Vitalik Buterin (Zoom EthereumBA: Ethereum en 2021+)

Transcript

per [Music] local Krypto atsas found

PR Def Gra

for thank you thank you so much for for being here this is H super super important and significant for for our community uh so thanks a lot for taking this time and uh you can go ahead and take it from here okay now I'm much and notes everyone um so I will um switch to sharing some slides now uh so there we go are you seeing it okay uh so what I wanted to uh of talk about is just quickly go through and

what H um what happened recently with the ethereum beacon chain launch kind of what that means what has been accomplish accomplished so far and kind of where we are in terms of ethereum ethereum adoption ethereum technology how those things kind of connect to each other and what this means in terms of of how where we needs to go and in the next year and in the years after that uh so big exciting news of the year right the ethereum proof of stake Beacon chain launched uh so this is part of this many year long process that started in I would say even

2014 of trying to of upgrade the basic and technology that the ethereum chain ass sets on and so trying to switch from proof of work to proof of stake um trying to add more scalability through things like charting trying to make the consensus more secure and just trying to really kind of push the limits of um you know what what a well-designed blockchain can actually do um and like basically just push the bar much higher than blockchains as they existed know five or six years ago so research um on proof of stake and sharding has been

happening for more than six years a development of this system that we now often call eth2 has started about three years ago really around the beginning of uh 2018 um we had this kind of big event in taipe in about March 2018 where we brought a lot of the of researchers and the development teams together and then another one in Berlin shortly after that where we agreed on some more details of the road map and then it it continued from there so but now as of two weeks ago the beacon chain the this new ethereum Pro of State chain actually exists um and some statistics actually some the participation in this in the

beacon chain has been much better than um even we expecting so now there's actually I believe more than 1.5 million eth more than1 billion dollars of eth um has been deposited by people who are be who wants to become validators um likely about 4,000 unique participants um so including everyone who put in the minimum which is 32 eth and then in um trying to look um at um addresses um on these one side that deposited to try to figure out who are the unique participants trying to see who contributed more um so over 4,000 participants actually fairly

decentralized and need about 30 of the largest participants together to reach 51% so sign much more decentralized so far than even like say the E Bitcoin or the ethereum proof of work chains have been um and about 99% online so people aren't just depositing and forgetting about it they're actually actively participating running their nodes and doing quite a good job of it so it's worth you of talking about know what the beacon chain does um and what it does not do uh so basically this is part of this complicated process right to try to move ethereum from the proof of a work

ecos system that it is in now to a proof of stake system and this is a complicated process proof of stake or at least our implementation of a proof of stake tries to do a lot um there's a there's a great thread by Jeff Coleman on this actually if you go to on Twitter and you look up techn crypto he has this big long thread in a fairly fairly recently where he goes through like basically you know why we spent so much time creating our own proof of stake instead of just copying like what exists in like say piure coin or NXT for example so it's a complicated process it's a very complicated new system

with you know just as proof of work itself as a complicated system um with new assumptions and so we decided to kind of take this path where instead of upgrading all of ethereum to proof of stake all at once we G of start the proof of stake chain first and then we have a later step that we call the merge where basically once the approv of stake Chan has been running for some time time and once it has been der risked for um for some time so once we have some assurance that it works in the real world then we kind of move up a work over to it uh so chain comes to uh comes to consensus you can deposit your e you can stake um there's this uh Pro very

strong property of economic finality that it provides so basically if once a block is published on average after something like 13 minutes the or well maybe more like 15 on average 15 minutes the block is finalized which means that there's just no way it's not even possible for the block to be reverted and there's a very strong guarantees of Security even before finality like even after um say two confirmations so about 24 seconds right so strong properties um now the E2 effort is not just about staking it's also about charting and scalability and other things and those

things have been researched uh they're basically all in different stages of development now but much of the kind of comp really complicated Tech and the features needed for those uh stages to be um to be able to happen is already included um right so like the concept of of Committees of random sampling that we need for charting for example Le is basically already in the beacon chain so it's not going to be a difficult to add but things that we do not yet have right so we do not yet have charting um we do not yet have the ability to process transactions on the prod stake chain we do not have like client support um we

also still have the proof of workchain right so most or or all ethereum applications today are still on the proof of work chain and they are um the proof of work chain is not going away at least until this event that we call the merge happens where and all ethereum applications get moved over to the appr of State chain so what's left I think it's worth kind of talking about what's left in the context of kind of what the goals are and just why they matter in the context of you know the ethereum ecosystem where it is today right and it's worth talking about the ecosystem here for a bit right the ecos

system has um been very successful I think in know 2019 and 2020 a lot of exciting applications are starting to see just be used and there's all of the decentralized finance stuff um there defi um maker Dow for example like other kinds of stable coins starting to happen uh there's a Unis swap The onchain decentralized Exchange and some other ones like balancer for example have been very successful and have grown just immensely in the last year uh in particular uh ens continues to be used

um nfts a lot of these of kind of less Financial applications some kind of art some kind a virtual reality a lot of really cool things happening um and poap also a very cool um prediction markets also you know there's there's been a lot of adopt of kind of people getting excited about and using prediction markets and this recent us election that's in our kind of global world has become Global News uh so and most of the most of that is also happening on ethereum so a lot of like really

interesting of things um happening and also just a lot of adoption a lot of people coming into the ecosystem and especially as of you know we're seeing with of media attention um starting to happen prices um you know Bitcoin obviously hitting its all-time Highs but ethereum not too far behind there's a the community is kind of growing right and you can see it in a lot of ways and there exists a I think a very large chance that the community will become 10 times bigger maybe more than 10 time more than 10 times bigger can we say 100 times bigger over the next few years and

people are really going to start using blockchain applications of different kinds just on a very large scale right like I think if this next kind of wave of growth happens we'll have you know regular people using um you just reg cryptocurrencies or stable coins for payments or to store their money um we will actually have um you know things like ens be integrated into a lot of mainstream applications um we um would have of nfts used for a lot of a lot of different things um potentially some kind of these blockchain based um you know identity and reputation and credential

systems will be kind of plugged into um you know mainstream applications that people use across entire countries and we just need to be ready for that reality from a technology point of view right so this is I think why we've been focusing so much on the scalability right because uh so first of all you know we for we want to uh just reduce transaction fees so that non-wealthy users can participate right so today transaction fees are frequently even just to send ether something like like $1 or greater or greater than $1 and this is not okay right and I've said since 201 15 that this is not okay and that we

need to be working very hard on make making sure that the there's more space for transactions so that we don't have this Dynamic where everyone is bidding for a very small amount of space so that only very few transactions can get in um we want to make blockchains usable for Noni non-financial applications so uh $1 maybe you know wealthy people can accept it if it's for kind of financial transactions but even for for non-financial transactions people are used to things being much more close to free um just make sure it's possible to support hundreds of millions of users at all right right now um on the ethereum Chain there's only space for about 15 to 45 transactions per second depending on

what kinds of transactions they are and that is just not enough for hundreds of millions of people right like if you do the math um 15 transactions per second is about 470 million transactions per year and so what that would mean is that if there were hundreds 470 million users each one of them would only be able to send one transaction per year so they're just isn't enough space and so you know we need scalability for all of those reasons and also just a preempt centralized scaling Solutions right so if there is demand someone will find a way to scale and if there are no decentralized forms of scaling then everyone will just move their coins onto exchanges all of the activity will

happen inside of exchanges and we'll see just a big growth in these applications that have only a very small connection to the actual blockchain and where most of it is just being done the same way as before so for all of these reasons I think um increasing scalability is a very important challenge um next goal is light client support so you know red we want to red reduce kind of infer risk um you know so reduce this kind of centralization that we've seen you know sorry you know yeah I know you know infura does great work but like come on you know your logo looks almost exactly like the Chinese character for King like how much more Central I skin you get um so know we want to

uh just give we want we want it to be Poss meaningfully possible for people to interact with the ethereum environment without needing to make those kinds of sacrifices and rely on a yes centralized third party for their um to um make make it possible to use the blockchain for them right and uh so in fur risk we want to just reduce the need to rely on centralized providers includ by giving kind of resource constrained environments so phones um browsers embedded kind of iot devices ways to independently verify the chain and like

what like clients do is they let you do that they let you verify what is the current head of the chain um what is the state of a particular accounts particular transactions without needing to verify the entire chain yourself there is kind of a light client in the current proof of work chain but it's not very uh good and so in stake we we wants to use this as an opportunity to make like CLI support much stronger and finally the merge right so all ethereum moving on to the proof of stake chain and we've I think in ethereum have been strong Believers of proof of stake for a long time and so you know we believe in proof of stake because of its

better Energy Efficiency proof of work just wastes a huge amount of energy uh proof of stake increases security so a lot of U good things I think come out of ethereum just moving on to being a pure proof of stake system so all important goals um and these are kind like I think the main things that we want to be focusing on right so medium-term going scalability road map so basically right now we're over here on the left so 15 to 45 transactions per second just of the ethereum layer one um 1,000 to 4,000 transactions per second if we add

rollups so there's been a huge amount of work on rollups this year this big scalability solution um rollups for simple payments and rollups for exchange are already here so looping diversify ZK sync um all three of them have been actively running on mainnet for some time uh zync has been used inside of gitcoin Grants so a lot of good work for payments and for General evm applications um optimism has been pushing hard they're in their very final stage stage of their test net uh get getting ready for to kind of start their um main net launch process so rollup scaling for General evm applications is

also here so this gets us to 1,000 to 4,000 and then going even higher up to 100,000 transactions per second rollups together with so kind of stacking on top of um E2 sharting and so a lot of scaling is going to happen over the next couple of years major progress so here is kind of the road map in a simple form right so you know the beacon chain the the ethereum proof of steak chain has some launched uh so a huge amount of progress there but there's still all of these pieces left to go right so we wants to add like client support So doing that fairly soon um we want more work on rollups so not just rollups for for payments also rollups

for General evm contracts want all these rollups to be much more and of decentralized them than they are right now um on then we want to see Shing um and that's something that there's a lot of work going on to right now now then on the existing uh ethereum CH the proof on the proof of work chain side right there is the Berlin heart Fork that's coming up fairly soon then uh EIP 15 59 this big economic upgrade um that a lot of people are excited about a lot of progress there there's a test Network there are implementations there was an amazing report by Tim Ruff Garden doing an economic review of it and then

there's eth 1.x right where eth 1.x is do work on just making the existing ethereum design more scalable so both of these things happen and then after that the final goal here is to kind of combine together E2 and the proof ex ethereum proof of workchain and basically take all of these existing ethereum applications and just move them over onto eth to right so for application developers everything will continue on look like almost exactly as I did before except instead of running on proof of work it will entirely run on proof of stake this is my hope for the end of 2021 so have all of these pieces

all of these pieces in the middle finished um it's unlikely that we'll get um the last two uh finished within 2021 but we hope to see test nets for both uh so you know I think if the ecosystem works hard then we can get most of kind of the rest of the rest of this process done and if we do that then you know it'll be great we'll have much more scaling we will have um know light clients um potentially the light client support for the proof of stake chain even before the merge could be used to provide better light client support for the proof of workchain which I think is very interesting um more scaling um better e better economics for sending

transactions uh so a lot of kind of good technical things that kind of set the stage to make the things that the exciting things that we hope to see happen at the application layer um actually possible right so etherum Community is growing a demand for ethereum is growing a lot but we need to scale in order to meet that demand and this is happening um I on Twitter I published a much more complicated road map so if you wants to look go look at that you're very welcome to so finally kind of how can you help right I think there's a lot of ways to kind of help in this transition uh so a lot of things I

think are going to be happening in in 2021 um and a lot of different kind of pieces coming into place a lot of things happening oh just over the course of a year in sequence of in parallel so on the proof of Stak side just run a validator know if you have 32 e just participate uh in the Pro stake system yourself um staking pools so work if you have less than 32 eth um you can join a staking pool there's a lot of interesting work happening on decentralized out of trust minimized staking gos right now so help helping to improve the user experience for solo

stake um for solo staking as well just generally a lot of work to do to make it easy for people to stake regardless of whether they're staking by themselves or whether they are part of a pool um rollups so rollups are this I think really important short-term scaling solution so you know Shoring will be ready in a year or or two years but we have scaling challenges now but rollups are almost here or for payments they're already here now but the the technology being here does not mean that adoption is here right and so we still need to make uh deps and wallets roll up compatible if you accept payments accept not or accept

e um not and erc20 is not just on the base chain but also inside the major layer to so inside luing inside ZK sync um other EQ things participate in the research Community there's a lot of uh activity happening there um participate in the E2 spec design Community um some less technical things that are important also just kind of translating and interpreting the existing knowledge and materials to different audiences and just helping the broader public understand what's going on so there's a lot of people whether they're just in the public or whether they're like Banks or companies or just

different um you know entities and potentially government um that are looking to just understand what's going on in the blockchain ecosystem and they might even be uh building applications but they need to understand like what is going on in terms of scalability and what options exist if they want to build an application that processes more than um you know a few transactions a day right so rollups exist shorting exists um just help people understand you know what what is going on in those areas and like how they can take advantage of both of those

things and just like do what you can to to support adoption of the scalability technologist among those already interested in crypto right so scaling is just going to be ex extremely important and it's both happening next year and it's going to be very necessary next year uh and so making sure that we actually get the benefits of scaling in time before um you know the crypto space blows up by another factor of 10 and kind of transaction fees without scaling and start going up to like the tens or hundreds of dollars and then it becomes much more difficult for people to do anything right but if we help get people into rollups if we help people kind of

under set up their applications so that when shorting comes they will be able to take advantage of it then we can I think solve all of those problems so let's do that thank thank you thank you so much uh Vali for your talk that was uh that was amazing and and an enormous privilege to have you here uh in in through MBA so thank you very much um we know you you are kind of tight with time so we have like two or three questions uh if you're up for them um I will hit you with the first one so we have a question here by

Santiago his question is when if 1.0 becomes a Shard will it stay as a proof of work chain processing transactions for e 2.0 or will it switch to uh proof of State as well you're muted oh I I can I I just realized I can unmute myself sorry I just I muted myself out of habit just because I do that by default and then I forgot that I can could unmute myself um no so this good question very important as so proof after the merge proof of work is going

away completely right so what's it's the if it's not the ethereum chain that's becoming a Shard it's ethereum applications that are moving over and being kind of and becoming part of the pro of stake chain right so basically the state the ethereum state so kind of account balances a smart contract code smart contract storage all of those things are being moved over all up once from the proof of work chain to the ab the the the proof of stake chain um I mean whether it'll be a Shard or whether it will just live on the beacon chain not decided we're actually leaning toward putting the eth one directly on

the on the beacon chain at least at the beginning um so it will be the pro the proof of stake system that will then uh process transactions and um that will basically manage kind of updating the updating the state of the ethereum system so no more no more proof of work um you know no more um um actually a lot of the a lot of things um on the eth onean side are go um that are kind of very high level are are going to end up going away and being replaced by e two things awesome thank you thank you very much um we'll go with another question by Mariano who's a a great guy that we

we all know here in at the Meetup uh his question is all efforts poured into EIP 1559 and account obstruction are they going to be available on eth 1.0 or will be launch on eth 2.0 I'm hoping that there'll be the that they will be available on the on eth one even before the merge and then when the merge happens everything that was available before the merge will of course carry over um with a couple of exceptions so like for for example um if you use the block hash op code for Randomness then that will then then that'll stop working in group of stake but like with VAR small exceptions um everything everything that's available

on E one will just continue working on eth two and yeah e559 I think will very likely be available significantly before the merge um so we're hoping uh it will be on mainnet next year um and like not and not next year as in December like next year as in much much earlier than that um and the count abstraction as well um though with account abstraction there's still kind of more choices in terms of exactly which um which approach we'll take okay excellent uh are you up for one more question sure cool so uh we have another question here uh which is um could you tell us

more about polinomial commitments Zaki star evm and CBC Casper sure so I can briefly kind of go back go go through all of those for about one minute each so so polinomial commitments right so Pol pols are an extremely important kind of mathematical piece in all of these in a lot of these fancy cryptographic protocols so including ziki snarks including um some of our data availability stuff and the reason why they're powerful is because a polinomial is a mathematical object that can store a huge amount of information right so polinomial is like you know 5 * x to the 243 plus 11 time x 242 plus blah blah blah right so there's a huge

amount of information encoded in a polinomial but the you C but you can describe relationships between all of all of those pieces of data using polinomial equations right so if I say you know a plus b equals c where a B and C are all pols what you're really saying is that like hundreds of numbers are all equal to each other the a the the X to the 10 coefficient of a plus that plus the X to the 10 coefficient of b equals the X the 10 coefficient of C same with every other coefficient right so mathematicians or cryptographers love pols because they let you express huge amounts of data in one single thing um and polinomial commitments are you can

think of it as being kind of like a hash of a polinomial right so you can use a hash to take a big piece of data and kind of represent it with a small piece of data except they have the property that polinomial commitments let you do math right so they they let you check that like for example a plus b equals c or a times b equal C and they let you check some other things and so polinomial commit basically polinomial commitments like you can use you an equation with all you can make have an equation checking polinomial commitments and that ends up representing an equation that simultaneously checks huge amounts of data right because pols are huge amounts of data and that basically

is kind of the key to the magic for why things like Zeki snarks are actually able to um you know be these very short and VAR simple proofs that c that that can check these very complicated statements that involve these really um really large computations and why you can verify these proofs much faster than doing the computation itself so I don't know I that's like about as kind of accurate or simultaneously kind of advanced and and understandable as I can be in one minute um but you know very cool math um so ZK snarking the blockchain I so obviously I mean ZK snarks and zero knowledge proofs

um have been um making huge progress over the last year uh so and I wrote a couple of blog posts like on plun and some of the recent protocols uh so the um are very they've been very powerful recently um become and the most recent thing that's happening is it's BEC people are starting to see if to uh make uh ziki snarks or Z Starks to zero knowledge proof just general purpose computation right so I know the Labs team has been starting to work on this the starkware team has been working on this and for a long time it was thought to be

impractical because it's too inefficient but those assumptions I think over the last half year are really being kind of overturned and challenged and so I think within a couple of years it will be possible to just make a general report proofs of evm computation and that will just be very powerful in terms of making like things like charting possible um so excited about that um CBC Casper basically just switching over from Casper FFG to a different K sensus algorithm it has some interesting properties so one of those interesting prop properties is that like it I mean first of all it's just fundamentally

simpler and it's likely to be more secure because it's simpler because FFG has these two separate pieces right where one piece is the lmd ghost that does a kind of block block Fork choice so if there is one block and then another block that both point to the same parent then they you know how do you choose which one um figure out which one to choose um and then there's a Casper FFG which finalizes blocks and these are two separate systems kind of and the interaction between those systems is complicated right so Casper CBC does not have these kind of two separate systems that uses LM lmd ghost for everything and so it is potentially much uh simpler and more secure it also has this interesting property that there

is no hard two-thirds the threshold right so like for example even if only 55% of participants are online it will still be able to get a degree a small level of finality where you need 5 perents to get slashed where and 5% getting slashed is still a huge amount of money and being lost by an attacker right um in order to kind of in order to revert that chain so in instead of having this kind of hard threshold between is it finalizing or is it not finalizing Casper CBC just kind of makes it much more fluid and kind of allows for a much more kind of gradual transition between the two and it just generally performs much better in the case where a larger portion of the

participating nodes are offline fantastic thank you thank you very much for that thorough explanation in in in a couple of minutes uh thanks a lot we're out of we we we out of time we we don't want to take more of your time uh vitalic so um I just want to like ask you one more thing like is there any anything you would like to to add before before we let you go just you know whatever it comes to to your mind no I mean I I I feel like I kind of went through a lot of that in uh you know like my last slide right so just kind of putting that back back back on very briefly um you know there's a lot

of different things happening um in the ecosystem uh there's a lot of progress in the ecosystem and there's a lot of progress on the technology and they're both kind of needed like the the technology is needed to make the ecosystem possible and the ecosystem is needed to make the technology useful uh so they're very valuable and there's a lot of things that you can do to help and and even if you're a non-technical person there's a lot of things that you can do to of help bring other people um into the ecosystem um to kind of help different ethereum applications um so there's just a lot to be done and I'm excited that all of you can be part of

the journey thank you very much again that uh that was amazing Bal so again thanks a lot for for being today with us for taking this time we uh we are honored and super happy so uh now we'll take a a little break of five minutes if you want to stay you're welcome to stay if you if you have to leave just go ahead and and and leave and then I'll switch uh to Spanish now so thank you thank you very much again thank you

price you know