the order of a thousand orders per second. And I think it comes back to what's that goal? what are we doing here? And I think we're not afraid to just, like, throw away everything we've built before and move on to the next, at least what we think, has the potential to be, like, 10x-type technology. And also, like, if we're wrong, then I'm, you know, I'm not too proud, hopefully to say,

well, like, okay, we'll just, like, go back to Ethereum and, like, go back to Starknet or go back to optimism or, like, whatever. We'll always build on the best possible technology. And we care about, like, the best possible technology. on this like two to three year time horizon, not like a 10 year time horizon because we're willing to throw out everything we built before and just build on the next 10x thing. Yeah, I think no one would ever accuse your team of being overly cautious, Antonio. I think you guys are definitely pushing the envelope here.

And we do want to talk a bit more about the tradeoffs a little bit later, specifically this decentralization piece. Before we get there, this is probably the second or third time you've mentioned that you personally think this is actually a risky decision. I want to hear you get into that. Why do you think this is potentially risky? What do you think the risks actually are? Yeah, it's a great question. I think there are a couple risks, probably the biggest risk is it's just like really, really hard to build. It's like there's so much stuff to build. It's like we got to build like the base level blockchain. We got to build all of the different modules into the base level blockchain that, you know, knows what a deposit is, knows what a transfer is.

We have to build this off-chain orderbook network. We got to build our own Oracle network. We got to either integrate or sort of like build bridges to different chains. We have to build sort of like the alchemy, like indexer level equivalent of this chain. We can't just use like the graph. We can't just use like alchemy to make a really easy user interface for our users. We have to build our own mobile apps. That's sort of like an unsolved problem in defy.

obviously we have to build like a website. So it's just like there's so much to build. I think that's probably the biggest risk is that we've just bitten off a lot more than we can chew. And here we are actually. Like we started this project from a research perspective about the beginning of this year. And we committed to open sourcing the finished protocol by the end of the year. And that's actually an extremely aggressive timeline. Like if you look at the timelines of pretty much anybody building other chains that are out there,

It's usually like two to five plus years, and we're saying we can do it in one. And not only will we do it in one, we're committing that like we're going to build this amazing product, not just this kind of base level protocol that other people can build on later. So it's a lot to build. I'd say probably a couple other risks. You know, there are potential risks with the technology. Like one of the things that we're obviously betting on to improve is bridges. And we can dive more into that later.

we do think that's improved quite a bit. The one thing I'll say on bridges that at least buys us a good bit more time is DYDX is a derivatives exchange, not a spot exchange, and that actually matters quite a lot for this decision because the one thing that a derivatives exchange needs from a bridging perspective is one collateral token rather than like you need every single like asset that you want to trade in a spot way to be on that chain. So if we kind of reduce the problem, at least right now, to for one, there's a single collateral version of DYDX like there is on V3, it means we only need one really good decentralized kind

collateral option on V4. Currently on DYDX, V3, that collateral is USDC, and we're in talks with Circle and Center to potentially get like a native USDA deployment as well, which I think would make the collateral, at least, as decentralized as it is right now. We don't have to go on a tangent and get into the fact that like USDC itself isn't decentralized, which I'm fully aware of, but it would at least kind of maintain the kind of level of censorship resistance that we have for our collateral on V3. So those are probably the two biggest things.

Like maybe we've just like bitten off way more than we can chew from an implementation perspective. And then maybe the technology doesn't increase fast enough. And then the last thing I'll say is like maybe I'm wrong. Like maybe like the ecosystems that are built up around a lot of these blockchains, like from a community perspective, are really important. Maybe like the it's just important. Like people want to be on Ethereum. They like really believe in Ethereum.

They like, you know, use all their uniswop and different apps on Ethereum. And they just want to be on one chain. And like the people that are on Salana, like really believe in Salana. And they just like want to build, you know, use products that are on the native chain that they care about. I think that's wrong personally, but I could be wrong. So I think that's like another risk. I'd say those are probably the biggest things. And Tony, do you guys have to like rebuild all of kind of the wallet infrastructure as well?

So I can't use metamass with this thing. What am I going to be using? Or can I use metamask? Yeah, it's a great question, actually. So you will be able to use metamask and every wallet that you're normally used to using. But we do also have to rebuild the wallet's interface. And we have to like rebuild wallets. Like, what do I mean by that?

We've actually done something similar to this for DYDXV3 as well. When you use DYDXV3, there's a separate key pair that you're storing, which is like a stark public private key pair that Metamask doesn't know how to handle. Similarly, for DYDXV4, there'll be like a Cosmos, if you will, public private key pair that MetaMask won't know how to handle. We've effectively built a wallet into our products, which is a whole other like class of complexity. But that lets you just connect to DYDX with any wallet that you might want to. the use. But I think I sent out a hot tweet about this at some point, too. Like, there will be no, like, using DYDX chain, like, won't feel, I think, like using any other, like, blockchain that you're used to using. Like, normally if you go to a different chain, it's like,

okay, you've got to, you're shipped off to some random bridging website that seems kind of sketchy. Like, maybe that works and you can figure it out. Like, maybe it doesn't. Sometimes you have to get different wallets that you may not be used to using, like the current one on Cosmoses like Kepler. If you're trying to use something like a Solana for the sake of arguments, there are great wallets, but maybe you have to download Phantom or something, and that's different from the MedMass wallet that you usually use. On DIYDX before, we care, again, about the product experience. So there's not going to be any random like downloads that you have to do for other wallets. There's not going to be any random bridging websites. Of course, there still will be the concept

a wallet and the concept of bridges. Basically, what I'm saying is we want to build everything full stack into D-YD-X. We want to build like the best possible bridging U-I into the product. So it's just like, okay, I click deposit on D-YDX. Yes, behind the scenes, maybe we're, you know, going through like a bridge or something like that. And we'll be transparent about that, of course. But fundamentally, it's all within one product experience. So again, there is quite a lot to build, but I think that's the way you build the best product. So Antonio, let's say, let's say the community does move forward and say something like DYDX, the token is the designated staking token inside of the DYDX app chain. How many, you mentioned this a few times, decentralization of kind of the order book. That very much depends on a diverse

decentralized set of independent validators, independent stakers. How many independent validators are you hoping for in this new app chain? Is there going to be, obviously, you're going to have to start paying like fees, issuance, maybe, block rewards if you're going to fund it that way, or transaction fees at least to this new validating staking class. This is a new entry into the DYDX ecosystem. How does all that work? So how many stakers, how many validators, and how are they going to get compensated? Yeah, this is still something we're researching, to be honest, like how many, what's the

upper limit and what's kind of the tradeoff on tendermint between the number of validators and the performance of the chain, especially. when we toss in what's the off-chain order book system look like. Roughly, we'll probably start with somewhere on the order of 50 to 100 validators and continue to build up from there. And I think that's more than decentralized enough to start. And I think if you look at something like the evolution of Solana over time or the evolution of some of the other layer one blockchains,

normally it starts at about that level like 50 to 100, sometimes less. If it's a janky blockchain, but usually you can bootstrap the validator is fairly easy. And then it continues to grow from there. And then I think the most important thing to realize is it's almost like less about the number of validators than it is about the distribution of the L1 token itself and like who the holders are. Right. Because the holders are the ones that are actually staking to the validators.

I think people sort of has this misconception that in that like number of validators is extremely important. And I think it is important, but I think it's not the only thing that's important. Let me maybe go through an example of that. Like in tendermint, tendermint basically assumes that at least two-thirds of the validators are honest. So then if you look at that, you might be like, well, okay, hold on, Antonio, you just said there's going to be like 50 validators.

If you're telling me that only, you know, only one third of those validators need to be dishonest and then like something bad could happen. And that's like, you know, 15 people or something. That's like not that many people. It seems like it's not decentralized, right? But what would actually happen in that situation? So there's kind of two failure cases for a tendermint consensus. The first is more than one third of people are dishonest.

And what can happen in that case is the chain can get halted. They can't actually like double spend. They can't like steal your money or anything like that. But the chain could get halted if let's just take like the most conservative case. There's 50 validators. If like, you know, 17 out of those people, 16 out of those people are dishonest. Yes, like 16 people could freeze the chain. Yes, that's the case if there are 50 validators.

But what would actually happen then? You would probably go back to the community. You would probably be like, okay, like the DYDX chain is frozen. What should we do about this? And we were like, okay, let's just boot out these like 16 validators that for some reason decided to freeze the entire chain and instead take to new actually active validators. So yeah, degradation of performance. And like clearly that shouldn't happen.

But if you like game theory it out the whole way, it's not just like there's 15 people and they can steal all your money. There are other like failure cases for the, in the worst case, if, like, two-thirds of the validators are dishonest, they could potentially steal funds. But one of the cool things about Cosmos and Layer 1s in general, and one of the kind of arguments I was making on Twitter is I think a lot of people have a misconception, well, maybe not the misconception, but sort of like the conviction that different Layer 1 chains are definitely less secure than call it smart contract systems built on secure layer 1 chains.

I would argue that is maybe the case, but not necessarily the case. I think there are tradeoffs when it comes to security as well. Because the chain, what you get if you're building your own layer one system is you get sovereignty, right? You can fork. You can like, you know, do different like token votes to make things happen. And I think that's a really powerful thing. So like, okay, let's just say like absolute worst case like DYDX chain gets taken over like two thirds of the validators conspire together.

to steal everyone's money, what happens? Like, that could happen technically, but probably, you know, then we would be in a situation where it's like, okay, DYDX chain has gotten like taken over. And then probably the community would come together and be like, okay, this is like, is clearly like not what should happen. Let's just like fork the DYDX chain into like a DYD, version of the DYDX chain where like that attack never actually happened. Again, if you're just like game-thearing it out.

And that's something that you can't do on a smart contract system, right? if you're built a smart contract and there is a vulnerability or something and all the money gets hacked, it's like, that's it. Money is gone. It's not quite as simple as this because there's also like bridges and implications for what happened to all like the bridge funds on your chain. But there are a lot more options to recover from attacks if you're on a sovereign chain. So I'm not going to sit here and say that like the DYDX chain is going to, you know, have as many

validators and as many nodes running on it as Ethereum or potentially some other layer one chains like lawn or maybe even what like different layer twos for Ethereum will be able to achieve once they decentralized their sequencer networks. But again, like if you think about it from a pragmatic like product perspective, like, okay, what do we actually care about? We care about like the people's money not getting stolen. We care about people's money not getting censored from a trading perspective and you like game it out, like all the different things that could happen.

I would argue this is at least in the ballpark, if not better than from a decentralization perspective, a lot of the other potential options that could be built on other chains. Antonio, you alluded to earlier how it's going to have to be the community that determines whether it's the D-Y-D-X token that becomes the staking token of D-Y-D-X chain. And I think we can just go ahead and assume that they are going to choose that token. That would be like the logical choice for them to make. I don't really know what other token that they would choose. So granted, that is the choice that they make.

that adds in a bunch of extra complexity into the overall system, right? Like, we need to pay validators. And so kind of Ryan alluded to, like, well, we could pay them in transaction fees. So maybe we could pay them in USC. But if we followed the mold of all the other blockchains out there in the world, you would just pay them in issuance, like new DYDX issuance. But then there's a question of, like, well, how much issuance? Like, how much inflation are we going to have in the native DYDX token?

Which just begs the question of, like, DYDX needing a, its own monetary policy. And so, like, where is this? Like, there's a bunch of unanswered questions about, like, how this would work. And especially with, like, well, how many, how many DYDX tokens do we want to, like, on Ethereum, you stake 32Eth on DYDX train. How many DYDX tokens do you stake? Have you, do you have any idea about how some of these questions will get answered

over time? So, first of all, you're exactly right. There are a lot of questions like that. And I think that's a really excellent line of thinking. I don't honestly have all the answers right now, or do I even have the power to just, like, dictate what exactly we should do. But that is something that we're going to, that is one of the challenges we're working on, at least advising the community on what we think would be best over the next year. So I think there are a lot of open questions, to be honest, about the products that

we build. And this is certainly one of them. Like what happens to the layer one token of the chain? Is there, as you say, inflation? Is there issuance? Like what percent of fees should be paid to validators versus like stakers to those validators and so on and so forth? And what is that mean for the fundamental value of the layer one token and, you know, if DYDX is that token at some point, what does that mean for DYDX? There's a lot of questions. I would say right now, like, who knows, right? Like, even on a lot of these different, like, layer ones, even on Ethereum, right? Like, people still don't, like, exactly know or, like, have really good reasonings for why does, like, ETH have value? Like, how do I think about, like, the value of ETH long term? Like, isn't ETH

just going to keep inflating, like, forever? like yes, I know there's like burning of eat that goes on with the EPA and all of that. But there's a lot of questions, right? And I think the one of the things that we try to do at YDX is just like steal all the best ideas from other people, right? Because that's the best side. That's the best and the easiest way to come up with good ideas. You just see what's like working for other people and you do that. So I would imagine we would just take like the best possible practices from other chains and try to adopt that and like reason about how that should be put onto a system that is generating revenue.

Like there are trading fees that are going through the system, like, where should those goes? They go to validators. They go to stakers, et cetera. How much goes to which? I don't know. But I think that's something that will spend a lot of time figuring out over the next year or so. And I'm looking forward to the community member who writes the DYDX, the triple point asset

thesis on the DYDX token. One last question on like the, if there will be a bunch of validators, you said like maybe 50 to 100 to start. but the idea is to decentralize this thing. There's an interesting relationship between a decentralized network and a high-performance order book-based exchange, where the more validators you have, the more latency, orders are going to go through the network

and propagate through the network. And I think when we have such a high-throughput exchange, that latency adds opportunities for people to front-run that latency. And I think there's now if D-YDX is going to be its own layer one blockchain, there's M-A-V to contend with. Have you thought about how to approach this M-E-V question when there's a ton of volume going through D-Y-D-X in very fast transactions,

and now it's its own, like, L-1 validating network? Yeah, it's a great question. And, like, first things first, like, I don't know. I don't have, like, all the answers. Of course, we're thinking about it. Yes, these are issues. But I think just, like, zoom out, and then I promise I'll, like, actually talk about the real question, too.

But if you, like, zoom out, like, nobody in D-Fi has figured this out yet, right? Like, nobody has figured out, like, how to fully combat MEP. Nobody has figured out how to have, like, really super low latency, DFI platforms. So maybe zooming back into us, like, let's at least as a, like, first step, try to at least make sure it's not worse than all the other, at least like the best other DFI product that's out there from like a latency in MEP perspective. And I think we can get there. I think we can take a lot of, like, one of the cool things about tendermint is that you know the leader schedule. This is a bit technical, but the way like leadership elections work and a lot of proof of stake systems is they're known beforehand. So again, let's maybe

take a simple case where there's like the three of us are the only validators. You would know that for the next 10 seconds. I'm the validator. Then like you're the validator after that for the next 10 seconds. And then you know, you're the validator 10 seconds after that. And in that way, and this is actually something Salana does pretty well. So like we want to take some really good learnings from what they've done in terms of reducing latency in a pretty big way on their chain. You can sort of optimistically just send the transactions to whoever the next leader is. So they don't necessarily just have to propagate throughout the entire network. And that can be a way to reduce latency. There are a lot of different approaches that we're considering like that. But I think that's the

general like high level thought on how we would reduce the latency. And then things like MEV. I think one of the interesting thoughts that we had on MEV is, yes, there certainly is MEV. There's like a lot of weird stuff the validators can do with reordering transactions, with like messing with the order book and stuff like that. They obviously still can't like steal people's money or like make people make any trays that they don't want to make. But they can reorder things in the same way that if you send in order to uniswap right now, it could get picked off. One of the interesting thoughts that we had on the DYDX chain is that the incentives of the validators are a bit different, right? It's not just like there is this general purpose chain that has all this stuff running on top of it.

And like no one thing running on top of the chain is significantly important to the chain. You know, like some Ethereum validator is not really going to care about some like random Dap your friends with launched yesterday. It's like not important to the success or failure of Ethereum. So therefore, you should just go like ham on the MEV side if you're a validator and extract all the value that you can. But on the DYDX chain, the validators actually in some sense are intertwined with success or failure of the network. And ideally, like the stakers of two those validators have a vested interest in the success or failure of the network as well. So I think from that perspective, like they should want, they should not necessarily want to do things that degrade the network. they still may do that if it's still like if they do the math and they're like,

okay, well, maybe this will like hurt the chain this much, but I can make this much profit out of doing it. And like, yeah, absolutely they'll do that. But one of the things that you can do is you can at least make MEV attacks a lot more transparent. So you could be like, okay, like again, the three of us are running validators. Let's just look at the average fill price of the validators. And we'll say that, okay, Antonio looks like his validators.

they're not getting that much like price impact, like the pricing is really good whenever he proposes a block, when you propose a block like, oh, like, hey, wait a minute, like the pricing is always like way worse than when you propose a block, like what's going on here? And then like you can effectively just make like the stats around Mavie a lot more transparent. And it's sort of like a second order way of combating it. But you can get to a point where, you know, you could propose to stakers of the chain.

You know, hey, look, like here are the validators that likely are performing. a lot of MEV, like, let's de-stake from them. Potentially, there could be, like, penalties in an extreme case. And so I think the point I'm trying to make is it's, like, a bit simpler of a problem if, like, the validators of the chain are, if there's, like, only one thing running on top of the chain, there's not, like, a general purpose amount of stuff running on top of the chain. And that makes the problem a bit easier.

It certainly doesn't, like, make it go away. But I think it gives you more tools for tackling that problem. I definitely do take the point that having. a standalone blockchain that does one thing and one thing well can constrain the MEV problem down to like community consensus and the community can say hey this was legitimate MEV and this was illegitimate MEV and the fact there's only one application on the chain makes that a little bit clearer I do take that point however I will also say at the same time you said at the beginning of this question that like you know no one really knows how to tackle MEV no one really knows about the

future of MEV etc even on Ethereum and I want to push back on that because like you Yes, the MEV seems to be this unsolvable problem that always, like, even after you solve one part, like there's another part of MEV to solve. You just kind of push it off into a different direction. Now you have to go solve that thing. However, I will say that there's, of all the chains, and this is something that, like, I generally critique other layer ones for is that no one really seems to be tackling this problem more directly than the Ethereum ecosystem.

We have, like, the flashbots that was born out of Ethereum. We have Ethereum in the future roadmap. We have things like Proposer Builder separation. Both Arbitrum and optimism are generating their own strategies, their native strategies for harnessing and capturing fair M-EV and redirecting that back into the ecosystem. And this is generally the critique of or the thought of staying on Ethereum or leaving Ethereum as a whole is that if you stay on Ethereum, you get to draft in some of these tailwinds. So these innovations kind of are sticky to the EVM standard and why people generally always want to like fork. the EVM, because then you get to have all those tailwind benefits of like, if somebody solves a problem in one direction, like, you get to draft on that. And so like, the high, high level question,

Antonio, is, is DYDX prepared to have to face all of these headwinds, all these MEV unsolved MEP problems all by itself without drafting on like some of the research that's going on on the EVM side of things? Yeah. So first of all, I totally agree with you. I really appreciate the Ethereum community, right? Like, I've been building on ETH for the past five years. And I do think Ethereum has a much better builder's mentality and like, let's actually solve these problems mentality than a lot of other chains. So, and there have been some great things built. Like you mentioned flashbots, like some other things. I think those are not a one-stop shop to fix everything, but they do have really good tradeoffs and give you some strategies to combat it. I mean,

it comes back to what I was saying before for Ryan's question, I think. Like, what are the biggest risks for DYDX? It's we have to develop everything ourselves. Again, like we don't have to do everything from the ground up. Like we could look at, oh, okay, hey, like, look, on Ethereum flashbots is working super well. Like, can we work with the FlashBots team or can we adopt FlashBots somehow to work on the DYDX chain?

Like, would that even make sense? And we'll, like, adopt all the best possible ideas. But there is a lot more for us to build. And I think that's the fundamental tradeoff. It's like not just we're like uniswop or something like that and not to pick on uniswop. I could say this of any Dap. And we're like, oh, well, now like FlashBots was built. And that makes the Uniswop trading experience way better.

And we didn't even have to do anything. Like, that's amazing. And you hear people talk a lot about composability on Defi. I think that's something you guys and a lot of others are really excited about. And I think that is something that's really unique in Defi. And we've effectively not 100% given that up, but we've. trade it off against composability for scalability and for decentralization. And I think one of the things that's important when you're building a product is you're really

honest about the tradeoffs that you're making, at least with yourself. And then ideally publicly with the community as well, like, yes, we trade it off against composability in some sense to get like decentralization and scalability. And we also trade it off against like, okay, if we're just going to build everything in a vertically integrated way, there's so much more for us to build. But I think if you look at a lot of, again, like we sort of compare ourselves more to really successful tech companies or tech products and protocols rather than just defy. And I think if you look at the evolution of a lot of different really successful tech products throughout the years, you see them go through this evolution where they become more vertically integrated over time. It's like, okay, we like build a great product on top of, you know, like Twitter or API or something like that.

Now we should go down one level and like own that API for us. Like we started with just a smart contract. And then we were like, okay, well, we need to own like more of this experience. So we build like our own like more of this experience. Like we already like built that. We need to own more of the experience. Like this general purpose thing isn't built exactly the way we wanted to. So you just see things get like more vertically integrated over time.

And yes, that's harder. And yes, that's why normally like bigger like tech products have the bigger tech companies behind them or engineers that are building that. But I think it's like not that ridiculous, actually, to just try to be more vertically integrated if you just look at the evolution of tech products over time. And that actually is a perfect segue into my last question before we go into a break for sponsors. There's a theory out there that this migration to a DYDX, the DYDX chain, a cosmos application-specific chain, and importantly, the decentralization of the block production is one part a technological innovation,

technological improvement, but also just simply that app chains are a strong legal improvement, as in if we can separate and decentralize the block production, that's like, as you said, kind of like the last big centralization property of DYDXN's current form. And once we decentralized that, especially after when you say you're building this very verticalized system, when I hear a verticalized system, I'm thinking like, you know, a trading engine like gargantuan, like the best trading engine of all time. It's extremely verticalized and extremely robust, but it's got this one pinnacle of like centralization,

which is the block production. And so there's this threat out there from the Twitter accounts, E Leon, Elion, that says that validiums, basically what StarQuax is, where we are, what DYDX is leaving, are the perhaps the superior scaling technology

from a technological angel, but they are not the superior legal technological solution. And so from a decentralized, conceptualization perspective, D-Y-D-X gets further legal protections by being further decentralized because in the current setup, too few entities are producing the proofs that create blocks. Would you say that was a motivating factor in the migration from its D-Y-D-X in its current state to D-Y-D-X chain? Yeah, it's a good question.

I won't get too much into the legal stuff right now, but it is obviously something that we think about. I wouldn't say that was a major factor in the decision. Again, I think it goes back to building the best possible product. If you think about the definition that I sort of gave for what the best possible product is, though, it does include this level of decentralization. And I think that comes back to like what are all of us even doing here in D5. Like is decentralization important to people?

And the question like, let's be honest with ourselves, like we don't know yet. Like we think it is to us as like users, but we don't know if it's going to be really important to products on sort of a global scale yet that certainly hasn't been proven out. Or is it just this thing that like some nerds are really excited about because now they can be financially sovereign or whatever? You know, obviously, I think it will get to the point that it is really applicable and is the better way to build products on a global scale because it does have much better quality of access. It does have non-custodial features. It is much more composable. you can plug other technologies into it really easily.

And I think that that is the reason sort of like subscribing to that narrative is the reason and sort of also like being honest about where we were at on that narrative with DYDXV3. Like UIDXV3 could never like, you know, be at a point where it was fully decentralized because I've had this like centralized bottleneck. We really need to get to a point if we want to benefit from a lot of that stuff in the way that a lot of other people talk about like, you know, a Hayden for Uniswold. or something like that, where the whole system is decentralized. And I think that that's something that's important from a product perspective for a variety of reasons.

It's like, well, okay, like, why do you care about like a quality of access? I mean, I guess it could be legal, right? If your like state is trying to, you know, take away your bank accounts or, you know, whatever they're trying to do, I guess that's like arguably legal and therefore, like, devised answer to that. But I think there's a lot of reasons to be excited about decentralization. and we just like really wanted to start playing in that ballpark where we were fully decentralized and not sort of like you know like half acid anymore with this kind of hybrid centralized

decentralized decentralized system Antonio believe it or not we haven't even gotten to the spicy questions are you ready for the spicy questions yes I'm ready all right guys we're going to get to some spicy questions but before we do we want to thank the sponsors that made this episode possible there is a brand new staking feature in the ledger live app today we all like staking the assets that were bullish on, and now you can stake seven different coins inside the Ledger Live app. Cosmos, Pocodot, Tron, Algorithms, Tesos, Solana, and of course, Ethereum. With Ledger Live, you can take money from your bank account, buy your most bullish crypto asset, and stake that asset to its network, all inside the Ledger Live app. Through a partnership with Figment,

Ledger also lets you choose which validator you want to stake your assets with. And Ledger is running its own validating nodes, offering a convenient way to participate in network validation, and it even comes with slashing insurance. Ledger Live is truly becoming the battle station for the bankless world. So go download Ledger Live. If you have a ledger already, you probably already have it and get started securely staking your crypto assets. The Layer 2 era is upon us. Ethereum's Layer 2 ecosystem is growing every day and we need bridges to be fast and efficient in order to live a Layer 2 life. Across is the fastest, cheapest, and most secure cross-chain bridge. With a cross, you don't have to worry about the long wait times or high fees to get your assets to the chain of

your choice. Assets are bridged and available for use almost instantaneously. Across bridges are powered by Uma's optimistic Oracle to securely transfer tokens from layer two back to Ethereum. A token proposal is being deliberated as we speak in the Across Forum where community members will decide on the token distribution. You can have your part of Across's story by joining the Discord and becoming a co-founder and helping to design the fair, fair launch of Across. If you want to bridge your assets quickly and securely, go to across.t.to bridge your assets between Ethereum, optimism, Arbitrum or Boba networks. The era of proof of stake is upon us, and Lido is bringing proof of stake to everyone.

Lido is a decentralized staking protocol that allows users to stake their proof of stake assets using Lido's distributed network of nodes. Don't choose between staking your assets or using them as collateral in DFI. With Lido, you can have both. Using Lido, you can stake any amount of your ETH to the Lido validating network, and receive STEeth in return. SCEEth can be traded, used as collateral for lending and borrowing, or leverage on your favorite D-Fi protocols,

all this without giving up your ETH to centralized staking services or exchanges. Lido now supports Salana, Kusama, and Polygon staking. Whatever your preferred proof-of-stake asset is, Lido is here to take away the complexities of staking while enabling you to get liquidity on your stake. If you want to stake your eth, soul, or Matic, and get liquidity on your stake,

go to lydo.fi to get started. That's L-I-D-O-F-I to get started. Hey, guys, we are back with Antonio from D-Y-D-D-X, making the case for why they decided to abandon Ethereum and go in the direction of D-Y-D-X's own app chain. And I think in the first part of this, Antonio made the case very well and optimizing for those four features that he mentioned. UX features themselves, decentralization and security.

Antonio, I want to talk a bit more about the trade-offs in this piece. We promise we'd get a little spicy, right? And so here's kind of a take. I know you were talking about this being an optimist. towards making D-YDX more decentralized. But I actually think there's a case for the reverse being true, that this is increasing the centralization of D-Y-D-X. This is a tweet thread from David Phelps,

and I'll read a few quotes. I thought they were good. D-Y-D-X's chain will be able to do all sorts of cool shit to protect himself that would seem criminal to the crabby defenders of trustless, immutable code is law, Old Guard, of which I feel like I am one, a crabby defender of trustless immutable code's law,

Okay, and so examples of this, validator set in D-Y-D-X, it's a own app chain, it'd roll back transactions, vote for various validators, block thieves, maybe censor transactions, the worst case scenario, two-thirds majority, so to actually steal funds. Hell, it could take time off for the holidays, says David Phelps. When every project has their own chain, a mutability is no longer an intrinsic property but a social construct. Communities can roll back. I want to read that again. When every project has their own chain, immutability is no longer an intrinsic property, but a social construct. So to me, a crabby old trustless bankless maximizer, this is kind of a problem because immutability is the entire point, isn't it? And if we are just giving a set of validators, call them neo-banks. I don't know what you call them. A distributed set of people who control the order book, you're giving them access to all of the MEV, the order flow, the ability to censor transactions, aren't we reverting from the system that we left? And why is this decentralized in the first place? In fact, if they're voting based on some sort of token weight, isn't this similar to a shareholder vote? And so aren't we back to a corporation, like basically a bank, you know, structured shareholder corporation where the shareholders get to vote on what transactions go for it and what don't, what the business does, and what it doesn't,

doesn't do. Respond to that for me, Anthony. What do you think? What do you make of this? Am I just being a crabby old, salty immutability old guard maximalist here? Yeah. So first of all, I generally agree with you, to be honest. I think you make a lot of good points. And I think it comes down to what do you care about? Like, do you care about, like, is immutability decentralization? Like, yes. But then the question is what other forms of decentralization are there? And are they worth doing? Like, is it just like the same thing as like, okay, let's go use Robin Hood? And I can like buy Robin Hood shares on the public market. So like theoretically, I control Robinhood or something. I think it's pretty different, right? And I think there are different tradeoffs for using an immutable

system with one that is not immutable, but at least as controlled by the community. So if you're using an immutable system, you can 100% know that code is law. Like there's no way that it can be changed. Arguably, if you go all the way down, to the level of Ethereum nodes or Ethereum like validators, that's also pretty much just social consensus. But I would argue that it's a lot better like social consensus than what likely any one app chain is likely to get to. Like Ethereum has a long history at this point. So being like, oh, you like got hacks for $500 million. Like it sucks to be you. We're not rolling back the chain and like code is law. Like Dow hack aside, which I think was fine,

given how early it was. So, like, okay, I'm not going to sit here and say that, like, 100% like code is law, but on most D5 platforms, like code isn't law. And that's okay. And the platform should be transparent about this. Some code basically is law, like on uniswap, and I respect them for this. Like, the code of the smart contract can't be changed by governance holders, at least for their current smart contract.

But for other platforms, like even GYDX today, like, you know, not to call anybody out, but other, like, aspects of, like, compound or, like, AVE, and, like, a lot of other DFI platforms that are out there, effectively, the token holders can upgrade the smart contract, which is basically the same thing as controlling, like, the base level of validator set. So I think that there are tradeoffs of having full immutability, like, arguably, there's less, like, credit risk or whatever for using the platform, because you're not really relying on what some decentralized network or set of people is going to do, but it also, like I was saying before, it gives less sovereignty to the

platform and makes it that if there's a problem or you want to do an upgrade or something like that, there's just no way to do it. So I don't think anyone is objectively better, and I think there are tradeoffs to it. Obviously, we've sort of more chosen the camp that the system is upgradable. Yes, there are a lot of arguments as to like why this might, this simple case might not exactly be the case, but arguably we could get to a point where like the token holders are the ones controlling the entire platform, and that's how the platform works. So then, okay, like second question, like, is this just the same thing as a public company, like I buy shares in Robin Hood?

Like, is like DYDX different from like Robinhood at this point? And, you know, not just DYDX, I would argue any DFI platform that is controlled by some set of participants like token holders. I would say it is pretty different because you get to a point where there is code that's in the blockchain that does enforce that at least these participants can control the network in the ways that is publicly defined. Like if you buy a share of Robin Hood, like, it doesn't really matter what you think. Even if you buy like 10% of Robinhood, like Vlad's probably just going to be like, no, we're going to go like strike this deal with

like Citadel or whatever. And we're going to do payment for order flow and like, damn it, I don't care what you think. So I think it is like a pretty different construct that exists in like a public company because it's not laws that are enforcing the social contract and the social structure. There is code that enforces like that social contract and like this social structure. So I think it is pretty different. And I think the other thing I'll say is that it's just much more efficient. If you can get to a place where you're doing liquidity mining, where you're giving like stake to the participants of the network, the participants of the network ideally are like, you

know, holding that stake or participating in governance, that's just like a much better cycle and sort of like virtuous cycle, I would argue, than a public company. Like, you know, you're not getting like Robin Hood shares as you're trading on Robin Hood. You're not like literally getting to votes on like what markets Robin Hood should add. And empirically, that sort of world doesn't exist in in the current world of just regular old companies. But I think it could exist, even in a platform that is community governed. So I think it remains to be seen how this stuff plays out. So again, I'm not going to sit here and say, like, yes, it's the same thing as an immutable smart contract. But I think it's pretty close to that.

And also, I think there is really something valuable about community governance. Got it, got it. Yeah, I understand what you're saying. I do wonder if decentralized sequencers. Yep. Goodbye, Ryan. Under the assumption that Ryan just got rugged by his internet, I will take the next question. So one of my questions for you, Antonio, is you earlier,

talked about, well, if we do our own native D-Y-D-X chain on the cosmos ecosystem, well, then there's all these bridging risks for getting assets over to the D-Y-D-X chain. Ryan's back. But Ryan, I'm going to go ahead and finish my question. But that doesn't matter as much because we only have, we only really need a one good collateral asset, which are using USC, which is a trusted asset anyways.

But all the USC I've ever touched has all been on some Ethereum chain. and whether it's like from all the Ethereum bridges out there, I can easily get into DYDX in its current form because the bridge ecosystem exists and I can just get into DYDX with like just one or two transactions. Are you worried about losing a bunch of just like possible client customer inflows into the DYDX chain simply by just like not being as adjacent to the Ethereum ecosystem?

It's going to be more, there's going to be more friction getting USDC over to the DYDX chain. Have you thought about this? Yeah, we absolutely think about it. And if again, you should go back to, okay, I believe Antonio, he cares about his product, like let's build the best possible product. One of the really important parts of any product is onboarding. And what does onboarding mean for us? It means actually getting your money onto our system. So we do think about this a lot. And I would say this is, and I mean, I just talked about this. And one of my previous answer is like one of the biggest risks in something that we're still thinking about. I think we can build something that's really good from a product perspective. If you think from a product perspective, what do we actually care about? We care about people getting their money onto DYDX from the places they have their money. And we care about this for our users. So, okay, who are DYDX as users? They're traders.

They're like really advanced traders. They're probably people that are trading on like FGX on Binance. Like maybe some of them are trading on Uniswap. So the two things that we care about are actually onboarding to DYDX, first of all, and probably a bit more importantly even from centralized exchanges, because, okay, let's just be honest, with where the world is at right now, like most of DYDX's users, at least, like the active traders of the world have their money on finance and FDX. And then, of course, we also care about people getting their money onto DYDX from Ethereum. So there's kind of two states of the

world, like there's the world where there is like a native deployment of USDC on the DYDX chain or any Cosmos chain, maybe just a quick side note. One of the cool things about Cosmos is that between Cosmos chains, there's this really cool, actually decentralized and actually very secure bridge called IBC, where it's easy to actually move funds between Cosmos chains once it gets on any Cosmos chain. So like if any Cosmos chain gets USDC deployment, we could pretty easily and pretty securely use that. And we're working on that. I'm optimistic that'll happen on the timeline that we care about. But even if it doesn't, then we sort of fall back to using some sort of bridge. And I think there are good bridging solutions that we can use right now. And I do think there are a lot of

people working on bridges, like nomad, like Axler, to name a few, that we are in active conversations with. So again, like, I don't have all the answers here and it is a risk. But we will build a really great product experience around whatever the actual bridge is into the DYDX product itself. But I would say probably like the most ideal product experience for us, which again, like we're not every debt that's out there is like, okay, you go on like finance, you go on coinbase.com, you go on FTC and you're like, I want to withdraw USC. Okay, there's like a drop down menu. I'll just like withdraw to the the DYDX chain. And of course, we need to like business develop our way into some of those relationships. But I think that that is something that's achievable on on the timeline, maybe a year

or so from now that we care about. Do you know, I have another, just a follow up on kind of the idea of non-native assets on the DYDX chain. That's another thing. That's another thing. thing I feel like I'm going to miss, right? It's like I remember the original DYDX, it was on main chain. You could do a lot of spot trading with ETH, right? And so it was nice because ETH is a crypto-native asset that has no external dependencies, right? And the newer versions of D-YDX, I know a lot of the perpetuals are built with USC. What is USC? Of course, it is an IOU for some dollars in a bank account. So it's a much more trusted form of money. It's not like a crypto-native form of money. And I guess the challenge with

D-YDX app chain is it can't really have a crypto-native form of money, if that makes sense, right? So if we move ETH or some Bitcoin, let's say, to the D-YD-X chain, always has to be wrapped. There always has to be some dependency. There has to be a bridge or a custodian in order to get the funds there.

Even if it's through IBC, it has some dependencies, some security dependencies from other And so we lose that crypto-native aspect of having a decentralized exchange. And I guess, you know, not that it's necessarily the wrong product decision, of course, D-Y-D-X knows best. But I almost feel like a little bit the way I felt with, you know, when Maker moved to multi-collateral, it's like, oh, we're doing something new, which is awesome, but we're also

losing something. We're also losing something that was fundamentally bankless and fundamentally crypto-native. And maybe once again, this is the decentralized. maximalization maxi in me talking. But like I regret that we are losing that aspect of what D-Y-D-X has been, which was a bit more of a decentralized crypto-native exchange asset platform. And where I see D-Y-D-X moving is a bit more towards, again, kind of the not completely, but a bit more towards the Robin Hood, a bit more towards the FTX and that side of things. Do you regret that loss at all? Or is this just basically, would you say,

hey, Ryan, you know, it's cool that your decentralization maxi. Not everyone is. This is what the market wants. We have to go where the market takes us. Yeah, I would mostly say that. But I also would say that I agree with you. Like we are in some sense losing like the full like extremely like, you know, maybe arguably like nation state like decentralized version of like,

okay, let's all just have like BTC and Eath and let's make sure there's like tens of thousands of nodes on this network. Like yes, absolutely. it's a trade-off. But in some sense, we have to give the market what it wants right now. Yes, there are like decentralization and security trade-offs of using bridges, kind of talked about that being a risk, talked about us being at the forefront of technology to be able to mitigate some of those risks. But it is a trade-off. I think if you actually look at what the market wants, I would say probably the best example of this for us isn't even a defy product. It's Bitmex.

Bitmex did offer this product. They did offer, yes, of course, a centralized exchange, but one that was totally custodied and collateralized by Bitcoin. And there were no other collateral options available. And then, you know, that drove a huge rise to Bitmex's perpetual platform. And then Binance and FTX and some of the others out there offered the same platform in terms of trading perpetuals, but they offered it with stable coin collateral. And this turned out to be like the real killer use case for what people actually wanted

to trade. And consequently, if you look at the volumes of the platforms now, and I think this is a major factor contributing to this, it's almost all on Binance and FTCS. So market is spoken, like, this is what they want. They want to trade synthetic assets on cryptocurrencies using stablecoin collateral. But even if we think about what does the market actually want right now, if you look at like an FGX or a Binance, one of the really cool features that they have that we don't have actually, and we want to have at some point is what's called multi-collateral. So you could, this is sort of similar to like multi-collateral dye or something if you're thinking about it through a defy lens.

But it's basically you can come to FTX, you can come to D-YDX, whatever, with any type of collateral that you have or at least one of the supported ones. And usually this is a bunch of different types of stable coins, like could be die, it could be USDC, could be ETH. And you can use that as your collateral. And you can still hold it and you can trade. So that's like the ideal product that we want to do. build. I think we're a little bit of a ways away from getting there. Likely, the DYDX chain will launch

as a similar product to what exists on DYDXB3 right now in terms of just being single collateralized with multi-collateral being a feature that's added in later. And then, of course, that'll mean we'll have to start thinking a lot more seriously about what bridges are we using, like what's the product experience for that? How do we make sure it's secure? I don't have all the answers there. Hopefully the world will be in a better space in like a year and a half from now when we start tackling that kind of stuff. But I think in the meantime, we just have to focus on building the best possible product and giving the market what it wants.

But, you know, again, like bringing it back to the highest level, like what is our goal? Our goal is, yes, to build the best product that we can in the next year, but really to build the best possible product that we can five years from now, 10 years from now. And I don't think it's possible that we achieve our goal of becoming one of the biggest exchanges in crypto before that time horizon. And so at some points, like, you know, this sounds a bit hand wavy, but I think it's sort of the reality of the situation. If you're like a builder and you're a CEO, it's like, you got to take like a bit of a leap of faith and be like, well, okay, like the scalability stuff has to be solved for crypto like in 10 years from now.

Or like what are we even doing like some of this bridging stuff. Like it's very possible. It could be solved. If it's not, again, like we'll move to a different technology. But like there's some reasons to believe that could be solved in a pretty high quality way on like. that sort of time horizon. But okay, here we are in 2022. All we really know is what the state of the world is now. And we kind of know what it will be like next year or two and move towards that. Antonio, thank you for coming on on this extra long state of the nation, saying a few extra minutes

so we could get all the questions because we knew we had a lot for you for you. And I feel very educated and informed about what to look out for next. Just one quick set of questions before we let you go. I know you have a hard stop here. How many people on the team, like how many engineers do you have working on this problem? And like, do you have any sort of like roadmap for us? Like when, when can we expect the YDX chain? Yeah. So right now the DYDX core team is about 40 people with roughly half engineers. We have pretty much most of our engineering team working on this problem at this point. So have about 10 engineers working on V4. And we're looking to open source V4 by around the end of the year, which like I said, is a pretty aggressive timeline. But,

we're optimistic about going out and executing on that. Awesome. Antonio, thanks so much for stopping by. We really appreciate what you're building in the space and everything that DYDX has put together thus far has been absolutely fantastic. And so we're looking forward to seeing the next chapter. And yeah, I definitely appreciate.

I guess where can people stay tuned to find out more about this? Yeah, absolutely. So people can go to DYDX. dot exchange, which has all of the links to our socials and everything. If you want to tweet at me on Twitter, I've been trying to respond to everybody who's asking me legitimate questions. So at me if you want. And the last thing I'll say is, thanks, guys.

I really appreciate the great interview. I was tweeting this, but I told everyone bankless was the first podcast I wanted to come on, especially after this announcement, because I knew you guys were going to give me a hard but fair time of it and appreciate all the good questions.