Unchained - The Chopping Block (feat. Anatoly): Solana's Second Act, SVM vs EVM, & Extension Dilemma - Ep. 724

Episode Date: October 24, 2024

Welcome to The Chopping Block – where crypto insiders Haseeb Qureshi, Tom Schmidt, Tarun Chitra, and Robert Leshner slice into the juiciest topics in crypto. In this episode, the crew is joined by special guest Anatoly Yakovenko, the mastermind behind Solana, for a deep dive into the network’s evolution and future. The conversation kicks off with a look at Solana's rapid rise from its rocky early days to its current status as a blockchain powerhouse. They debate the role of memecoins in Solana’s ecosystem and whether they’re here to stay or just a passing craze. Next, Anatoly opens up about Fire Dancer’s impact on the network’s scaling and performance, sparking a broader discussion about vertical vs. horizontal scaling. Finally, they explore Solana’s ambitious push into the mobile space with the Seeker phone and its implications for the broader crypto industry. Packed with expert insights, bold predictions, and a few surprises, this episode is a must-listen for anyone following the future of blockchain. Listen to the episode on Apple Podcasts, Spotify, Overcast, Podcast Addict, Pocket Casts, Pandora, Castbox, Google Podcasts, TuneIn, Amazon Music, or on your favorite podcast platform. Show highlights 🔹 Solana’s Rise: Solana’s journey from early turbulence to becoming a blockchain powerhouse. 🔹 Memecoin Frenzy: The explosive growth of memecoins on Solana and their impact on the ecosystem. 🔹 Fire Dancer Impact: How Fire Dancer could revolutionize Solana’s scalability and performance. 🔹 Vertical vs Horizontal Scaling: Is Solana’s vertical scaling approach stronger than Ethereum’s rollups? 🔹 Seeker Phone: Solana’s ambitious push into mobile with the Seeker phone and its potential disruption of the app store landscape. 🔹 ETHOS Phone: How does Ethereum’s new phone compare to Solana’s mobile vision? 🔹 Network Extensions: The introduction of network extensions and whether they align with Solana’s long-term vision. 🔹 App Chains: Could the rise of app-specific chains lead to fragmentation or strengthen Solana’s ecosystem? 🔹 Developer Experience: The challenges of Solana’s development environment and their impact on innovation. 🔹 Ethereum Rivalry: With both networks advancing, how is the competition between Solana and Ethereum shaping up? Hosts ⭐️Haseeb Qureshi, Managing Partner at Dragonfly  ⭐️Tom Schmidt, General Partner at Dragonfly  ⭐️Tarun Chitra, Managing Partner at Robot Ventures Special Guest ⭐️Anatoly Yakovenko, Co-Founder & CEO Solana Labs Disclosures Timestamps  00:00 Intro 02:56 TCB’s Criticisms 05:12 Evolution of Solana's Ecosystem 17:36 Solana's Competition 22:05 What Is Solana's Moat? 28:31 The Role of Atomicity and Composability 34:11 Network Extensions 44:20 First Mover vs. Second Cohort 48:44 Future of Asset Types and Content 52:49 Developer Experience and Token Implementations 1:00:38 Solana Phone vs. EthOS 1:09:16 Thoughts and Criticism Learn more about your ad choices. Visit megaphone.fm/adchoices

If you look back at the history of Solana, it was very not obvious that Solana was going to have a second act, right? Like, most chains don't have a second act. Like, they're in the spotlight for a period of time. Wasn't to me at all. What's that? It wasn't obvious. I mean, in a way, it made no sense to me why it came back. But it also, no, there was no doubt in my mind that we were going to survive in some weird way.

I just didn't know it was going to be, like, relevant. as relevant as it is today. Not a dividend. It's a tale of two quond. Now, your losses are on someone else's balance. Generally speaking, air drops are kind of pointless anyways. Unnamed trading firms who are very involved. I like that ETH is the ultimate pump.

DFI protocols are the antidote to this problem. Hello, everybody. Welcome to the chopping block. Every couple weeks, the four of us get together and give the industry insider's perspective on the crypto topics of the day. So quick intro, first you got Tom, the DFI Maven and Master of Memes.

Hello, everyone. Next, we've got Tarun, Your Brain and Grand Puba at Gauntlet. Yo. And then special guest today, we have Anatoly, the Tsar of Solana. Welcome back, Anatoly. Hey. If you remember, that was actually your title the first time that you were on the show,

so I wanted to bring that back. And I received the head hype man at Dragonfly. We are early stage investors in crypto, but I want to caveat that nothing we say here is investment advice, legal advice, or even life advice. Please see chopping block that X, Y, Z for more disclosures. I put out a call about this episode because I wanted to see what people wanted us to ask you because you were actually on the show two years ago.

Two years ago, of course, was a very different world. Two years ago, if you remember, this was in the pre-FTX collapse era. It was like summer of 2020. This was like, I think, before Luna collapsed. So it was still, we were all in our collective childhoods as an industry. And that was the time when, you know,

Solana was falling over. There was a lot of stability issues. There was still, you know, there was a big conversation about fee markets of whether it makes sense to finally capitulate and start adding fees on Solana. And so I was listening back to that episode before coming here to this one because a lot of people, when I asked them, hey, what should we talk about with Anatoly? I kind of thought people would be like, oh, you know, talk about the Solana phone. Let's talk about, you know, meme coins or something. Most of what people want us to talk about is they want us to grill you. And so there's a, there's a tweet by Austin, Austin Federa, and said, use this opportunity

as a deposition. I'm tired of Salana friendly episodes. Treat him as a witness and make your hardest case against Salana. Let's do it. Should, okay, thoughts? I mean, I don't know, I don't know, Tom and Tarun, I feel like, so let me, let me give a little bit of context. Let me give a little bit of context because this has been kind of a recurring theme on this,

on this show. So back in 22, we had a number of episodes. were very critical of Solana. I remember there was one episode in early 22 where I think like the key excerpt from the show is where I compared Salana, like the Salana Mempool, or not Mempool, whatever, transaction market

as akin to breadlines, where people were just sitting around waiting trying to get information to go through. That's stuck in my head. Did it really? Yeah. So there's some like most Soviet trauma kind of thing? Yeah.

And that had impact on the design decision. Really? What happened later? No kidding. I find that very surprising. Okay. But that was like a zinger that came from the show when Solana was having trouble with, like, managing all the defy activity that was happening in lieu of having a fee market.

And so we had a number of shows where we criticize, you know, so let me run through this. Because there's a litany that I've recounted once before on the show when Mert was here. So what did we criticize Salon about before? One that Salana insisted on having no fee markets for a very long time until it was basically forced to capitulate because of the rise of defy. Obviously there were all the critiques about SPF, FTX, kind of being the sort of vassal chain of an exchange, obviously that was not very long-lived.

At the time, like in 22, there was a weak culture of open source on Solana. Very few of the top projects on Solana were open source. It was also hard to be open-source in Salana because even if your source was open, And you couldn't actually verify that the builds on Solana were equivalent to the open source code because there were non-deterministic builds. That is very different today. But at the time, you couldn't really tell what you were interacting with. You couldn't even see the code on Block Explorers on Salana at that time. Builder experience was pretty painful.

Obviously, there was all the stability issues. There were a lot of people, including us, critiqued the performance claims of Solana, of talking about transactions as equivalent to vote transactions, which normally are considered more kind of metadata. within most other blockchains. So those are most of the critiques that I made of Solana back in 2022. Today, now, you know, I think from Austin's perspective, it's like, okay, now we have this Mickey Mouse kind of a positive attitude about Solana.

I think a large reason for that is that all of those criticisms are no longer things that I critique Salana about because they're all changed. Literally every single one of those things has really significantly improved. Not all of them 100%, but pretty much every single one

is very different than, what it was in 2022. And that's part and parcel of why I don't consider myself to be any more a critic of Salana the way that I was in 22. I think Salana has really come a long way. That worries me. Why is that our secret is that we absorb the negative feedback. We internalize it.

Not just me, but everyone in the ecosystem. We love the negative feedback. I have a quick, shillah, mother, L1 right now. Yeah, things change. And they improve. But this is how people iterate. We called it fire-driven development. Is that like sometimes, like, if you overthink and over-design a solution, you're like,

it's like tightening. I don't know if you ever change a tire. You don't want to tighten all the bolts at the same time. You tighten one bolt, it screws it up. So, like, kind of leave it open to failure to see how it's going to fail. And then you look at it and change it. Blockchains or tires? You heard it here first.

Building a complex system is hard. because you just don't know all the ways it's going to be used, all the ways it's going to fail. And if you like, the fee markets is a great example. Like we didn't have a solution for fee markets that wasn't a copy of Ethereum's. And we saw the problems there and we're like, we don't want the same problems because you have a global fee spike. Then all this parallelism is useless. Like it is literally like, what the hell is the point? Might as well build Deltus or whatever. Right. So sometimes, sometimes, like doing the wrong thing at the right time is like really, really important in the journey of

product iteration. If you can afford it, if you do the really, really wrong thing, you have a safety violation or stuff like that, then that's bad and that could be catastrophic. So you kind of, you need to know what things are like, what kind of failures are survivable and you can learn from them. And it's really hard to know in advance, right? On some level, you kind of have to get lucky that your failures are forgivable failures. And on some level, like, if you look, if you look back at the history of Solana, it was very not obvious that Solana was going to have a second act, right?

Like, most chains don't have a second act. Like, they're in the spotlight for a period of time. What's that? It wasn't obvious. I mean, in a way, it made no sense to me why it came back, but it also, no, there was no doubt in my mind that, that we were going to survive in some weird way. I just didn't know it was going to be as relevant as relevant as it is today.

Yeah. And I guess the thing is, you know, if you look at the way that Solana figured itself, right, originally I remember the story you also used to tell about Solana. It's a Solana is supposed to be the Nasdaq of Blockchains. And if you take that North Star seriously, okay, Solana is a Nasdaq of Blockchains. That means obviously that, one, it's going to be used primarily for, Well, by the way, wasn't it NASDAQ on the blockchain, not NASDAX? It's not NASDAQ speed.

It's blockchain and NASDAQ speed. All of us got it wrong. Okay, fine. NASDAQ and blockchain speed. So, of course, the implication of the NASDAQ is that, one, it's going to be, you know, very high throughput. Two, it's going to be primarily financial activity happening on, on Solana. And three, it's going to be serious financial activity happening on Solana. Right?

Like, all three of those things are implied. by that framing, right? Instead, Solana became first kind of a, you know, first it became serum, which was the closest thing to like, okay, this kind of sounds like what you mean by NASDAQ on the blockchain. But then we went from serum to then, okay, the dominant application on Solana is NFTs, NFT trading, which is like has none of those properties, doesn't really need to be super fast, isn't really very serious, doesn't, doesn't, you know, it has a, it obviously has a, flippancy relative to NASDAQ that you probably weren't expecting when you were going and building

the system. And then now the dominant application on Solana is meme coin trading. And most of that meme coin trading is not happening on order book dexes. It's happening on AMMs, which is also like the cheap, the peak state contention, like exactly the wrong way to build the NASDAQ relative to, like, I remember the conversations having with you, I think it was on the previous show where you're talking about like, yeah, you don't want that. You don't want AMS. No, no, but they are horrible at state contention. But there's state contention if they're the same assets.

The fact that you're adding new assets means that you're creating, it's, it's that you have to be very clear about that. The state content for AMM is if you're going for the same asset, right? But here, it's like I'm creating new assets all the time. Like that fact that the asset creation rate is faster than the contention rate is very important. Yeah, it is bizarre. So how do you think about this NASDAQ of blockchain thing, given how Salon has evolved?

Like, again, I would wish to say that labs or anyone even like that worked for me, it was even remotely responsible for any of this. We just unblocked founders. We had no idea what they were building. And the way that you unblock him is like price and features are the two main things. And then content. But like content is what the founders create some mean coins or whatever. But like people at labs or whatever,

or Helios, they see a customer that's trying to do something and it's too expensive or it takes too long or whatever and they go fix it. And once they fix it, that problem scales out and it kind of blows up. And what people want to do is trade meme coins. I want to trade digital assets. They want to have fun. A lot of the stuff is basically just activity that's fun. I see it as no different than like loot boxes in iOS, right?

Like you have a $100 billion gaming market that iOS services, they take a 30% feel. of that that's the vast majority of iOS revenues from those kinds of apps. The serious apps that people think about that Apple is indispensable for is Uber or whatever, because that's the magic that I have a logistics pipeline created in the moment's notice from any point around the world to me anywhere I am. But like Apple takes like what, 2% cut of that fee or something if people use Apple pay? So it's weird. The joke that I make is that the number of rational markets is countable, but the number of irrational markets is uncountable.

So eventually the- This is like your fifth diagnosis joke this month, buddy. That side is going to be bigger in size than the rational size. It's just inevitable. So like here we are. Also, I like to remind myself that when Steve Jobs shipped iOS in the App Store, he had to look at the top 10 list and see the fart apps in this beautiful thing that he built and be like okay this is what we're shipping with I think it's actually a very good sign for platforms

when they take on a life that was like not predicted by their creator like honestly if you look at a lot of the ideas and apps that like Vitalik gets excited about they're kind of crap it's like d-id shit or like reflexer and just like like kind of like useless garbage or like they're like bad product ideas versus the stuff that people actually use or like Instagram actually when it started to get really big, it was not latte art pictures. It was like memes and people posting their market places and the shit they were selling on Instagram. And like, you know, it's like, okay, this is not what it was really designed to do, but people are sort of hacking it to do the thing they wanted to do that the founders didn't really imagine. Yeah. I mean, the other thing that was

interesting was like reading all these, there's this kind of an interesting divide that I've noticed between sort of like Ethereum Ethereum world and Solana worlds, which is in the Ethereum world, everyone says it's like 1975 for the internet. In the Solana world, everyone says the 90s because meme coins are like porn taking up all the bandwidth. You know, like, we had to build the fiber to like make the porn faster,

which I think is like the same as like having to have enough capacity to have more meme coins. As weird as that is the same. is the fart app, right? Like the porn and fart app thing are like the same. EBOM's World. This is like, I don't know if somebody's going to figure out it to take Ebom's world and turn into YouTube. This is the stage that I feel crypto's in. But the founders that I talk to that are in that space, like are making a lot of money and revenue, but they have no intention of like quitting. And they have big dreams. Like they want to build like next.

gen financial institutions for zoomers and the next generation, they feel like this is a way for them to go and onboard those users and build those things. So like, I mean, that's great, right? Like, it's hard to, yeah, hard to predict how that stuff is going to play out. But if you, okay, so if you think of this as eBOM's world, it sounds like you are implying that this is a transitional phase in Salana's journey. But EBOM's world never went away. It did, but it stopped being the cultural center, right? Like YouTube and Instagram and all these things sort of became these loadstones of how

internet culture was formed. Do you think the same thing is going to happen to meme coins with respect to Solana? I think, well, I'm very bullish in crypto. So I think there will be the next trillion dollar networks or whatever protocols will be crypto-based or not going to be corporations. Or at least I think there will be some that are crypto-based. and those just by sheer scale have to serve a very large audience. And that's mid by definition, right?

Just by the size of it. I don't think the weird stuff is going to go away. But I think you're going to start getting into like, I don't know. Like all the ideas that we try, it's going to be like loyalty points or whatever. None of that stuff has panned out yet. But people are trying it. And I think they're just a little too early. Like I, my guess is that like, you know, within five, 10 years, if every business in the world has some service that can manage your wallet and create assets and consume assets, they're going to start with these experiments and they expect all their customers to have wallets and stuff like that.

They'll just become part of their day-to-day business activity. And then we'll see like that more mid-level content where, yeah, I'm getting like Taco NFTs from Taco Bell and I'm converting. you know them but it's loyalty points right like with a little bit more flare or whatever but we'll see like i i think the space will mature and then get bigger simply as crypto gets bigger i honestly don't think there's going to be a ton of new protocols or a ton of a ton of new asset types i think we've probably discovered 95% of them nfts and coin and tokens cover the vast majority of things you can do and then AMMs or whatever, all the different markets for them already exist. All the ways you can

remix them already exist. It's all been kind of proven out. I will say that the way you said that reminds me of the famous patent officer in the 19th century who said like everything useful has already been invented. So that's a big, big statement to make there. 10 megabytes is enough for everyone? Yeah, exactly. Paul Krugman talking about the internet and like, oh, you can't do anything with the internet. You can't do it a fax machine. I don't know, I don't know that history of technology bodes well on people saying like, yeah, we've discovered everything. It's pretty much done. Pack it in. We'll see. We haven't discovered how to scale these things.

Wait, wait, wait, wait, wait. Okay. We haven't, we haven't done our inquisition as, as, yeah, yeah, you guys are being too nice. We're in the process. We're in the process. Okay, okay. So speaking of scaling on the last show, you claimed that, so this is 2022. 2022, you claimed that you did not think of the Ethereum or the EVM ecosystem as competitors. You thought your only competitors were Aptos and suite. Okay, fast forward today, two years later. Would you revise this statement or would you say that's still correct? Base is probably the closest competitor now, I would say, when looking at activity.

I think basically how I break this problem down is you can convince. compete on, you can have like, your killer thing, components of PMF is price, features, or content. And Ethereum's like was weakness was price and Solana had a killer price that just Ethereum could never get to. So on price, there was a killer. Price doing transaction fees. Yeah, just fees to users. And then feature set, I think it's arguable, which has better feature set. But I think honestly, the, just the integration.

between a standard, like, there's weird, weird things that we didn't intentionally create, but happen to have been like huge advantages in SVM that ended up being, I think, partly killer features. And you can kind of deal with them. And I think the base guys are smart enough to figure it out. So that was part of it. But like, content is what Ethereum had the huge advantage in because they have all these legacy, I don't want to say legacy, all these great OG assets.

on the trade. With network effects. You cannot move to Solana, and there's no way you can do that to give the same effects. So that network effect is the content that was the killer thing. And when you have that,

you can increase price to as far as the market. When you said content, you don't mean like social media content. You mean like assets. Yeah. Stuff for people to do, right? Like what is unique about Ethereum is the assets issued there?

And I think now that's shifting over somewhat to Solana, one, meme coins are new content being created every day. We saw that with NFTs. Even if the Solana NFTs didn't hit the same valuation, it's just the activity was astronomical, like in terms of users and stuff like that. Base, I think, is the biggest competitor that I think of to the Solana ecosystem. And I honestly do look at the other alternative at once because the disadvantage that the Ethereum ecosystem still has is they're like everyone uses Mattermask.

You're like you can't move as fast in feature set and stuff like that. You're kind of like moving, your features are moving as fast as the slowest part of the standards body, which is like not great, right? I think that generally tends to be a while in advantage because you can have more scale because everyone's in the same standard, you're kind of losing a bit on features and stuff. So, okay. So you think of base as being the biggest competitor. Base, of course, is a roll-up sponsored by Coinbase or launched by Coinbase or whatever. Do you and Aptos also are doing really well.

Like, they're on the rise. They've cut up on features and price, right? Like, I don't even if the problem that they have is that the Salana had such a huge difference on those from Ethereum. there was a 10-order magnitude improvement now the improvements are marginal and not enough to like be the killer part right if you improve salonah's fees by 100x you might get some weird use cases that go run there but they're not the the major use cases that people care about and then it's not enough of a differentiator dropping freeze from a fraction of assent to a fraction of

a fraction of a cent is just not economical economically relevant for most use cases and there's just some fees like to this goes down later to my theory of where value accrual happens in these networks, there's just some fees that are, you cannot get rid of. And it boils down to database hotspot problem. If you have state contention, somebody has to go first. And no matter what you do, that is a fee that somebody pays, either it's through a backroom deal to a validator or, you know, a priority fee, right? Right. So, okay. Comparing yourself to base, base, of course, is a roll up. And one of the question is actually that Mert posed is what is Salonah's moat in a world where every

blockchain is fast, scalable, and cheap? And I think this connects a little bit with the rise of this new generational blockchains. So we see things like Aptus and Sui being, using the move VM. You also see the rise of things like Monad, things like Mega-E, which is a roll-up, and of course, base, which also has very high performance, very low latency because of the the sequencer set up. How do you think about how Solana continues to push its edge in a world where the EVMs and the Move VMs are catching up? And even SVMs and Solana Forks.

Somebody could take the Fire Dancer Code Code Base and run Solana. Is there a Solana for? I think there's a couple that are small, but... Are they live? Yeah, there's one called Zen. Yeah, there's a bunch. There's like a long tail of little ones and I think there's some bigger ones in the works. My theory there is, again, content is like when everything is marginally the same on features and price,

YouTube is as good as features and prices Netflix as TikTok. Content is what starts dominating. And this is where I think incumbents have an advantage. Like Ethereum, if it could have parity in features and price, it would just grow through natural growth of like crypto market. But because they kind of gave that up to the L2s, I think this is where the Achilles heel of the Ethereum like roadmap. So you think that you think the Ethereum Scali Roadmap was a mistake? This is the big question. Like if I was put in charge of the influence in Ethereum ecosystem and somebody said that you have to beat Bitcoin,

I would probably make the same design decisions. If you have to build a world computer, I would make the Solano design decisions. The thing is, I don't care about competing with Bitcoin because I can't, like, I don't think it's an engineering problem. I think it's just a beautiful accident that it exists and let it be. But I have my engineering skill set throughout my

life was into making things faster and lower latency so I can understand the world computer, understand how it makes money, I can make it faster. So like that's just the easier path for me personally. Like for me to try to go beat Bitcoin, I would have to have like that mine virus of monetary premium. I just don't have it. Yeah. I like that for Bitcoin is a beautiful accident. So Bitcoin is kind of like the slurf of blockchains. That's like anachronistic. I'm trying to connect with the meat point, guys. That's an anachianic analogy.

Give me some room for this. Give me some room. Like if I had to like academically get into it, I'd be like, okay, well, making sure that there's only 21 million Bitcoin is the primary use case of Bitcoin because this is what scarcity is its number one thing. So I need the cost to validate scarcity to be as low as possible. So that means the smallest possible chain to do that and push everything else away. Right.

So that would be like the engineering path that I would go under. But I'm not a believer in like that there needs to be another better form of like you're only going to get marginally better at solving scarcity questions than Bitcoin. And I'm like there's nothing to gain there. Like how am I going to beat Bitcoin by 10x doing better scarcity? But isn't that also true for Ethereum then? Like, I mean, it sounds like from your answer. Bitcoin. I think it should analyze.

It's like I think you should make the analogy to programming languages because that's like it's like a cleaner analogy. And I feel like Fortran is like this thing that like sits every time you install any math package. There's five million lines Fortran run, right? Like it's there. And no one gets rid of it because no one wants to rewrite that shit. And that's like the scarcity thing.

It's just like there's no, you can't really beat that at its own game, but no one's like writing new stuff in Fordrick. I can beat Ethereum at latency and throughput, and I can be marginally competitive with all the other L2s. As an engineer, I can see a path there. So then to me, that's like, that's easy. That's almost like, no, it's people you can hire L7 engineers and I'll go punch out the code and go through Grafana and solve all the problems. To answer Mertz question, there's basically like two long-term things that I think are going to happen is that one is once feature price are marginally the same across all the ecosystems, content is going to matter. And then you have incumbent advantage and simply the network effects that people talk about in Ethereum now exists in Solana from people that launch products here get benefit from all the other products in the network. And those I haven't seen translated between L2s.

This is probably again, one of the Achilles heels and the Ethereum roadmap is that there is fragmentation that makes it harder for like a meme coin launched on pump to get revenue in Jupe and Radium and for that to benefit all three companies, right? Like you have to move all three of them into the same L2, but then like how's that marginally better than them all just existing at Salon or launching there. So you kind of run into this. Okay, so let me push back on this concept, because this is only that I remember in the last episode,

you also mentioned that one of the foundations of why Solana works and why, at that time, there was more of a conversation around horizontal scaling versus vertical scaling. And of course, Solana is kind of the poster child of vertical scaling, whereas Ethereum was classically horizontally scaling, originally through shards, now through roll-ups.

So the story for vertical scaling is that, well, with vertical scaling, everything's on the same machine or everything's on the same like kind of logical substrate and therefore you can have atomicity and composability and all this stuff and here's my claim

is that atomicity and composability is overrated and over time what we have found is that there's actually not that much stuff that people do that's like really really atomic right there's like flash loans and all the stuff that we imagined in the past that like really needed

atomicity and that without that it just wouldn't work. And the reality is that most people when they're interacting with blockchains, they click three buttons. Actually, there's not that many things. It's just like one button and then everything happens. And if one of those clicking three buttons is like, well, I move from this chain to that chain or like, you know, grab a lock on this other chain and then do this thing, from a U.X perspective, I've become more skeptical of the story that composability slash animicity is the killer use case. What's your You're not wrong. I think just a single transaction that can lock a bunch of stuff and execute atomically. There's like, it's arbitrage bots that care about this in Jupiter because it can

create routes that are very, very efficient. Anytime you add like a synchronous costs there, you just can't do these kind of, this kind of routing. So how much that matters, I think is less important than being in the same computer and literally just the APIs and everything else just being so easy for founders to like go do this stuff because like in the early days of pump or whatever they were literally just moving using a wallet to move shit from their bonding curve into radium and to do that between chains would be a major pain in the ass and then having users then switch chains and stuff there's just so many so much plumbing that you have to solve there, that it becomes not the path at least resistance.

And being all in one state machine, the path of least resistance, just go to use your next door neighbor, right? That is like literally right there. I think there's a lot to be said there. Whether that like, because you can do all the stuff even between all the ones. Warm hole is good enough. This is what it ends up being is that like the bridge going through Ethereum, because it's data validating with broad proofs or ZK is no better than warmhole for the end user for the vast much 99.999% of the cases, right? What do you think of then like the relationship between apps and chains? This is the other sort of recurring topic is, you know, sort of like audience capture

or platform risk for applications where it's like, yeah, you get big enough and you spin off and you do your own chain. And, you know, it's like people who are trading on pump. It's like, okay, they actually think they're trading on like Phantom or Photon or Trojan. They don't know that they're on Solana on this thing. It's just like, yeah, we're training coins. You know, what's to stop them from bleeding, which is kind of a trend that we've seen in in other chain ecosystems?

I think the chain needs to have enough bandwidth. So the extraction is basically so low that it doesn't make rational sense to lead. Like if you're... Well, I think it's not so much, hey, this is a... you know, rent-seeking relationship with the chain. It's more, yeah, I got a premium if it's pump chain instead of pump the app. And like, that's kind of the kind of incentives that we've seen, I think, in part why people are leaving to start app chains. And internalizing the MEV.

Yes, that too. I don't think the MEV can be that big of a revenue source compared to a chain that can charge a percentage on each, on the value transfer. Like, ultimately, they have the most control about value capture, and the MEV is going to be, like, a fraction of that. And, like, so I think it's going to be a better split than Apple and their 30-70 split. My guess is closer to, like, 10 to 90-10. But, like, the goal of the chain is to create an environment where it is as low as possible. And in my dream, the apps have the biggest validators because they're capturing part of that MEP anyways. And they can easily just like do that.

They run a validate and tell people to delegate there and they earn a return. So like in some ways, like if you care about MEP, you can just go boot up a validator and get halfway there just from that. I think you're going to see a bunch of apps creating their own chain simply because they're, there is valuations that give them a multiple, but all that stuff collapses to where one or two survive. And then you're like, shit, I'm on my own. I don't know, I'm enough users. So like that's just part of the normal cycle.

You have like investment dollars that tell everyone to unbundle. And then a whole bunch of stuff fails 1 and 20 survive, and then everything bundles back. That's part of the natural cycle. The advantage that, like, Salana as an ecosystem has, is we're only focused on bundling always. Like, it is one giant state machine. We need to increase its bandwidth. Everything needs to be as integrated as possible.

If the story is that Solana is one giant state machine, there's been this, there was a kerfuffle, I think it was like last month or a month before about network extensions. Exactly, which is this kind of loose rebrand of L2s, but also like L2 like things and or, non-SVM virtual machines running on Solana. So given the way in which Salana has been very critical of this horizontal scaling story in EVM land, how do you think about how network extensions as a concept fits into that story that you just told? The goal for the L1 is to increase its bandwidth

to the point where the price is so low that the stuff that's off-chain is just not really designed for running in this composable state machine. Like Pith is a great example. Like they're 1 to 2,000 TPS in their Pith network. The price that they want to pay for recording those Oracle updates for transaction is so low that it's not just like a business worth hunting for.

Right. It's not like all the Solana L1 engineers need to like bend over backwards and test the chain at 10x capacity just to win that to get like a few extra dollars per month or whatever. So that's the goal of the L1. It's like increased its bandwidth to the point that the stuff that's off-chain is priced so low that it's irrelevant to the actual value accrual to the L1. And those things that are like just, this is a database for this particular game instance. Great.

Like there's reasons to do that. You might want to like have that record and validity and all this other stuff and do some interesting games around like correctness of how everyone played, whatever. That's a business model. Go hack it out. But like, does Solana L1 need to increase its scale to go move each one of those game instance transactions to the L1? Like how much value it would it bring?

Like pretty minimal. But this is very different than let's have OP and Arbitrum create their own general purpose VMs with all these. the same defy applications, all the same kind of things and super chains to wrap them all together and basically compete with Solana, right? So the goal for Maine net is to not allow for that to happen, right? Increase its bandwidth. How do you not allow it? Like if somebody says, I'm going to create a network extension that is a fork of Salana, that bridges back to Solana, and this is for, you know, meme coins? Absolutely. Here's all the code to fork it. Here's all the support. It's all part of the

the same open source community, go do that, have a ball. But that's not the strategy. L1 is going to increase its bandwidth to make it so there's no rational price difference. Like the users are not going to get a better experience in this other thing. So there's no reason for them to switch over. There's no reason for you to run that business there unless there's other more, less engineering reasons. Maybe you think you get better valuation capture and your VC's tell you go, launch your own L1.

or whatever. That's fine. That is like, as a founder, I'm like, go take a shot at that. Maybe it'll work. Maybe not, right? But that's not the same as, like, Ethereum L1 is at capacity. We're not going to increase L1 bandwidth because we're afraid of losing homesickers. So go move all of your users and business into L2s.

Literally, like, bankless guys were saying L1 is for, is not for users. It's for L2s. never going to say that. That's never going to come out on my mouth. I'm always going to say that Ethereum Mainet, I mean, Salonamainet is always for users, right? The primary goal is to have

40 in the flip. Okay, all right. So in the spirit of deposition, so your claim is that, okay, well, Salon of Mainnet is always for users. The fees on Salon of Maynet will never rise. So by counter to that is that that's a function of supply and demand. Once you have fee markets,

it's not up to you whether or not fees on Salana rise. Fees on Salana will rise if demand outstrip supply. Salana is already at the point where network congestion has been pretty consistently high, especially lately. Actually, we had just another all-time high of activity, I think, over the weekend on Solana. And the stability of the network has been much stronger than what it was a couple years ago,

but there's still a lot of network congestion such that many transactions fail of just normal people trying to send transactions through RPCs. Now, this hasn't resulted in high fees. You can actually look at the stats. Like go to validatorsapp.com, and that is an uncontested transaction that is just no, it's not touching any congested state. And those are priced at like medium, like medium fees going through state to stake weighted QOS. They have zero failures because they're not touching hotsup.

Right. But if someone's trading something on Jupiter or someone's trading a meme coin, they're, they're experiencing. that, okay, a lot of times transactions fail? So those fees are inescapable to some extent. There's obviously engineering reasons to make it better, but there is a limit to how far they can go because you have a single piece of state that's globally replicated, that you have low liquidity,

high volatility, and many different users trying to touch. There's no way to fix that problem. And the only way to resolve the breadlines is to have people pay priority and get ahead, and that person gets to trade, but their trade pushes everyone else to failure because their slippage setting is set too low. And the only way we can improve upon that is,

go to 200 millisecond blocks, multiple concurrent block producers, all of the stuff you have to do anyways. Like an L2 will not solve this problem. They can hide it with like priority lanes to a single server or like effectively what NASDAQ does, right? You have your 10 jumps or whatever that pay. for like Ethernet cables of the same length. But there's no, it's a fundamentally a physics information problem.

You have to synchronize all the state and get everyone to go submit it. Like if you ran an OTC desk for any of these meme coins, your fee from spot would be much higher than everyone's slippage settings for you to take those orders at size. I will say one thing related to this, that is sort of the thing I think that really kind of changed my mind and opinion on Tlana from early 2022 to now is just like looking at the evolution of the scheduler code base within the client. So the scheduler, if you look at how it's evolved based on demand and based on adjusting to some of the events that have happened and how it's more tied.

to the MEV part of the systems and in Ethereum, you can see that by coupling MEP with how you're doing scheduling can actually generally load balance your contention a lot better, have a lot lower contention. Whereas I do feel like one problem of being completely agnostic to that in the way that L2s are, is that you kind of fragment your MEP in some weird way, such that it actually, if I add up all the MV, there's actually like worse losses for users. across all the chains in Niagara. So I do think there's, you know, one of the things that's, and you see this in Sway and Aptos too, right,

where the scheduling is like much more complicated. And I think like there is this one kind of annoying thing in EVM chains, which is when you look at how they deal with scheduling in parallel and how they look at the interaction when there is state contention, it tries to take a stance that's completely agnostic to what type of contention you have. But if you're agnotic, there's no free, lunch. Like, if you're agnostic, then you're going to have this fee problem, right? There's sort of

like uncertainty principle there. If you're, if you're perfectly agnostic, then your fees will have to be perfectly elastic. You can't have both fees bounded and your kind of like contention service bounded. And I think if you look at the evolution of the salinas scheduler, that seems to be a principle I would say I like see as an outsider reading the code base, looking at how it's constructing this DAG, looking how it's traversing it, looking at how it's like doing the local ordering. And I think learning that lesson live from production failures is is a very different thing than trying to a priori design it. And so I think that's where I see the yin and the yang of the Ethereum philosophy or swanah philosophy. I don't know if that's it's kind of related to

this thing you're bringing up fee wise, but maybe it's a little more in the weeds. So it reminds me of a what's that essay? It's like a day in a libertarian paradise where it's like, you know, I call the police and like I haven't paid my subscription and to like shoot my son and then like my house is on fire but I need to like pay another subscription and like you know it's like you remove any sort of basic support and it's like oh you let the market figure it out and you know it works to a certain extent. There is a lot of competition in that space too so there's just a lot of input from because that's like the beauty of it is that is very revenue there's a lot of revenue to capture there so a lot of smart people are all trying to solve what is like effectively

a knapsack problem heuristic. How do I pack these blocks with all these different limits and contention between like all these things to maximize revenue? And they're forced to look multiple layers deep into the content of the transactions to do it. And I think as a sign that I think Solana's scheduling and dispatch are, you know, more attuned for MEP and kind of like optimizing experience for a user at the same. same time is unichane kind of the flashbot's design copies shreds right like they basically

have a lot of the stuff that makes it easier to do scheduling in parallel um starting to hit l2s but my point is that basically or not having that does have some costs right especially when you're going cross l2 there's no way around my like long-term answer to murd like the short to medium term is i think like there's content and scale and the size of the network, there's network effects, and that is the killer content that Solana has that I think is much harder to bootstrap now. Because one, we had, we got the opportunity to do it five years ago when there was just a lot less stuff. And now all of these next generation networks are competing with themselves, plus the survival, the grizzled survivors of the last cycles, right?

I think long term, though, my belief is that the kind of end state design for Solana is, what finance needs and it doesn't matter what later it is. I think it's multiple concurrent writers that are globally distributed like a large set of them that is so large that the latency to get your transaction included anywhere in the world

to go to the nearest block producers shorter than sending it to New York. This is kind of like my dream is that we basically built Spanner for finance with like trusted not trusted, but trust minimize atomic timestamps for all the important financial events and all the stuff gets zipped together and priced as efficiently as possible.

That's, I think, the end state for finance. And it doesn't matter if you do it in L2 or not. Yeah. You made an interesting point that your claim is that, okay, part of the answer to Merritt's question of, you know, how does Solana win in a world where everybody can catch up with respect to performance and speed? and scalability. The answer is, in part, we got there first.

We were the first ones to figure out that this was the thing that everybody really wanted, and we just delivered the fuck out of it. And I wrote a tweet a week ago where I talked about that crypto doesn't really seem to have a first-mover advantage. First-mover advantage, I think in general, is kind of a bit of a meme

in that it doesn't really exist across most markets. But I think what you see in crypto is almost like, instead of a first-mover advantage, it's not quite a second-mover advantage, so much as a second cohort advantage is that the first cohort, these are the pioneers who go out and make the first set of mistakes.

So this is the people who basically immediately followed Ethereum. And we're all, you know, if you think about like ETH Classic and I don't know, just people who are copying Bitcoin or counterparty or people who just didn't quite have enough of a difference

and didn't really understand totally what people wanted. And that second cohort, you know, of you guys and Avalanche and near, they were all able to really effectively realize what Ethereum got wrong and what the people who were immediately trying to innovate on Ethereum got wrong and like sort of just gradient descent toward, aha, here's what people really demand. And I'm going to give them as much of that as I possibly can. That second cohort gets to do the land grab because they actually

got the price signal from customers of what they really want. And that land grab is a lot harder to do in the third cohort because everything is way harder when people already have. an empire that you're fighting against. So increasingly, I think, like, if you look across many different verticals, whether it's, you know, blur versus open sea, or whether it's, you know, Ava versus compound, or, you know, all these things where you see the second mover ends up winning, they aren't, they weren't even literally the second mover.

In many cases, there were many other movers who came after the first mover. But they were part of that class that got to observe the first mover and iterate against them. And I feel like that's a lot of what Solana has done really well. If the things you're building are like in a sigmoid, you get to the point of the curve where now improvements are marginal to the user, and that's much harder to like Wintam that way. Totally. Although to your comments about new content, right, like if this is still EBM's world,

then, you know, there's still many networks that are created after that were like they had completely new content that wasn't, didn't get copied fast enough. and then they were able to create a kind of new niche moat, right? Like, you know, you brought up the YouTube example. I also think of Vimeo, right? Like Vimeo, even though it's small, has a much smaller pay, but like highly paying user base and like has somehow survived from that. And I kind of think like there's, it's not clear that you know,

everything that all the content that ever needs to be on chain is already known. Like I feel like that that's the one thing I will disagree. I didn't say content. I said asset type. I think everything will probably... That is sort of the same thing. I think that is kind of the same thing. It's going to be a token or an NFT. Like even LP positions in uniswap or NFTs, like... I just think it's going to be more like state is going to be less fixed and it's going to have more streaming like behaviors where... It's like the rest of the internet. You look at the 90s internet. It's all static content.

everywhere, right? And then it became dynamic content. I feel like there's some notion of an asset type we don't know of that is that transition, right? And that completely changed how people made websites that changed how you would be catching. Yeah, I'm going to take Anatoly's side here is that I don't think we are going to discover new types of content. Like if you think about streaming, like it wasn't as though streaming was inconceivable in the 90s, right? It was just hard. It just we didn't have the bandwidth for it. But like all these kinds of content, Oh, short form video, long form video, images, games. These were all concepts that we were readily available.

They were just bandwidth for it. But it wasn't just bandwidth, right? It was a lot of, like, routing, a lot of, like, other types of optimization, a lot of, like, local caching. Like, like, my point is, like, there's a whole system of optimizations, though. This isn't just very incremental. These are, like, very primitive interfaces, too. It's like saying, oh, the internet still runs on HTTP, you know, it's like,

well, yeah, but the stuff that people are using it for today was like not really imaginable, you know, 30 years ago. And so it's like, yeah, like prediction market tokens might be tokens, but it's not really how people kind of conceived of tools. Yeah, yeah, yeah. That's kind of why I'm like, I don't want to, I don't think anything like there's no new asset. Yeah, look, I, I too, I too can also call H-ton's on a bit string and shove it somewhere and send it somewhere, right? That's not the notion of content, right? The notion of content on the internet is actually the entire, you know, abstract thing. It's not just like the bytes that are sent, like how you interpret the bite. In the same way, I don't think the notion of a single token is like the end-all

be-all in the long. Sure. There may be kinds of tokens that represent things that are surprising to us that they end up on a blockchain or surprising to us that people care about them or that they're valuable. I think I agree with Anatoli here that I think this is going to be a relatively small pocket of what happens in blockchains in the next 15 to 20 years. Like the reality is that almost everything that we see working today. How much Red Bull is an old idea? Hold on. Let me finish my point and then we can

conceive a bed if you want to take one. Almost everything we see that works on chain that is any meaningful scale is an old idea. Dogecoin is old as shit, right? Stable coins, old as shit. Prediction markets, old as shit. These have all been around for a very, very long time. Now, like, the way in which

meme coins are happening or the way in which we see these individual stories playing out are surprising. The details are difficult to anticipate in advance. But is the idea that like, yeah, we'd be using them for real world assets and staple coins and treasuries and meme coins and da-da-da-da? No, all this menagerie of things that we might buy and sell on chain, we figured it all out a long-ass time ago. In the same way that people talked about there being TV on the internet and voice over IP and all this stuff, all the old ideas. You know, these were not like, oh my God, in 2006, we suddenly discovered that you could

put video on the internet and it changed everything. Binding curves, AMS. clubs, all those things are covered kind of most of the market. I think, okay, the reductivism to everything as a token or NFT is like missing a lot, right? Like the examples you gave, like stable points

and prediction. I'm arguing at a stronger level than that. I'm not arguing, okay, yeah, token. I'm saying literally the use cases for the tokens have almost all been articulated in the past. That drives the requirements for the chain, right? Like, this is kind of the biggest slack that we get is that,

Solana SVM development just sucks ass. It's a pain in the ass. You have to spend a lot of time onboarding it. But to a business to like those developers, they're not the important developers. I love them. They're all important to me individually, but they're the plumbers. But pretty soon they'll be AIs. That part I don't know about, but they're basically building Windows drivers. It's the people that consume that stuff in JavaScript, and as soon as that stuff is in JS, that's the developer experience.

It's like the company that takes a bunch of JavaScript code and glues it together. They don't give a shit if the developer spent two nights on Red Bull optimizing SVM, stack allocations or whatever. It's all already in JS to them. All that stuff is hidden in the con. So like the developer experience that matters is, as soon as you're done with smart contracts, it's like all that stuff is already baked.

And like the number of smart contracts you need to write gets smaller and smaller every year because there is already a token implementation. And when there were a whole bunch of weird things that people want to do with tokens, there's now token 22, right? Or token extensions that cover that tail. And I don't know when the next set of tokens will need to be written by maybe 2030. Right? Like just the amount of like featureization that you need to do in the driver level,

in the plumbing, gets smaller and smaller every year. And that drives kind of the design decisions and everything else that people need to do in us in the L1. Do you need more VMs? Do we need WOSM? That seemed like really important two years ago that we're going to need to support multiple VMs and all this other stuff. But that optionality. I remember that being a big story a couple years ago.

It's just, there's no demand. The entire 2017 ICO boom was about alternative VMs, if you think about it. There's like pretty much. Why being an ICO was like we're a VMs running sideways. Well, there was like supposed to be a move VM within Solana and all this other stuff and there was interoperable. I totally forgot like this. The team that got to Maine that was neon, so they have like an EVM implementation, but.

But that's like a virtualiz VM, right? That's not at the, that's not like a side, a sort of. parallel VM? What the 99% developers want to do is they want to create a token. And there's already an implementation of tokens that nobody even like forks and re-uploads like they do on Ethereum. They literally just use the API, create new token in JavaScript. And they're done.

And there's no need for a new VM. Like you're not going to like spend, you know, 30, you know, L5 engineers like, for two years to build a new virtual machine just to have that use case be another token. And that's, I think, kind of like a lot of the sub functionality and like all these other things collapse when the use cases become much more fixed. You have markets, you have tokens, you have NFTs. And what matters maybe is speed and cost. And like compression was like one one thing that would like seem to matter because there's like 500 million compressed NFTs now and in in like Metaflex, like Merkel trees.

But to a developer, like, once that stuff is done, they're like, oh, this is a cheaper API with, like, different markets that run it. And they go plug it into a JavaScript, and it goes, right? That's it. Like, do they care what VM that is? Do they care, like, if it's built in solidity

or like any other, none of that stuff starts mattering less and less? Granted, right? But my understanding, I mean, I don't really care. It's not something that I think is particularly strategic. But the vision behind it, I assume, at the time that it was originally conceived, was that if you can have these VMs running in parallel, side by side with each other, in this sort of segregated state space, then Salana can eat everything.

Solana can eat the VMs. Solana can eat all the move VMs. They're not even segregated. You can atomically move state between them in a transaction, can hit instruction one VM, like SVM. Right, right. So there's a sort of cable connecting them, right? But they're... Foreign function interface.

Yeah. The problem is like, it's like Windows. Could run JVM and every other VM or Linux, right? From, I'm an... I spent 20 years working in operating systems. I know I can build this 20 different ways. The thing is, there's just no demand for it. What people actually, like, weirdly what happened with SVM is, it's a very restrictive environment.

There's no interfaces unlike Ethereum. So once there was one implementation of token called SPL token, there was no second or third implementation that ever got any meaningful market share. Because everyone would literally just use the JavaScript command to go create another token, just call the same program. And in a weird way, that allowed for the success of things like Pump and these other things because when you're creating a bonding curve or whatever or a market, you can consume any instance of SPL token because you trust the implementation.

There's no way for implementer of that interface to screw up or maliciously code the implementation to go fuck with your customers, right? If you're the front page for the markets and you're presenting markets, you have to basically allow list every version of ERC20 to make sure that when people click, deposit LP capital, they're not getting screwed because somebody like made an implementation that steals their funds, right? By by the sheer like weirdness of SVM, that just doesn't happen because everyone just reuses existing primitives. So there's no, like there's no optionality that devs actually need. There's no need for another VM because what is it going to do? Is it going to build a better

version of token. Like, if there was literally demand for a VM to go do something, like, you could just code this up into one standard implementation that everyone reuses and that plumbing work is done as soon as it's in JavaScript. So weirdly, I think some shortcuts that we took by skipping all these things that made a lot of sense to toward, like, if you were to tell me, like these choices that you're making are actually going to be end up beneficial like three years later, I would not have believed you because as a operating system designer, a VM designer, or whatever, you want to give developers maximum optionality. That is just kind of like this developer instinct. Like I want my devs to have the most widest power tools possible. I want

them to be as plugable and whatever, yada, yada, that ends up actually being a net negative in smart contracts. You want to limit the contracts to be as as harder to modify, as easy to like basically, what we're trying to do is like make it easier for businesses to talk to each other with the least amount of friction.

And that means you're standardizing over-implementation that everyone agrees upon are correct and cannot be attacked. Okay, so let's switch gears a little bit and let's talk about phones because I know this has been a passion project of yours and I know you have a long, background working in telecoms.

So Solana originally, so originally there was the Saga phone. And actually, when we brought you on two years ago, one of the topics that we discussed was Saga, the original Saga phone, which was back in 2022. So a brief history of the Saga phone. Saga phone originally launched, had okay sales, didn't fully sell out. Then Bunk, the MemeQuahun on Sala launched and did an airdrop to Salaun Phones. All the Solan phone, original lineup of Saga phones sold out because Bunk was just so. Bonkers ran up so much.

And basically the phones were worth more than the retail price. And so a new lineup of phones has now been announced called Seeker. Seeker, Salana Seeker. And Seeker's going to be launching in 2025. And there have been 140,000 pre-orders, which is pretty incredible number. How many total sold of the Saga phone? 20,000. 20,000.

So a 7X jump in the total amount of, at least. in the pre-orders. These are full-price pre-orders. The pre-order is a full price of the phone. Right. So this is like 50, 60 million in total pre-orders, I believe. Yeah. So you're pretty incredible numbers.

We'll see USC too. We're like the probably the number one merchant for circle. Nice. Nice. And it seems like many people are anticipating that there will be more air drops and there's like an NFT that's not bundled with a phone. to like make it very easy to air drop to to phone holders. So in response, seemingly in response to this, just in the last week, we had the announcement

of an Ethereum phone called ETHOS. And this Ethereum phone, it's derived from lineage OS, a fork of Android, and it's a mobile device for Ethereum-onchained apps, runs a like client for, runs an Ethereum like client. And it's seemingly not gotten anywhere near the kind of reception that. the Solana phone has gotten. Tom, I don't know, you were making some comments. What's your take on the new Ethereum phone? Well, I mean, they call it like an EDC

because I think they're not trying to position it as a phone. They don't really sort of hype up any of the. So what is an EDC? Like everyday carry. They're trying to invent a new category, I think. And, you know, I think they're downplaying any of the sort of telephony kind of stuff. I think they've actually kind of, I don't know how much of it's actually shipping with the phone. Is this like one of those AI, like the rabbit?

A little like a Reddit? Shipping a phone sucks. The certification costs, like all of the stuff, like 20,000 units is a loss effectively. So basically you need to break like 100K units to not be a loss on the, just building out the hardware. So I can, we debated heavily on whether the saga should have been like a non-phone, like kind of a to skip all the certification, all those other stuff. But my feeling is that, like the prize, the biggest prize is taking a chunk out of meaningful chunk out of Apple and Samsung, a Google kind of ecosystem by building an alternative app store with devs and like, and you need this to be a daily driver kind of thing. So if you're going to go build

hardware, might as well, I mean, if you can afford it, might as well go, like, take the bigger swing. What do you, what do you think of the ethOS phone then? And it's all like, given My understanding is that Saga as well as Seeker have been your personal passion projects. I don't know if it's going to, like, it's, I feel like I want to give those guys a hug. Oh, oh, that bad. No, like, they're, no, no, not because of, like, it's just the journey of building hardware sucks. Like, it is so painful. And they're, like, I want them to succeed.

I never want to tape out a chip again in my life. That was like the worst experience ever. Like I, if they need any help with anything, I'm, I will happy to like give them all the support that we have. And then with all the, with their hardware journey, like whatever resources we can give them. Like, I think it's just such a painful process. And I hope they succeed because I think if you can find an opportunity to create a new device category, that's huge. Right.

Like that. But it's really, really tough because you don't just, the iteration cycles are so slow. low. And this is why, like, especially going for a new device type, your, the obvious question is why isn't it's just an app effectively? Or a like client. Or like, like, if you have a really good like client, can I just have a mobile app that only uses the like client and like, why do I need the, I get it for the T and the custom hardware piece. So like, how do you differentiate like the needs of the phone between those two? So I sincerely wish them like the best of luck. And, like, Like whatever, if they've listening to this, if they DM me or whatever,

I'm happy to connect them to all our hardware people and help them in any way. I think it's just going to be hard. And it's really hard to predict success or failure there. But for us, like the thesis remains the same is like Apple and Google are going to fight any kind of like open app stores tooth and nail. And it's going to be like active Congress effectively, I think, to really open them up. And even if they do, like they did in Europe, the fees are like, I don't know if you saw their docs. And like, if you link out to pay through the web, you have to pay up to 28% and then show them all your books and stuff.

It's just insane. It's definitely going to be a long fight before we have like the final version of all this. So crypto is disruptive because it can create digital scarcity. and that changes the revenue model for people distributing digital assets. So you don't have to be as reliant on this giant distribution channel to find 1% of your whales. I think the interesting part here is that you can literally give a whole bunch of your digital assets to your whales to move them over to a better platform with lower fees. That's the idea.

That's the thesis. Maybe we saw like a sliver of that with Bonk because what we saw. was people that receive bonk bought sold the bonk and bought the phone because they were getting the covalent bunk back that was the weird part that happened it wasn't that they were like

I'm gonna because like the arbitrage doesn't make sense you're paying a thousand bucks to get a phone a week later on this meme coin that is priced right now most people are not dumb enough to do that I mean you can short the meme coin right like so but effectively

that's what they did they already had bonged, they sold a little bit of it to get the bunk back. And then they got a whole bunch of other rewards. But we need that loop to actually work repeatedly and consistently. Have you considered maybe just keeping the phones? Like, just make the phones and then just keep all the airdrops? So this reminds me. Why would they still air drops?

There's no distribution. Just don't tell them. Just don't tell them is all sold out. Sorry, you didn't get you. We're finding you. Keep all the phones. You announce the addresses first? If you do, I will invest in this. I will invest in this. This seems like this won't work because like, why would you get the aerodrops?

The Bitcoin mining A6 of 2016, this was my first drug. It was this company that made one of the first A6. They taped it out and then they mined for six months before they sent it. That's the game. If I remember, if I remember right now, delay the delivery so long that you get the airdrops, sell them, and then give people their phones.

If I remember right, that was Avalon, maybe? Yeah. If we can get like a million units out with a developer ecosystem that is competitive in revenue to anything remotely

of that ratio of Apple and Google, that is such a bigger prize than picking up, know, meme coins in front of a steamroller. I don't know, man. Bunk has done pretty well. I don't know. We'll see. But all right, I respect that. I respect that. Okay, so we're running up on time. We were originally planning to talk about news. So we were literally planning to talk about Trump's failed World Liberty Finance raise as well as goat token. We've had no time. We spent the whole time on this

deposition thing. Well, first, I want to take, how do you rank that deposition? I'm sure you've been on a lot of depositioning podcast. It was it? What is the word? thing about Salana. What do you guys think is like the Achilles heel? Look, I'm telling you, I, I'm not that critical of Salana anymore. This was a, here's my biggest, here's my biggest criticism of Salana is I think you guys do not take criticism that well. I think you take criticism very well. I think most of the people in my biggest criticism is you can't take criticism. There you go. There you go. I mean, I think it's, I think it's very valid. I think it's very valid. Here's the thing. Solana is

Salana still feels, I think, internally like it did in 2023, right? I think internally they feel like a sort of... It's just like... No, no, no, no. What I mean is that they feel like a minority. They feel like they're sort of on the back foot. They've been through a lot of pain.

And I think they hold on to this mentality, which is very real, right? It went through a super tough, dark night of the soul through 2023. But Salana now is the big boy on the block. Right? Salana is fucking $80 billion. You know, it's a massive network.

It's a success story by any stretch of the imagination. But I feel like psychologically, this lot of community doesn't feel like that. They feel very like, you know, sort of high on their guard. And I think that could stand to improve. No, they should, every founder needs to feel like they're a wounded, hungry animal and winter's coming. Okay, I have a criticism that's not as dumb as Haseebate. because Haseeb is actually kind of stupid criticism, to be honest. I'm going to be very blunt.

It's like, okay, fine. I think FireDancer is a bit of a LARP slash useless. I actually think like Agave and the Gito client are actually sufficient. And the, I think it's awesome as an engineering project. It's awesome as an engineering project. I don't think it will change any developer or user's experience on the network. That's why. Absolutely not.

So that I agree with you there. But what it does is a second implementation literally means that the bus factor is not one. It's two. Just the probability of a catastrophic bug, like a geth bug, goes down exponentially. Finance can run both nodes. Coinbase can run both nodes. But keep the development cycles synchronized on new patches in response to events, like making sure all the people.

I think you really do slow down your development process this way. I have a hundred of us to do. Maybe in Ethereum. I almost certainly true. I almost feel like they should be able to ship faster because they know that another team also built this thing and the probability of them. Divergence between Reth and Gep is a great example of this.

But liveness is not as important as safety. You have so much experience in software engineering among organizations. When you have two organizations that have to agree on how to build something, of course it's going to slow down. There's no fucking way you're going to move as fast when you have a single implementation that you can kind of be benevolent dictator. Liveness is not as important as safety, obviously. I think the difference here with Solana is that everyone is like

okay with a liveness failure, like a future one. Like we're willing to eat that as an ecosystem, everyone is like internalize those to the point they're like if it means double the capacity block capacity or if it means no safety violations or whatever. So like be it right. Obviously trying to minimize them and do your best. But if there's a fire, like we need to double block capacity next week. And both fire dancer and agave ship implementations and three months later they diverge because they rushed this out. Like, everyone has kind of internalized that cost already. I think you split developer mind share, though.

And like, don't get me wrong, fire dancer from an engineering standpoint is very beautiful. I think it has, like, the best Reed Solomon implementation of any client anywhere in crypto. Like, if you look at that thing, it is like by far the fastest. It's very efficient. You can tell KJB wrote it. And I just like, but I kind of look at that. I'm like, okay, great. We've really optimized the wrong.

We focus all these really smart development cycles on redoing, like, forking and doing the same thing twice, effectively, right? And I get that there is a benefit, but there's also this, like, cost of future features you can add, but with that you have when you split diverse people's energy. My counter to that is, if anything, finance is showing me that

all this other stuff is marginal, but you can just like, like if the performance that we saw at like FireDancer demo, one million TPS, if that was on Maynett now, would anyone give a shit about any other protocol changes? Like at all, ever.

Or it's done. I mean, it doesn't matter if it's not. I want to see that in a realistic setting where you're getting DDoS and spam. Sure. I want to be clear. Those are all like day-to-day engineers.

problems, but like from a protocol perspective, how much do you care about anything else? Like, all these other improvements have become so marginal compared to just raw horsepower that does it even matter to change anything? I do agree that if fired answer is successful at achieving that type of number in production and it becomes a default client, it does become much harder to Merck's question of what is the moat to get around that. On the other hand, there's this problem that two clients could also slow down your feature development that makes it easier for developers to write new software.

Like, your devX might get worse because now I have to have to support two platform interfaces. And that is a devX was a point. WebX doesn't matter. Okay. Well, my point, guys, we're on time. This is, we should have started with this. That would have been like. Yeah, yeah, yeah, yeah, yeah.

What can I say? What can I say? You guys were rolling for so long. I was like quiet saying nothing for a while. All right. All right. We got to bring you back at some point. And we'll do part two of the deposition.

Until then, I want to say congratulations on all the success. A lot of respect for, of course, Slonne Ecosm and for you in particular. I think I said this last time to close up the show. But I think you are one of the treasures of the crypto industry. Not a lot of founders who've gone as far as you have. Not a lot of founders who've gotten as far as you have have have this kind of positive positive sum optimistic mentality, and I think it does a lot for an ecosystem that can sometimes be jaded and dark and, you know, misguided. So I want to say we all appreciate you for that,

and we welcome you back on anytime. Appreciate it. Thank you. And hopefully Raj doesn't, hopefully Raj does eventually realize this isn't the, the hater podcast, as he wants called it. Salonis, you think we're the hater podcast. So we got to, we're just got to own it. We just got to own it. Although somebody also did say we're Solano VCs now. So we kind of, you know, we're sort of a chameleon that way. Anyway, all right. We got a wrap. Thank you, everybody.

We'll be back next week. Yep.