podcasts
Collection
Collection for podcasts • 511 items • Updated • 2
speaker stringclasses 3
values | text stringlengths 6 2.51k |
|---|---|
A | Welcome banklessation to the frontier of universal synchronous composability. That is once again the topic that we are talking about here on the episode today. We got Brendan Farmer, who is the chief espouser of the AG layer out of Polygon, along with Ben Fish, the CEO of Espresso Systems, the shared sequencing marketp... |
B | Thanks. Great to be back. |
A | And then also we got Brendan, the chief farmer, the aggregator over at Polygon. Brendan, welcome to the show. |
C | Thanks. You know, I've never farmed an airdrop. That's like, so embarrassing to but you've collected airdrops. |
A | You just never farmed them. Yeah. |
C | Brendan Airdrop farmer would be a good. |
A | Brendan airdrop farmer would be hilarious. Yeah. You can give your first kid your middle name. |
B | Start with a no. |
A | Okay, so we've got two different camps here. We have a shared sequencer, and we have a layer two framework on the show today. We're going to talk about a little bit of a collaboration, a little bit, something you guys got cooking up behind the scenes. But maybe, Ben, you can kind of get us kicked off here. What is a sh... |
B | Yeah, so one thing is that actually espresso is not a shared sequencer, but it's more of a platform for shared sequencing. In fact, what espresso provides is a marketplace, or coordination mechanism through which chains can sell their or transfer their sequencing rights to third parties. Now, when a third party buys up... |
A | Brendan, since you're on the ag layer side, how do you receive the whole marketplace for shared sequencing? Just from your perception? Give us your reflections on this. |
C | Yeah, I think it's a really great idea. The way that we look at it is the AG layer is not an interoperability solution. It doesn't do message passing. It doesn't do shared sequencing. What it does is expose a shared bridge and provide a safety guarantee for rollups that are using that shared bridge that are interoperat... |
A | So I want to pin down some basics here before we start to get into what's net new. Brendan, you said the AG layer is not an interoperability solution, but maybe I'll throw something your way, is that it actually reduces the need for interoperability solutions. Bye. Of aggregating, consolidating a lot of all the liquidi... |
C | Yeah. So maybe saying it's like not interoperability solution is maybe the wrong framing, but I think you're totally right. A nice way to think about what the AG layer enables is it provides all these safety guarantees, but with a shared bridge and shared liquidity, it basically eliminates the need for market makers an... |
B | I just want to say, I think Brendan brings up a great point and that maybe we could get into what interoperability is, because there's different flavors of interoperability. And I think the crux of what Brendan is saying is that there isn't one of these elements that is a full interoperability solution. Interoperabilit... |
A | Yeah, that's exactly where I want to go. I want to first talk about the idea that the ag layer is meant to spawn further chains. Right. Which is going to create the need for what, Ben? I think you're talking about. Right. Whereas there's going to be a whole space between these chains. But the first and foremost, what t... |
C | Yeah, I think that's a good framing. Yeah. |
A | Okay, so we have the ag layer, then you have the spawning of many. I like to see these networks in physical topologies. Right. So you have the one central Ag layer, centralizes all the liquidity. It minimizes the need for a lot of, like, message patching, as you said. But then, Brendan, you said, like, well, it also. W... |
B | Just a question, I think since we're all having kind of a conversation about this, basically on the fly, when you say that ag layer provides the shared liquidity, are you referring David to the shared bridge specifically, because the way that I view it, ag layer is actually doing a number of different things. And one c... |
A | No, that's super helpful, because, like, I was totally naive to that, to that element. So, like, maybe we can actually talk about where liquidity lives and how it gets passed around and ultimately how espresso will integrate with that. |
B | First of all, I would say that the shared liquidity can take on different kinds of forms. There's various forms of this. So one form of it is when two different chains on the layer two of Ethereum have a shared bridge, which means that you move, let's say, eth into this bridge, and then that ETH can be used on one chai... |
A | Maybe. Can I try and, like, just re articulate that just for listeners? Because, like, sometimes, sometimes listeners are only listening with, like, 80% of their brain power. |
B | Absolutely. |
A | So, like, you have Eth on the layer one, and you want to get it to arbitrum. There's two ways to do that. You send it over to arbitrum. That's one way you can also send it to optimism, and then you can send it from optimism via another bridge to arbitrum. And now it's actually, it's in the optimism bridge, and now it's... |
B | And that often affects fungibility, so. |
A | Totally important. |
B | Good. So that's one aspect of, like, I think, what people talk about when they talk about shared liquidity, then the other thing about sharing liquidity is you could have, so you can always move assets from one chain to another. It's just very slow and might go through the l one. You can have liquidity providers that w... |
C | Yeah, I think that's right. I. Yeah, I would just echo what Ben said and say the way that we think about ag layer and the important properties that it provides with respect to shared liquidity are asset fungibility, which is the ability to, as Ben said, be able to move assets without relying on wrap synthetics. Anytime... |
A | Something I just want to parse out. Ben. It sounds like all of the cross chain intents, cross chain swaps, bridging, all of this stuff that's fulfilled kind of creatively, uh, by market makers. Like across, for example, it's a bridge, but it really just like, uh, auctions off the rights to a market maker to facilitate ... |
B | That is correct. And however, and with things like ag layer there, I believe is more that can be done. So the ability for two blocks to be constructed on two different roll ups at the same time, and for them to like as they're being constructed, basically communicate with each other through multiple rounds of messages.... |
A | Okay, checking my understanding here, Espresso has its product offering for the Ethereum roll up space like marketplace for shared sequencing. We can come and create the greater than the sum of the parts if we all agree to come to this common place to build our blocks and we can all get better execution. |
B | Yeah, you can sell your sequencing rights on the open market, and somebody who's also sequencing for someone else might buy yours because they can then act as a liquidity provider. Like across. |
A | Right. And that's like your basic basal level feature set that you're offering the layer two landscape. And it sounds like with the addition of the Polygon AG layer, there's like, ag layer exposes, like, more surface area for espresso to be, like a better version of itself because it has more optionality with some of t... |
B | Yeah. The existence of ag layer creates an even stronger reason for people to act as shared sequencers and participate on the. On the buy side of the marketplace. Right. Like, the marketplace has a sell side, which is roll ups, selling to third party sequencers. And it has a buy side, which is a third party sequencers.... |
A | And, Brendan, there are ag layer integrated or Ag layer enabled roll ups. Correct. It's like kind of binary you are or not integrated with the AG layer. |
C | Yes, yes. |
A | And so this infrastructure, the shared sequencing marketplace from espresso plus yaglia together, when you combine those things, like, I know, like, the idea of Justin Drake's, like, universal synchronous composability, like that itself, I think Ben definitely alluded this to. It's not like an actual, like, it's not so... |
B | Right? |
A | Like, it's not. It's a concept, it's ideas, a design principle. There's no actual binary of, like, you actually now have perfect synchronous composability. It sounds like these two technologies together get you, like, really, really far. Is that the main purpose for this? |
C | Yeah, I think it's exactly the goal. And obviously, I'm a little bit biased about what the AG layer can provide, but I think that there's a really objective case to be made that the ag layer is this shared bridge, minimal safety guarantee. Plus, espresso gets you universal synchronous composability. |
A | In terms of just, again, I like to look at networks and topologies, the tree of chains spawning out of the ag layer. The webbing between all these chains is this shared marketplace that espresso is bringing together, and that does like the super fast, low latency communication between all the different branches on the ... |
C | Yeah, I think that's a good metaphor. I don't know, Ben, if you have an objection. |
B | Yeah, I think the one, I'll sort of speak for Justin here, since he isn't here. But I think when Justin talks about universal synchronous composability, he's really focused on universal synchronous composability with the l one itself. And that's not something that ag layer enables, because the l one, you can't really, ... |
A | So zoned universal synchronous composability around the ag layer. |
B | I'm always afraid to use the term universal synchronous composability myself because I'm not entirely sure how it's defined and I may misuse it, but whatever it is, not to find a but this kind of synchronous like composability that we're talking about here, enabled by ag layer and espresso, is between l two s and l two... |
A | Yeah, yeah. My model for like, understanding how ethereum will achieve, like, global universal synchronous composability, not to add another word onto the acronym, is that like, well, all the chain constructions, the frameworks will figure it out internally first. So the polygon ag layer will figure out how to discover... |
C | Yeah, I think that's right. I think that's the most likely scenario for how everything plays out. But I will say that without naming names, there are some, especially forward thinking l, two ecosystems that I think are more amenable or open to the idea of using the AG layer, because they understand that it's like, it's... |
A | I think probably one of the reasons why everyone's still a little bit tribal about their preferred solution is that we're all talking about it, and we haven't actually seen them be actually in production out in the market yet. So maybe we can turn the conversation there. Brendan, how do people get their hands on the us... |
C | Sure. So the first thing I will say, I might get in trouble with the polygon marketing department. |
A | Nice. I prefer that. |
C | But the AG layer doesn't exist in the way that we're talking about it now. There are a few chains that are hooked up to a shared bridge, but all the things that Ben and I are describing about the ag layer enabling low latency interoperability and providing all these safety guarantees, that has yet to launch. And we're ... |
B | Yeah, well, maybe we can talk independently about your. Yeah, so I guess that's AG layer for Espresso. We're launching first, you know, production. We've been, we've had versions of Espresso and Testnet, you know, for, for over a year. And we just released our fifth Testnet called Cappuccino, that's supposed to be the ... |
A | Ben, can you kind of share maybe just how difficult or not difficult it is to integrate with the aglaya? I know, like, everyone is still in their engineering phase at the moment, but, like, with the product that espresso has ready out the gate, is this like, trivial to integrate with the AG layer? Is this like an engin... |
B | Interestingly enough, integrating with AG layer and integrating with espresso are sort of true two different things, but that touch similar parts of the code of a roll up. When a roll up integrates with aglayer, it needs to make certain changes to its CKVM contract. Similarly, when a roll up integrates with Espresso, i... |
C | Yeah, I think that's exactly right. For the AG layer, it's modifying certain aspects of ZKVM and what it's proving and what it's exposing. But, yeah, I mean, certainly Polygon CDK will have support out of the box. And I think for us, there's no reason not to support other stacks and other ecosystems. |
A | So is it really up to the roll up to do mutual integrations with both the AG layer and the espresso for the synergies between these two things to blossom? Or does espresso actually have to talk to the AG layer? Or is it once they're both integrated, just the roll up gets all the fruits of those mutual synergies? |
B | Yeah. So I guess to go in a bit more detail into what needs to change in a roll up contract to be compatible with Espresso, it's today roll ups tend to have one sequencer, and the roll up contract will only accept transaction blocks that were built by this one sequencer. So it's like a hard coded address that, that is ... |
C | Yeah, no notes from me. I think that's a good way to look at it. |
A | As you're summing some of this up, I just got out of a podcast with Anatoly, so I'm in a little bit of a Solana mindset. And, Ben, you said something that I want to go back to, where if we're building a block on a polygon chain or an optimism chain, and they're integrated with the ag layer, then you're able to co build... |
B | I'm honestly not familiar enough with the Solano ecosystem to maybe carry this analogy all the way through or follow it completely. I will say that espresso doesn't necessarily speed up the construction of blocks that's happening in these roll ups. The way to think about it, I mean, first, there's two components to esp... |
A | I think maybe it's about like who, who's the actual block builder behind the scenes here? And I think with Solana it's like, well, you got to be a pretty damn juiced up box in order to really do your role. And I'm assuming that that's also generally true of whoever's going to be doing some cross roll up block construct... |
B | Yeah, I mean, I will give one slightly spicy take, maybe, which is that. So I think ultimately all chains, whether they're l one's or l two s, will participate in some kind of marketplace for sequencing. And that's just my particular view. You know, we talk about sequencers on l two s and we talk about l one's. And it'... |
A | Right, right, yeah. |
B | And today they're generally the sequencer for Solana or for Ethereum is randomly elected from some set. But that's not the only way to do it. And likely a more economically efficient way of doing it is to actually sell the sequencing slot, not randomly elect from some set. That has benefits for reducing mev extraction,... |
A | Brendan, were you going to say something? |
C | Yeah, to sort of back up to those, like, the dynamics. Like, I think it's the case, like, it's important to emphasize, like, not only is the sequencer picked from among the validator set, but, like, the sequencer, like. Like that. That implies that, like, the sequencer, or, sorry, that validators must have, like, the s... |
A | Totally, totally. |
B | I think it would be. Personally, I think it would be really cool if we see a transition from l one to using the same technology that's powering l two s as well. And I could see that happening. I could see l one s changing just the way that they do sequencer election and also becoming things like ag layer aware. You cou... |
A | Right. |
C | Well, like a very simple, like, consequence of this is like, you start to ask the question like, why does every l one not settle to Ethereum? Like it's a serious question because, like, more assets are issued on Ethereum. The bulk of the value in the crypto ecosystem, excluding bitcoin, is on Ethereum, and it's from Et... |
A | Maybe we just gave some listeners a taste of the very, very far off future. But first we need to see the synergies between the ag layer and Nespresso produce localized, universal shared composability, universal synchronous composability. Then we'll get it to Ethereum, and then we'll get it to the entire crypto industry... |
B | Of course, we're interested in hearing from everyone, but first and foremost, developers of chains who are building either existing l two s or looking to build new l two s and looking for ways to be as interoperable as possible so that their users ultimately can enjoy the same experience that they would enjoy if this w... |
A | And Brendan, same question. Who do you want to hear from? Who do you want to talk to? |
C | Yeah, so I think, obviously chain developers to echo Ben. But I think also, for all this to work, there are a lot of different players that are required to really deliver an amazing ux. And so I think that's while providers, I think it's block builders. I think it's like intent solvers. There's sort of like a lot of. I... |
A | Pennyless nation, if you would like to talk to either of these two gentlemen, there will be Twitter linked in the show notes in the YouTube channel or wherever you are listening to this podcast. So, Ben, Brendan, thank you for coming on and showing us a little bit of the future. Appreciate it, guys. |
B | Thank you so much. |
A | Bankless sanction. You know the deal. Crypto is risky. Defi is risky. Layer twos, they're working on it. They're still risky, though. You can lose what you put in. But we are headed west as the frontier. It's not for everyone. We are. We are glad you are with us on the bankless journey. Thanks a lot. |
README.md exists but content is empty.
- Downloads last month
- 4