Tim Beiko & Justin Drake: The Sci-fi Roadmap to Ethereum

And but there's also some like really future thinking stuff. Uh so for example, upgrading to post quantum cryptography. And even beyond that, this is something that kind of only opened up in the last few weeks. Not only should we defend against quantum computers, but we can harness them to make consensus even more robust.

Welcome, Bankless Nation, to the sci-fi roadmap to Ethereum. You know about EIP 4844, you know about proposer builder separation, but you probably didn't know that the Ethereum roadmap does not stop at these things. It continues. Each component of the Ethereum roadmap has deeper frontiers than what you may be familiar with. 4844, PBS, these are just the first stops on a longer track of Ethereum's sci-fi roadmap. In this episode, we're going to explore these deeper frontiers to Ethereum with Justin Drake and Tim Baco, the two individuals who can probably see the furthest along the Ethereum frontier out of anyone. We're going to learn about what is statelessness and how it's helping Ethereum make the furthest reaches of the internet even more trustless. How Ethereum is going to absorb ZK Tech at the layer one level. How VCs are subsidizing Ethereum's layer one research and development efforts. What is an enshrined rollup and how is it different from all the other rollups that we're familiar with? And why does Justin Drake want it so bad? How all of this stuff is leading to Ethereum needing fewer and fewer honest actors to work trustlessly. And what does quantum computing have to do with the Ethereum roadmap? And no, it's not just that it breaks EC DSA signatures. And what are trustless mempools? Who do they protect? All of these things that I just mentioned are just the tips of the iceberg. You're gonna learn so much in this episode, so much that you're gonna need to put your gigabrain hat on, or else your brain might explode. I hope you enjoy this episode with Tim Bacon and Justin Drake. This is the sci fi roadmap to Ethereum, and we're gonna get to it as soon as we talk to some of these fantastic sponsors.

Make this episode possible, especially

Bienvenue, uh Bankless Nation. That is the only French that I know. Here we've got Tim Baco and Justin Drake, and we are at ETHCC. And this is one of the very rare opportunities where I can get everyone here on a couch together. We're gonna talk about the future of Ethereum. To me, ETCC is even though it's in the middle of July, it's in the middle of the year. It still kind of feels like one revolution around the conference circuit at the very least. It was the first conference post-COVID, and this is the third ECC post-COVID, uh, so two years. Uh and Ethereum has grown up a lot since then. We've had proof of stake now actually shipped. So that is in the rear view mirror. Layer twos are no longer hypothetical. They are here. I remember last year, ECC, it was the year of the ZK EVM, and this year seems to be the year of the ZK EVM mainnet. Uh uh a lot of a lot of things have actually come to pass, but there's still a lot of things to look forward to. And so I'm hoping this conversation can guide me and and the bankless listeners, all the the bankless nation. What is the future of Ethereum? What is there left to do? And so, Justin, maybe, maybe I'll start with you. At a highest level, at the highest level, how can you how would you place Ethereum in time in its development roadmap? Maybe a percent completion or like left to be done. How would you describe where Ethereum is in what it wants to become?

Right. So I think the the initial vision of Vitalik of having you know this this base layer which is secure and scalable. So we have the proof of stake and and the sharding is actually coming to fruition. But one of the things that's kind of surprised me as a researcher is that we we seem to discover something new every two or three years. And so the design space expands, but that's kind of exciting, right? Because there's more potential. So on the one hand, I want to say that we're,

I don't know, 75%, 80% of the way there in terms of completion of the initial vision. But the vision has grown in scope. And so maybe we're only 50% of the way there.

Would would you resonate with that, Tim?

yeah, and I think one thing to maybe add to this is um a lot of the new things that are coming tend to be related to like protocol adjacent uh phenomenon. So like MEVs obviously a huge one, layer twos is a are a huge one. So like if you think strictly from like Ethereum L1 as like a you know system, um yeah, I I think we're we're getting there.

Right. There's some quality of life things. So for example, account abstraction, uh, encrypted mempools.

Um

like these are things that we didn't necessarily realize that we needed at the very beginning.

Um and but there's also some like really future thinking stuff. Uh so for example, upgrading to post quantum cryptography. And even beyond that, there's something that kind of only opened up in the last few weeks. Not only should we defend against quantum computers, but we can harness them to make consensus even more robust.

I think hopefully in this conversation, if if I guide this conversation correctly, we'll start with some of the short-term concrete stuff. There are short-term things that are being worked on right now. Probably the Bankless Nation is pretty familiar with, things like 4844, PBS, the short-term things that are that are going to be included into the Ethereum protocol, you know, short in in Ethereum time frames, about a year, ass soon TM. Uh then there's some like stuff that's also being worked on that, you know, are middle-term stuff. And then there's some very long-term stuff. Maybe I'll put quantum computing in there. Uh, I don't know if ZKing, the entire Ethereum, is still a topic of conversation, but I want to open up that box. And so, like short-term, midterm, long-term, like, and long term is like the sci-fi stuff. Um, so I think that's how this conversation is going to go. And maybe towards the end of this, we'll just summarize everything and talk about in that fullness of time Ethereum, the version of Ethereum that will eventually, the trajectory is strong. We'll get there. What does that look like? Uh, but let's start with the short term stuff. Like, what would you say is the immediate priority that the devs are focused on the most in the short term?

Four and four four.

And I think beyond that, so um the next upgrade is called Denkun. There's Ford44, it's the main feature that's part of it. There's also a couple other things that we're adding. Um, so

On the execution layer, one thing that's notable is uh deactivating self-destruct. So this is something we've talked about doing for years. And it might be the first time we actually deactivate the feature on Ethereum. Um, trying to think, I don't think there's another case where we've done it. Um, I mean, we deactivated proof of work. That was a big one, but like something that like smart contracts and whatnot depend on. Um, so basically preparing this entire upgrade. Um, another cool thing that's uh being uh delivered with this upgrade is exposing uh the beacon uh state route in the EVM. So this means that the execution layer will be able to make like proofs and commitments about what happens on the beacon chain. Um, and it'll help things like make staking pools more trustless. Um so generally, yeah, this upgrade is what devs are working on right now.

And so this is four four is is the big one of course, and this is what we call dank sharding. Pro is it this is proto dank sharding.

All right, sorry.

Yeah, so the middle step between full denk sharding. Uh, and we already have layer twos on Ethereum. They're totally operational, but this is what I would put under the roll up centric roadmap for Ethereum. As in, we are making Ethereum, the protocol is making rollups a first class citizen. Can we unpack that nuance a little bit? What does it mean for rollups to be a first class citizen? And how does that change what Ethereum is for with with prototycharding?

Maybe I can start with pro lank sharding and you can talk about longer term like making roll-ups a first class citizen. But yeah, just to unpack bank sharding. So when you make a transaction on layer two today, um the layer two bundles your transaction and a bunch of other people's transactions and posts the data for that back on layer one. Um and this is why layer twos are kind of cheaper. So computation on Ethereum is expensive, but storage is cheaper. So layer twos basically say we're gonna do all the computation off-chain, put the data, the raw data from that computation back on Ethereum L1. And if you know there's a fraud or if you want to reconstruct the state on a ZK rollup, then you can take that raw data and like run through all the computation, but only if you if you have to. And this is kind of the trick that that makes layer twos keep a lot of the security assumption because the data's on Ethereum, um, but also be cheaper because you're not running every transaction. Um, so today when you make an L2 transaction, something like 90 to 99% of the cost is actually posting that data back on Ethereum L1. Um

And the reason why that cost is expensive is L1 stores that data forever. But that said, layer twos don't actually need that data to be stored forever on Ethereum layer one because you know with optimistic rollups, we have this seven day uh exit window. So this means that, you know, after seven days, no matter what the data is, somebody could have like done something, exited all their funds. So what we're doing with Proto Dank Sharding is we're adding, can think of it as temporary data storage on Ethereum layer one. Um so that this way, because the data is only stored for a couple weeks, um, we can charge much less for it. And this will reduce the cost of layer two transactions because now um instead of paying for storage forever, they're just paying for storage for a couple weeks.

And then Justin, if you want to finish this conversation, but like, what does that mean for Ethereum the protocol? So we have layer twos that have cheap data, very cheap data that that um that you know lowers the gas fees, makes them faster. Uh but again, the roll-up-centric roadmap of Ethereum, why are we prioritizing layer two so much? And how does that what does that mean for the protocol?

Right. So I'd say that Ethereum is prioritizing the friendliness to build rollups on Ethereum because you know

putting posting this data is is the main is the main bottleneck.

But and I guess one of the consequences here is that we're gonna be able to scale the number of users on Ethereum maybe from 10 million to 100 million or maybe even hundreds of millions.

And one of the things that I expect in the short term is we're introducing this whole new resource, which won't be like fully consumed at the very beginning. And so we might see dynamics similar to what Arjun did a few years ago, where basically the fees were so low that they subsidized the fees. So it's possible that for a period of time, maybe six to twelve months, we're going to be able to use rollups completely for free. And there's going to be this kind of this temporary subsidy almost that could act as a bootstrapping mechanism.

I guess one of the things that I've been thinking of in the long term is.

How can we

really uh have Ethereum native rollups? And we we call those uh enshrine rollups. And

um

what we're gonna do in the long term is actually get an enshrined rollup, one instance, by SNRCying the EVM. So if you take the EVM today as it is, every validator has to verify every single transaction. But once we've SNRKified it, all the validators need to do is just verify this one SNORK per block. And so it's automatically becomes a rollup. So the main net becomes a rollup.

Ah, okay. What does that mean for the other roll ups that also already exist?

So, what I expect will happen is that there will be kind of a nice compliment complementarity because the enshrined rollups will have best in class security.

But on the other hand, and one of the things that they enjoy is access to social consensus. So if for whatever reason there's a bug in the EVM, then we're going to go and fix it.

Whereas if there's a bug in one of this, one of the rollups, I know Arbitrum, then kind of their problem. And so the way that the the non-uh enshrined rollups fix issues is with governance.

And so really we don't need this governance token. And so this in a way makes it more secure because governance, not it's a way for flexibility, but it's also an attack mechanism.

Is this so if we're snark of snarkifying the EVM and we're getting similar properties to what you would get from a roll up, you can bundle up a many, many transactions and have them uh have them go through much cheaper gas costs and much easier computation.

What does that compet is that competitive with the current layer two landscape?

So the the layer two landscape, I don't think, will compete with enshrined rollups for a few reasons. One is on cost.

So you can do certain things with

application layer rollups that you can't really do with an enshrine rollup. One of the things, for example, is batching.

So if you are willing to settle, let's say, once every 10 minutes, you can take these bigger batches of data and then you have more efficient compression. Think of it as when you're zipping a file, you have better compression if you have more data to work with.

Another thing that you can do with the non-enshrined rollups is you can be very opinionated in terms of where the cash flows go. So, for example, Optimism, they can decide to fund public goods.

But we, you know, with the enshrined rollups, the only real public good that we can fund is economic security and economic bandwidth.

And then another thing that you can do with the application uh layer rollups is that you can have whatever virtual machine you want.

Um, you don't have to stick with the EVM. And so there's going to be a lot of advantages there.

And then maybe some other benefits could include services like pre-confirmations.

It's relatively difficult to do pre-confirmations with an enshrined rollup, but that's something that the application of rollups can do.

Where is this on the theoretical idea to EIP on the roadmap spectrum?

So getting to one base rollup is actually only the beginning, but it's like 90% of the work. And the reason is that we have to do the really heavy engineering to have these snark proofs for every single block. But then there's this really cool thing that we can do, which is maybe only 10% of the work, where we can expose as an opcode.

The verification of a snark proof for the EVM. And so that means that anyone can deploy an enshrined rollup at the application layer. So it's

it basically opens up the possibility to have as many copies of the Ethereum layer one EVM as we want,

all with the exact same properties as Ethereum Layer 1.

When will that happen?

My guess is that we're talking five to 10 years.

So did we accidentally skip into the mid to long-term roadmap of Ethereum?

Yes.

Okay.

But I'll caveat this that

the access the development of um snarks has always surprised me in a positive way.