Year in Ethereum | Merge & Scalability Summit

[Music] [Applause] [Music] thank you cardrick for uh what i'm sure was a great introduction um thank you everyone for being here i'm very uh pleased to be speaking at youth online this year my presentation is titled uh the year in ethereum uh two underrated developments um let's let's jump in uh so to start off um hi i'm josh um i work for the ethereum foundation and i've worked in the ethereum space for a while now uh previously i founded each global uh the company putting on this uh conference and also a company called l4 if you want to follow me i tweet at uh the address you see on your screen and you can also find my writing at uh stark.mirror.xyz so i each year um myself and my friend and colleague evan van ness uh write this blog post called the year in ethereum we've been doing it since 2018. and this year i've been thinking about what's going to be in the 2021 edition and i want to kind of share some of my thinking about this and in particular what i want to talk about are two big developments and two underrated things about them so as we'll see in a second some of the things that happen this year are very obvious but i want to kind of dig into what it might be not obvious about them the things that might be underrated or not appreciated and i want to try and convince you that these things really matter as much as anything else that that you see uh in your twitter news feed so let's jump into it number one the year of the nft i'm sure this is a surprise to everyone watching this that i rate this as an important development uh this year uh this is a pretty amazing graph right um uh far on the far left here we have the you know kind of a brief heyday of cryptokitties um which was uh a project in q4 2017 um and and you know since then not a lot of interest in fts at least in terms of the number of active you know people buying and selling them on ethereum or the you know market value of what was being traded uh a year ago today um it wasn't much different than it had been for the preceding two years a little bit of activity but i mean just comparing with you know the recent peaks uh it's really been an astonishing story of growth in terms of users and the value of the assets being traded now of course nfts are are not just on ethereum but ethereum does command most the market share at least in terms of the value of the assets and the amount of you know money changing hands um and then if these of course also expanded uh you know pretty substantially into other categories as well there's a a game called axe infinity built by a developer in vietnam and when you kind of rank the the the value of the their assets um uh that make up the accident infinity protocol this you know compares to some of the largest gaming companies in the world and the last like graph i want to share here is just the you know the comparison of the nft search interest for search results for cryptocurrency it's incredible to think i mean imagine telling someone in 2015 that you know in a few years um an application uh that will be built on ethereum a project that no one believed would ever ship or and even if it did that it would ever work would have a higher amount of search and trust than the entire category of cryptocurrency it's a pretty astonishing change and that should make us humble about what might happen over the next five years or even beyond now if you ask someone what an ft is they're probably going to say one of these things today you have profile pictures you have art you have ens names you have game items you have you know blog posts or crowdfunding on mirror there's a pretty wide variety but remember that nfts are just you know things that conform to a standard you can attach any kind of digital file or asset or anything to an nft token and we're really only beginning to explore what that might look like the other you know component of the story this year is the amount of mainstream interest um from celebrities and and big brands and institutions like christie's it's exciting and validating to you know have something that once was such a small niche thing as a crypto punk you know on display in christie's uh on jay-z's profile picture in times square it's pretty incredible right it's all fine um what i'll say about this is you know celebrity interest and interest from big brands it comes with a cycle um we also saw this you know in different ways in previous cycles where crypto is cool when the numbers go up and not cool when the numbers go down um so while it's validating to have a lot of celebrity interest or whatever it's not the most important thing um it's not the thing that really matters or that when we look back in 10 years we'll say really mattered so what does matter um here's the thing that i think is underrated about all of this uh the most important thing to me about the nft boom over the last year uh is this a new kind of person can now make a living on ethereum by make a living i mean you know make money support themselves build a life build a career out of um that's incredibly exciting so what do i mean by this one way of telling the story of crypto is through the expanding circle of people who can make a living from it from 2010 2013 uh if you were making a living on crypto it probably meant you were an investor you bought bitcoin the price went up you held good for you uh you made a living and maybe much more than that from 2014 to 2017 in this kind of next era you have you know notice investors uh but developers and also people that work for crypto companies so you know early companies like coinbase are coming online they're hiring people individual developers are launching altcoins or other layer one protocols we're seeing the first icos like ethereum it's you know another way to make a living and maybe much more for some of the people on this list from 2018 to 2020 things don't change all that much the details change right if you're an investor you have new investment products you have new ways of making money as defy comes online and so forth if you're a developer you have new kinds of applications that you can build and make money from and there's lots more crypto companies coming online and a greater variety but it's still like you need to be one of these types of people in order to make a living from ethereum but this year all that changed this year now for the first time you know people who are creatives people who are artists musicians writers and also gamers can make a living on ethereum um that's an incredible zero to one moment that happened this year and is not going to stop happening anytime soon now to earn a living using ethereum uh you don't need pre-existing capital to invest which a lot or even i would say most people in the world do not have uh you don't need to be technical which most people in the world are not you don't need to be into finance or investing or understand what a derivative is again because most people don't and you don't have to work for a crypto startup because most people can't or won't or don't have access to the opportunity all you need to make a living on ethereum now are three things you need to create something that has a digital form you need to build a community of people who value what you create and you need to use ethereum that formula gives access to uh you know making a living to to earn income to a huge proportion of people that did not have it previously and that to me is is the most exciting thing uh that about the nft boom of this year and remember this is not going to stop nfts are a blank canvas and we've barely explored that canvas if we can imagine you know a hypothetical graph of you know everything you can do with nfts it might look like this you know the the space of all possible nfts we've really only scratched the surface the the very kind of first smallest corner of what can be done here at a certain point nft stop being a you know an area of technical innovation and become one of social innovation what do people value what can they coordinate around and that design space is is absolutely huge and it's only going to continue to expand as more and more people get into this more and more people realize that they can earn an income from it so that's the thing that i think really matters about nfts this year so that's the first theme let's move on to the second the year ethereum scaled uh and this is not the year that ethereum finished scaling but i think it is the year that it really earnestly began to happen for the end users perspective um in 2018 i wrote a blog post called making sense of ethereum's layer two scaling solutions and at the beginning of that blog post right at the top i wrote the following um for ethereum 2018 is the year of infrastructure this was going to be the year when early adoption will test the limits of the network renewing focus on technologies built to scale the network now i didn't include a hard prediction in that because i didn't want to be held accountable to it but i do remember that i was very optimistic when i wrote this that within a year maybe two state channels and plasma would be live and being used to scale the world of applications on ethereum um state channels is live but it's got a niche interest or kind of a niche use case and plasma as we'll see kind of evolve into something else a few years went by not quite this many but let's call it around three and here we are uh things are already happening at more than three billion dollars is locked into layer 2 protocols where users are using those assets in a way that is performant but still secured by the core ethereum protocol security layer just to collect a few examples there's lots of layer two solutions out there but a few things that happened this year alone arbitrary has been on mainnet for a while but it opened up to everybody in august optimism is on mainnet since august but still permission zk sync has been on testnet since may 31 and starkx is used by a bunch of different projects including dydx and mutablex so rare all of which went live on mainnet this year and diversify which went live last year there's a lot more to learn about layer 2 protocols and all the various projects that are out there right now i do recommend l2b.com as an incredible resource to learn about them so the other thing i wrote in this blog post is that something that held up pretty well i think that we should expect that there will be new and unexpected layer two protocols that improve on the existing models we have today or make new trade-offs between the different competing interests and this turned out to be right on if we look at all of the projects listed on l2b.

com basically everything on this list the technology didn't exist three years ago plasma kind of you know did not turn out because there were real limitations that approach but it spawned the world of roll-ups which turned out to actually work and so if we remove the two plaza projects from this list everything you see here did not exist the technology did not exist when i wrote that blog post it's a real testament to both the strategy of the ethereum community of having a you know programmable layer that can be the basis for many different types of scaling solutions and also the incredible community of people who have been researching and innovating and taking risks and building companies and building these protocols over only a few short years so that's all obvious it's all great it's very exciting what's underrated about it what i think is underrated about this is that as these projects launch and we gain experience from them in production the long-term architecture of the greater ethereum ecosystem is coming into clearer view um and it's like we've kind of been on this journey and we reached the top of a hill and now we can see more clearly where we're going the journey is not over but now we have a vantage point where the landscape is a lot more clear a lot more obvious to us so i think the future looks like this the future is modular with ethereum at the core and i want to kind of run through what that modular future might look like and i should first caveat that this is not an official roadmap of any kind this is not the official view of the ethereum foundation this is my synthesis and summary of what i'm seeing in the ecosystem and also my adaptation of you know what i've learned from others writing and research two particular sources i should mention are of course the roll-up centric ethereum roadmap blog post from last year by vitalik and also more recently a whole bunch of blog posts from a writer named e paulinha on twitter who's a pseudonymous writer who's really produced some excellent stuff that has helped me see the future more clearly and help many others as well um so both great contributions from community members uh right here i strongly recommend uh you read both of them and special thanks to you pauline also for giving us some feedback on this presentation i really appreciate your time and what you've done for the community so let's talk about what the modular future looks like and let's begin with a simple model of ethereum and i want to kind of run through the the basics that makes this all possible and why we end up where we think we're going to end up ethereum has two basic functions it executes things and it secures things it has an execution function and a security function that are you know very intertwined but we can distinguish between them the execution function is um what processes your transaction right ethereum runs smart contracts the ethereum virtual machine or the evm you know takes transactions update state great this is the layer that apps and users interact with if you send a transaction it's you know handled by execution the other part of ethereum is the security layer um or the security function uh ethereum of course maintains consensus over the state of the evm and proof of work and proof of stake ensure that there's an extremely high cost to trying to mess with that state so you can be you know quite assured that once your transaction is finalized it's going to stay that way apps and users might not interact with this layer in a direct sense but without this layer of course there's not much point to the execution layer because it wouldn't be secure so here's our little you know box representing ethereum and here it is performing these two functions execution and security the problem is this simple model doesn't scale as we've learned you know we want execution to be fast and cheap because it needs to scale to the demands of the 8 billion humans who are eventually going to want to use it but there's a trade-off if we try and make ethereum fast and cheap in a kind of naive way it becomes less secure and without that security i mean we just can't compromise on it otherwise the execution layer is no better than a spreadsheet or a database held by some group of people who we don't necessarily trust for ethereum to be trustless to centralize all these things it needs to be secure so what's the solution the the outline the solution we've known for a long time which is that we need to modularize we need to have kind of separate environments that handle different parts of this need and you know this is the insight behind the layer two protocols i talked about in that blog post from three years ago and it's the kind of basic function of roll-ups and really what i'll be talking about here today is roll-ups because they are the dominant uh scaling solution that are now being implemented and used so the basic idea is this we have ethereum at the bottom and we have this other environment that is doing execution now ethereum still does execution uh in the sense that of course the evm still exists but where the app and the user like what they're interacting with where all the actions happening it's happening on this other environment and so we're going to kind of you know hide this box um just for simplicity's sake so the idea is we create these modular components that can still preserve the security guarantees we care about how is this possible um it works like this in a simple way um here's the execution layer and let's imagine that you want to use unit swap and so in this case uniswap is going to run inside um the you know this layer two let's roll up um and instead of just you know doing everything on main chain we're just going to store some data on the theory of mainnet we're going to take the information about the trades you're making the outcomes of those trades the state updates and so forth we're going to roll it up and kind of compress the data and put it onto ethereum maintenance the data we're putting there has two functions there's kind of two components to it one on the left is data that proves the validity of the transactions right a kind of a receipt that sure that shows somebody or can be shown to somebody to say like no the transactions that happened off of ethereum their valid transactions um you know they followed the right rules for updating state and then there's also just the raw transaction data itself of you know these state updates and what changes were made so let's talk about the thing on the left because this is at the core of the security that is being inherited here the reason that layer 2 can be secure is because of what's happening here these proofs of your transactions are what lets these transactions inherit security from ethereum and there's two ways that that works um one way is uh the optimistic way uh optimistic roll ups function as follows you are always able to prove later that your transaction is valid you don't run the proof every time we record state to the ethereum main net but we preserve the option of doing so later and so if there were a problem a challenge or something went wrong you're always able to go back to that data produce a proof and say nope uh i'm fine i can show you that you know this transaction actually did happen that it followed all the rules so you know that money that bob sent to alice does actually belong to alice now the other way to do it is a zk roll-up uh and the very simple description here is that we're using uh special zk cryptography moon math as sometimes called to create a proof that transaction was valid and this is still scalable or saving us time because this proof is much much smaller and cheaper than just doing the whole thing on ethereum in the first place so we're still getting scalability gains from this so that's the the proof side of the equation the other part of this is uh the raw transaction data so what's going on there um we need this for a few different reasons excuse my slides the raw transaction data has to be stored and available somewhere for a few different reasons users need to know what data is in the block right if you are using unislop on layer 2 you need to be able to go check somewhere you know what was the update what was the outcome of my trade you know where is this money now you need the you know the information about what has been happening to be available somewhere and you need to be able to rely on that the other nodes that might participate in the roll-up network to you know produce the role of blocks also need that data to produce their own blocks and optimistic roll-ups need the data in order to produce the fraud proofs that would actually secure your transaction um this brings us to the third function of blockchain infrastructure we've talked so far about execution and security but we also need data availability uh and data availability is basically the function or the layer that stores transaction data and it's needed for all these reasons that i already talked about for security uh to enable decentralized rule ups where there are kind of multiple entities that produce blocks and also just for users to know like what's going on we didn't have to kind of talk about it especially before because in this simple version of ethereum ethereum already provides for the kind of data availability that's needed because you know full nodes store this information so it's able to be checked by wallets by block explorers so users can get access to it and so that other nodes can produce blocks but once we're in this environment of having modular components we need data availability that scales a bit further and so there are some new requirements so this is the third thing that our blockchain infrastructure needs to provide so three things execution security data availability or you know esd for short now as we kind of get here um this is really kind of giving us a better picture now of what the future might look like and what the different components of a modular blockchain world will look like now of course we have this kind of version of you know here's what ethereum does and here's what layer 2 does etc but we can also just think of it more abstractly um i mean forget ethereum for a second forget layer 2 forget everything you know these are the three things that users need um to have you know useful blockchain infrastructure they need a security layer they need an execution layer and they need data availability and there's different ways that we could you know build an infrastructure around these things to enable them all to work in concert and when we think about what the future looks like we can think about well what projects or technologies are best placed to be the most successful solution for the different components of this world so let's talk about that a security layer we probably don't need too many of these um and the reason is that well it's actually better and more secure to have a small number of security layers that have a lot of pooled security instead of a whole bunch of separate security layers that are kind of carving up the available security in terms of you know assets and economic security so right now really only ethereum and bitcoin could play this role because they are just you know orders of magnitude uh more secure than any other level one blockchain um and bitcoin can't do this stuff uh bitcoin just cannot run the zk proof's needed for zika roll-ups it just cannot run the contracts needed for optimistic roll-ups and so it just can't provide this function but ethereum can so probably the security layer for uh you know the world's blockchain infrastructure is going to be ethereum execution layers we can have lots of those um there might even you know what we're seeing in the market now is there's lots of competition between different projects as they compete on you know how they use the security of ethereum whether they use it at all the the execution environments the technology available to developers all sorts of things might allow these to compete with each other and we are seeing that in the market now now some of these will be layer twos meaning that they will actually you know use the security of ethereum as a roll-up and others might stay separate or independent um for you know whatever reason but they're still serving a purpose for the broader ecosystem to enable users to you know use these applications uh and now we're gonna fill in some of these names you know this is what we see today there's a whole bunch of layer two projects that are using ethereum for security and then we also have projects that might not use it and are either side chains or separate l1s or something else there's still bridges between all these things so you can move assets onto them but they're not using the security of it so here's you know what the kind of uh execution environment landscape looks like today and it could grow to include many more projects or consolidate in some way in the future we don't know what that looks like because we haven't got there yet then of course there's data availability and here too will probably have some competition one of those data availability layers will likely be ethereum eventually the plan is to add to ethereum data shards that will enable a like really like a large amount of data availability that is secured by ethereum to be used by rollups and others and then there will probably also be other data available layers that are you know separate and able to be used by anyone and there already are assemblies on the market right now and so then we'll have this right well some of these projects that are layered two projects will use data availability on ethereum others will use a separate solution uh some might even let users choose which provider they want to use or change between them as as needed by you know what the user wants or what the application needs and maybe someday these separate layer one execution environments might decide to use ethereum as a security layer it is you know if it continues to be far more secure than their own security they're able to provide then it might make sense to specialize on being an execution layer and leave the crypto economic security piece to ethereum and make it easier for your users to you know benefit from that and all that comes with it so this is you know what you know link might see that this is what's in the valley ahead of us uh or it's at least one potential version of the future uh and at least for me like thinking about this and seeing projects actually launch and what they're actually doing and and what solutions they're actually using makes this picture a lot more clear it kind of gives us an idea of what the kind of basic components are that the world's blockchain infrastructure needs to provide and how these all these projects might relate and i think this will become increasingly clear to more of the community and more of the market and projects will start to see themselves as fitting into a certain piece of this landscape right now there's just all these different you know blockchains layer 1 side chains l2 whatever and i think most people don't really understand how they interact they're all just kind of separate little islands that make different claims and have different relationships but no one can really make sense of how it all fits together and i think something like this um is becoming more clear and it will make more sense so you know this is the the beginning of you know what might be considered the greater ethereum ecosystem of you know how these different protocols uh will interoperate how they will support each other how they will collaborate as you know one large community as opposed to many separate communities so that's what i think uh is the underrated thing about these rollouts uh not only is progress happening but it's progress that lets us see further uh into the future than we could before i want to remind you of course this is not an official roadmap this is not the official view of the ethereum foundation this is my synthesis and summary of mostly other people's ideas predictions and guesswork so take it seriously but perhaps not literally thank you so much for your time i really enjoyed being here today giving this talk and i hope you enjoy [Music] [Applause] [Music] you