Can you hear me all right?
Should be a fun little chat.
Yeah, I think we can just get started while things are kicking off.
But first off, congrats on the launch of layer 0v2.
It's been about a week or so now.
Yeah, I mean, I think post-testnet we're pretty confident.
I'd say the larger congrats here is the cluster's rollout going so well.
What was it like getting testnet out and then mainnet just this past week?
Any fun stories or surprises?
No, I think with a bunch of other launches, there were some crazy stuff.
But this one was like we launched testnet and then just didn't change or do anything for a month.
It was just more audits, more everything, but nothing changed.
So it's a good sign, I guess, evolving maturity to have it not be a mad scramble or anything breaking.
Yeah, that's a very, very rare feeling.
I'm sure I'll get to experience it one day.
Yeah, I thought we could kick things off.
It would be cool to just start with a quick overview of v2.
I think you're best paced to do that and then dive into clusters, the idea of universal applications.
What we've built here and what more could be out there.
Yeah, maybe to kick things off, do you want to quick feel about what layer 0 v2 does compared to v1, what you're excited for?
Yeah, when we announced v1, I said I know asking anybody to watch a 10-minute video is an act of aggression.
And so for v2, we did a 15-minute video.
So if you really want to know a lot more about it, you guys can go and watch that 15-minute video.
But no, v2 really, there's a huge number of changes, but the primary changes is this decoupling of security and execution within the network as a whole.
So what it means is applications can just have way more modular and easier security.
You can use anything through DVM adapters.
If you wanted to use IBC as a verification method within Cosmos, you can do that.
If you wanted to use wormholes, getting a lot of news today.
If you wanted to use wormholes validator set as a verification method, you could.
The network is totally agnostic to what the security piece looks like in there.
And then execution is now decoupled.
So no liveness risk, no anything tied to the executor.
So it's just like a much more modular extensible system.
It has all the same properties you would want of complete immutability, censorship resistance and permissionlessness.
But I think is what rails should be built like.
And I think so far has been awesome.
Yeah, no, I've been following very, very closely, obviously, since along to public test map.
But also, also before then, there are some really cool features that we couldn't find anywhere else.
Like being able to automatically natively swap gas tokens, kind of embedding, embedding DVNs of DVNs within your validator set that are very, very powerful.
And I think kind of let you launch with a basic easy to use configured security set and then upgrade that as needs be as time goes on.
So yeah, amazing work on that front.
And then, of course, we launched just just a few days after that, this past Thursday launch clusters, which is the first universal name service that lets you bundle up all your wallets, your ETH, your Bitcoin, your sole wallets, Cosmos and Cosmos and Cardano and code coming soon.
Using basically able to take any action from any chain and routing it to a single global state, which we're calling kind of a universal app.
Doesn't matter if your liquidity and your wallet connections on ETH, if it's on L2, if it's on another L1, you just get a seamless experience.
And so this chain abstraction has been super popular since go since go live over the past week, we've seen about this over 15,000 names registered and 4000 that 4000 of those messages have been routed through layer zero hasn't met a single hitch, which is a bit of a bit of a bit of a miracle.
These are the hardest are the hardest in front of it, right? But that's been performing seamlessly. So yeah, impressive, impressive work on that front.
What are your love to see it?
What are your thoughts on the idea of universal applications and global state chain abstraction more generally?
Yes, I mean, this is this actually this was another driving factor of like why we built V2 is that when we built V1, it was just like a very the world was very EVM centric at the time.
And we in our design patterns were very EVM centric. We just spend most of our time building in solidity like we had not spent a bunch of time really thinking through some of the interesting nuances as you move to other VMs.
And so as we got more familiar with Solana with Aptos move with sort of all of these other ecosystems, it has, we just learned a lot, we learned a lot from that process.
And for like the entire time, like for the last two years, we've just been being like some like somebody has to do this, you know, there has to be like from from a UX way.
It's just it's just such a nightmare if you've ever used any other bridge or anything, it's just just total nightmare to need to understand address mapping across every space, all of the domain to for like users to make mistakes on their end like it's just so obvious that this was needed
and like a key piece of just like consumer UX and just making the space more usable. And I think the structure that you guys did on basically just like passing right just this just this nice like primo slash is super cool.
I just think it's like that that is what I wanted to always think about, you know, any of us who have ever, I don't know, my kid on his computer, I don't think he ever clicks on the hard drive.
He just says like apps, the games that he's used to playing, but like any of us who ever use the computer prior and I didn't think about like system passing and just in general, like, I don't know, it's just the way that my brain works.
So I really love the layout in the system.
Thanks, appreciate that. Yeah, very much trying to work backwards from what's intuitive here, you've got a lot of systems, I think that technically, if you really stretch your brain and suspend disbelief can be considered to be multi chain applications.
Stick an arbitrary text record that that encodes arbitrary data. And sure, this is a, this, this can be a universal codec.
But in terms of what's simple and understandable, and you bring it in front of an average person within 15 seconds that clicks it makes sense like the.
Yeah, the clustering feels far more intuitive to me than subdomains you got primo primo is you. You've got a lot of detail you've got primo slash penguins primo slash ordinals primo slash defy.
And so on. So we've, we've worked really hard to make that a cohesive appealing consumer experience, one of the one of the fun things we did was put together an animated video pipeline, so that everybody who registered their own cluster gets a full 20 seconds
video kind of detailing detailing detailing their wallet addresses their names their profile picture, how it how it fits in, and a ton of people have been sharing that on Twitter, it's been it's been fun to see.
So, yeah, appreciate that and love your love love yours your penguin I think is distinct distinctive, you've sparked a trend with the wizard hats.
Yeah, wizard wizard hats are elite and I downloaded and shared my video as well. So I like it. It's a sweet animation.
And also, yeah, it's really interesting. How do you encode kind of generalized this applies to the bridge side, but we've also run into it on the outside.
How do you encode like a generalized format for all for all addresses, you've got either the ends of 20 bytes, Lana, Lana 32 bytes.
Bitcoin has eight different legacy formats with various forms of adoption. You got it, you got to decompress so doing, doing that well I think that's the standard for the whole space.
Maybe let's, let's dive in and talk about a, the kind of a hub and spoke replication model, and how that can help apps include including us but also others replicate replicate global state across chains, while still getting good.
Good consensus guarantees.
Oh, yeah, I can. I need a brief overview overview and then would love your would love your thoughts on that as well.
Let's do it. So yeah, the key, the key question people ask about clusters all the time is what chain is done.
And the answer all of them is not is not very satisfying because what if two people by the same name on on two different chains at the same time, like is is polygon or avalanche faster.
What we've come up with is this. It's a it's a computer science term hub and spoke. What essentially means you can kind of think of a think of like a solar system, where you've got one, one circle, like the hub in the center, and then a bunch of other other circles
And then you can you can take any action from any of the any chain, any of the planets, any of the spokes, and then it sends a quick message to the hub.
The hub then orders finalizes handles have handled things like who bought this name first, and then relays all that information back out to the to the spokes into the planets at once.
And it's kind of a neat way to inherit, inherit security of the hub, no matter where you're starting out and deal with deal with things like reorg risk effectively.
Yeah, we've we've been we've been using that with, or we're going to be rolling that out with clusters for really efficient, not just not just an issue multi cross chain initiation, which we have today, but cross chain state replication.
So this can be not just an off chain tool that used and used in composable apps too.
I'm curious, Brian, if you've seen anyone else, anyone else taking this approach or how you've seen people tackle the global state problem.
Yeah, yeah, I mean, I think there's there's a general sort of paradigm shift as people start to develop applications that have a state that's like sharded across a bunch of different chains and have to deal with a synchronicity, right.
So one of the nice things when you're on a single chain is just like every possible person doing a transaction every every action sort of like knows the current state of the world at the time of that, when you think about even a small cluster of three chains.
You know, to two people on two different chains both want to buy primo slash or want to buy whatever thing they're trying to buy like who gets it right.
How do you settle a dispute? How do you deal with race conditions? There's all this sort of stuff that happens or a bunch of things are happening at the same time, and eventually you're going to have collisions.
So there's, there's a couple of different ways that you can handle that. One is that there's there's a sort of single source of truth, which is one chain that everybody basically needs to go in and ask for permission from.
So that keeps a total ledger of all, you know, every, every record, so every, every registration, let's say happens there.
So if I want primo slash, then I go and I tell this chain and it says, great, nobody's registered it yet. You get it. And then the next person coming in who communicates with that chain slower, they don't get it.
That's just very clear. Whoever ever talks to it, whenever it's written in that ledger, that's that sort of state of truth.
And then the other much harder way is like, if you wanted to try to maintain like a global state being replicated everywhere, you still could have like a single source of truth that's pushing an updated state, but you're gonna have a little bit more collisions for latency,
and you're gonna incur like a really high cost of needing to write to all these chains.
You can imagine this chain needing to like, push out every update, it gets to Ethereum from all of these other chains, like that's going to cost a lot versus Ethereum, just like querying ad hoc when it when it needs to know something.
And so I think it's like the intuitive model. I think it's what a lot of people move to when they when they have something that needs real unified state, I think there's, you know, you can do something like Stargate, which is like there is no each chain has its own state.
And you can give guarantees of the system as a whole, like credit allocation is sort of has this trade of solvency and you know, the global state will meet one certain thing, but no individual node needs to actually know or care about global state.
But it's much harder when you're trying to register something or buy something or you like need that permanent state.
Yeah, it's very interesting different approaches for both contentious and non contentious state. If I'm if I'm bridging a token from A to B, then nobody else can really front run me there.
But if I'm doing a token swap, or I'm doing a name registration, then you then you do start to have these contentious issues.
So technically, in the old model pre Stargate, they could front run you because people would either use both they use wrapped assets where you couldn't but otherwise they were just using naked pools of liquidity and so like, you could go and be like, I want 500k but by the time your message got there, there isn't 500k on that chain to give you anymore.
So I want to like the novel things of Stargate was needing to do this thing called instant guaranteed finality, which just guarantees that each credit has it. Each chain has its own state and can guarantee on each pathway that you will like you will have enough when you arrive.
Without that there there is some contentiousness which which would make experience suck a lot.
And so let's talk security stacks for a bit. The, the one has been a little maligned for has been unfair unfairly characterized and maligned the v2 is far more powerful in terms of being able to configure sets of arbitrary security actors, arbitrary
What are the. Yeah, what are the approaches that you might recommend for like a very, very simple app that doesn't maybe need nation state level security and an app that does.
Yeah, so we've always, we've always taken the stance of the protocol should be like pretty unopinionated so the most obvious example of this is just like listen when you're an application coming from layer twos to layer ones, like you need to decide do you want to bypass the roll up period and go faster,
or do you want to wait for the role up for the, you know, fraud proof here the seven days and so the every there is no reason why the protocol itself should tell you you must do one thing or you must do the other if you as an application say I only want to wait one block from this chain or whatever like if a reorg happens your application suffers the rest of the network doesn't suffer nobody else eats any pain you don't affect other applications it's entirely your choice and so we really have tried to take that approach of just like everybody sort of
can decide and the industry will sort of coalesce around what best practice are and what what this should look like and what each application has demands for security wise and so everything you can do in v2 you could already do a v1 it just wasn't easy enough so v2 makes it much easier I would say for a low like a just a vanilla application, you're probably going to want, you know, at least, you know, let's say two dvns or depends on if you're I mean, if you're really dealing with like no value, you can just go a single one, but if you want to
you don't want to choose a couple their defaults right now the defaults on most pathways are either layer zero in polyhedra zk like client or layers or labs in Google Cloud, but there's about 20 15 to 20 dvns that are that are rolling out now. And then like I talked about you have the, you know, adapters. So I'm going to be able to use all the l2 native bridges, going to be able to use third party validator sets, you're going to be able to use all of that. So I think that the real thing people need to consider is like how
many do you have? And do you want to run your own? And I think that's probably the most interesting part about all of this is if you are dealing with something that is hyper security focus. So like Uniswap right now, they just do governance, right? It's just like turn on the fee switch or not. But you can imagine where Aave or Uniswap, we're actually doing something that's like exposing the underlying pools are exposing, you know, tens of billions of dollars. What they probably would want is they want some backstop, they want some external. So maybe you want to
be able to native bridge and you want to use the other sort of most secure dvns that have economic security tied to them. But then you also probably want to use yourself. So like Uniswap could actually recreate, you know, they can basically make a dvn that is effectively their governance of this delegated governance. All of these delegates effectively can comprise this this dvn. It can be stake weighted by how much uni tokens they hold, they're the most aligned with the protocol itself. And that's always a backstop saying,
even if everybody else colludes, every other dvn we picked, every the entire network, every person is malicious, we can always veto it. We have last rights, we can't independently do anything malicious without everyone else being corrupted as well, but we can veto everything. And so I think there really is just like a super wide spectrum when it comes to that.
Yeah, I love that. It feels like for credible neutrality and a protocol that's really be a hyper structure that lasts a decade plus, you have to be able to let people configure their own setups on board, on board with the quote unquote student package, get a get a feel and then upgrade to nation state level security when time is right.
Thanks for that really detailed overview. There's so much rich richness and depth and the two I'm sure the world's going to see is as more and more apps roll out on it.
Shifting shifting gears a bit to clusters. I'd love to kind of get your half half insider half outsider take on how on what it what excites you about things. How did you kind of see that see the name, see the namespace fragmented world beforehand.
What do you what do you feel like the biggest needs are here and what is what what about clusters fits into those
Yeah, for us, it was when we were thinking about building EVM to Solana. So just like Ethereum Solana, we thought the experience was really brutal of just
needing to have, you know, needing to manage all of these different address sets and know like, I need to know, hey, Fubar, you are like
x and y and z and like all of these different addresses, depending on like the chain or VM that I need to send you something on and there can be
this process where it's like continually need to ask continually need to do everything like what a consolidated namespace does is I don't need to think about it right if you have like a reasonable mapping to how you handle these
addresses. I'm just like, oh, like Fubar slash eat Fubar slash solid, whatever, whatever you have this internal mapping and I can now just send without thinking about it and know that you will have updated like if you lose your keys, if you don't have control of that address anymore, of one of these, if you lost your solid address, like you'll, you'll probably update it, right. And so the kind of the burden shifts from meeting to ask you every time and have all of this friction and be like, hey, what is your address again? Is this still a good address?
For every setup to just know and have everybody know on that. And I think it's way nicer for I mean, you think about your it's it's airdrop season right now. So everybody is, you know, after dimension, you know, all the pudgy penguins got this, this airdrop and you have all these other things coming out. And so, like, how much easier is it for a protocol to think about who who do I want to, like, reward or incentivize who are the users who I care about and and not needing to go and like manually
track down addresses and like maybe you know who they are on on Twitter or like some social proof tied to it. Like now it's much cleaner where it's like, oh, like primo slash penguins. Like I say, I know who this guy is on Twitter. I see he's like one of the huge holders. He does whatever x and y. Like, I know there's leaderboards and all these other stuff that a lot of projects do. But I think it's just so easy to consolidate all of that to make this combination of like consumer ease and social proofing
and just all of it just just ties together so nicely. I don't know. I find it super interesting space.
People will will spoof token transfers to you now that look like your address better or not. So you have to kind of go to this off chain off chain third party backup of of how have I labeled my addresses. And most people don't have the most don't have the bandwidth or for even if they have the bandwidth, they don't have the desire to waste their day on these menial tasks. If I could send from FUBAR slash E to FUBAR slash soul and have that automatically recognized by by a bridge.
I mean, that that will be the case as soon as we get these integrations rolled out rolled out. I think that's it. That's a huge UX unlock.
And on the I think that universal profiles are an interesting use case of like you've got endless endless points leaderboards and whatnot nowadays. Ideally, these are ideally we solve the civil problem and people can just can just link their stuff together.
For unified profiles on on open sea on blur on on the on a nazuki collector profile on new web three gaming and whatnot. But you also brought up a super interesting point about kind of universal communities you want to you want to target you want to target the penguins say
how do you it would be good choice. Yeah. Why would why would you even do anything else you want to target the you're on target the wizard penguins with with a Superman or super penguin. What's he named G man pudgy. Yes, with a wizard hat and a pudgy man cape and it I think it'd be very clean to be able to go in and just get a unified ad get a unified address list of all the pudgy.
And their communities. What are their addresses. What are their soul addresses. What are their what are their cosmos addresses with the recent dimension drop. That was I mean thankfully thankfully it was a dual cosmos chain. So you had both cosmos addresses and the addresses. But a lot of people got stuck like trying to get it out. And if you're trying to do an airdrop to the pudgy penguin community on Solana. There's no good way to do that today.
Ideally, that's the sort of problem. That's the sort of problem we can to make NFTs truly on the chain. Then I think clusters has a large large role to play in that as well. No, don't need to go track people down on this cord anymore. You can simply scrape some on scrapes scrapes some easily accessible on chain and API data.
Yep, 100% hopefully a lot less people getting getting fished and I delegate helps with that a bunch but like still happens all the time.
Yeah, I mean that delegates all the subset of these use cases, but ultimately there is. Yeah, delegate delegate helps the case where you're claiming from a cold wallet into a hot wallet.
But there's a lot of cases where that delegate was built in its initial instantiation to be carefully scoped to a single chain. It was actually so we've got I mean we're on 18 different AVM chains at the moment.
But if you if you delegate on the then you have to go re delegate on Solana. It's actually that precise UX issue that led us to thinking about and working with layer zero is actually you need, especially for consumer friendly and consumer facing things.
You need truly global state not not 18 replicas but but one one thing for all your wallets for all your chains.
Yeah, yeah. Yeah. Well, awesome. I think we can keep this keep this short and sweet. Any final thoughts, thoughts, parting words, topics you want to go over before we wrap we wrap things up.
Actually, I should know this, but I but I actually don't when when do the when did the cluster auctions finalized when when I kind of when it kind of be certain that I own primo slash.
Yeah, so we're wrapping those up in max four weeks, considering we really want to target a target a fair distribution period here, make sure that first come first serve didn't like take, take your primo slash.
But I think it may, I think it makes sense to wrap those up sooner rather than later. And so we're also exploring kind of rolling rolling rolling closes while we finalize the full full clusters launch and full clusters architecture.
Well, but yeah, it's excited for it. Yeah, should have lots of first first bitter boost and whatnot from sharing as well. And hopefully the hopefully the penguins too.
Great. Well, thanks so much for thanks so much for hopping on. I think we can think we can end here and
Yeah, great, great to learn more about layer zero. Thanks for sharing your thoughts on on clusters and have a great day everybody.
Anytime. Thanks so much. And yeah, see ya.