Unchained - The Chopping Block: Why the Azuki Elementals Drop Was a Big Flop - Ep. 512

Episode Date: June 29, 2023

Welcome to “The Chopping Block” – where crypto insiders Haseeb Qureshi, Tom Schmidt, Tarun Chitra, and Robert Leshner chop it up about the latest crypto news.  Listen to the episode on Apple Podcasts, Spotify, Overcast, Podcast Addict, Pocket Casts, Stitcher, Castbox, Google Podcasts, TuneIn, Amazon Music, or on your favorite podcast platform. Show highlights:  why the Azuki new NFT collection 'Elementals' caused such a controversy among holders why Robert bought an Elemental and what was he expecting who the Michael Milken of NFTs is, according to Tarun how BitGo 'pulled a Binance' on Prime Trust how Prime Trust, a crypto custodian, collapsed and was put into receivership whether what Prime Trust did can be considered fraud and how the situation highlights DeFi's 'long term advantage' why the majority of blockspace is EVM-compatible what the problems are with current Layer 2 rollups such as Optimism and Arbitrum why rollups are so 'political' in the Ethereum ecosystem what makes Arbitrum better than Optimism, according to Tarun what shared sequencers are and whether they will be the future in crypto Why Haseeb is not so bullish on the app-chain thesis Hosts Haseeb Qureshi, managing partner at Dragonfly  Robert Leshner, founder of Compound Tom Schmidt, general partner at Dragonfly  Tarun Chitra, managing partner at Robot Ventures Disclosures Links Unchained: Polygon Cofounder Proposes Upgrading PoS Chain to zkEVM Validium The Block:  Azuki NFT collections ends Elemental mint with $37.5 million Fidelity preparing to submit spot bitcoin ETF filing: Source CoinDesk: Azuki NFT Prices Slide 44% After Creator Releases 'Basically Identical' Elementals Nevada Files to Place Crypto Custodian Prime Trust Into Receivership Azuki’s thread on its situation Learn more about your ad choices. Visit megaphone.fm/adchoices

Not a dividend. It's a tale of two-quan. 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. D5.8 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 we've got Tom, the DeFi Maven, and Master of Memes. Next we've got Robert, the Crypto Connoisseur, and Captain of Compound. Then we've got Tarun, the Gigabrain, and Grand Puba at Gauntlet. And then finally, I'm a seed of head-hike man at Dragonfly. So 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 blocks at XYZ for more disclosures. So we were just freaking out about this new Azuki drop that has got the internet completely lit up at the time that we're reporting this. Robert, please explain for us, we, okay. Robert, please explain for us

what happened with this Azuki drop. Can you just walk us through the storyline here? So Azuki announced that they were going to be doing a new collection. And they did this by surprising everybody who held Azuki's with some like token, a bean NFT. And the bean NFT would become a new NFT called an elemental. And they decided that they were going to give out a bunch of elementals to existing Azuki holders and then sell a bunch more. And what happened was they've raised a ton of money by

selling these new elemental beans pre-reveal to people in a Dutch auction. And obviously, almost every Dutch auction in crypto history sells out instantly at whatever the starting price is, because everyone is so excited. I don't know why anyone uses Dutch auction mechanisms anymore. It's a facade. It's silly. Just set the damn price it'll sell out. But each one of these sold for two ether, and they raised something like $36 or $40 million selling these new elemental NFTs. And everyone was excited. They're like, oh, my God.

Like, you know. As you add, so people, after the Dutch auction to the initial holders of the beans, which is this previous kind of low value Azuki collection, it was supposed to also go to a public auction. So people in the public could also buy it. But it never got to that point because it got sold out immediately by the, by the bean. which made the community already very upset about the way this was managed. That's right. And then there were some issues with the website, and it took a while to be able to reveal them. But the long story short is once they were revealed, the big reveal that like

everybody was waiting to see what these cool new NFTs were, well, they're shockingly similar to the original collection of Azuki's. In fact, I challenge pretty much anyone to go on to opency or blur or, you know, your favorite. NFT platform and compare Azuki's to Azuki Elementals. Basically, it's the same collection. And so, in essence, what has occurred is they've, I think, like, tripled, you know, the number of Azookis out there. Yes, they have a different name, but they're Azuki's. They're basically identical. Tom's flipping back and forth between Azookis and Elementals, and you won't be able to tell the difference. And there's a couple of differences, like there's new, a couple new traits that the

elementals have that the original Azuki's don't have. And there's like slightly more frequency of like cool, you know, elemental things like fire and lightning and earth and, you know, all this stuff. But fundamentally, they're Azuki's. And they basically use this as a way of like, you know, massively increasing the size of the collection. And so everybody's up in arms. The price of Azuki's dropped like 40 something percent pretty much instantly upon this reveal. And crypto Twitter is captivated by this new mechanism. called, you know, sell more of the original NFT. You know, there's two main lessons from this.

First is Robert retiring from NFTs was a sci-off. Total scam. Clearly, you are still very plugged in. No, I'm an observer. I may dabble here and there. I will, for disclosure purposes, state that I do own two elementals, one that I got for free and one that I purchased below the Memphis. Okay, very good.

The second thing is, and I think I saw someone sort of hinting at this, but Azuki was maybe one of the only big of the large cap NFT projects not to do a fundraise. So maybe this is like they needed cash and all the other. Oh, they were trying to do a fundraise. They were trying to do a fundraise. They were going out to market alongside. I don't remember. But did they get it?

Did they close? I think they did. But also, I don't think they needed the money. Like, I think, you know, they were making a bunch of money on royalties and even primary minutes last year. So I feel like it's like NFTs. They're all trying to figure out their business model and how this whole thing works and looks sustainable. Like, do you try to turn this into a media business or like a gaming business or is this a pure art play and you try to make money on royalties or you sell plushies, I guess, if you're penguins? but I feel like the trick here, like the, really the, the grail is like being able to do continue,

continue as new primary issuance. Like being able to sell 40 million dollars worth of JPEGs is like fucking crazy. And in theory, you know, depending on the market or more true historically, you're also getting some residuals and some secondaries, you know, royalties on a secondary sales as well. And so it's all kind of a question of like, well, how can I keep my community going like by throwing events or doing some giveaways? or doing something to sort of allow you to do new primary issuance. And like everything is almost sort of a game around that versus any of these other things, in my opinion.

I agree. I mean, the most concerning thing is that, you know, the standard playbook for an NFT project, like regardless of if it's like media or whatever, is make more NFTs. Like that seems to be like the go-to instinct for almost every single project that's successful. And, you know, there's various degrees of success. and credibility in that approach. But almost every project's instinct is like, make more NFTs. You know, it's almost like, you know, there was projects very early on that were like,

oh, what do we do next? We'll make another token. And like there's projects with multiple tokens. Like I feel like the instinct is just make more NFTs. And I think that's the wrong instinct. I think like to succeed, you know, project will be better off instead of making more NFTs and making a more complex ecosystem and blah, blah, blah, you know, figuring out things that you could do for the existing NFTs, not just make more NFTs.

I know this show gives a disclaimer about not providing any investment advice, but I would love to know what your thesis was for buying one, because I don't understand, you know. Okay. What were you expecting them to look like when you were buying the elementals? I mean, honestly, I was expecting them to look like something different from a lot. Azuki's better than beans. I don't know if anyone out there's an NFT collector, they launched a collection, you know, at some point called beans. And the beans, I think, are the most ridiculous NFTs ever made by like a blue chip-ish collection. Yeah, they're not very cool. It's like very hard to imagine somebody making a bean as their PFP. No, I personally think that mutants are extremely uncool. Like, Bored ape was, you know, awesome and they made mutants. And they're like, obviously, like, very aesthetically unpleasing. They're like not cool. They're kind of offensive. Beans were kind of like the mutants of Azuki. They were like, silly. looking and kind of stupid. And why would you really want beans? It was almost like a joke that they were playing on their own collectors. Like, oh, you're going to want beans, right? And like the beans

were these little, like, silly things. And, okay, it was a massive divergence from the original collection. So I just assumed incorrectly. I think everyone on Twitter assumed incorrectly that by releasing a third collection, the third collection would not be Azuki's. It would not be beans. It would be like some other thing in the universe instead of just being more Azookis. I just, I really struggle, one, that they didn't understand that if you are trying to maintain the exclusivity of your brand, that you don't just make more of the same thing. If Rolex decides all of a sudden like, hey, we're going to create a hundred thousand of these this year just for the hell of it because, you know, whatever, we need some revenue.

That is such a bad decision, although it's kind of obviously going to make money right now. It basically torpedoes your business. It's hard to imagine that they didn't understand that. And I guess the other side of it is, it does seem like a bit of a failure of imagination that they're just like they're just more anime characters that are also in profile looking to the left. You know, like there doesn't really seem to be that much. Like you mentioned, there's like, there's a few other animals. There's like raccoons or cats or something.

There's a few animals. Yeah, most of them just look like ordinary. I would not know those are not part of the original collection if you had not raised it to me that, oh, no, these are elementals. These are different. I can't believe this is turning into an NFT podcast. I feel like we have to be that NFT podcast now, now that NFTs are not cool anymore. I agree. We have to embrace the least cool thing at all times.

That's right. That's right. But what was so notable about this collection was that people were talking about how this is the revival of NFT primary sales because this is the largest NFT primary sale in like six to seven months when NFT volumes have collapsed and volumes are super low and this feels like a dagger in the heart of NFT collectors I'm just like why would you ever buy another mint you fool?

You missed out on the ladymaker of Fumo mint which is another notable primary primary issuance. How much did that raise? I don't know an aggregate but they're going for like 40K now but it's it's like a it's very different. They're like plushed animal or,

uh, yeah, stuff animals. You gotta pull one out. All right, all right, all right, all right.

Okay, let me describe. Can't do an Ft show without you showing us what you're showing us the JPEG. Yeah, look at that. What way,

what? Anyway, they're, they're redeemable for, for an actual flash. Oh, oh, oh,

oh, I see. So it's really like a collector's item. Correct, correct. And they, they sold on a bonding curve. So, it's,

Unisog style. It's Unisogs style. It's Unisogs for Malady plush toys. Yeah, I see. Okay. Got it. Do you own one time? I did not. I do not. I miss the mint and too far up the bonding curve now. You're no longer as cool as I thought you were. Sorry. Wait, wait, wait. Can I ask a more opinionated question? I was watching some documentary on Michael Milken the other day because, you know, I just forgot. I forgot he got pardoned or something recently, which I thought happened a long time ago. So I was just like, oh, let me go revisit some financial history. So the real question to me is who's the milken of NFTs to you? I have my answer, but I'm curious what everyone else's answers.

Hold on. Before we get there, my first question is like, is this what, like a Friday night with a girl? Is this what you like watch a documentary about the history of like financial engineering? Sometimes you just got to where you got to. Okay. Got it. Cool. Sorry, Robert.

Go ahead. So I was going to ask Tarun, since I've never thought about who. the Michael Milken of NFTs is. Turin, who is... Machi. Ah. Machi Big Brother is a, in some ways, has very similar style, uh, aesthetic in terms of the types of equivalent junk bonds, I guess, that exist in NFT land.

It's neither a compliment or an insult. It's sort of like a liminal sort of thing. Yeah. How would you describe the, the milkiness, you know, if you were to be still a down? I think it's just that like, no, like, no, like, no, like, no. one else ever sells a ton of things and then immediately goes and takes out loans on the same thing and takes leverage and buys them back. It's like, it just like reminds me of the Ivan Boski and and Milken like back and forth where like Boski would like manipulate some price and then Milken would like sell things into that price manipulation. And like, you know, it's like kind of amazing you can people seem to be doing that directly with, you know,

And so that I made me, maybe that's too niche of an answer. But, you know, I don't know enough NFT traders. I just know him because he's the most famous one. I got to say, I am so outside of every aspect of this conversation. I'm going to vote that we end the NFT section for the show, but we're going to bring this back. Hey, hey, hey, hey. You know what now that if he's not cool anymore, we should open every show. If anyone has their own notion of who the milk in this, please tweet at me.

I'm curious. Yeah, tweet it to Roon. Please tweet it to Roon for your milken of NFTs. Okay, so moving on to more traditional finance, prime trust. Very much in the spirit of the times, there's another big failure slash, you know, spontaneous combustion in the crypto realm, which is this custodian called prime trust. So I knew of prime trust. I never really knew that much about it. I knew it was kind of out there somewhere in the world.

So last week, the story was that they were going to get acquired by Bitco. They were in some kind of distress. It was very unclear exactly what was going on, but there were some rumors that things were bad in a bad state over at Prime Trust. Prime Trust, to be clear, is a fairly small custodian. I think the three biggest custodians in crypto

are probably Coinbase, BitGo, and Anchorage. So Prime Trust is a much smaller player than any of those. Obviously, there's also Fidelity in NASDAQ and those other folks who are launching custodians, but Prime Trust was always there. So BitGo was going to acquire them, and they sort of pulled the Binance a little bit of they said, oh, we're interested in acquiring. We sent them an LOI.

they kicked the tires and they said, oh my God, never mind, we're not acquiring these guys. And that caused anybody who had money at Prime Trust to be very sad. Now, I'm not totally clear who has money at Prime Trust because everybody seems to claim they don't. So I think it was rumored that TrueUSD had money at Prime Trust, but then they've disclaimed having any money at Trium Trust except now they've seen they have $26,000 at Prime Trust, which is basically nothing. So just today, the Nevada financial regulator basically said that, Prime Trust should be put into custodianship and it's now reported exactly what went wrong at

Prime Trust. We have a window into how this custodian ended up collapsing. Basically, they ended up at one point, so initially they had their own crypto wallets that they were managing on behalf of customers. They ended up moving over to Fireblocks and having Fireblocks manage all their wallets, except they had this concept of legacy wallets, which were basically their old wallet system, I guess, and for some reason, you know, fireblocks didn't have certain flexibility they wanted, and so they were creating throughout 2022 some of these old legacy wallets, using their old system, but it turned out these legacy wallets were not being persisted or something.

And so they essentially were creating these wallets on their own interfaces, but they did not actually have the private keys to these wallets. And so they were telling people, ah, yes, we created a new wallet for you and your Bitcoin is over here, according to our internal ledger, but they did not have any of the crypto.

And so when they realized this, they were like, oh shit, we don't have the money that we keep telling our customers we have. Let's go use our cash deposits, which are being given to us buy customers who are depositing Fiat in their custodian. And let's use that to buy the crypto that we don't have to give people withdrawals and kind of keep the party going.

So this looks like just a total complete circus. So they're down. So their liabilities, they have about $85 million in crypto liabilities and about $70 million in, sorry, maybe I'm getting this money. What's the cash liabilities? The other way around. it's $85 million in cash liabilities and like 69 mil in crypto liabilities. But the crypto is almost one to one.

And it's almost all Audius tokens, audio. The rest is like some Bitcoin and stuff like that. But yeah, they only have 3 mil in cash and, you know, so they're like 82 mil short. And so, yeah, it's like where do the money go? Maybe it was all lost and sort of trying to buy back the crypto that they lost. But also, I mean, they're custodians. They should have insurance. So like, the whole thing seems very goofy to me.

I would be amazed if insurance pays out for something this retarded, like this stupid. I mean, like, I cannot imagine how an insurance doesn't be like, yeah, sure, that seems like that falls within the bounds of your insurance policy. I mean, actually, this is a good question, though, because I thought insurance was actually still hard to get for a lot of custodians. I'm sure, I'm sure. So I'm sure also their deposits are like, you know, they have like. Maybe they just like went way over their cap and they're now not now not. covered because they went over their total. I mean,

I also think they don't cover fraud, right? This is fraud, basically. Right? They are, they are. But it wasn't fraud when they just lost it. That was like business practice insurance. I don't know. But like when they, when they fucked up the key management

themselves, that wasn't fraud. That was like incompetence. But like that was an accident. That was an accident. Eredically, that could be covered. Right. If you then, at the moment, at the moment you realize it, you tell your customers and you're like,

yo, we don't have the money. We don't have your crypto. Yeah, we threw your, we threw your crypto in the trash can, but we'll buy some for you. At that moment, yes, that could have been, the fraud claim could have been averted at that point. But when you keep the show running,

when you realize and you start buying crypto with your cash that belongs to customers, because it's a custodian, right? This is not their money. This is not like an exchange where, these, you know, it's all basically in an omnibus account and, you know, who knows really what the bankruptcy rules are. For a custodian, the rules are very simple. For a custodian, it is not your money. It is the client's money, period, full stop. So this kind of thing is, you know, taking their

money and using it to buy crypto's theft. Yeah, it's trading your customer's assets without their awareness of it to try to make it all back. I mean, that's just crazy. Well, to be, to be fair, Also, like, under what theory of making it all back, you know? All of the 2022 stuff was that. Yeah, to be clear. So, like, at least FTX, there was, like, some theory of how you made the money back, right? FTCX at least was like, oh, I'm going to, like, go long my shit coins and, like, I'm going to trade it. I'm going to do these brilliant yield farming techniques.

With these guys, I'm like, why would you, why would you do this? Why would you not, why would you think like this isn't going to immediately blow up in your face when you just have three million in cash versus 85 million? in liabilities. Now to be clear, these are all relatively small numbers in absolute terms. This is a small custodian. So we can kind of point and say, ha, these guys are idiots. And audio, so their crypto liabilities, you know, the 70 million in crypto liabilities, obviously a lot of it's already been drained because people heard all the instability. And so they were pulling their money out. So this might be a pittance compared to what they, what they previously were custody. I don't know. But the crypto tokens, as, as Tom mentioned,

61.5 million out of the 69 million in liabilities are Audius tokens. It's like 30% of the audience supply, which means that the non-audius crypto is like $8 million of like Bitcoin and ether or something. So this is like a very, this is very small amounts of money that these guys were But also, also, I think the true USD thing was kind of a little bit weird also, right? Because true USD is one of the, they were kept a large portion of the true USD supply at Prime Trust. So, yeah, I don't know what's happening with True. USD because like I think you know some of these rumors around insolvency came out like a week ago

and then there was a big you know true SD short but then finance is also like incentivizing true USD like not charging fees on the pair so I like don't know what is actually happening it wasn't all at prime trust so I think I think that's basically what what people are basically yeah I think they claimed they took their money out at some point quite a bit quite a bit ago before even any of this recent instability they took their money out then they said after prime trust went under that they had $26,000 at Prime Trust, which is, again, like a rounding error for true USD. So, yeah, all in all, kind of a clown show. So now the Nevada regulator is saying, okay, let's put this thing under receivership and shut it down.

Like, why is this still operating? I don't know. But, you know, again, this is such a small custodian in absolute terms. Like this is probably, you know, like the big custodians are probably 100 times bigger than these guys. at least at this point, I don't know what they were at their peak. But just goes to show the importance of diligence. And, you know, and also the important thing of just knowing what you're, knowing the competence of your business leaders.

This is one of the things people don't give enough credit to for Coinbase. Corpiz has been around for, how long has it been? Like eight years, nine years since Coinbase's inception. And they've never gotten hacked. And that is incredible. an absolutely incredible record of cusseting that amount of Bitcoin, being a honeypot for that many years, and never getting hacked. Like, and people don't, that is one thing I feel like people do not give enough credit to

CoinBits 4 is just having absolute world-class obsec and robustness in the way they do things. That's part of the reason why they're so trusted by the industry. Agreed, but the bigger issue here beyond not getting hacked, because in essence they got hacked, I mean, they didn't get hacked, they lost the assets. But like there's the assets go into the black hole and then there's the cover up, which is the consequence of it, right? Like if Coinbase somehow had assets go into a black hole, which it always could, right? Like there's no guarantee that any custodian can keep all the assets completely safe at all times. If the assets go into the black hole, you know, you're basically saying you would trust them not to cover it up and lie about it and fraud it.

And so the real problem here that I see with prime trust is not it's hard to save. because it's extraordinarily hard to safeguard assets, right? Even as a custodian, it's how they handled it, which is like the truly like disappointing thing. Like it's so hard to safeguard assets. It's hard for individuals. It's hard for businesses. It's hard for custodians whose job it is. Like it's just hard, right?

It's hard for exchanges. How many exchanges have lost assets? Mount Cox lost, you know, lost assets, right? Like Quadriga probably lost assets. You know, it's like there's so many instances of people losing in crypto assets because keeping them safe is extraordinarily complicated. And from there, it separates into the good guys who own up to it and the bad guys who cover it up and dig themselves deeper, trying to make it all back somehow.

I agree with you. Obviously, the thing that needs to be condemned is that they were hiding it and covering it up and trying to run this kind of Ponzi-like structure. This is almost literally a Ponzi scheme where they're taking deposit. it's from new people and using it to pay out folks who had their crypto black hole. The thing though that, like the core root of it is

that these guys had no business custody in crypto if they could not figure out how to make sure that when you send the money, that there's a private key associated with the address receiving it. By the way, this has happened once before also. Not even the one-on-one. I don't know if you remember there was this like steakhound thing, which was like an early-eat-staker.

Right, which also got locked out of their fireblocks assets. BLS key shares are much harder than the private key. But if you're going to be doing them, then know what a BLS, know how the fuck a pairing works. I do not. There's no way this has anything to do with BEL. No, no, no, but you have to understand how to maintain them, make sure you understand which share to keep, which one to dispose of, which one is trusted setup.

Like, if you fucking can't understand what a pairing is, like you should not be managing key money using it. Yeah. No, that's right. And look, I mean, ultimately, they kind of made the right choice in moving to fireblocks and saying, okay, we're going to outsource the technology part of the custody. to fire blocks, which is, wow, great move. But then they second guess themselves.

They're like, oh, wait, no, we need this other feature. Who knows what it was. And that caused them to just completely screw over all their customers and themselves now. So, anyway, my takeaway from this is I always go back to first principles. This is yet again, one of the reasons why DFI has the long-term advantage. Because if a DFI protocol somehow black hole assets, everybody would know right way. And there would be no ability to cover it up. There would be no fraud next steps. There's radical transparency, and this would not happen with a smart contract. That is true. But, okay,

there's another conversation to be had here about self-custody, right? And that custody is hard. That's part of the reason why people want to outsource their custody to fireblocks or to Coinbase or to whatever. It's extremely hard. That's why I gave all my crypto to Tarun, and he keeps the private key safe. It's still safe, right? You're not supposed to ever answer, though. Yeah, that's the beauty of trusted third parties. They solve all. No, I mean, I do think I hope that stuff like this forces people who are aiming to be

custodians to not to actually understand how cryptography works. I feel like the problems are a lot of people who are like, oh, this is going to be just like, you know, running a random software company or like, oh, this is going to be running like a trading desk. I don't have to think that hard about the cryptography stuff. Isn't it just like call this library and hit encrypt and decrypt and like it works?

And I don't have to think about what other fucking files it's generating, how I generate the entropy, dot, dot, dot, dot, dot. And then like, you know, I think there's a lot of people who are very over-eager and think they already know that stuff without realizing they don't, which I think was also kind of true

in the FTX case too. in some ways. I feel like it's been a while since we've actually had like a large amount of crypto get black hole, like the parody wallet stuff was kind of like the last thing that comes to mind. But beyond that, like it's usually almost always hacks, right? Somebody's stealing it. But like very rarely does it actually get stuck somewhere because someone lost access or lost the private key or something.

Wasn't there, winter mute, didn't they do something? Oh, with the multi-sig? Yeah. Was that black hole or was that hash? No, that was someone else was able to get their address and take the money. Yeah, they, like, deployed the NOSSA-safe to optimism to the address that they should have deployed to and on the main net. And so, yeah, sneaky. All right.

All right. Anyway. But, wait, wait, wait, it was, but the thing to Robert's point earlier, that Wintermute thing was observed almost immediately by tons of people, right? Like, it was not like, you found out a year later. This prime trust thing has been going on for a while. right? It wasn't like two weeks ago they started doing that. Yeah, this was last year. Yeah, exactly, exactly.

Throughout 2022 that they were doing this. Listen, every Ponzi scheme starts with someone like losing a little money and being like, oh, I'll make it back before anyone finds out. That's like Bernie Madoff. That's like every, it's like half the Ponzi schemes. The most hilarious thing to me was like the, in 2021, when people were overly bullion, anytime there would be a tiny market downturn, people would say that make it back in one trade meme.

And like, I sort of think they really. I thought it was a joke and then now I've come to realize two years later that all these people took that shit seriously. A lot of people did not get the joke. I thought it was a joke. I like can't believe. Yeah. No, but here's the other thing too, to ruin your earlier point about like custodians should be scared of this industry because it's hard and getting one little thing wrong can mean basically the end of your business or loss of massive amounts of customer funds. It's one of the reasons why, you know, when we started on the investing side,

basically like we, you know, looking at custodians, because there were, of course, a lot of custodians back in the day that you could conceivably customer your digital asset side, and now they're even more. The number one thing is just how strong is your tech team? Like how strong is your paranoia? How strong is your OPSEC? And that's one of the things that, you know,

despite the fact that there are all these, you know, great stalwart financial firms that are now getting into crypto custody. for us, one of the reasons why we default to the coin bases and the anchorages of the world is that they've been doing this for so long and they take it so seriously. Like for them, it is absolute, you know, of prime importance to never underestimate the task that's in front of you, which is keeping your crypto safe, not messing it up yourself and not getting hacked. And it's so easy.

You get it wrong once. This is also why I'm quite negative on some of the traditional finance custodians doing it themselves. I think they're going to end up just buying someone because they like are not going to be able to. They all seem so overly optimistic of like, oh, we're just going to hire like five people and do it. And I always like, good luck. But the one thing that's funny about it is none of this stuff is very static, right? Like if you look at the choice of like public keys, even through the Ethereum merge, like understanding like the changes to the curves, whether you should be expecting the same public here or not, whether you should be expecting like the,

the signatures to look the same, all of that stuff in this space changes much faster. You know, like in Web 2 land and definitely in finance land, everyone always tells you the following added, which is like never roll your own cryptography, never make your own cryptography, like just use some other library. And cryptocurrency is the only space where actually every fucking chain is writing its own new thing because they were actually actually implementing some new cryptography that had never been used before. And if you, the user, aren't cognizant of that in and of itself when you're doing it, you're just going to completely fuck up. And like, I think like somehow Tradfai people have this like, we already know this.

We're so smart attitude towards this. And I don't think they appreciate like the nuances in these systems from a math standpoint, from a CS standpoint, from an engineering standpoint. There's like so many layers to it. It's not just like. press and crypt and decrypt in the library, which I don't know. Sorry for my little bully pulpit, but Tommy, you were going to add something? Yeah, I was going to say, I think obviously having a great tech team is important and, you know, all as equal better than not. But I think, you know, independent of that, I would actually much rather take a team with like an insurance policy from Lloyd's over a team as a great tech team, but no insurance any day. And so I think there is something

to just like falling back and having some sort of. sort of recourse because it is sort of the... Nobody's insuring custodians for like the entire assets. No, not yet. Probably not even for like 20 years. Not until like the science of keeping the assets there is like soft. But this gets to my point of like I don't think we've, this industry is not like stationary cryptography.

Like every single time there's like some new signature scheme. There's some new encryption scheme. And like that's most of the advantage. that's mostly the advances in this space that are not found anywhere else is that. So because of that, it's like, I don't think we're anywhere near some crystallization point, except for Bitcoin keys. Like, Bitcoin basically is not changing. Like, it's plausible that if you have, you know, a mega player, like a fidelity,

that they could potentially self-insure and basically say, like, look, if you want an insurance policy, you pay extra and we'll use the rest of our balance sheet to insure your deposit. That seems plausible to me. And maybe that is a structural advantage that somebody like Fidelity could, could, could lord over a coinbase or an anchorage or a bit go. But right now, I mean, for those of you who are not in the industry and don't realize this, it is almost impossible to get insurance in crypto.

Like underwriters just do not know how to underwrite crypto. They don't know what the risks are. There's too little history. The blowups are big and dramatic, but then like a lot of things are just fine for a very long time. It's very difficult to underwrite this space. So, you know, ask anybody who's in crypto trying to get insurance, mostly you're just going to get the door slam in your face. or you're going to get crazy, crazy premiums,

which for a startup, you know, just don't make sense, right? You can't be paying, you know, half your runway and insurance premiums. It doesn't make any sense. So that will someday change. I don't know if it's 10 years, five years, 20 years. I hope it's not the longer side of that. But the trad-fied players who are very big and have different, you know, diversified businesses, I could see them gaining an edge there

in basically saying, look, we're going to self-insure. But, yeah, Lloyd's is not going to be insuring a custodian anytime soon. So, okay, let's move on. So it was a little light on news this week, so we thought we'd check in with a little bit of a different angle and do kind of a broader examination of the state of crypto. We wanna talk about layer two's,

because there's a lot of conversation going on right now about layer two is one of the most dynamic areas within the broader crypto smart contracts ecosystem. So we wanna talk a little bit about the state of the market and some of the interesting things that are going on in the layer two world. So I'll start with a little bit of exposition and then we'll just kind of kick off a broader discussion.

So L2s, for those of you who are not aware of what an L2 is. An L2 is short for a layer two. A layer two is basically anything that roots its trust in the layer one, but creates sort of another layer of blockchain block space that you can interact on that's not directly on the layer one itself. So the way that I like to analogize it is that a layer two is kind of like a skyscraper that creates more real estate on top of the underlying city. So in the case of layer two's mostly today we're talking about roll-ups,

although lightning on Bitcoin is growing quite a bit. Actually, there's a lot of news about that, but let's mostly focus on roll-ups. So roll-ups, there are today, call it four main players. There's more now, but we'll say that the big ones today are there's Arbitrum, which is the leader. It's got most TVL. I think they've got like $5 billion plus.

They've got a lot of daily actives. So Arbitrum, huge player. Then there's optimism. Both of those are optimistic roll-ups. And then we've got ZK Sync, which is kind of newly on the rise. They're growing very quickly. which is a ZK roll-up and then there's Starkware, which is not EVM compatible, unlike the other three.

They have their own totally separate language and ecosystem. So there's a few things that are going on right now. And I'll have the first topic of discussion be about layer threes. So layer threes have kind of become this meme that's becoming increasingly popular. Everyone's announcing their new layer three. Arbitram has orbit chains, ZK Synch is hyperchains,

StarkX, they have validityms, which they've been kind of advertising for a while, and they're some of them obviously already out there. And then optimism, they have, what does optimism call their layer threes? Is it hyperchains as well? Superchain. Superchain, super chain, sorry. I forget the prefixes.

No, no, no. But the reason that you might is that one of the newest super chain entities used to call everything they built hyperstructures. So, you know, the Zora. So Zora recently announced that they're not. launching their own OP stack, which is a fork of optimism. And they're building a super chain on optimism. So in layer three, simply put, my understanding, Terun, correct me if I'm wrong,

my understanding of a layer three is that basically it's a layer two on top of the layer two. So you build a skyscraper on top of the skyscraper. But there's a fast exit so that basically if you want to go from the super mega skyscraper on top of the other skyscraper all the way down to ground floor, you can take one elevator and go all the way down. The skyscraper analogy, I feel like, doesn't quite recourse as well. I thought it actually recursed quite beautifully. How many times have you seen someone build a skyscraper on a skyscraper?

Other than like that failed hotel in North Korea? Yeah, yeah, yeah. Go to the Middle East, you're going to see a lot of very interesting structures. So, yes. So that's basically Layer 3. Layer 3 is becoming more popular. And the Opie stack has become kind of a meme at this point, is that lots and lots of people are now building on the OP stack.

One of our portfolio companies called DARA, which is like a roll-up as a service company, facilitates people launching their own roll-ups. Base, of course, is an OPEC stack roll-up. Zora's OPEC-Sack roll-up. I guess these are all going to be hyper-chain, or sorry, superchains now. What do you guys think of this layer three concept

and the proliferation of the O.P.S.ac. You guys are, I should also caveat. You guys are investors in Appelada, I believe, Dragonfly's not. I would say, you know, in the same way that in the 1999 tech bubble web van was a good idea, but

like wrong timing, wrong execution. Arguably, all of this stuff actually existed in 2017 under the name of plasma. And like a lot of the exit games and stuff are the same. But like people didn't understand how to

implement these things and where the interfaces were and like which things were necessary components. In the same way, hopefully the thesis says, that cell phones were necessary for these types of businesses that didn't work in 1999 when no one had cell phone.

And now Ethereum is sufficiently strong to be this base piece whereas it was not development-wise. That Ethereum was sufficiently strong. Why Ethereum? Well, because I mean, you, A, to use as a DA layer, to use it as the final exit layer

of like your assets are really stored there, you need to like build all these contracts for doing that. It was very hard to do that. in 2017 bordering on impossible. And I think a bunch of different other things in Ethereum ecosystem, developer ecosystem wise have made it, but made it a lot easier. Whereas in 2017, you could like see the idea, but like the idea of implementing it

was impossible, which is why you saw like the amisa goes over the world kind of try and fail and stuff like that, right? There are so many people to try. I would characterize it differently. I would characterize it differently. So I think in 2017, so this plasma, for those you don't know, those who came in in the last three years, probably never even heard this term. So plasma was kind of the predecessor of all the layer twos that have taken place,

you know, the optimistic roll-ups and all the roll-up stuff. Plasma was basically a very primitive version of a roll-up that had a lot of problems around data availability. Basically, they didn't really have an answer for data availability, and they didn't have good solutions for general computation. So a lot of the initial ideas around plasmas were that you could transfer UTXOs on plasma, and then the exits were a gigantic nightmare, but as a user, you had to track your own UTXO

and then manually exit it. And it was a whole thing that was just very unrealistic for users to be able to manage all this. So Misa Go was one example. There was Lume. If you remember, I think Giorgos used to be at Lume, and that's how he first became,

you know, Mr. Giorgos, Mr. Lair 2 is from all his work there. So there were some folks that were really trying to launch these plasmas, but they weren't really working. And I think the biggest thing that was holding plasmas back is that there wasn't really a good answer for general computation.

So you couldn't do a turn complete or do EVM, right? And that's ultimately what we've seen is now massive, massive product market fit is EVM. If you cannot do EVM, you're just not going to get use, right? People do not want to build. I mean, there are obviously some, you know, kind of application specific blockchains like, you know, DIDX and so on that do have some traction.

But for the most part, the vast majority of block space is EVM block space. That's what most people care about. Majority of used block space. There's a lot of not used blocks. There's a lot of empty blocks in some places. That's true. Yes. The majority demand for block spaces for EVM block space.

So I think the, the, the, so all of this, I think, I think the reason why this happened, one was that we needed to actually develop the ideas behind roll-ups, which took a while. There were a lot of false starts. So, like, optimism, they started with the OVM, which was like a kind of different, like a shim between EVM and like their own internal representation or whatever, and then they abandoned that, they kind of scrapped the whole thing, and then just did what Arbitrum was doing with the direct EVM compatibility. So long, short, has been just be EVM compatible. That's number one rule. And then the second thing also is that optimism is still not

really done, right? Like the OP stack doesn't have fraud proofs. And almost all the roll-ups are still highly guarded, right? Like they're protected by multisigs, you know, the upgrades, upgrade process right now is still not being done through decentralized governance. Arbitrum does have fraud proofs, but the fraud proofs are, they are whitelisted, so only certain parties can actually submit the fraud proofs. So the reality is that, you know, for OP stack in particular, you know, for base, Zora, whatever, these things are not even really roll-ups

because there's no, there's no fraud proofs, right? So there's no, you are not getting the guarantees from Ethereum without any fraud proofs. Right now it's just basically side chain that spits out data availability on some other layer. So I, the point I'm making is that I think it really is about R&D. I don't think it's about, No, no, no, but they were sort of coupled, right?

Like Ethereum's roadmap at that time with trying to do what, like, sort of full sharding, like I guess some people like near did, right? Like it was just very different in the sense, like some of the design decisions being made were made for that use case. Whereas like right now it's like very clear, okay, we need to like make storage blobs cheaper. We need to like do all this stuff to make roll-ups better, right? But a lot of that just came. I agree it's R&D, but the R&D, but the R&D,

actually was initially all at the base layer and getting and then that. But what has Ethereum even done that facilitated roll-ups in the last three years? I mean, making developer tools so you can do the verification contracts. Like, there's like lots of little tiny things that are annoying that like I would say that like 2018, 2019, you could not implement. Right? 2019, 2018, you could implement you thought you could implement compound, took a lot of work. You had contract limit sizes. You couldn't, like, you had to do all this stuff. And the logic for a lot of the L2 stuff is actually quite complicated, you know, comparatively.

I'm just saying, like, I think that there was quite a bit of Ethereum core development needed and a focus not on sharding at the base protocol and a focus on this kind of like, almost like, distributed sort of form of that. Now, yeah, again, I agree with you that. It remains to be seen that if we get like, how do you deal with fraud proofs, how do you deal with data availability? And so, yeah, data availability for, I guess, the listeners who don't understand. I know we kind of tossed that term around like three times is the idea that the base chain

where your assets are on is the one that's really like providing the data for these other chains. So you're posting the data both to the base layer and to the other chain, such that if you wanted to exit or there was some type of fraud, which is what a fraud proof or fault proof kind of shows you, then that way the layer one can say, okay, well, you can exit. You can take your assets out. Like something wasn't working there. And like, so your assets are still tethered to the main initial chain.

A lot of the earlier designs, to see the earlier point, kind of were focused just on. doing payments, right? Like Bitcoin's Lightning Network is just payments and doesn't really have contract functionality or, you know, putting conditions or covenants on things. But, you know, I think the idea of like having tightly coupled shards that were each an application versus this kind of you can add your own and you only have to, in the average case, you don't have to interact with the expensive layer one very often. There was a lot of things that needed to go, right? to get to this point.

But I think the question to me is, like, what's the level of interoperability between these chains? Because, you know, in Cosmos, you do technically already have the interoperability. You just have zero demand for block space, basically. And the question here is, like, there seems to be some demand, like, I guess base. And Arbitrum, obviously, has some of the most highest amount organic demand. Optimism has a lot of demand from sort of incentivized, demand. So the real question, I guess, is, you know, how much do these things need to talk to each other?

And if there's 20 of them, like, are they really going to be all 20 users? It's still going to concentrate to the ones that have the most liquidity user apps. Well, I think it goes to the ones with the most liquidity and user apps and reason to be there. Why wouldn't it? Like, what else is going to For sure. But we're kind of have this like boom right now, right? And people who are like, I'm going to build my own app chain, but now it's called a, you know, X roll up, like OP stack roll up, arbitrage roll up, you know. Yeah, I think that that is a lot of it, frankly, is people want the own chain. But like, you know, normally if you're like rolling your own cosmos chain, you can make an argument around, oh, well, here's all the reasons why this chain is going to be

optimizer, tailored to my specific use case or these validators are going to be perfect for my specific thing. I kind of struggle with that a little bit more with, with rollups. Like, I don't even know how much customization we've seen amongst some of the OP stack deployments that have come out so far. It just seems like there's less customizability. And so it's like why this thing versus just doing it on sort of the main L2. So one other element of this that I think is interesting to talk about is Polygon. So Polygon, they kind of sit in this weird no man's land where they're sort of there in L1, but they also have other products underneath the same product suite.

So they've got ZKEVM, which is there. there's ZKVM, which was built by Jordy, Jordie Bealina. And they recently announced that they are transitioning, they're going to be transitioning the sort of Polygon main net, which is ostensibly a Ler1, into being a ZK EVM Validium. So Validium basically is a, it's basically a roll-up, a ZK roll-up

that has the data availability totally off-chain in some other place. It's not, you know, the bridge is on Ethereum, but the data availability is somewhere else. So there, it's cheaper to run a validity than to run a kind of full ZK roll-up. And the interesting thing, so there was, you know, Sundeep,

the founder of Polygon, was just complaining that people treat them like the ugly stepchild of the Ethereum community in that no matter what they do, the Ethereum community, like always find some way to poke a hole in it or to tell them that they're not, you know, they're not real, they're not really decentralized,

they're not really, whatever. I'm curious, how do you guys see what Polygon is doing and how it fits into the layer two story? Because they're definitely, one thing is very, very clear is that they're not Ethereum Cool Kids. They're not the anointed. They're not blessed by Vitalik and all these other folks. How do you contextualize that into like the everything that's happening in the blockchain world? I mean, I don't think you have to be anointed. I don't think you have to have like some social proof.

You just have to have a reason for something. something to exist, right? And I think it goes the other way where if there's a reason for something to exist, some roll up, some app chain, some whatever, then later people will say, oh, it's here for a good reason, it's anointed, or like blah, blah, blah, blah. Like I would say to any builders out there, you know, screw what people on Twitter say if it needs to exist, it should exist. Yes, that is true to an extent, but like the roll-ups on Ethereum are unlike

a lot of other technology races in that they're very, very political. Particularly in Ethereum, roll-ups are super political. And I think part of the reason why rolls are so political is that it is deeply tied with the long-term roadmap of Ethereum, right? The sharding to roll-up transition was a big deal, you know? Exactly, exactly.

The original story was that Ethereum was going to transition to a bunch of shards, like a bunch of little baby Ethereum, that would all talk to each other. That was the big vision of how to scale Ethereum. And then they gave that up, because they basically said, look, doing much of roll-ups is gonna be easier, okay?

So, and you know, you guys will be able to move much faster than we as Ethereum can. So let's just embrace the roll-up vision that instead of a thousand baby Ethereum's, there's gonna be a thousand baby roll-ups on top of a thousand Ethereum's and we'll just be like the substrate on which all these skyscrapers will get built.

Okay, in that vision of the world, it really matters. It really matters who Ethereum embraces as being like, yes, we will support you. We will, you know, make the op codes that you ask for cheaper. We will convene our blob storage around your particular ideas. We will tell people, like, there are a lot of things

that are not so subtle above the way in which Ethereum is influencing the outcome of the roll of wars. But the op codes that they've rolled out benefit everybody. We do, no, no, I, you're really talking about two kind of separate things. The EIP process for making changes to Ethereum is extremely politically motivated, politically generated. But arguably, Arbitrum's organic kind of very natural success come from the fact that they just, like, built the most reliable roll-up first, right?

Like, arbitral. Totally, totally. Which shows the chasm that they had to overcome, right? Like, optimist, they were so far ahead of optimism. But my point is, like, technical sophistication can, like, get you past these political. That's kind of the beauty of crypto, right? It's like, you built this thing that, like, was. better for oracles. And so then the purpose protocols are better. So then they kept brought there.

And like that was like completely unrelated to theorem politics. But what you're acknowledging to ruin, what you're acknowledging when you say that is that optimism was the one that was preferred by the Ethereum. But I'm saying in spite of that, it didn't change the outcome. The outcome was like you built the best thing, right? Like whoever built the thing that found the users. Imagine imagine if the, imagine if the Ethereum foundation was like actually arbitrum is better. And we love arbitrum and we're going to like pose with them for photo ops, we want to tell everybody that Arbitrum is the way that Ethereum is going to scale. And Vitalik mentioned Arbitrum in every single blog post.

Optimism would have no oxygen. They would not be able to breathe from how dominant Arbiturban would be. But if they need some technical advance that was really good for applications that had users, like in the case of Arbitrum, making it having all these features such that perps were actually much more reliable there than the perpetual futures, defy protocols were more robust than they were on optimism, then, you know, if the opposite thing happened, I still think the same outcome happens because people who are, the developers who are building these apps are just looking for particular properties, right?

It's like, why do I choose AWS or GCP as my Amazon Web Services or Google Cloud for my cloud hosting? It's like, oh, maybe they have some particular feature or like they have a particular type of GPU and like it's reliable and I can get it and like I'm always able to what what exactly is that difference between arbitration and optimism well I think the optimism bedrock upgrade kind of fixes a lot of the things that people complained about but like you know the block time reliability wasn't as good like the latency the latency reliability the way the message passing to the end user was for for MEP stuff was just better I think it was you know they kind of like the learning path of going to

from OVM to different things. So my point is, like, there's a sense in which Arbishop did really get a lot of these engineering decisions right at the time that there were these applications that really needed them and had this amazing organic growth story, right? Like, they really did have organic, non-incentivized via tokens growth. And so, and that built up their ecosystem. So my point is, I think the roll-up wars are, there is some real politic, no doubt.

It certainly shows up. But I do think your technical capabilities are the most important thing eventually. If you find you make the technical capabilities that are exactly what the developers at that time are demanding. It's going to be hard to compete with that. Now, I think your point about Polygon is Polygon was sort of started as early, and they were effectively their own L1 that happened to be an EVM, right? And then, like, now they're kind of backing into it. So it's just a different gestation story, right?

Like, they didn't start purely on the mothership. They kind of started with this thing that used a little bit of substrate and a little bit of cause, like, you know, like they have like their code base also doesn't look like we are purely in ETH. You know, like, and I forget what that law is. That's like how your code base looks reflects the communication pattern at your company. There's like, there's like a, you know, I'm not. And I feel like that's actually true in blockchain. So it's true in roll-ups too.

It's like how your code-based looks is a reflection of your contributors, community, and the applications that want to build on you. And you should strive for that to look like the thing you want. And I think that that will be the ultimate determinant. Anyway, I'm done ranting. Tell me where to say something. Well, I think Maddox started it out as like a side chain, right? And so they've always been like close to the Ethereum ecosystem.

and they've been trying to figure out, I think, like, it's, there was not an answer for years as to, like, how you actually go about building a real business and company like this. And the answer, you know, for a while was it would just jerry rig this thing and kind of make something that looks like what people want. And it's like, yeah, it's probably if you, if you don't do that and you try to have some sort of, you know, purity test, you probably just die because it's straight up not functional or not even possible to do what you're trying to do. So it's like, you're navigating this, this, this, this weird balance of trying to find something that has immediate PMF, but also trying to build for this, but also trying to build for this. long-term path, it's not, I think, as simple as, you know, just, you know, building to some pure R&D vision. I hope that, you know, it works several years down the line. PMF being product market fit. So I really like that point, Tom. And I totally agree with it is that in a way, being able to just build tech that nobody wants and nobody uses and just like hang around and wait for the world to change is a very, is basically a privilege that can be afforded to you in an environment that's a wash with capital. Polygon, you know, like, I mean, the one thing

that I will give them a lot of credit for, they came from India, obviously the entire team is Indian, and a lot of their usage is outside of the first world. That's where the kind of their stronghold is kind of non-first world users. And they had to scrimp and scrape for everything that they got, you know.

And I have a huge amount of respect for that. It doesn't appeal very much to the Ethereum of intelligentsia, but because, you know, it's not this sort of pristine, like, wait, why didn't you, why didn't you start as a roll-up and immediately blah, blah, blah, blah. And they're like, look, we try to build the things that people want it. You guys are building the things that each other said that each other should want.

And nobody was using any of your shit, you know, two, three years ago. We were building the things that people actually wanted. People wanted block space. They wanted to have, I don't know, whatever, all the stuff that Polygon has launched. They obviously have a large product suite. And you guys said you wanted Web 2 kind of traditional companies and brands to partner and do stuff in Web 3. well, okay, we're doing that. That's exactly what we're doing. That hustle is what I really

respect about Polygon, but it is certainly true that they don't have this kind of, you know, pristine kind of evolutionary origin story, the way that they were, the way that so many people in the kind of Western technology world would very much prefer them to. So, okay, last thing I want to talk about on the, um, Deroon, you mentioned, with respect to what was going on with optimism versus arbitram. One of the, one of the things we've been seeing a lot of just as VCs, getting a lot of pitches about sequencers, shared sequencers. So okay, what the hell is a shared sequencer?

So let me see if I can come up with a, okay, I'm gonna use a skyscraper analogy for this. So in, in, so when you're running a layer two, there is some entity that is basically deciding the ordering of all the transactions in the layer two and they're kind of doing the work of processing stuff. they're kind of the CPU of that mini blockchain. You can sort of think of this as like the leasing office of the skyscraper, right? You need somebody in there who's like kind of doing the work of allocating all the rooms or whatever. And so a shared sequencer is this idea that you can have multiple skyscrapers

that are controlled by the same leasing office. And the advantage of this, there are a few advantages of this, but one of the nice advantages of this is that you can basically do a, sort of do a deal, do like a, you know, a transfer, a cross-chain transaction, a bridge, or some kind of cross-roll-up MEV. So basically I want to do this and that across these two blockchains. And if both them cannot be done, I want to cancel the transaction outright. So you can get some like atomic guarantees across multiple blockchains or multiple roll-ups.

You can do that through one leasing office, right? You can be like, I want this room in this skyscraper and this room in this skyscraper. But if I can't get both, I don't want either of them. That can only be done if you have, you know, one party to negotiate with. So these are these sort of shared sequencers. These are becoming kind of a new hot thing. There's also decentralized sequencers, which is kind of the story for all these guys that not only are these sequencers going to be shared across multiple roll-offs,

but also they're going to be decentralized in some way today. They're all pretty much centralized. How do you guys think about decentralized sequeners? Is this an interesting thing to invest in? Do we think it's going to be big? Do we think that only small roll-ups are going to have shared sequeners, or is this going to be the new future? I'm sure Tarun, you have very strong opinions.

I can see you smiling. I think the funny thing that we learned is like a lot of companies we invested in last year started saying they were doing different things and then all of a sudden converged on, hey, we're all going to do shared sequencers, which I thought was like funny. If you, when you see like the market pivot into something, there's some sort of a least idea level demand for it. I think it's very hard to get right. your goal is to be as like lightweight as possible without just making another blockchain that's intermediating between all these chain. But yeah, the idea, like the dream is that,

you know, one of the most beautiful things in crypto and especially in defy that you don't really have in normal finance at all is this concept of a flash loan where you can borrow capital and your loan only exists if your activity or arbitrage or whatever your function is able to return profitably, then the loan happens.

And it's sort of a way of like guaranteeing adamicity, you know, guaranteeing execution in a particular order of the sequence of transactions. But it's also a way of doing sort of like risk-free lending because it really is risk-free, right? Like it only executes if it, if these things, conditions, covenants,

work and are profitable. And obviously, as the blockchain ecosystem got much more complicated post-2020, people always want ways to do the same thing. They would like some guarantees of that form. And there's always sort of this inherent problem where, like, if you have two machines that are independent, they don't quite have to respect one another. And you could kind of think of, you could, if you think of these chains as nations,

the idea of the bridges or bridges are like when you go to you go at the airport and they like check your passport and they're like nope we're denying you there's sort of the sense of like you're moving between regions that are different they have different rules and you maybe in that new region your assets will have different kind of covenants or lack thereof added to the goal of the shared sequence is somehow to pick some subset of things that are free trade agreements at least that's how I like to think of that. It's like guaranteeing animicity or these two different regions is a little bit like NAFTA. It's like we agree that these two things. And I think. Okay. I like the leasing office better, but I'll take NAFTA. I just think the fair trade versus not fair. These are very Tarun-esque way.

Fair trade versus not fair trade is like, you know, like the bridges have to handle all possible things, handle all the tariffs. Fair trade is like, whatever, we guarantee you this will always go across. And so I think like there's a sense in which. that's sort of where it's hard. It does have the same, it has some of the same problems of like, how do you remunerate the shared sequencer for doing this action of making it feel like a fair trade zone? And how do you make it so that it's easy to verify on both sides? But yeah, there's clearly a ton of interest in this. I think there's also a lot of debate over

this, you know, I think people in, you know, the land where people don't want things to be asynchronous like Solana, they sort of view the shared sequencer as like, well, aren't you just going back to what we're doing because you're like guaranteeing these admiocity guarantees that we always guarantee because everyone's on the same chain and there's no separation of church and state. And on the other hand, you have you have people in Ethereum who are like, hey, we're a bunch of OP stack rollups. We're likely sharing a lot of assets on Ethereum that are being used in our chains. Can we make some extra guarantee? on both sides.

And that's sort of this fair free trade agreement type of thing. So the first thought that I have about shared sequencers is that, so one, it feels to me like, I mean, it's interesting, but it feels a little bit bike sheddy. Like it feels like this is, the amount of interest and attention and energy going into this problem feels a little bit overweight. It's like actual importance to user experience of these things. Like the reality is like optimism stack does not have fraud proofs, right? Like that seems like a really big problem.

And but but so much so much brain power is being expended on how to create shared sequencers. It feels a little bit like a kind of misdirection of just because it's intellectually interesting. The other side of it of course is that like the the the problem with shared sequencer is also the kind of the noisy neighbor problem, which is that if you are an arbitram or you are you know kind of optimism main net, you're you got a lot of shit going on. You're very full, right?

you don't necessarily want to be adjoined at the hip to, you know, 20 other roll-ups that are also going to be kind of sending their transactions to the same sequencer. That's, you know, you've got the sort of the passport, you know, stamping office. If you have, you know, 20 security lines going to the same passport office, it's just going to be more of a shit show. And like you're going to have latency problems, congestion problems, all that sort of stuff is just going to wear down on you. And so the biggest roll-ups, I have to imagine, are going to be fairly selective in who they, allow to share sequencing with them. And then these folks who are going to be on shared sequencers are going to be folks who are like, well, I really want to be on a shared sequencer with Arbitrum because that's kind of,

you know, that's where all the action is. But Arbitrum doesn't want me to be sharing their sequencer. And so there is kind of a little bit of like, you know, people want to be part of a club, but the clubs they're going to be invited to are not the clubs they want to be a part of. I think, though, if you have any desire to decentralize the execution you effectively have, you kind of have to have this kind of sense in which anyone can do it. Yeah, I think that's kind of the bigger point people make. I mean, that's always like kind of the classic criticisms of L2s.

It's like, haven't we kind of just gone all the way full circle when we have like it's just a single computer processing all the transactions? And like, you know, we have one point that, you know, does go down not infrequently and can censor transactions. And so it's like, yeah, it's not as bad as, as. you know, sort of having arbitrary state writing capabilities, but there are sort of, you know, downsides to having sort of a single sequencer. And so I think that's kind of more the angle that, that I guess I think of this coming from in terms of benefits. But to your point around like

performance, like it feels like we're so far from being at a point where, but decentralized sequencers and shared sequencers are two like totally orthogonal things. Sure. I think one will cause the like, I can't ban. I can't. I can't say I can't join the club problem. And then the other one will kind of be like where most the income is. Wait, why would, why would you say like if, if, if, if, if, if, if, if, if, if, if, if, if, if, if, if, if, if, if, if a CK. Sink has decentralized sequencer, member in one, like, they, they can still decide with decentralized governance. No, you can't join our club. I mean, theoretically, the shared sequencer itself should be allowed to be joined by anyone who's a sequencer member in one.

Like, if you had a decentralized sequencer. I can't give you any of the shared guarantees unless all the pools of possible sequencers is overlapping because otherwise I'll lose some of those guarantees. So I think that's like if you actually are decentralized, then anyone can always give these evidence of guarantees. You can't be kind of... So I'm saying the desire to decentralize a sequencer

will cause the shared part to not be as clubby as you might think, is my guess. Okay, let me understand your mental model here. So you're saying, let's say that optimism, let's say OPSEC, Optimism A Net, moves to a decentralized sequencer, not a shared sequencer, okay? They moved to decentralized sequencer. You're saying that once they have a decentralized sequencer,

then anybody will be able to say, hey, I want to add myself to the shared sequencing conduit so that I'm also being entered into the same decentralized sequencer. Why would it be the case? Why wouldn't it be the case that the plurality of the people who are sequencing for OP stack would say no. We strictly only want to have one blockchain that we're sequencing or we want to have three or we want to have five and we're not inviting you to that five?

I mean, I think if they're doing a free trade agreement with Arbitrum, like they're like sharing, like in order to give those out-dimistic guarantees whenever to any application that requests it, to have that, like, liveliness for that, they effectively have to let anyone in the Arbitrum sequencing pool, who is sequencing at that time, be providing the shared guarantee on the leg on the other leg, right? Like, so I think, I believe Arbitram already has, does have this implemented for having multiple sequencers rotating,

like for epochs, it's fixed epochs. So, like, unlike... Oh, I see, I see. So if it's your epoch, you could, as long as you're also on the Arbitrum, Like it's your epoch on arbitralm and your epoch on optimism, then in that moment that you line up on both rotations, you can be the shared sequence. Yeah, that's that's kind of where I kind of think if they really are going to decentralize, the shared part ends up having having to inherit that property. Kind of.

Okay. I see what you're, I see your point. Sorry, Tom, I interrupt you. No, no. I was pretty much, I think it's more like it feels like we're so far from sort of the performance required for these things to be sort of. mainstream and successful to then consider, you know, busy performance degradation for the sake of decentralization that, yeah, I don't know.

It just also feels like a thing that, yes, we should get to in the roadmap, but like not as as sort of critical. I kind of agree that like it's a little bit web vanny in that you probably need applications that are really demanding it, right? I think first. Yeah, yeah. I mean, the amount of times I've heard the hotel and train problem as an explanation

You know why we need this is telling that right now this is more an intellectually interesting problem than it is like a, you know, the most important thing. Well, maybe, maybe if we step back from this infrastructure kind of consciousness, what are the applications that you think you're most excited about that are coming in this format, right? Because there's a ton of them coming out. In, in this format being like wires cross. Like, yeah. Well, I think we've established that like I am not as bullish on app chains as you are. I mean, look, I think DoIDX moving to Cosmos is really interesting. I'd love to see how that plays out.

I think right now the DoIDX, they have a lot of confidence that they're going to be able to pull people away from Ethereum and move into their own infrastructure. I think there'll be an interesting experiment to see how robust this sort of app chain thesis is for a product that has really strong brand, really strong. user base really strong metrics like do idx yeah what are you guys seeing are there things that you guys are excited about that you think are interesting app chain opportunities that a no i'm hearing silence well let's continue this one on the next show okay all right all right we're over and i think i think uh we're we're starting to flag a little bit okay we're a bit over time so we're gonna call it there but interesting times in l2 land uh will this is a story that will continue checking in on as it continues to evolve. For now, that's it for this week. Thank you, everybody.