Current State and Future of Second-Layer Networks│Panel│ETHDam 2024

Um next up we have a discussion about L2s. Um I am going to moderate this, but I also have to preface it with I I I think this is actually a very important topic. My personal belief is that L2s very much lead to a lot of fragmentation and we're very much in an L2 like absolute war zone right now and then you have all the old L1s and all the craziness. So I'm very interested to have this conversation with some people that I've spent some time with. Um not one of them.

And so I'm very excited. Are you guys excited or what? Yeah. I I'm going to I'm going to try and be neutral though. Um so uh this is going to be on the current state and future of second layer networks.

We've got Scroll, ZK Sync, and StarkWare. So I'd like Oh, I didn't check this. [ __ ] Uh Togrul. How do I say it? Togrul.

Perfect. Do please take a seat. Round of applause. As well as as well as Joel and Victor. So please take a seat.

Welcome. Welcome. Um Now I'd like each of you to give a very short introduction of of yourselves um and also maybe a short opening statement about what you think about the current state of uh L1s and then we'll jump into some questions. Oh, [ __ ] L2s. Good good catch.

There's a sign. At least they're sharp. Okay. All right. Someone else go first.

All right, I'll start. Uh my name is Victor. I work at StarkWare. Uh we're building StarkNet. Um I really like olives as a fun fact.

I think everybody should give a fun fact about themselves. Um and the current state of L2s, I agree with you that there's this thing of L2 wars, but at the same time we shouldn't be we shouldn't act in our own little bubble. We should act within the broader Ethereum bubble because we're not a fragmented community. We're just one big Ethereum community. Nice.

Nice, he said. Hello. Is it working? Okay. Hello.

My name is Togrul. I mostly shitpost on Twitter and complain about things. Sometimes pretend to do research, but I actually don't. I just complain about others. And uh yeah, I I don't think uh I I think there is an L2 war, but we should make memes, not wars.

That's what we should do. And a fun fact about yourself? I I don't There's nothing fun about me. You chat [ __ ] That's the fun fact. Hey, so I'm Joel Valenzuela.

I work for Dash, which is a digital cash payments focused cryptocurrency. And um Hey. And a fun fact about me, uh there's there's a few to choose from, but um yeah, let's just say I I haven't been using fiat currency pretty much at all for 9 years. So or coming up on 9 years this year. So That's fun, right?

Not so fun when you're doing it, but How do you go to the supermarket then? I do. I I walk with my own two feet. No, but how do you pay for stuff? That's a secret.

Yes, I mean you should've been in my talk yesterday, but yes, um there there's there's ways of doing that, but feel free anyone to ask me after this. Um basically I use an L2. Just kidding. Cool. So on that note, what what in your guys uh qualifies as an L2?

Just to do some definitions. How would you explain it? I I I was just on Twitter before I went on stage saying that Optimiums are not L2s. So I I I would say an L2 is a protocol that extends an underlying layer, obviously, but it has to have a trust minimized bridge. So what I mean by that is the security assumptions of communicating between those two layers have to be equivalent or less than one out of n where n is unbounded honesty for safety and for liveness, it depends.

I'm still not sure because like certain people are arguing whether Validiums are L2s or not purely because like the safety uh the liveness assumption is requires stronger honesty. But like I'm still not sure. I would say that what we care about is safety and liveness is kind of we can we can adopt it via economic um assumptions. Okay, so there's two kinds of L2s sort of. In my mind there's what I call uh like a L like technical L2 and a market L2.

So the market L2 is something that the market uses as an L2. So Litecoin is Bitcoin's primary market L2 where Litecoin is used as a substitute for Bitcoin in payments by most of the top payment processor metrics much more than the Lightning Network, for example. So that even though it's a completely distinct chain the market has kind of called that an L2 sort of. But for the more technical definition I've never heard that before, actually. I've never heard someone refer to Litecoin as an L2.

Yes, well, it's it's serving the purpose as an L2. Even though technically it is absolutely not. Yeah, of course. And so technically I would say some sort of protocol that a different layer that in some way is inexorably tied to the base layer. Like for example, it has to write to as a settle to the base layer somehow.

There has to be some sort of um I guess interaction with validators or there has to be something You're saying it doesn't have to be code. It can be some other likeness. Are you Are you talking about the first category or the second category? The market one or the The market one, I think. Market one, I mean I've actually heard it doesn't have to be code at all.

In fact, I've heard some Bitcoin maximalists unironically seriously say that ETFs are an L2 scaling solution for Bitcoin. And that's a cope. make that into a That was not That was not a That was not a joke, by the way. It was That's a cope. That's not a joke.

Yeah. Well, so one man's cope is another man's joke. An L2 for me is uh is an entity that takes all of the computational burden off of Ethereum and Ethereum is more of this like settlement layer. It's like the base of the base. And it's really tough to do a lot of work on Ethereum.

And so for me an L2 is an entity that tells Ethereum, "Hey, I'm going to do this for you and all I need to do is provide a little proof that summarizes what happened elsewhere." And this proof is um what connects these two entities and it should be the most secure proof. Um Yeah, in order to to take off the computational burden out of Ethereum in the best way possible. Thank you very much for that differentiating opinion, which is nice. Um You're all right in your own way.

Um What's more important in the L2 space? Is it general purpose scaling or is it specific use cases? I guess or aka app chains in in that regard. Um Yeah, any of you can take the lead on that. I think it really depends on a specific scenario.

So like for certain apps I can see them being an L2s purely from the perspective of like owning their own block space, but not being general purpose. So for example, I've been hearing a lot I I know this is going to be a bit of an alpha, but I've been hearing a lot of chatter about like certain projects being L2s building L2s on Solana because they realize that they just want to own the block space. It just makes sense. And then for example, for Ethereum most of the L2s are more from the perspective of scaling. So they are general purpose and they are there to scale the overall throughput of the ecosystem.

Well, just to say or the combination of scaling and privacy is then I think that's an extra layer on top. Yeah, you you can you can do that. Aztec is for example working on that on on on like a combined scaling slash privacy L2. Yeah. Just to reme- So the question is what's more important in the L2 space, general purpose scaling or specific use cases?

I forgot myself, so. Uh sorry. Yeah, sorry. I misheard the question. I think it's ideal if you're a general purpose chain and at the same time you allow for one specific use case like gaming, for instance, and at the same time DeFi projects.

Um I think that's yeah, probably it's the ability of being super flexible um and catering to a lot of people to exactly what they need. Yeah, my personal opinion is purely for the use case is mostly. I think if you have a I guess a general purpose L2 that people are trying to use, that kind of becomes the L1. It's like the vampire L1, right? That becomes the next thing.

Whereas on the other hand um at some point absolutely it's it's not always easy to shoehorn every single use case into one exact technical solution. So um so for example, if you have like a social media site and every single post, every single comment is some kind of a blockchain transaction, there might be some trade-offs you have to make for people's basic base money that they don't have that kind of compromises one or the other. And so, something like that would make a lot of sense. And this is something I've sort of gone back and forth with some like Lightning Network people on is um unlike most, I guess, critics of the Lightning Network for general scaling, um I do think there's a use case for Lightning. I do think it it can do things.

It it it's optimal for some things. But, ultimately, the market is showing what it's optimal for, which is uh settlements across exchanges and custodial platforms like that. And so, for example, if you have your Coinbase user, you can send for pennies, let's just say, to a Binance user or to Kraken user or whatever through their large managed liquidity pools. Sure, that's fine. That's not a general-purpose use case.

Or just very specific kinds of things. Or like if a social media thing like, say, Twitter wanted to run purely on Lightning. Every time you tip back and forth, it all goes through their one big Lightning hub and all that. It that's a lot of Token symbol Token symbol Degen. What was that?

Token symbol Degen. Yeah. So, basically, I I believe that the whatever assumptions, like trust assumptions or whatever you use for the the chain should be for the general user, general purpose. So, if if someone says, "Oh, I want to use Ethereum cuz I want my smart contract to be immutable, trustless, whatever." All those kinds of things have to apply.

And so, if you have that Well, but I got to use Unfortunately, I have to use Polygon for my NFTs today because I I'm too poor for Ethereum, right? And so, unfortunately, all those things that made me want to use Ethereum for that, I have to sacrifice a whole lot of those to do something that's doable. So, maybe all NFTs should be on a L2 and none on the main chain or something. But, the fact that you can do both and it's just like pyramid of wealth of like you get all the You would not do well in a room surrounded by old Nuls guys. Yeah.

Uh but you used the the the the your second definition of L2 here, the market L2 because Polygon is not really an L2. Yeah, well then they might have to reframe L2-ish. L2-ish. Definitely. They used to claim that they're an L2, but not anymore, I think.

I think they got bullied enough. Yeah. What do they claim to be then? Do you know? No, Polygon now I think they just refer to them as a side chain.

Like it hasn't So, Polygon zkEVM is an L2, but like Polygon POS, the one that the majority of the people use is a side chain. Okay, let uh just for the audience members, can you define very quickly the difference between a side chain and an L2? But, we want to keep this really short cuz we're going to move on from this. Side chain doesn't basically inherit any security guarantees from the base layer, whereas L2s inherit certain security guarantees from the base layer. But, they achieve the similar goals.

That's In a way, but you can say that about any protocol. fair's. Um okay, so as the guys I'm going to go move to the Slido soon. Probably well, a longer If there is a poll, please let's go nuts for a poll in a minute. Um The So, the title of this talk or this panel really is the current status and the future of of L2s, right?

We can put the panel up. Let's go nuts. [ __ ] it. Yolo. Um is We'll get back to this.

Guys, focus. Get distracted by the screen. It's for the audience. It looks interesting. God, you would not do well on TikTok.

Um So, what is the current state of of L2s in your opinion? I think there's something that have been used. There's there's a whole generation of L2s that are now coming online that arguably lack adoption. And I'm just genuinely interested in what where you think we are now and then kind of a little bit of a forward-looking statement in terms of of what you think's going to come. I think it really depends on For example, if you look from the Ethereum perspective, it really depends on whether you want to look at the protocols that will eventually become L2s that are not definitionally L2s now.

Like, for example, the majority of L2s that you currently use. Like, technically, there's only one true L2 on Ethereum right now. It's called Fuel version one. And it has $8 on it. But but yeah, but but everybody else have Optimism, Arbitrum, Scroll, Stark L2.

Oh, that's a fantastic protocol. And somebody actually tried to check whether whether if you like propose an invalid state transition, whether you can steal those eight that $8. And actually, no, they they got slashed and they lost like 1K or something on that. So, yeah. So, the cost of the attack was like 100 times more than the actual funds inside that chain.

Yeah, but the rest of the L2s are of the L2s are eventually going to become L2s, whether it's going to be like in half a year, a year, two years, whatever. And I think if you look at them, you can see a lot of things uh being done both on the protocol level and on like use user from the like things from user perspective. So, for example, like Arbitrum, Optimism have like massive TVL, a lot of apps, like a healthy ecosystem. Same with Base. StarkWare has a lot of unique apps that are being built specifically because it uses a different VM to everybody else.

And then if you look at the older major ones, they are innovating in some way, whether they're EVM equivalent and they're exploring how to like optimize the EVM. So, for example, Arbitrum has Stylus, uh which they refer to as EVM plus. Then Polygon is researching parallelization, etc. And then there are also new VMs that are just coming. There's Fluent that has WASM.

There There's Movement that is doing MoveVM, etc. So, there are a lot of things that are happening. And I think in a year or two, uh once the tech matures to the point where we can actually call those all those protocols L2s, we're going to have like a very vibrant ecosystem. I think that all the L2s at the moment are in this like work-in-progress status. Um sure, we all have our own prover, we all have our own sequencer, and we do things differently.

Um but if the end goal is really to decentralize all of your components, let's say you want to build like a network of provers, um that's still not the case for most of the L2s. Um so, I'd consider, yeah, like the whole L2 industry is still a work-in-progress thing, even though we have really cool apps and liquidity coming on these L2s. Um it's definitely something that is not 100% finalized. Like, it's not like your final product. So, it's always improving, you will always find like new ways to innovate the chain.

Um Yeah, that's how I would define the status quo. Cool, thank you. And I I'm interested in your opinion on this the on these this question from a usability perspective because that's what you care about. Yeah. So, hit me.

Um Well, for me, there's a a massive currently massive fragmentation kind of problem. And I definitely think that for all the for the end user who just wants to use the thing, um yes, if they all decide to go on Solana cuz it's simple. Right now, they will have like What is it? 75% transaction failure rate or something like that. So, there's those problems.

But, over the years, especially if you're a long-term user, um I was wanting to be part of the Ethereum ecosystem, but just obviously priced out of L1. And then the next big thing is choosing an L an L2. And so, it's like, "Well, do I use Optimism? Do I use Arbitrum? Do I use again Polygon, the L you know, whatever you call it, the economic L2, whatever?"

But, it's still presented, whatever the menu of options are. And then now there's Base. And I don't really know um I have not known historically. So, I picked Polygon because it was like the number one, like the biggest at the time. And I thought, "Okay, well, this is the this is where I get the most bang for my buck."

And then I kind of feel like I chose wrong. Like, maybe I should have gone for Arbitrum. But, now Base is absolutely killing it. But, then I I do feel like I'm just kind of a Coinbase customer at that point. And so, at one wallet, for sure.

Yeah, which I don't I'm not a Coinbase customer or a Base user. Same thing, right? But, um at some point, you know, okay, now there's a new L2 that does things better. At some point, it's very hard to keep user retention long-term if they have to sort of bounce around of which one to do. And if you do get everyone on one, what trust assumptions does that have?

What security does that have relative to the L1? Is that then just a Like, if everyone's using Base, I don't like that world, right? Maybe I'm being harsh on Base, but it seem seems like Coinbase runs you know, the company, which is under a lot of pressures from governments and things like that right now. It's like one company that's kind of under, you know, a lot of pressure to kind of control everything. But yeah, so the the user, especially if you want to use like one of the things that's the most it's like a lot of this is done from the perspective of let's make things easier for the user.

And paradoxically, it's the exact opposite of Oh yeah, let's they don't have to do the fees and do the gas estimates. Let's just do something cheap. So, like an L2. And oh, they don't want to worry about native tokens. So, let's do a stable coin.

And so then I click on click on the little menu of like, oh, I'm going to send and then there's how many stable coins on how many networks and it just it's so easy to get like the wrong one. It's like, okay, I'm going to pay this invoice in, you know, I see the stable coin. But then you got the wrong stable coin or the wrong network and now you got to flip like swap back, but then whatever network doesn't support this and it just it's chaos and it's a nightmare and I don't know how how there's so many users already on all these networks. let's this is a this is a verbal response that's requested. How many of you enjoy making a transaction in crypto?

I I said verbal verbal. Yeah. Who here [ __ ] hates making a transaction in crypto? I Yeah. I I hate it primarily because all the wallets suck.

I've never I've never managed to find a good wallet that does everything that I need it to. That's after Justin's talk. We get to roast the wallet people. It's going to be a long afternoon. Okay, so I guess on this topic when we talk about we asked a poll earlier and I think it's nice that we revisit this now because it's it's it's about how do you make things more usable.

So I want us to actually all stand up and we're going to stand here and we're going to read it and we're going to read it out together. So are ETFs a Bitcoin scaling solution? And this is what the audience thinks. 34 members is quite a lot and so argue argue with them if you want, but I think it fits the market definition of an L2 in your thing. I would say when I was coming on the panel to discuss I caused this.

Yeah. Yeah. I would say that when I was coming on a panel about to discuss L2s I wasn't expecting We can sit down. expecting to be discussing ETFs on Bitcoin as an L2 scaling solution. Really?

Why not? I don't know. I'm just it's it sounds a bit depressing. I'll be honest. Like to build the most cypherpunk thing possible and then to scale it via ETFs.

So so here's an interesting thing. We're at a conference about privacy and the reality is is if someone if someone buys a Bitcoin on chain, suddenly everyone's trying to figure out who they are as an institution. There's a plethora of tools out there to help people do that. Alchemy Intelligence being one of them that is the most yeah, crowdsourced um and it allows institutions privacy. And I'm just not I'm not defending it.

Let me be crystal clear. I'm not defending it, but it's one of the features that enables them. So I think what do you think about privacy in L2s built in fundamentally easy to use like just what's your thoughts on that because we're at a privacy event. Could you just give me an example of how you would envision like for you how would you see the ultimate usage of privacy in an L2? Not necessarily for managing asset like crypto assets per se or using them like for example generating meme coins definitely not that, but very much for like business transactions or if I'm trying to actually sign a contract with a counterparty and yesterday Oasis presented the idea of deploying a smart contract inside of a secure enclave and their privacy in and the and what goes on in the inside is provable or the privacy out is all maintained and that's not a shill for them, but that's just basically saying like it seems like privacy is getting closer to being real.

I'm not saying it's there, but um yeah. I think the best way to like enhance your privacy on an L2 is to use your dog's DNS uh thing the domain for everything and then you you're fully private now. I'm joking. I I think you can improve certain things. So like I I I think like if you take like for example an EVM equivalent uh chain you're always going to be leaking certain data especially with certain use cases even if you try to completely privatize it, but you can do a lot of things.

So for example now that smart contract wallets are coming along you can do certain things where you can like for example uh introduce stealth addresses to the smart contract wallets which would obviously not guarantee full privacy that the amounts being sent will still be seen, but that's a good starting point etc. So I think there are certain things that can be improved. Uh but in general I just don't see this happening aside from Aztec and like a few other L2s that are specifically focused on privacy this happening anytime soon purely because like people are focusing on making their existing code base mature and optimizing the existing things. I think once that's done then we can have a more serious chat about privacy and how we can introduce privacy there, but at the moment I feel like it's a bit too premature. I challenge everyone in this room to try and disprove that statement over the next 24 hours.

Like I think if there ever was a building for the people that can hopefully inspire that to happen sooner it's going to be here. So I just want to get to the audience questions. So sorry if you want to say anything else on this then Yeah. Just real quick. I just think that privacy is extremely important and especially once you do things that matter and you it's a lot of playing around on the EVM ecosystem right now.

Once it's serious, once an NFT isn't a a picture of a frog and it's a your house deed for example, I think you're going to need privacy. So I don't know if it Well I don't know if it should be baked in, you know, I don't really have a comment about baked into the L2 natively. Use Railgun in the meantime, please. Uh what animal did you just mention in your head? frog I'm sorry.

Is that Well, the first question is why is Tog a duck? It's not a duck. It's a wazey. It's a fictional Come on, people. Everyone should know it.

Like the penguin, right? I think a penguin for me. Almost. It's a fictional creature that lives in a fridge for 40 days. I'm not sure why.

That's the lore. If anyone wants to ask about it you can do so after ask about that you cannot do it after the after the event. Um Dash which is dark coin which is X coin equals Insta mine Insta mine scam. Oh, no. You can I 2014 called it once it's references back.

I haven't heard this conversation like 7 years. Is that is is it that recently? It's been like 10 for me. Okay. You can also have that fight outside.

What are the exciting recent developments in cryptography for scaling? Jolt was just announced like a few days ago which is like a new prover. But I I think an ooh. Ooh. Yeah.

I I think I think there there there's a lot of stuff happening in parallel. The the like it's sometimes hard to keep up. Like one day you talk to someone from ZK sync and they're like, oh yeah, we came up with this cool thing to optimize things and you're like, oh [ __ ] why didn't I think about it? And then you talk to somebody else within our team and then they come up with something else. So there are a lot of things happening and I think uh we're still not there, but I think we we have a few years of this break uh neck neck breaking speed of zero knowledge proof improvements because we went so far in the last 5 years.

I remember 5 years ago when like doing a range proof would like it was like super expensive on a normal PC and now with like Plonky3 for example you can even compute a full uh validity proof for for an EVM instance on your normal PC. Don't drop the mic, but We're we're looking into circle Starks now and it's just kind of an extension to Stark proofs. It was done in collaboration between Polygon and StarkWare and they released a whole paper on it around a month ago. Um So that's pretty exciting and now we're working into integrating that into the current StarkNet prover. Cool.

Yeah. I mean hate to go back to that, but I think one of the most exciting developments in L2 scaling is the the more people Bitcoin ETF that scales Ethereum. Sort of. Yes. Well, I mean but is the concept of the economic L2 in the market L2 in people's minds where people say, oh, I use this instead and instead of I think a lot of the L2 argument comes from I need to use this chain and then I can't use this chain, so I'm going to not use it while still using it sort of.

And I think when people have the mindset of oh, I do my this on this chain and my this on this chain even though there is a certain amount of fragmentation that happens between juggling five chains or whatever, at least it gets people in the mindset of I'm just going to use the thing that works and then the L1s have to compete directly with the L2s and then eventually it makes the L1s better or makes the L2s more like enshrined L2s or whatever come to the top or whatever. So just generally speaking I think the the concept of economic L2s in people's minds is kind of an exciting thing cuz it puts market pressure on the right way to scale. I I think I think like there's an advantage of hopping between five chains because you get airdrops from all five chains. Uh yeah. Okay.

Yeah. Yeah. Which by the way that's a missed opportunity outside is the when there's the food trucks is call it the airdrop station. So, you get free snacks. Should I go there and ask when token?

I think at this point it's proven that airdrops do not create long-term retention. I think that is a statement that I'm quite We'll see how that evolves. Um yeah, TBD. Question for Scroll. Why is it so expensive?

Call the devs, bro. I I I was like I was like I'm going to come on stage and at least and people are all going to like pester me here about how expensive it is. Uh we're rolling out EIP-4844 and like upgrade in like 10 days for 14 days, something like that. So, it's going to get much cheaper. And yeah, and we have like uh compression algorithm coming that will drop the prices by another of 50 to 60% on an average transaction.

So, it should get much cheaper in like a couple of weeks. Okay, well, there you go. Uh I'm going to ask for one last closing statement that you think is maybe a fun fact or something about L2s that everyone here should know. Uh let's keep it short, fun, punchy, funky, whatever you want. I'll start, I guess.

Um every L2 is someone's L1. Boom. Controversial. Uh I think I what I learned today is that the ultimate way to scale Ethereum is to launch an ETF. Hey, audience said yes.

Audience said yes. Gary, if you're listening, hit me up. LFG. Um I'm going to close it on I'm going to come back to the whole bubble thing and that there is no real war. Um we all feel like we're in our own bubble and we want to all protect our own tech because we're all innovative in our own way.

Um all the L2s are great. We're all different and let's not make war, let's make love and this and memes in this Ethereum giant bubble. Is is Ellie here? Ellie here? Wasn't that such a nice statement?

But we we also hate each other passionately. Yeah, I I we didn't go into the whole like I'm going to steal your [ __ ] debate. Um so, please give a round of applause for this panel. Thank you very much. Thank you.

Thank you.