Solana as NASDAQ 2.0? With Max Resnick and Logan Jastremski

Transcript

We still have another thousandx to go. The goal is to get more volume than NASDAQ every day. On the average case day, we're surpassing NASDAQ. The goal there again is equal and access information to that data. It's about creating fair markets. Max put out a post with a a nuclear target of Amsterdam that just end the Salana network. What are we talking about here? Can you even theoretically get that information onchain cheaper and faster than a centralized NASDAQ? Welcome back to the rollup, your home for technical talks and open conversations that shape the modular future. Nothing on today's show is to be construed as financial advice. You can see our full disclosures

list at the rollup.co/ventures. Sit back, relax, and enjoy today's episode after a quick word from some of our awesome partners. Redstone is DeFi's modular oracle. Supporting over 100 clients on over 50 chains and offering 1,000 assets. Redstone is the leader for Bitcoin staking and yieldbearing assets. They offer the most robust customizable price feeds available in any model. Join the Redstone exhibition today.

[Music] What's going on, guys? Welcome back to the rollup. Today, we have Logan and Max sitting in the same uh closet together in Miami. They pay 5K a month to share a closet. Uh this is what happens when you build on Salana. So, uh guys, welcome to the rollup. It's uh your first time, Logan and Max. Good to have you back. Coming out swinging, Andy. Good to be back. Lo, Logan, perhaps a quick intro about yourself uh for our audience. I know that you have an awesome podcast which I'm a fan of. Um but how did you get into the space? What's your kind of main priorities and who's Logan? Sure. I I'll keep it brief. Uh I think we have

more fun things to talk about, but ultimately started out my career in the Bay, ended up at Tesla running the back end of the supercharging network on the software side. got really into crypto um really in 2017. Thought I was smart, lost a lot of money 2018 as with most people. uh but got really interested in DeFi and DeFi summer really early around what was happening with composability but I paid $1,000 for a unis swap and I thought ETH is probably not the future of France and started looking at really what was kind of like the underlying metrics to or kind of like more of the physics aspects of scaling blockchain.

um kind of taking that Tesla first approach of first principles and the more that I kind of like dove deeper and deeper I just kind of became high throughput build um Sono was the first high throughput chain uh so ultimately started the podcast just to talk with smart people one thing led to another ended up starting frictionless capital where now we're really focused on I'd say the high throughput blockchains were loud vocal supporters of Salana and the bare market after FTX Claps wrote an article about kind of fighting the FUD at 11 or 12 bucks. Um, but yeah, I I think at the end of the day, what we really want is just applications at scale. Um, and so we kind of pivoted

outside of the Ethereum ecosystem because we didn't think you could actually make cool things at scale or it was going to be much more difficult. Um, so yeah. Yeah, I think that's a common goal here. Uh, we want applications at scale. Um, and uh, yeah, welcome to the show. Max, you've you've been on before, but I don't think we've you've been on since your migration away from the Ethereum ecosystem and towards the Salana ecosystem. Maybe you could talk like why and where where you are now. Yeah, so uh, I was pretty vocal in the ETH community for a while, partly just because I thought ETH was not scaling.

And like Logan, I believe in applications that have more than 13 transactions per second. I think that's kind of the goal that everybody has. And I think the rollup road map was not my cup of tea. I wanted to scale to L1. I was a really strong advocate for that. and ultimately realize that no matter what arguments were put forward or how smart you were or how good your arguments were, there was basically no way with the structure of Ethereum that ETH could ever scale to even a fraction of what Salana is today.

And that's not even to mention what the goals are for Salana, which is 100x, a thousandx from here. And so I just kind of like said, "All right, if I want to continue working on blockchains the way I think they should work, then ETH is not a good place to do that." And happened to be chatting with some of the folks at Salana and in particular at Anza, which is the we develop the main client that runs Salana today because fireancer isn't on mainet yet. And it was a good fit. So decided to go over there and work on bringing my vision of what blockchain should look like to life in a more realistic setting where some of my

ideas might actually get into the chain. Yeah. Yeah. I think uh in our last conversation Rob called you a super villain. So I think it was a bit unfair. I think you're more of like a uh like a I don't know. Rob, what do you think? I'm pretty sure Max agreed with my take that he is an Ethereum super villain. I mean, it was funny. Uh my uh co-founder in the fund, Diego, we we got called all sorts of names at Salana at the bottom. And like I I think for me as like kind of the longer and you're in the space, the more you kind of just like tune out the noise. And so I think it's bullish that Max is getting called names now

because I mean he's kind of on his arc to do badass [ __ ] And I don't know, you kind of have to be a little bit of a mover and shaker. And if people aren't uh calling you names, you're probably not on the the correct track. If you want to do good things, make friends. If you want to do great things, make enemies. That's kind of Yeah. Well, I never really understood why I was a super villain on Heath because if you look at what they're all saying now, it's exactly what I was saying nine months ago. I just gave him a bit of a warning that that this was all coming. Uh I said, you know, Salana is the first real competitor to ETH. While ETH still had 80% of transaction volume and revenue and everything and was doing fine on metrics, 6 months later, it was

surpassed by Salana. They all laughed at me for saying that, but it was completely correct. and and within six months it was obvious and and now they see what I saw. I just happened to see it earlier than them and now they all agree in many cases they're trying to do the same things I was advocating for. There's a new EIP for uh having minimum validator requirements which was something that I got in a ton of trouble for talking about for some reason and now it's mainstream consensus opinion. So uh in some ways like when you're way early to something, it can be really uncomfortable for people because they don't see what you see. But I think now everybody's starting to see and that's a

really positive direction for ETH that everybody's waking up a little bit. Although I don't know if it can save it. Max, I think you already know this, but like it's all it's all love. Like I think actually a lot of the things that you were doing were actually the right things to call out at the time. Super villain, maybe not the right term. Maybe more of like an anti-hero if if that's the right term. like you're you're calling out the things that like no one wanted to hear but like that's the first step in actually solving the things is you need awareness into what the problems are right so that you can then go and solve it Ethereum people were I think a little bit a little bit sheltered and jaded at as to like any kind of competitors didn't even want to recognize the competitors in the room

but you you were just kind of calling those things out and then ultimately I I mean you can only shout into a void for so long right, without people really listening to you. So, you ended up rather than fighting against the trend, you ended up going and and going to a a place where they actually, you know, supported your beliefs and your your long-term vision for blockchain design. Yeah. Yeah. Exactly. And so, I mean, I'd like to talk about the social layer a bit later because I think there's some really interesting comparisons, but let's just get to the meat and potatoes here of the Salana thesis and kind of where we are at. So, um, you guys said there was a recent podcast with Uma from Succinct on Bankquist talking about native rollups.

It got clipped. It got shortened and she basically said that this one computer thesis is dumb. Now I I don't really want to rehash the modular versus kind of monolithic conversation here, but perhaps a more uh refined question is if we are to uh agree that um you know in this hypothetical future there's going to be Salana will be the home of all of the world's activity or a lot of the world's act activity. Um is is is this a future that you guys are aligned with for the entire space meaning that there is just one chain? It is Salana and everything else that is being built will

either have to come join or kind of figure something else out because it it's going to be like a very much winner take all scenario. Or perhaps you you could explain the nuance of why Max called this one chain thesis and I quote [ __ ] on Twitter like two two or three weeks ago. Well, I I didn't call it Uma called it [ __ ] I think but I would say like called it dumb. But I think you said something on Twitter like two or three weeks prior, but it's it's okay. It's not about he said, she said. I think that it's important to understand the nuance though of of of this kind of IBRL thesis as well as this one chain versus many chains. Well, I guess I would say I don't really think that a single blockchain is equivalent

to a single computer. It's actually not what a blockchain is. It's a network of computers. And the first version of blockchains like Bitcoin, we had computation where one of the replicas would say, "This is what I want you to do to advance the state." And every other replica did exactly the same work. So it's like worse than doing everything on one computer. It's like doing everything on 10,000 computers. Each of them replicate each other's work. And that's not necessarily how blockchains have to work. It's how blockchains work today. It's how Solana works today. It's how ETH works today actually. Although there are some advances that are coming that make it so that one computer can do work and the others don't have to do it.

And that's probably where blockchains are headed. But that doesn't mean that there isn't a ton more juice to squeeze in the model of one computer does a bunch of work, all the other nodes verify it. Uh I mean hardware is really advancing still at an incredible pace and it has for the last you know 40 years been advancing at an incredible pace. Bandwidth as well. We're learning new technologies like turbine that allow the replicas to communicate with each other faster and better and more reliably. We have new consensus protocols that let them come to agreement faster. And the code on the VM is getting better. Everything is getting better. They're all multiplying and there's really no way that you could possibly have any idea what the limits

of scaling are when somebody fixes a bug in Agave and we get uh 100x increase or sorry we get 100% increase in performance right away from that one bug that was like five lines of code and there's way more like that and there are other that are that are not in the code. So, I just don't understand how anybody could be so confident that this uh one blockchain thesis cannot scale. I mean, we know it can scale to 1 million TPS. Is that not enough? Well, maybe we need to go to 10 million. Maybe we need to go to 100 million TPS. Whatever the number is. I just don't know how these people have any confidence that the advancements that we're making are not going to lead to that situation. I think if you maybe take a step back, the thing

that kind of originally got me excited about the Ethereum vision was in 2017 the narrative was like the world's computer and really along the lines of like banking the unbank and that was really like a product vision that I was excited about because you could actually provide services to different parts of the world that were either very hard or just like not really possible. And I think over time because of kind of some of the design choices Ethereum made, even as I kind of like highlighted in our intro with like the unis swap thing, it just became more and more expensive.

And so for me, how I kind of focused on kind of the high throughput blockchains and why I got got excited about the high throughput blockchains was really just what are blockchains actually good for? Like if you take this step back and say like, "All right, we're all excited about all these systems. We can nerd out on all the different like design details, and I'm sure we will during this podcast, but what are they actually doing?" And today, the best thing that they're really good for is trading and moving value from point A to point B.

And so I think what we discovered like fairly early on and got a lot of flack for it was bandwidth was the actual bottleneck between these blockchains. Uh for a while Ethereum, even Bitcoin really had this ethos that you had to run uh these computers with relatively low bandwidth requirements and relatively low hardware requirements. And they did that for kind of decentralization kind ofish reasons, but they refused to kind of budge on some of those um I would say like inputs. They weren't very flexible with those requirements. But I mean, if you really

kind of break it down even further, a transaction is just a certain amount of bytes and each blocks is a certain bout of bytes or megabytes. And so there's a certain limitation that you can actually do with how many transactions in a block. If you literally double the block size, you can double the transactions theoretically that you can do within the block. And how that's represented in the physical world is having faster internet connections between nodes. So you can propagate that transactions to the rest of the quorum. You can agree upon that information and kind of like make state changes. And so I think we've always kind of had this idea that it wasn't kind of like modular versus integrated

necessarily. It was more like low throughput blockchains versus high throughput blockchains. And then you could kind of get into like some of the nuances, but you had to essentially go to high throughput blockchains to actually push the scale um to actually make blockchains more useful for kind of this trading and payments. Yeah. So Andy mentioned this IBRL, increase bandwidth, reduce latency. This is kind of the mantra of high throughput blockchains and I think we have a very like consistent foundation for this um across all these ecosystems generally like you know we believe in in blockchains as a tool for commitments um and moving around uh information and

value and generally like the best technology out there for markets and so you know Salana has been building this vision of putting the financial system onchain and to do that it needs to increase bandwidth reduce latency I I want to kind of go way way back. Like Logan, you mentioned you were around in 2017 and this feels a lot like a convers like kind of a a contentious topic that was around in 2017 with like increasing the block size on Bitcoin. You had Bitcoin Cash. Ultimately, you know, Bitcoin Cash was created. There was a bigger block size. They reduced latency so people could use it to spend money on coffee, spend Bitcoin, you know, for everyday payments. But that's Bitcoin Cash. And then you know the canonical

Bitcoin is still Bitcoin. And so that came at the expense of of essentially decentralization. So you guys true though that's just not Yeah. the case. Like look at Mon. Look at look at the specs for Monadab and it's like 32 gigs of RAM. It's like a gaming computer that you can find at Best Buy for $1,500 off the shelf. That's their node spec. And it turns out the the order of magnitude difference between a Raspberry Pi, modern Raspberry Pi from today and a $1,500 desktop you can get at Best Buy is more than the order of magnitude difference between say notes back and what we had on Salon or on Suie or that was like the sweet spot like if

you really cared about decentralization it was let's have kind of a gaming computer from the last two years that you could find for $15 $100, you're going to stake 32 ETH, which is still much much much more than that. And that should be what's required to run a node that participates in consensus. And once that happens, you have so much more capacity. And if you go a little bit further out on the curve, you can juice a little bit more capacity. But at the end of the day, it's like there's way more available in between Raspberry Pi in the Sahara Desert with 10 megabit upload and gaming computer in rural, I don't know, Virginia that has 100

megabit upload that has 32 gigs of RAM has a modern CPU. That difference is way bigger in terms of orders of magnitude than the difference between Salana node and this really feels like the dimension that like ETH people feel like you know there is there is like they're not budging right on increasing the node requirements and that's what's limiting their throughput. Salana you guys increase the node requirements you get more throughput. I I I mean I I've had this conversation I feel like a hundreds times now and the the thing that it really boils down to is just how much should the hardware actually cost and then how fast of internet connections should you be the argument for the

longest time and and honestly why we have always said that Ethereum's probably not going to really be able to cross the chasm was more just because of this ideology. It's not because they didn't have really good engineers. They have very smart people like Max called this out for a long time. It's just saying that they're like unwilling to budge and like revisit those hardware requirements. But the argument was that the hardware requirements have to be low so that people would run nodes. And what we saw on practice, I mean, I think Ethereum mainet for like full nodes that are actually running. I haven't checked this data in a long time, but generally it kind of like oscillated between 4500 and 6,500 full nodes. And the thought process is in my mind if you 10x reduce

the cost for the barrier of entry say 3,000 instead of 300 if the true cost for people to run these decentraliz decentralized systems is actually the cost if you reduce the cost by 10x you should have 10x more of the nodes but we just didn't see that in practice. I don't know what the latest like Salana numbers are generally I think they're like 2500 to 4500. So like either 50 to 60% of the full node count of Ethereum at any given day and the hardware costs are more expensive. And so that like delta of like what people perceived as like hey this magic line in the sand and we can't cross this never really made sense to me. And then it like as we'll

probably get into like in these systems you can't eliminate physics. You can move like the high data availability from like Ethereum to Celestia or to like some alternative DA layer or you can remove the execution from the base layer to an L2. But like that physics still has to happen. You're just moving the pieces from point A to point B. And the thing that we've always appreciated was just like the cleaner approach of integrating them all. So you're not like reliant on like these 10x teams to have to coordinate. Yeah, you could argue that decentralization like is or isn't important, but what you guys are actually saying is that it is important and this architecture allows

you to have your cake and eat it too. I would argue that. Yeah. I mean, ETH ETH execution on L2 is not decentralized. So, if you care about decentralization, and I do care about decentralization, it's very hard to be building on ETH today because the the LT thesis fundamentally is not designed to be decentralized. And in fact, they said, you know, we're basically going to give up on decentralization of block building of sequencing and rely only on decentralization for verification and for finality eventually and security.

And I think that for me is not an exciting place on the trade-off curve because you probably could just trust that the New York Stock Exchange is going to work pretty well on almost every day. If you really wanted to disrupt New York Stock Exchange or NASDAQ, you need to do decentralized sequencing of the blocks and having like a centralized sequencer. Even though it may scale your chain, yeah, you can scale the chain by running a centralized server that runs the VM. Okay. and you run it, your mega ETH, you have 1.5 terabytes of memory on your box. You pay $100,000 a month in server bills and you can run a really, really fast VM on a single server. But that is not the same

as a blockchain and it does not have the same guarantees. It does not have the same economic properties. And so we'll see which one wins. My guess is that when you centralize all of the power into one entity, there's just no way that that entity doesn't start to extract value in the long run. And eventually systems that have middlemen that extract a bunch of value are going to lose. And that's just true of any system. I maybe to build on that briefly. I mean, in my mind, I always kind of thought of like L2s as almost like data compression because you're doing like batch transactions to the layer 1 to use that block space more efficiently. You're taking I don't know, this is like an extreme example. You're

taking like one megabyte and trying to like compress that to like 100 bytes. Like that would be interesting. But with the L2s and like Mega ETH, like I had them on the podcast. It was a really fun one. We did a very technical kind of conversation. What I think people kind of like fail to do from the engineering standpoint generally is like push things to the extreme and kind of think about if this was truly successful, how would this look? And for me, if you have a single sequencer L2, what you're starting to see even with arbitum and some of the other L2s for trading purposes is specifically is people colllocate to that sequencer to get

information better than other people. And to me, like we see this as Max kind of like pointed out with like the New York Stock Exchange. the New York Stock Exchange, they literally give everybody the exame same exact cable length to the box so that no specific highfrequency trading firm has an unfair advantage and it's only model base. And so if you push the L2s, the single sequencers like Omega ETH of the world to that extreme case, you're going to see the same things where you're going to have to have like everybody in the same data center colllocated with the same cable length. And to me, like I that doesn't excite me either because we already have that in the world. And I don't think that's actually like creating either better price discovery or a venue that I

would say global powers around the world would actually want to trade on. I mean, it may be true. It does. Like Max put out a post with a a a nuclear target of Amsterdam that just end the Salana network. What are we talking about here? Well, it turns out that the uh the Salana network is actually more decentralized than I thought when I put out that post. I just looked at the know if there's only about 15% of nodes in that data center. So, okay, if you did launch a missile, you have to hit like two other data centers. Okay, but like with many things, we there's a discrepancy and this this often happens in Twitter debates where it's there's a discrepancy in time frame. So, we're we're talking about single sequencers

today. Of course, the the rebutto from the uh L2 world would be like, well, we're going to build these out into uh more decentralized systems that uh don't forego the same performance properties. Whether that's possible or not is basically yet to be determined. That's kind of the beauty of the space um is that you're able to kind of see. You may not think it is, but in the in today's world, um the the L2s as they are are are clearly operating with a single sequencer that does the ordering. But of course, the the the properties of the security and the decentralization come from um the base layer. They come from data availability layer and they come from Ethereum. Again, we can we can debate that. Other thing that I find a

little bit misleading from you guys is whether purposely or or not I I I I assume all good intentions is me on Salana is the worst of any blockchain. I am constantly sandwiched and I am constant there are users who are getting meed all the time on just small memecoin trades on large trades and this is going to what you guys are calling reev real economic value. So which is then being used as a metric to put the valuation of Salana in some sort of like economic modeling. So you guys are saying effectively and you know kind of fill in the gaps where I'm wrong that ME and that being closer to sequencers and

having this edge is wrong and you don't want to support that. Salana has a lot of me on it by means of uh certain groups having whether it be information location or just overall transactions that are meant to be sent to only a certain subset of validators then that that that revenue or economic activity is then being calculated in like a positive way for the chain and then being used in economic modeling as some sort of valuation metric. So, I don't agree with the MEV or REV as a means of valuing a chain. I I think long-term L1 MEV is is kind of silly. I think a lot of the applications will want to, you know, obtain that themselves. Um, and so

I I find it a little bit misleading if if we're worried about sequencing L2s with a single sequencer, but the the rev and me situation on Salana currently is is quite bad. Well, first let me address the point that Salana has more me. Yeah, Salana has more because Salon has more activity than any other chain and that's what drives me right now. Uh the other thing is like yes, ME is very clearly bad and I don't argue that we should continue along the current path that we're going on. This is actually a consequence of the design of the blockchain. It's very fundamental. You have a single leader. That single leader controls all the execution in their entire leader window and naturally there's going to be a lot of extraction

that happens there. And that's true on any chain. It just so happens that Salana has a lot more activity than any other chain. So, uh there's a lot more me there. It would actually single L2 because you don't actually rotate the leaders at all. It's just a single. No, it's just first come, first serve. Uh I mean it it depends. Some of them don't do first come first serve. Um there's I think there's a couple variations that people are playing with. Arbitum has time to optimism and base have priority ordering which is another form of auction that extracts a ton of me almost all of base revenues from me not from base fees. uh it's not the case that these things have no me and also my

point is more in the long run how do you stop that centralized sequencer after it's gathered all of the network effects and all of the users and all of the applications and then it decides to turn on something like time boost that arbitum has to start extracting money and and for a while New York Stock Exchange sold collocation they sell data products they sell all these kind of little advantages that they can only sell because they have centralized access. And my point is like yes, MV is bad on launch day and that's is the number one thing I'm focused on fixing with with multiler is fixing me. But I can see a world where we can fix me in a decentralized system. I don't see any

world where in the long run we fix the kind of me that happens when you have a centralized party that sequences everything. Because in the long run the profit maximization ends up taking over like the people who the Greek people get in the room and they move out of growth phase and they move into shareholder value maximization phase and you know the Dow token holders are no different than these type of people either and they vote on proposals like Arbitrum's tide boost which is like clearly bad for the long-term health of arbitrum but is good for the short-term profits of the Arbitum val and that is like exactly what is going to happen to every single centralized sequencer that has any kind

of traction and the ones that don't have any motion it will not happen too and that doesn't matter. So the we're kind of circling around this this idea that I wanted to bring up because I I've been seeing more and more about it and I think you guys are basically explaining your viewpoint pretty clearly. It's around this idea around global data synchronization at the speed of light. like could you explain what that means and what use case that is important for and like my opinion is that I mean I mean I think you know my opinion it's kind of like okay well is this not better done across many many many blockchains and achieving much higher scale but yeah just what is that what's the use case why is it important and

what does that mean so for my mind I think going back to like that original Ethereum vision from like the product focus. It was like the world computer and Ethereum started with that kind of northstar and has now kind of deviated. Salana being the first high throughput blockchain totally always said he wants to build decentralized NASDAQ. And if you think about this from a market standpoint and actually building a better product globally, what would you need to do to actually make that product something that not only people in the United States would trade on but in Japan, London, everywhere in the world?

One part of that and one very key part of that is informat access to that information and making sure that that information is synchronized as quickly as possible and there's no kind of withholding of information to all the parties so that you have that data as readily as available as any other party. And the goal there again is equal and access information to that data. It's about creating fair markets. And so to me, in my mind, having kind of this like synchronized information at the speed of light, it's catchy. But from the product standpoint, why it's important is because everybody needs the most real-time information as possible. And it kind of goes back to that like single

sequencer. The reason why people want to colllocate to a single sequencer is because there's that information asymmetry if it's not actually des decentralized globally. And it's kind of funny like with L2s like as soon as you try to like add back in consensus you run into the same bandwidth problems that the L1's have where you have to do the data dissemination from like point A to point B and then it returns back to a bandwidth issue. And so for me it it's always just about yeah synchronizing that information and you need to do it fairly quickly. Yeah. I guess like to piggyback on your point, like in a world of Ethereum killers like Avalanche and Phantom, like the blockchain, not the

wallet. Like these were all kind of conceived to kill Ethereum. Uh that was not the case for Salana. Salana was not conceived to kill Ethereum. It was conceived to kill NASDAQ. It was conceived to become a decentralized NASDAQ. And in order to compete with NASDAQ, we need to have better prices than NASDAQ. So that's a struggle because we're a decentralized blockchain. They're a single server. So naturally, they're going to be able to do performance better than us. So we need to do something else different that differentiates us from NASDAQ in a way that's better. So like what are prices in a an exchange? It's like the spread.

That's basically a measure of what the prices are. And if the spread on Salana is higher than the spread on Binance or on NASDAQ, then even uninformed traders will go to NASDAQ because it has better prices. So for me, this whole uh multiler synchronizing information at the speed of light, it's really about how do we get information from around the world on chain as cheaply as possible to protect market makers? Why why do we want to protect market makers? Aren't those the guys that are like manipulating movement price or whatever?

No. Like market makers are the guys who put out the quotes and we want those quotes to be as close as possible to a fair price so that people get fair prices and cheap prices on Salana better than they get on NASDAQ. Well, there are two types of market makers. There are the makers and then there's the takers. The takers are people who bring in new information and they trade profitably against the market makers. And the spread is made up of two things. One of them is how many retail traders who want to go back and forth do we have? Those guys make money for the market makers.

And how many takers do we have and how much private information do they have? That is a loss for the market makers. And and so the spread kind of balances like the more losses you have from these takers, the the wider you have to be. The more retail flow you have, the tighter you can be. And that's like the whole equation here. And multile is about about how do we get that information on chain as cheaply as possible without resulting in a loss for market makers so that market makers can quote tighter and then we can beat NASDAQ. Can is that even theoretically possible? Like forget about in practice but can you can you even theoretically get that information on chain cheaper and faster than a centralized NASDAQ? I mean I mean the goal would actually be

instead of doing like price discovery on a centralized exchange and doing the ARB on a blockchain, you'd do the reverse. The price discovery actually happens on the blockchain and you can do ARBs on like the centralized server, I think. Yeah. Um so just to kind of like raise a point here, Hyperlid is a far better trading experience than anything else on chain right now in terms of everything. All they've done is basically say I mean okay that's my subjective take agree or disagree. I would agree that it has a better trading experience. I just don't agree blockchain because it's not okay but here's what I'm here here's the point right I would also say that a single sequencer L2 maximizing like using ref

using a custom database something like Celestia for data availability or something that's not Ethereum basically like the most juiced out L2 possible put a CLB on it will also be really really really performant so if like you guys are are arguing Salana is this decentralized version of of the two things that I've just explained or at least in some capacity aiming to be a NASDAQ, right? Which is trading, etc. But how will you ever compete with those who don't care about the trade-off? Like if I don't care about the trade-off, then I'm just going to say, "Fuck it.

I'm just going to do it." And your system is not going to be as good as mine. Users won't care. And you're going to be stuck in this kind of like middle ground of like it's decentralized, but we're also really fast, but not as fast as that thing because they don't care like we care. Like, do so. Yeah. Ladies, gentlemen, anons and dgens, it is stable coin season brought to you by your favorite, the best in stable coins, Fracks USD dollar stable coin backed by the best in the business, the Bidd by Black Rockck and Superstate. This is stable coin season. It is the yieldbearing stable coin by Fracks, coming soon to a chain near you.

Innovating finance global institutional and onchain deep liquidity to drive capital efficiency. Now maximize yields on institutionalgrade liquid staking with me. Unlock trillion dollars of capital via Fitbitcoin. All of this secured by Mantle's treasury of billions in stables and ETH capital pools. This entire ecosystem is coordinated by a vast network of seasoned VCs that are aiming to create the next generation of banking and financial technology. This is mantle. The issue is that if you give

all of the power to a centralized entity, they may behave well for five years, but once they have enough network effects and they have all the activity and they have all the flow, they start to monetize and they start to do things that extract and eventually inevitably the things that extract result in higher spreads. Yeah. But but but Max, what happens in like a competitive landscape is that another centralized actor comes in and undercuts their pricing. It's not a decentralized party that comes in as the cheaper faster alternative. It's another centralized one that comes in and does it cheaper and faster because the other one is taking shortcuts and and retaining profits.

if you guys I I sometimes it like it's easy to get into like the nerd likeness of all the crypto aspects but at at the end of the day like we have to actually build better systems like today you have a single sequencer that's called the New York Stock Exchange like they have that is just not true like that is just not a good comparison but I think we are we are talking about like which one is the cheapest fastest system because that's where the users are going to go that offers offers the best prices like now that you have to say okay if that is what is currently offered today how do I make a better product and I think what Max and I are kind of articulating is

instead of just having one sing single sequencer whether that's in the US Canada Australia wherever how would you actually make that system global okay because if it's in a single jurisdiction maybe another part of the world doesn't one to trade on that because hey, maybe they don't have collocation access. Maybe they think you're rigging the market. Whatever it may be, if there's only one computer and it's not kind of in your region of the world, maybe they don't want to trade. I think what we're personally excited about is the idea that you can actually make trading global 24 subtle 247 and instantaneous almost with settlement and you can get

to I think in the future par or as close to as par as possible as the centralized venues where you can do global price discovery where that information is propagated and where it's actually more fairer and equal access than the current systems that we have. And I think there's I I think you have to have like these very high throughput blockchains. I don't think you can have these single servers or like kind of modular blockchains of the world because you and like the engineering kind of just ends back up to that high throughput system if you actually want it to be like truly decentralized. I would also say these systems have huge amounts of network

effects. You cannot start a competitor to the CME to trade grain futures today because they're all trading on CME, right? But if you had if you combine the CME and NASDAQ and New York Stock Exchange and decay and London Metals exchange, you combine all of these into one market, that is a huge amount of network effect. And the level of better that you would have to be as an exchange to disrupt that would be much higher than the level of better you need to be to disrupt NASDAQ today. NASDAQ is like a 30-year-old system that I mean like the computer systems are 30 years old at least and it has very little changes because of the regulatory environment.

We know now how to make a better exchange. In fact, you mentioned Hyperlquid. They do something like this which is you basically allow cancels to have a head start over take orders. This is what protects makers. We know how to do that in a decentralized environment now with multiler. And the goal is to do that on Salana or on any of the other chains that are trying to do multi stuff is to do it there and hopefully first pick off NASDAQ and then start building up that advantage because you're a global chain has markets all over the world has nodes all over the world. In any period, any one of those can contribute to the network and can provide information. Then you build up a huge network effect that can't be disrupted

by a new centralized sequencer that tries to do the same thing you're doing but slightly better. It's not just the New York Stock Exchange. Like it's London, it's Japan. Like it's global. Like I I think like people I mean like the tagline is blockchain or at NASDAQ speed, but it's all the trading venues worldwide that would collapse into kind of the single unified standard which is the L1. And I think that's kind of like the exciting vision. Yeah. As a as a Salana holder, um you guys have a good uh you guys have a good opium. Uh uh All right, guys. We're going to a thousand boys. reality exactly and not

too much hopeium on my end. Okay. So again, uh global data synchronization at the speed of light comes comes through by tightening spreads, offering global markets 247 uh doing it in a decentralized fashion and doing it at scale. the the current argument is effectively in today's landscape uh L2s with single sequencers u do not achieve the same properties of decentralization nor ability to acrue network effects um because of the uh multi- multiaceted process to basically get to a decentralized version of themselves as well as the incentives to to do that uh being kind of uh construed with people who want to make profit um more or less

Um, yeah. I I think that's directionally correct. Again, like I I think it's helpful to think in like product perspective. Like if if you literally look at like highfrequency trading today between London and the New York Stock Exchange, people set up microwave towers, they set up private fiber lines, like they're trying to use Starlink to like do these arms because that information is valuable in London and they're trying to lower the spreads to Max's point to make pricing more efficient for everybody. And there's a race to get that information from point A to point B. And I think the thing that we're excited about just the high throughput blockchain architecture is the best architecture to make these spreads tighter. And things with

like0ero that are using custom fiber lines to help even route packets faster are kind of pushing forward that mission of making sure you can synchronize information quicker. I kind of think in the end state these high frequency or high throughput blockchains are just replacing racing to compete against high frequency trading firms at the end state. Um and it's going to be instead of just NASDAQ, it's going to be global. Like that's the big thing like because London has like it's kind of like all the trading venues around the world are kind of like their own L2s today. They have their own standards. They don't really talk to each other. you have to interconnect them and like you got to

convince people to trade on URL2 versus the New York Stock Exchange versus London. But if by having a single blockchain, a single high throughput blockchain that acts as the single unifying standard between all those interconnective tissues that people opt in to trade on. And to Max's point, if you have the most liquidity, if people start trading there, and that's just where price discovery actually happens, I think that's where it gets really interesting. So, so what you guys are describing, Seth, there's like this information frontier and if information discovery happens quicker on the blockchain because you guys are able to transmit the information

from Tokyo to to London faster than the NASDAQ is. NASDAQ might actually have tighter spreads, but those spreads are wrong because the in the information hasn't arrived there yet. and to the maybe it's better to talk in specific about what is going on on NASDAQ when new information comes in because suppose something happens in Japan that causes a like if you knew that information you would think that the S&P should trade lower like for example Japan just announced new rates for the central bank and now you need to go adjust your US rates right so that information travels from Japan to New Jersey goes onto the NASDAQ server and usually what happens

is that somebody who is a taker has that information. So the taker goes in and they wipe out all the liquidity because they have alpha about what the Japan rates are that nobody else has and they just clean out the market makers and and that has to be priced into the spread on NASDAQ. Like it's not like NASDAQ is going to have better prices if they can't deal with this problem. They they're not going to because every second there might be something crazy going on in Japan. A tornado just happened in in Rwanda, something like this. Like, oh, Russia just invaded Ukraine. We need to go adjust.

Everything like that has to be priced into the spread. And if you can't deal with it and NASDAQ cannot deal with these jumps from around the world today, you're going to have wider spreads. Even if your system is very performant and fast, it's all pricing in some information volatility. Yeah, you're just pricing in the fact that something crazy might happen and they might trade against you and you're going to get picked off. An example I like to use is for those who like follow football. It's like if you're watching a play and it's like third down and seven, there's a ton of variance in terms of the outcome of the game that rests on that one play.

And so if you're in the betting markets and you you're competing your market maker in that market, you need to be courtside so that when something crazy happens like the the sack happens and the the drive gets stopped and there's a turnover of possession, you need to be ready to like pause all your trading, like pull all your quotes and readjust. And they actually have people who just sit courtside whenever something crazy happens, they just like press a button and pause the quotes. So that doesn't happen on NASDAQ today. So if you get caught off sides because something crazy happened like Trump almost got assassinated or whatever like you just get picked off and you have to put that

in your spread. And the way that we're building for the future of Salana is a system that allows that information to come in without wrecking the market makers and and that way the spread will be tighter. Makes sense. Um and I appreciate the in-depth explanation guys. I mean, I I didn't expect to to to take like 20 25 minutes on that one, but I think it was like it's like is this better on a rollup or L2 or or not? I think that you you guys have made your point very clear. Maybe I can just put that in a bow, though, because I mean to me like it's it's that information that's really propagating around the world that is the valuable piece. Um, and this is why like I don't think like L2s, they're fine.

They're cool technology. If you're batching transactions and then propagating to that to the rest of the world, it's cool. Maybe instead of propagating one megabyte, you're doing a 100 bytes, but that information still has to travel around the entire world. Like that information is the important part and it has to go to the entire world. If you're doing the batching of transactions, that's fine. And to your point, Andy, earlier you asking like why can't you just do this across L2s is because different L2s have different markets kind of like different exchanges around the world and they don't really communicate between each other and then you have these arbitrage opportunities.

And so it's literally that to me that information that's literally synchronized that is the like very interesting part of blockchains for trading which is why you kind of need one giant high throughput blockchain where all that information is syncing from all the nodes around the world. Yeah. I mean the the the big question is like will Salana get to 1 million five million TPS before any other kind of like rollup based ecosystem figures out how to create and deliver very high throughput chains that have extremely native interrupt like native interrupt built in and then it's it's like who can achieve kind of but they will not get to

native well I think like there's Ethereum and there's other like rollup ecosystems. So Logan, you kind of mentioned like it's a different product. So I have an entirely orthogonal question like is there a use case for Ethereum andor its rollup that like is a fit for a use case that Salana is not fit for? I don't know. I I've always kind of joked like high throughput blockchains can always do anything low throughput blockchains can do. And like they're faster, they're cheaper, they're lower latency. Like I mean again this is like a very crude analogy but once you go from dial to broadband like engineers can build more interesting [ __ ] and like today like Max said at the beginning of the podcast like we still have another

thousandx to go. Like the big jump in my mind between low throughput and high throughput. We put out this like chart uh in the beginning and we got a lot of hate for it. kind of a common theme of being early but around showcasing the different data propagation of just like how big blocks were divided by the block time for each of the like kind of popular blockchains at the time. It would be great if we could have like real- time C like API to like measure this, but there were just kind of a snapshot for that point in time in history. But the most like uh I mean Ethereum was doing less than 56 kilobytes per second for contacts.

Dialup internet was 56 kilobytes per second and it was doing like 1/5if um it was very very low amount of data. Celelesia was like extremely small. uh like there was a bunch that were just like not really targeting high throughput and Solana was the first in my mind that really went from like this kilobytes of throughput to megabytes and that was a thousandx difference in performance increase and at the amount of data that you're propagating and to me the exciting thing is going to be going forward how do you get from megabytes to gigabytes which is another thousandx difference in performance and so what we're talking about here in my mind again going back to that data synchronization just to have like a

common thread here is you have to propagate large amounts of data especially if it's financial data to make sure everybody has that in the world and the little bottleneck from a physics standpoint is your internet capacity to send information from point A to point B to synchronize it. Yeah. So basically Salana can do everything Ethereum can do except Salana can do it faster and better. I had a feeling you were going to say something like that because that's basically saying that Ethereum is going to go and copy that. like if that's directionally and and I think we kind of hinted at it earlier in the podcast, but the only thing holding Ethereum back from doing that now is its ideology.

That's all that and like just Yeah. and the immutab I mean there's like there's head first into the ideology wall. Yeah. I mean there's there's the core properties of the Ethereum L1 that are significant in terms of its it you know whether you think it's a meme or not. the security or the censorship resistance etc. um that I think needs work and that's like that is where like I think institutions will will find Ethereum interesting but they will they will understand the vision that you guys are laying out for Salana especially after usage. I couple questions on the Salana social layer. I mean, today we have Fidelity just filed for a Salana ETF. So, I think that's going to go

through pretty soon, some of those. Um, but also there's like Pangu ETFs going on these days. So, I'm not really sure the state of like the ETFs, but I hope they all go through for the sake of uh of the market. Um, but following that that Trump launch, uh, that was probably the maximum amount of attention one could have on the blockchain and Salana held up decently well. uh there's you know some like failed transactions or people kind of having RPC stuff. Um but where do we go from here next with the Salana social layer? Like I think that we had quite a bit of um you know you know the LA Vape Cabal or whoever you want to kind of talk about. There's like this very like just keep clicking vibe

on Salana, right? Just like one more and you got it like very like financial and and that kind of energy. Where do you guys think you want to take the Salana social layer? And what do you think about the idea that Salana is no longer an underdog when it comes to be to being a kind of social ecosystem where it kind of was to Ethereum four or five years ago? I guess I would say the goal is always to produce real value at have real activity on chain and in my mind like meme coins are not that. On one hand, we have still very significant revenue from places like Pump that are like like the app itself is is generating a lot of

revenue like still brings a lot of usage and events like Trump and Malay and Melania are doing the job of dog fooding the chain so that when we have real assets trading on chain like we won't have the same issues. We fix fixed a bunch of bugs from that Trump weekend uh on theuler and um we're working on hardening that specific area of like the things that broke that weekend. So I guess I would say like it's really good to have some kind of activity on the chain. It's a permissionless blockchain.

We cannot control whether or not you launch a scam on the chain or not. And that's true of any actually decentralized blockchain like the alts you were talking about. They can always censor whatever scams they want. fact basis number that's true think but I think it's it's so weird to like watch people be like oh why why is like Salana like have meme coins like because we're permissionless decentralized blockchain anybody can deploy whatever the [ __ ] they want like so that's what's happening and people want to trade meme coins and gamble and that's I mean but do you think that the culture is reflective of that in a poor in like a negative way I don't think it's negative I mean I I I really now think the biggest dis disconnect is between kind of like the old guard of

like hey we're going to do everything on chain everything has to be on a blockchain versus kind of what like I would say the high throughput blockchains especially Salana today is focused on is trading volumes and so if you look at it in that lens then the Trump coin did 33 billion in trading volume on a single day if you compare that to NASDAQ that's like 10 to 20% of their trading volume and then like I I think about it literally as like we are testing these systems. So in the future if someone in Australia wants access to do a IPO of their business that's only doing 30 40 $50 million of revenue and I

say only but like to do an IPO in the US you have to have way higher revenues like it's an insane process. If we can open up that entire world's kind of capital markets instead of just being US-based, allow people to have access to credit, get loans, do global commerce, these are the things from a product standpoint that actually excite me and to so to see Salana do 10 to 20% of NASDAQ's trading volume, I'm like great, this is a awesome first step to what we eventually want to get to, which is building kind of these glo global decentralized trading engines where you're literally synchronizing the

information around the world and so yes sure Salana is like maybe in mind share flipped Ethereum now and that's cool but to me unless it's actually a better product going back to like why I got into the space I don't really care and so right now there's a lot of experimentations but the product and the vision has been only becoming more and more clear and to me That's the kind of next frontier which the entire ecosystem is working towards. I think it would be a big deal for an Avalanche or Phantom or one of these like ETH killer chains to have flipped Ethereum. It was a nice uh Twitter moment, but that was never the goal and we haven't achieved our

actual goal which is decentralized NASDAQ. Like you said, we had, you know, 10% of that volume on one day, but the goal is to get more volume than NASDAQ every day on the average. case day we're surpassing NASDAQ and we're still a ways away from that. The work continues. We're going to 60 mil CU from 50 mil in the next agave release and the plan is to be at 100 mil by the end of the year. So I would say we're just still doing the same thing at least in the core like dev community and the folks at foundation and the validators and the the really like core people who are really have been here for a long time that I talk to a lot. I don't think

there's any change in attitude. There's no like curse of success where we like go and celebrate and like get a bottle of Dom or something and and this is why we're not doing y parties here in Miami for ship week. Yeah, we're ship week. We're we're doing like the way that we do conferences is like it's just a co-working space. Everybody comes down and works together for a week, but they all have their monitor set up and their desk set up and they just work from there and you get to chat with everybody. But that's like how we do conferences. It's not like a can yacht party fiasco. Ma Max pivoted ship week real quick. Went from cruise ships to shipping products in a matter of moments. Um, cool guys. I mean, we're we're at about an hour. I I would Oh,

no. No. I'm not done yet. Here we go. We had another one. So, that inflation proposal didn't pass. Um, what happened? and like in salon inflation is quite high still which makes it tough to be like a really good collateral asset. I mean according to Tarun we had a really good podcast with him and John Charb talking about kind of like the moneyiness between ETH soul and some of these other crypto assets and Salana kind of struggles to be in this kind of moneyiness category or or is still kind of getting there which is partially a meme. One is very volatile um it's it's extremely volatile. two, you can get a decent rate just staking it with with a validator. So, why take extra smart

contract risk? And and then three, there's there's been quite a bit of push back around like the Salana DeFi ecosystem not really having grown with the broader Salana uh space. So, with with regards to the inflation situation, what is your guys take on that? And uh why is it manageable and why is it not a problem? I mean, I I wouldn't say it's not a problem. like I think we still are looking to address it in some form. Um we put out that initial 228 proposal. It was quite aggressive. It would have resulted in a very significant reduction in inflation. Uh it had about 61 and a half% yes and like 38 12% no. So it was positive but not positive enough to be

something that passed where we have a 66.67% six seven% quorum threshold uh of twothirds because we want basically it's a signaling vote that says the validators will adopt this if it's a a future agave release and we don't want the chain to fork because people don't adopt the client. So that's the point of these signaling votes and I think what we got is a signal that said uh the validator community wants something that was probably more gradual, maybe a little bit gentler. Uh there was a lot of people who voted no who said we are in support of reducing inflation but this proposal is too fast and too aggressive. And so the next step is we're going to wait a bit because that was a big contentious vote. And to get

everybody on board, we'll we'll do something that makes sense for a lot of those people who voted no but said they wanted to reduce inflation. And we'll come back in a few months or so with with something that is amendable to everybody and hopefully get a broader buy in than we saw in that proposal. But I think part of the thing that gets lost is like it was there was overwhelming support for this. 40% over 40% of the active stake voted yes on this proposal and only 75% of the active stake even voted. So uh this was like a strong signal that a lot of people in Salon want inflation to go down. I would say we're not trying to do it because of moneyiness. We don't care about moneyiness. Moneyiness is stupid. It's a complete distraction. The reason we're

doing it is for a few reasons, but the biggest one is that issuance is a leaky bucket and it's costing the network hundreds of millions of dollars a year for no reason and tax burden that we don't want it to cost the network. And so it's really just a efficiency change for tax purposes that helps the network in the long run because we're not emitting, you know, $300 million of soul straight to the US government every year. I mean also if you the whole moneyiness thing like Ethereum carried over a lot of the ideologies from Bitcoin. I mean the hardware requirement thing was really one of the big things but also the moneyiness was another thing from

Bitcoin. And if you look at it I mean Ethereum really pumped to like 100 million plus or billion I'm sorry FDV crazy valuation. And so people were trying to say like, okay, these are new technologies. This is a new asset. How do I value this? And out of kind of like a justification for the very high FDV that Ethereum had, it's not digital gold as Bitcoin is trying to be, it's trying to kind of play in the smart contract standpoint. They said, okay, it's moneyiness or the ultrasoundness. And I to Max's point like I I just don't think money is like something these smart contracts should be focused on. And this is why like the REV metric and I know

Andy you don't like it but it points to something saying this is able to be real yield potentially passed back to the consumers of the chain or stakers where it's not just net inflationary. it's actually generating some value and being given back to the chain or the chain stakeholders that can make money. Um, and then you can run like the whole DCF like as Treadfi people traditionally do and try to assign like some value to this. But I talk with like the large asset allocators in the world and Andy, you're mentioning like they're going to use Ethereum. They run the same stuff.

They they listen to this type of stuff. They're not like, "Oh, Ethereum's more decentralized. Let me use it." They they want to use the cheap and fast. They want to build excellent products, and they're trying to actually value these things from like a normal method. They're not evaluating money. This I think, yeah, I mean, this is a really good point. I think part of the community has been puzzled by why the L1 premium on tokens is so high. And one theory I have is just L1 tokens are the only tokens where the revenue of the product is directly tied to the token.

Like every other token like look at unis swap for example has this issue of a split between labs and the dowo and there's a very complicated relationship there and investors are very spooked by that. on an L1 there is revenue from being a leader in the blockchain and you run you you get leader slots directly proportional to your stake. I mean this is part of why proof of stake is way better than proof of work. Another reason is that you don't end up using enough energy to power all of the country of Austria for a year just so that you can send a few Bitcoin transactions. But that's like one thing about L2 L1's that is like

very nice for people buying the token is hey if this thing actually has a path to real revenue then a DCF model does apply. There are tons of protocols with great revenue where if they launch a token it would not be clear that DCF revenue like revenue model would apply because it's actually not clear that the revenue will ever translate into the value of the token. and L1's like immediately solve that problem. Look at all the L2 tokens and you'll see why not having this is a problem because they're just there's nothing to do with them.

You can't stake them. You can't earn leader slots. You can't do anything. It's just straight down only because all you can do is sell. Yeah. I mean, this is the Yeah, this is the conundrum that I think a lot of the space finds itself in. I hope for the regulatory clarity to provide a better environment for this distribution of revenues, etc. Um I just I just don't know if modeling. So when I spoke to one of the guys from Franklin Templeton Denver um and I'm sure a lot of the other institutions feel this way.

They don't really all they want is enough of a information to underwrite an investment right and they want to see okay what is the network effects of this? if this works over five years, what are the key points that we could model out and see this is how this is going to grow. Like they don't they're not even looking for like real sustainable business right now. They literally just want to see like okay if this network grows in adoption like what are the metrics that we that we should track and like how can we make a model out of this so so that like if we do underwrite investment and it's not going to plan then we can like you know change course or add more seller whatever and so you know I think just like generally having a model works and I think that's

what that's where L1's are at and yeah a lot of these L2s are just kind of DeFi apps or some of these are are are struggling there and so expect that to continue also just very excited to see how the inflation question is handled. I think a lot of people misunderstand the fact that we need uh early like a lot of proof ofstake networks need inflation in in their you know first couple of years uh to be relatively high for security purposes for bootstrapping but salon is clearly reached that point where as you say Max it's now just becoming like we're just paying validators for work that other validators would do for less right and and I mean depending on the state of the me rev on Salana going forward there's also more revenues to be shared there um so I'm excited to see

that uh get solved and I think the market will respond well to that. I'm all out of questions, guys. It's been a really fun conversation. Um we try to approach these with quite pointed questions. Uh and this was set up in that way on purpose, but Rob and I got soul bags. We're we're excited about the future of Salana and we're leaning on you guys to get us totally on the pod. So, thank you guys. Thanks, guys. Sounds good. I'll call them out. Cheers.