Defending Beamchain & the Ethereum Roadmap | Justin Drake
35:59 Share

Bank's nation, I just got off my flight back from bangkok from dev kn, where we all had a fantastic time throughout the week of dev con talking to all the nerds learning about the future of the theory um and about all the ways that the serum is changing the world and on the last day of depth on the day after that on saturday we host the bank less summit which had ten speakers all giving their most interesting, their most fascinating talks possible that is what I charged them with and all of the speakers rose to the occasion.

We've recorded all of those talks and we are going to be releasing those talks on the bank less podcast feed and on the youtube slowly over the next few weeks. You are about to go here just in drakes talk, which was a reflection of his talk at dev con beam chain talk, adding in some additional much needed context that I think he wish he had to introduce the first time when he introduced the beam at dev cunt, along with many other facets of the syrian road map. There's also Peter ban walken burgs talk, which are you going to come later as well as argent talk from connect, some of the most interesting, fascinating talks that i've listened to in this space.

We got a ton of positive, amazing feedback from the bank of lammers. So thank you to all the speakers who worked so hard putting their talk together. The audience loved that we had a hacked house.

I'm looking forward to the next bank less of IT whenever dev connect is announced. But meanwhile, while you wait, i'll serve you up all the talks from this year's bank less of IT, the first ever back with summer. So let's ahead and get into that first talk from Justin drake right now.

But first in one minute to talk about some these fantastic sponsors that make this show possible if you want a cyp to trading experience, bact by world class security and award winning support teams, then head over the cricket in when the longest standing and most secure crypto platforms in the world crack is on a journey to build a more accessible, inclusive and fair system, making a simple and secure for everyone everywhere to trade crypto, to crackings. Intuitive trading tools are designed to grow with you, empowering you to make your first or your hundreds trade and just a few clicks. And there's an award winning client support team available.

Twenty four have to help you along the way, along with they hope range of educational guides, articles and videos with products and features like crack and pro and cracked in N F T marketplace and a seamless APP to bring all together. It's really the perfect place to get your complete crypto experience. So check out the simple, secure and powerful way for everyone to trade cyp ta with your a complete gainer or A P go to crack ash back to see what to rim of free IT is becoming a super debt where you can do everything cross chain all in one place.

There's solving U. X. Issues by building products across every chain ecosystems, wap sparring and lending nf perps and more, all seemingly cross street, instead hopping tween countless sites and handling multiple s just use magic ve time inked to the meme coin faster.

Plus M E tok in is coming out this quarter from the M E foundation. If you use any of their protocols like the N F, T platform, you can claim some. So check out magic eating today and want them become a super dep.

And the home of web three in real time, you can get there at magic eaten dot. I O ticket started launching a token, worried about the I O S. Talk a token for five minutes to learn how to save hundreds of thousands of dollars for you team and your investors.

Now if you do in an initial consultation, tokyo will cover the cost of your token valuation, which is what you need for your launch. So don't wait. Reach out to the team right now at team at toko dot com, that team at T O K U dot com. Just a drake is going to make you very bullish. Just a drake.

Yeah so in being stepped watch by uh eric and max front front by eric um and the background by max that see how IT goes. So i'd like to share some takes about the road match. Um and just as a point of this clam, these are mistakes.

These are not official or road map takes when you think like that. So one of the things that as a researcher is getting me really, really excited is that I think we can take but is possibly humanity most ambitious distributed systems road map and complete the whole thing within the next five years. And this is kind of what I wanted to try and communicate to an extent, uh, three days ago, with the the beam chain.

And basically what I am suggesting is that we take the hardest items in the road map, those that are very, very difficult to do incrementally in the short term and kind of batch them uh in the in a single folk. And there's been um kind of different pieces of feedback. On the one hand, i've had amazingly positive feedback from the depth and um you know many, many people detail the email address, many people message me on telegram and on twitter and things like.

And today, i'm kind of very pleased to share of, somewhat astonishingly, that we have five new teams that have signal interest in building a beam client. And so from the the perspective of debs, um I think there is a lot of excitement to prostitutes in this in this new journey. But one of the things that have happened is that on crypto to twitter, the the feedback was much more in particular.

This is possibly like my most hated slide ever. Which is basically saying that you know, it's going to take roughly four years from twenty twenty five to twenty twenty nine to ship this extremely ambitious folk and people want things in the next four months and the next four years. And basically the I think the attitude of critter twitter is that we're onna enter this critter of darkness.

We have light at the end of the tunnel, but we have a five year tunnel, and it's just going to be full darkness during that period of time. And I kind of want to explain that this is not the case. There is not going to be a massive void here between today and five years.

That is going to be a ton of stuff happening, a ton of stuff happening. Now if firm l one is made out of three sub layers, so there is the execution layer in purple, the data layer in yellow and the consensus layer in in Green. And what i'm going to do is kind of go through them won by one and try and share a some of some of my takes.

But I think the key message that I want to share here is that the performance of the serum over the next few years is going to increase incrementally, and it's gonna be night and day. We're gone to improve all the performance metrics that are key to um to the layer one. So what are the performance metrics for each for execution? The key performance metrics is get per second.

How much can we through the E V M for data? It's bites per second or blogs per second. And one of the key performance metric for consensus is uh latency like how fast is IT take for transaction to be to be confirmed and and over the next few years, we're gona improve these three um performance matrix. Now for execution, the magical tool that we're gona use is Z K um and one of the tweet that kind of reflects this bullishness ers that I have is from john give us. John gibbs is one of the co founders of succeed behind the S P one Z K V M.

And what he tweet is that we're going to prove the entirety of the failure um on these bad boys and these bad boys are like four failure machines, like four basically big, big server machines and he believe, is that enough to prove the entire ety of the film? And on one piece of interesting meta data here is the number of three weeks this tweet, which is extremely, extremely bullish, got two retweet. People don't realize what is coming here.

And so what um myself and various others at the family foundation are doing is building a new website is not live yet but they will be called if proves that ogg and this is basically a block explorer for Z K vm. So we're gonna invite every single Z K V M vendor to prove l one E V M blocks and for each block to provide the proof and disclose how much time I took them to generate the proof and how many sense IT cost to generate the proof. And then basically the idea to show the world that these Z K vms are good enough for for, for production and also create a little bit of a friendly competition between the Z K V M with with various leader boards so that um we we'll be able to see uh which ones are the best.

And ultimately what I think will happen is that this is going to kick start um A A process where the whole film chain is going to be Z K proven by many, many different zk vendors. Now one of the things i'm excited about smoking fine, the entire a chain evm chain is that I won't have to run death. So I run a validator at home and I have to run deff, and I only have two terribles of story.

And it's a massive pain when I have to recent. And that uses too much space. IT might use too much RAM or CPU.

Well, very, very simple. I won't have to run gift at home. I can run basically the Z K version of my execution clients, for example, Z K gift instead. And what that means is that I will consume the zero giga bites of storage, and they will use the zero percent of my CPU. And zero percent of my RAM is going to be free to basically run an execution client.

And so I think in a couple of years time, we're gone to be in the position, as they said, where the running an execution client is going to be optional as a validator. And in order to achieve that, we need the snuck proving to be fast enough, right? We need the snuck proof to arrive to the attested soon enough that the a testers can constantly vote and know that the chain is indeed valid.

And so for that we're gona need sub slot Z K vms. We're gonna need zk vms where the proof comes within one slot and that means within twelve seconds. Uh, and I think that's going to happen by the end of next year, by the way, all of the days that I say take a with a grain of salt, all of this is just a attentive tive strawman timeline.

And then another thing that is going to make this state of affairs really good is actually a proposal by a the next speaker and then Robinson to delay the state route by one slot. So if we delay state routes by one slot, then now we have the full slot in order to be able to prove the the validity of the state routes. Okay, that's all great.

Now another thing that snakes are you gonna do in the in the road map is a kind of reshuffle and and maybe pivot the the stateless approach. So right now, well on the vertical, um no approach. And basically I think what's going to happen is that we're gona move towards uh snuck base state lessen ss.

And so one of the the cool things that has happened in the last few weeks is basically the deaths coming together and saying, hey, maybe we can skip vehicle le and go all the way to the n game in one go, which is to use bine remarkable trees. And that has two advantages. Advantages number one is that it's it's much more snow friendly and two is that it's post quantum secure.

And so all of this is the accelerating the um the the the the the road map runs notification. Now I promise that I that you know we're going to see massive, massive upgrades to the performance metric. And the one of the key performance metrics that we have today is the gas limit, thirty million mega gas per thirty, thirty mega gas per block.

And I think we're gonna grow that a hundred x we're in the next few years to three gig gas per lock. Now how can we do that safely? One of the key things to realize is that we would need all of the execution clients to be able to keep up with the three gig gets and this kind of two strategy strategies.

Number one is that you know you a home vue data and you no longer run gf, you know you run the the Z K version of your execution uh client and verifying a block that consumes three mega gas or you know three mega gas doesn't doesn't matter because IT is always takes one million second to verify the proof. And all the other approach is that if you want to naively rix everything, a very beefy computer, and that's something that maybe the big Operators can do. The big safety feature that we need chip first before doing these, these big tanee gas limit increases is to have something like multi dimensional IP one, five, five, nine.

And the reason is that the only thing that we can increase is the the gas limit for execution. So we can't have blocks be too big. So we need to have a separate limit for cal data and we can't have the state grow too fast.

So we need also a separate limit for state growth. So once we have these three different limits, we can have the first to stay relatively small and reasonable and then basically pump the raw execution gas limits. And by raw execution, I mean just things like arithmetic Operations, like additions and and beautiful icons and things like that.

Now not only can we scale the E V M horizontally and and grow the, the, the gas limit by all of magnus de, we can also scale the E V M horizontally. And the idea here is what I call programmable execution shouting. Or another way to say IT is programmable native roller.

So once we notify the h one V M, we can expose within the E V M A Z K E V M precompiling um and and this is something that any developer can use to spain a new instance of the evm. And I think this is something that can be done, you know maybe four years down the line. Uh before the before the the the the five year deadline comes.

Okay, so that was execution. We can essentially have unlimited uh gas per seconds, right? Because we have unlimited number of of E V M instances and even individual E V M instances will grow dramatically uh in in fruit.

Now what about data? The story of data is actually fairly straight forward. Um we have roughly one folk player um so the folk items are the one with a little red circles. And every time as a folk, I imagine that we're gonna on the increase in a number of loves and we're going to see an exponential increase more less so too exiting every single time.

One of the big unlocks is going to be peer us where we start doing data sampling, where the validator is don't have to download all of the blows, but they only have to download the smoke um you know chAllenges. And then the way that we do sampling is also going to improve with the two examples here. But overall, the story is that we're going to have exponential over the next few years of the data, and that's going to allow for many, many more transactions on chain.

And not only are are we gonna go two hundred and twenty eight blocks per block, but we can go way beyond that. And the reason is that we have a new the law, which is basically the increase of the natural increase of bandwidth time rough on the order of roughly fifty percent per year. Now here in the in in in the data section, I want to correct or I want to share at least my opinion on on, on, on various things that that that people say that I strongly disagree with.

So a bunch of people say da is abundant. I I don't think da is abundant. I think da and by da I mean like the highest quality kind of affair um grade D A is is is is not abundant.

And the reason is that if thurium limits itself to home internet connections and uh what bank shooting does is that it's it's a multiple er over the number of internet connections and the multiplier roughly one hundred x so what we're talking about here is providing a hundred home internet connections for the whole internet value. That is a tiny, tiny amount of supply. And D A is extremely css in the grand scheme of things.

The other take that I disagree with is that D A is, is, is fungible. It's a commodity. So you could take A F M D, A and just replace IT.

We have celestial D A, and is the same thing. IT doesn't really matter. And the two big reasons why IT doesn't matter is shed security and share synchro. So the whole point of um the block chains is at least in the context of defy is to try and build these money legos that are composer and in order to for for one money lego to feel comfortable composing with another money lego, they need to know that this other money lego is extremely secure, as secure as as they are and and is going to live for the next decades and centuries. But if you use some really weak uh external D A, then now suddenly you're going to lose share security.

But then maybe the the bigger reason, and this is something that is going to play out as we have more and more tools for composition service, is this notion of synchronous composition order in order to have best in class, compose border between money legos, we need to have shared sequencing. And the problem with da like salesa is that IT introduces a new sequence CER right like select is itself a blockchain with slots, and each block is proposed by A, A, A, A new sencer. And so now we have the affirm sencer and the the selected sencer, and we lose the opportunity to duce increase composition order.

And so I think people are going be willing to pay a big premium to consume A M D. A. And then the final reason has to do with the the native rollup that I was talking about IT turns out that um the precompiling only be consumed if the input data um to the to the roll up is is on a firm and has to be available um so that anyone can produce the proof for the validators otherwise there's an attack where someone can try and split, split, split the chain.

Oh, that is going to be yet another reason why I think if M D A. Is going to be hugely valuable. And then finally, which which is the culmination ation of these two takes is that I think we're going to see crazy high aggregate data fees, right?

People say that da is going to beef, you know uh, is not going to be uh uh uh a source of revenue revenue that we're not going to have high fees. And I completely disagree that and this is something that I shared in my latest A A bank less podcast. I think we're going to potentially see billions of dollars per day of fee revenue from the fees.

And that doesn't mean that the individual fees are gonna hind. The individual fees can be substance, but the aggregate fees for the whole internet value in terms of da consumption can be as high as billions of dollars per day. okay.

So we're dramatically improving the performance of execution, dramatically improving the performance of data. What about consensus? So a lot of people talk about, you know, having all so fast slot times, right? Why not? We have one second slot times or four hundred minutes, seconds at times.

And my take care is that slot times very soon are not gona matter as much as they matter today. And the reason is that we're going to have three confirmations, and we're already starting to see this right. If you use orbital, for example, you have two hundred and fifty second, two hundred and fifty minute second.

Sometimes IT doesn't matter that the l one uses twelve seconds at times. If IT were to upgrade to one second first time, that wouldn't make any difference whatever ver to the users of arbitrium. Same thing for uni chain unit chain also has two hundred and fifty million second, sometimes faster than Selina, by the way.

But that doesn't mean um that uh well that maybe a reason for not not needing to prioritize small times. And I think the same thing is gone to happen for bed roll ups. Even if you use the l one um decent sencer for your roll up, you won't need to uh to have a uh you will need to pay the cost of of of of slow U X.

And my my my rough take basically is that um within a couple years the vast majority of transactions are gonna be confirmed and they are gona be reconfirmed much, much, much faster than even the the slow duration on on the serum. Oh sorry, i'm a one a now the other big upgrade that you know could happen to the film layer, a consensus layer, is state capping. And what state capping does that? IT improves another performance metric of consensus, which is the security budget.

Like how much money do I need to spend per year to secure my uh to provide security for for the consensus later. And right now we're paying many, many billions of dollars, roughly five billion dollars per year to secure a arian. And if the Price of eve would turn x and we potentially be uh be paying tens of billions of dollars uh per year.

And so what what I think could happen in terms of improving one of the the other key performance metrics s is this notion of state capping. And this is this would be an incremental upgrade that would happen away before the beam fork at the very end. Now in order to do state capping, we need to solve the problem of M, V.

And the reason is that if we have low insurance but high MV than almost all of the income is going to come from M. V. But the M V has a problem.

It's extremely spiky. It's extremely volatile. And so we need to either smoove the MV or get rid of the MV somehow. And this is exactly what a test to proposal separation does.

A test to propose a separation basically opens up the MV market to the to the broader ecosystem and and and and IT segregates MV away from the from the validators. But in order to get um A P S, we're gonna need we're gonna need inclusion lists. And so see here, there's a little bit of a dependency in order to dramatically ally improve this, we're going to need this.

And to get this, we're going to need this. And so the good new year is that we already have an p fossil IT won't go in the next fork because all of the items have been scheduled for the next week, but I believe it'll go in the one after that. Now another big thing that i'm excited about, the consensus layer is not just performance but it's health, right? Because ultimately, the way that I think about IT and you know here i'm i'm stealing this mean from someone else, which is that the l one is going to compete with bitcoin and the l two are gona compete with sana.

And so the l one is to be as healthy, as secure, as robust uh as possible. And so um one one of the the key things that we need to do to achieve this uh this really strong foundation is to make the l one very, very friendly to solar validators. And A P S is a massive step in that direction.

And the reason is that today to be a very competitive validate, you have to worry about many things. You have to worry about timing games. You have to worry about MV Spikes soon you're going have to worry about the infrastructure to provide free confirmations. You need to pick, relay and seeks that all of these worries will disappear. And your main duty will be to just a test and provide and build decision's.

Now the other big improvement is that um that should happen incrementally is this notion of one if validators right today, in order to become A A validator, you need thirty to if this is what I call whale taking, you need to be a whale you need to have at least thirty to eve, which is one hundred thousand dollars to participate a validator. Why don't we reduce that by one and a half orders of agnes ude? And in order to get the one if validators, it's a very similar story to to stay capping.

There's like some ground work that we need to do ahead of time. So what we need is basically the validators to start consolidating um because this basically two ideas to get one if validators number one is what's called orbit and the one is called two ted staking. But in both cases, we need some amount of consolidation.

And in order to get consultin ation, we need um to increase the maximum effective baLance. And this is coming in the next week. And so you see here, we're planting seeds because we have this this long duration view.

And one by one, these things kind of play out. And then we get to to read the benefits are from from our investments. And then the final thing that we're gonna dramatically improve on as a as a solar solar staker is this notion of fever staking once we fully snake fied the consensus layer and the, uh, execution layer.

The mean is that you'll be able to stake on your smart watch, right? You won't need the laptop. You won't need a rusby pie. You'll need like the most minimal amount of infrastructure.

And so when you combine these three things, uh, fish taking, then staking and father taking, what I expect will happen is we're gona see a rena sce of solar validators. We are going to see tens of thousands of little, little validators coming up, and that's gonna extremely healthy for A F M L. One and is going to make IT the most robust change in the world.

And so that is the summer slide. So much is going to happen in the next few years. And um you know this is kind of a cold action table to be bullish about the future, just the long term future in five years, but also the short and medium term future. Thank you.

We got time for a couple of questions. Couple questions where raises their hand first. No questions. One question, one question.

You talk a little bit about a and and do you see that being A A centralization risk to the three mental? Okay, fantastic question. So and this is a recurring question.

And so I am I hope that you know, as a community, we can try and and and get through this. So aps, what he does is that IT opens up block building to the broader market of sophisticated actors. And so what we should expect is that the most sophisticated actors will be building the majority of the blocks.

And that sounds terrible, but the good news is that actually self building is going to thrive in that context. And that's thanks to the context to the that's thanks to inclusion lists. So with inclusion lists, we have sixteen validators at every single slot that are gone to be self building, the alist and what is in inclusions? St is basically a list of transactions that have to be included by the sophisticated builder.

So gonna take the sixteen inclusion list, we are going to take the union of them, and that's going to be the transactions that need to go in the next block. And why? Why is that extremely positive? The reason is that I expect the vast majority of transactions that ultimately go on chain will be included in the inclusion list.

And so the shift of mental model that I have is that the almost all of the block building called at ninety five percent of IT is gonna happen by the inclusion list and the validators. And then the builders are just going to do you a little bit of polishing at the end. They're going to ad maybe one transaction at the very top to do some prompting, maybe one transaction at the very bottom to do some book running.

They gonna read all the transaction a little bit so that they can extract a bit of M, V. But ultimately, the bulk of the building is going to happen. And I distributed manner from very unsophisticated in the week builders um and yet the mean that I have is that the the validators provide the raw material for the blocks for the inclusion list and the buildings are merely polishes, right they they merely kind of polish things here and there to produce the final block.

One last question before being max nic on over there over there here we go. I'm coming.

Hello um I wonder if you asked any application developers about feedback um for your old map and what was their feedback and what kind of um things you included there that help application developer specifically. right. Good question. Um I mean, i'm i'm very much in touch with with the community through a following critical twitter and coming to to to to uh to conferences and what not.

And to be fair, I haven't gotten uh that much negative feedback other than the the basically two of them one is the fragmentation problem and to is you know the scale ability problem on the on the regression point of things which is basically um I guess eric s talk just before me. Um I think we have a very good road map and this is something that i've i've been talking about for for all of this year, which is this notion of you know the united chains of a firm that comes from shared sequencing and and real time proving. And in order to you achieve a firm wide share sequencing, my my proposed solution is is is based sequencing and baseball up.

Uh, and you know maybe this is an opportunity to have some sort of of a debate wave, eric, or discussion with him because I think he's kind of too barish on the fragmentation issue, I think is going to be solved in in in in the in the coming months and and years in terms of the the scalability. You know I I guess one of the the common complaints is that um you know on solana, for example, I can I can trade meme coins no ten tens of times per minute and it's extremely fluid. It's it's extremely cheap and I can't really do that on this theme.

Well, three confirmations are going to dramatically improve. The U X. Is going to be faster than solana. And in terms of the fees, this is where you know increasing the the the amount of da comes to play.

Now if the only thing that you're doing is trading meme coins and you don't even need if if they are M, D, A, you can just use some very cheap D A, like the last ja or egg and da. And so what I expect will happen is that mean coin trading in twenty twenty five is gone to be. Cheaper than solana and faster than sAlina. And so there is a possibility if you know the meeting point builders want to come to to a firm to basically migrate in twenty twenty five once we have uh the the infrastructure in place, especially with the reconfirmation .

round of reply here. Just drink, everyone. Thank you.

New projects are coming online that the mental later to every single week. Why is this happening? Maybe it's because mental has been on the frontier of later to design architecture since I first started building mental D A power by technology from and d because users are coming onto the mental layer two to captures from the highest yields available in defy and to automatically receive the points in tokens being accrued by the three billion dollar mental treasury in the mental reward station.

Maybe is because the mental team is one of those helpful teams to build with giving you grants, acquitted these support inventer partners to help booth p your mental application. Maybe is all of these reasons all puts together. So if you're a dev and you want to build on one of the best foundations in cypher to or your user looking to claim some ownership on mandeep apps, clink is officially the preferred wallet of blankness and is the one we use any time when you want to transact on train with.

If you are on your brother er or on the go, you just wap wallet. IT makes IT easier than never to swap anytime, anywhere, use your wallet to transfer und directly from the tops, and tries exchange and thousands of tokens across the here and over ten other chains like base arbitron and optimism and delivers deep liquidity, fast execution in reliable quotes with zero gas walks through unis, wap, x and when IT comes to security, you can rest easy knowing its back by unit swap x one of the most trusted teams in defy their code is open source and independently reviewed. So you know, it's protected.

So why wait? Download at the union all today on chrome, IOS and android, and to claim your free uni dot eth username directly in the mobile war, start swapping smarter with unsound, centralized, held, all the power. But now the if I back, you enter the unlimited era with D Y D X. Unlimited features, instant market listings where you can listen, trade any market instantly with liquidity, experience trading like never before with all the trading tools, you know and love and and then there's way to automatic quality and wheel. Just deposit some U, S, E, C and start earning yell immediately is so simple, and D Y D X unlimited comes pack with so much more leverage prediction, market prospect futures, instant deposits and there's even a revient trading rewards program to get rewarded for every time you make a trade to go head and check out D Y D X on twitter, D Y D X to get notified once D Y D X unlimited launches.