Matter labs Mini-summit | Full Session

hello everyone welcome to the first ever ZK mini Summit event we are so happy to have all of you here and to give you some background ZK mini Summit came from a few brainstorming sessions with the idea of how we can give back to the ZK Community today for like 3 hours we are going to have different topics with thought leaders of the space but to set you up for Success we are going to start with a presentation or like a book reading of zero knowledge proofs explaining like I'm five and for you to follow along under your seats you will find a bag and in the bag there is a book yep that's it Dustin you got it so I would encourage you to open the book and get ready for the first session let's invite to the stage Marchin who's the chief protocol architect at matter labs and who's the author of the [Music] [Applause] book okay thank whoa thank you folks uh so yes as Al mentioned please check under your book this book started a little bit as a joke when I was trying to explain zero knowledge proofs to one of the co-workers let him be unnamed for now and something started as a joke very quickly actually turned into like hey maybe we should make a real thing out of it and that's how this book is created so as the title says this is Zer knowledge proof exp explain me like I'm five so it will give you some Basics to start but you're going to need to learn a little bit more before you can actually create a real circuit so just to get the expectations right having said that over next couple minutes we're going to go through the book I'll try to do the Reading part there is not much text in there it's most for mostly for children so it's it's mostly focused on images at the same time there are some questions please do not hesitate to ask I think we going to have at least one or two mics and also there will be also questions aail after afterwards so without further Ado let's get into the book and let's explain you zero knowledge like I'm [Applause] [Music] [Applause] five so imagine Alby so forget about Z knowledge forget about blockchains for a second just let's go back children let's go imagine a zoo it has many animals for Zoo knowledge we actually called a zoo a circuit and the animals are Gates now imagine we also had imagine we also had boxes basically three types of boxes there are boxes that have a hidden number there's a hidden number in a wooden box these we call the variables we also have glass boxes they also hold the number but for glass boxes you can actually see what's inside we call them public inputs and then we have stone numbers these don't change these are clearly visible you cannot actually modify what's in there and these we call constants now we had a set of animals and now each animal has a special power animals can look inside the closed boxes and then depending on the type of an animal they can be happy or sad for example we have equal zebra so equal zebra is happy when two boxes are have the same value now if they don't have the same value equal zebra is sad now the important thing here is you don't know what's inside the box but zebra does so animals can also look at the glass boxes they can also look at the stone numbers and as I mentioned each type of the animal has a different mathematical expression tied to it so we have not greater than crocodile and the crocodile is happy when the right number is bigger than the left number and this is where the first part of zero knowledge comes in is you don't know what's the number in the bottom box but you know that the bottom crocodile is sad so whatever that number is it must be smaller than five now the important thing to remember is you don't see the values but crocodile does now we can have far more complex animals let's say the the spider can actually look at four objects at the same time is the top it's the if spider if the top top box is zero it would like to have the left and the bottom box to be equal if the top box is one it wants the right and the bottom box to be equal so look the spider is happy so here's the question to all of you adults in the room can you guess what number it's in the top box yes no okay so in this case we call spider selection gate now there are PL the zoo has a lot of animals right we have beavers for division squirrels for modules cats for additions and many many more and we can put all of these animals into more complex setups basically creating complex circuits if all animals are happy this means that the circuit is fulfilled the thing is that we don't need to see the vales in the boxes but we can just look at the animals and what we say is if they are happy it's good and we we can almost take a picture of the animals remember the picture will not reveal what's inside the boxes it will show you what's in the glass box it will show you what's in the stone numbers it will show you the happiness of animals it will not show you the contents and we can call this photo of all these happy animals a proof so basically proof becomes a photo but enough talking let's actually build a circuit to solve the real life 5-year-old problem how to make sure that my sister split the candy equally so here's the problem your mother left some candy your sister decided to split it between you and herself how to make sure that she did a good job so here's the a simple circuit that takes care of this so mom's candy is on the left you don't know how much she left to you but what you know is that there is a squirrel on top there's the modulo to there's the beaver at the bottom there's a division by two there's some sister candy that you don't know there's also the caterpillar ad on top so if the circuit is correct if all animals were happy then whatever values you got should be correct but let's take a look if we actually zoom in imagine your sister put numbers in these boxes we don't know what numbers she put but the one that I'm getting is not correct the beaver is sad she did something wrong actually if you spend a couple more minutes for adults in this room spend couple more minutes you can actually see that there's a small problem with this but I'll leave it as a homework this is actually means that sister gave you more than her but anyway you you can take a look and try to prove it to yourself as a homework but now if we go to the next slide this is the positive outcome again we don't know how much candy mom left we don't know how much candy sister got but if all animals are happy this means that the two candy that we ended up with is the right value now in practice you probably heard about different ZK systems now the way you can think about them is like different animals you've probably heard about Cluny about Halo think about them same concept animals Gates whatnot they are just different just different types and there are lot of different provs cuz remember the proof is the photo of happy animals and there are different cameras that you can use to take the photo you probably heard about snarks Starks fries KZ GS that's where it comes in that's the technology that's used to take the picture so next time someone is telling you oh I have a snark with plony you're like oh sure so you use this camera with these animals got it next so now that we know the basics let's talk a little bit about how these circus are created so you can imagine that there are zookeepers so zookeepers they create new animals let's not get into details how they do that let's say they just create new animals and they create the scenarios they figure out okay these animals with these boxes now now we call this these zookeepers circuit Builders or crypto people and there are also visitors so once circuits are created the visitors arrive and they can actually put numbers in the boxes so if you're a visitor you're coming in you're putting a number in each box and then you take a picture of the happy animals if all animals are happy this means that the values that you put in the boxes were correct again you don't have to reveal to to anyone what values you put animals are happy values are correct and the only thing that shows up on the photo are all the happy animals and the public inputs and this picture is what we're calling a proof now there can be mistakes right a very common mistake is sorry there can be a mistake that zookeepers do for example imagine they put too many animals this is what actually called over constrained and in this scenario you cannot come up with the numbers that make all animals happy at the same time like look at this example what kind of values can you put here to make both crocodiles happy answer you cannot this is called over constraints and you will not be able to generate a proof for such invalid C circuit there are also other bugs like a missing animal if you want to learn to earn a million bucks find a missing animal I am not joking the examples of different bounty hunting for ZK proofs are all about finding the missing animal that if it's not there you cannot act you can actually put fake values in the boxes and all animals will be still happy because again if one is missing you will not be able to to get that right so this is a very bad this is a very bad scenario and we call this under constraint and in systems it can lead to invalid State transition being accepted invalid State being committed to blockchains so when you try to explain to your kids what do you do at work is I'm looking for missing animals honey now real circuits have thousands upon thousands of animals so looking at the picture to see all of them happy or sad would be very slow that's why we have Zeke Zeke is a special animal Zeke can actually look at the pictures of other animals and be happy if all of them are happy and be unhappy if at least one of them is unhappy so we call Zeke recursive proofs or aggregate proofs has a special power so remember it can look at to at a picture with lots and lots of animals and just be happy or unhappy depending on their whole state this means but now the tricky thing is that Zeke one Zeke might not be enough because can look only at thousands but with millions of animals and huge Gates we can actually have zeks looking at other zeks meaning we can have zeks can look at the picture of themselves so we have zeks looking at zeks looking at animals and that's what we start getting into recursive proofs in practice for systems like ZK sync we can have Zeke looking at Zeke looking at zek at animals to like four levels out of curiosity so that's that's why at the end we have one zek to prove them all one zek on top of the circuit to prove that if he is smiling then all the zigs below must be smiling and all the animals at the very bottom must be smiling and this means that the whole state transition all these animals that we had at the very beginning all these small transitions they must all be correct so this checking that Zeke is that the last Zeke is smiling is what we call proof verification and this is what's done on ethereum this way the only thing etherum has to do is to look at one Zeke and that's it so if you want to learn more if you go to this link you can actually see a web page with both the book and interactive quizzes so you'll find an animals in there and you can put things in the boxes see if they are happy or sad and try to solve some of them so with this I want to thank you all thanks for coming I hope this was useful I hope that it's as I said it's a little bit different approach to how zero proofs proves work you're not an expert yet but please wait for two more books Eli 9 and Eli 15 and then you will be there so thank [Applause] you yeah we're close enough hey everybody my name is Nick I'm here to present as Al be introduced uh dumbass crazy ideas for onchain daps right so little intro about myself my name is Nick I'm a software engineer for matter laabs uh here's my socials and website down there I am relatively new to web 3 about six or seven months in and I'll sort of talk about that as I go through my presentation today for those who want to follow along this is a QR code to download these slides we're also going to post these online available after the presentation uh but if you want to follow along now feel free to scan this leave it up for a few seconds cool so what am I going to be going through today I'm going to talk about why did I do this presentation and then I'm going to go through some of these ideas at the very end I'm going to wrap it up with some takeaways that hopefully we can all walk away with so why did I do this presentation right kind of goes back to I had first joined matter laabs and I had $100 worth of eth I wanted to figure out a way to spend that money I looked for a bunch of different games that I could play on chain I am very passionate about gaming I wanted to see if I could support the local developer Community if there's like an indie game that was available I thought you know what I would love to give them my money so I started going and looking through some of the daps that are available and honestly they were not great most of them just ended up not working or if they did work were just nft shills it wasn't fun and I just wanted something more out of it so I started looking Beyond gaming I started looking at other kind of daps that existed out there they were mostly finance-based or finding ways to swap tokens or trade nfts and it really felt like there was this lack of innovation that existed it wasn't uh something that I you know really felt passionate about giving my money to and I I wanted something fun right I wanted to spend my money and I didn't know how or where to so before going a little further into the presentation I want to talk about my mental model about how I've come to think about web 3 so for me I think about web 3 three and specifically smart contracts as kind of like AWS lambdas so for those who don't know AWS lambdas are this uh serverless uh Computing functions that exist hosted by Amazon and they're super convenient and easy to use in the smart contract world we have these calls and transactions for something like an AWS Lambda what that really equates in my mind to is you know every is a get request you're getting information every transaction is a post request you are making a change to the data that exists on chain the difference here is who pays for making sure that these Services exist in the AWS Lambda world it's whoever is paying for the Amazon account your AWS account right you get builded at the end of the month smart contracts you have to pay up front you don't pay for the calls the get requests but you are paying as a user the transaction costs those post methods now there are differences right with an AWS Lambda you can invoke them with HTTP requests with events or even through a scheduler that's not the same with smart contracts you can execute them or start a a smart contract function from an HTTP request but there is no way to call a smart contract function from an event that has happened on somewhere else on chain there's no way to just start and execute a smart contract function based off of a Cron job right it needs to be initiated by the user another limitation is you don't have the full access to the resources like you would with an AWS Lambda in AWS Lambda you have access to potential gigabytes of RAM but in smart contracts we're talking about kilobytes right you have to live within these sort of restrictions of the current system right now and that sort of gets me to the the next point which is I'm I'm looking at these apps and we have this sort of uh uh constraints that are built into the system these limitations but you know from my experience as as an engineer I've always you know seen that these limitations often Foster creativity and I'm not seeing that but let me sort of talk about what do I mean by limitation Foster's creativity so there is this amazing YouTube Interview by ARS Technica they sat down with Andy Gavin who is the co-founder of naughty dog this is the team who created the uh B Crash Bandicoot series and they wanted to go through the war stories of what it was like to develop the original Crash Bandicoot for the PlayStation before the PlayStation ever even got released right this game launched at the same launch day as the PlayStation so for them they had similar restrictions and their mentality their philosophy at the beginning was no cycle of CPU no GPU or bite of memory would go unused they would find hacks they would find tricks to get to every single piece of resources that they could whether they were supposed to or not so one example of that was uh limited memory access that's something that I had just talked about with ethereum and what was the case for the PlayStation was the PlayStation only had 2 megabytes of onboard memory on the console now the disc of the game had about 640 megabytes the game itself was about 60 megabytes so most of it was like 90% empty but the problem and the constraint was really this 2 megabytes of ram because the CPU and the GPU of the PlayStation could only ever access the ram it couldn't access the data directly on disk right it was it just wasn't possible with the interfaces it was too slow it takes time to get data from the disk down to the ram so what did other companies do they lived Within These sort of boundaries they built game levels that only had had 2 megabytes worth of assets so the original PlayStation games the majority of them were very small they had to fit in 2 megabytes but you know what did Andy do and his team they basically found a way to update the entire custom memory schema where they had 64 kilobyte chunks and as you're walking through a game level it would hot swap chunks of data that you would need directly from dis so in essence by building out this hot swappable memory you now had access to much larger game designs much larger game levels and they were doing things that no other game could at the time because it forced them to right so I ask you know the DAP developers that are out there what if what if you didn't have these constraints that we have currently with ethereum what can be done so that's gets into the first crazy idea the what I'm calling z64 it is what if you could have an onchain Nintendo 64 emulator is that even possible right so where does this idea even come from where did I like come to this like yes I'm passionate about gaming but why the Nintendo 64 and one of the things that you know I had mentioned before is this idea that anytime you interact with a transaction it's coming from a user right so video games today whenever you are trying to save a game typically you walk through and it's just Autos saved as you go along right that's not how it used to be whenever you play these older games on the Nintendo 64 you had to open up the menu you had to choose save you had to click yes and you had to wait as the game is being actively saved on memory and then you wait for that transaction to complete and then you go about your day people were just used to that that is the flow that is the limitation that they were used to that is very similar to what we're seeing right now with ethereum and transactions so what can we learn if we look at the you know Nintendo 64 and the I guess the the game building model of these sort of similar interactions and restrictions so I basically started from there and it turns out it's kind of possible the first thing I realized is that there is a web assembly um emulator for the Nintendo 64 so it is called n N64 wism by the way if you go look it up on GitHub but but it allows you to fully in a browser emulate the Nintendo 64 and play games in the browser I was like all right awesome all I have to do is make a website and host that onchain with this JavaScript cool make a website with boss GG all right now that's fully on chain all right that solves that problem I now have an emulator that is fully on chain that I can pull up through the browser next problem what if I need to save the game at any point in time well it turns out to save a game is about 16 megabytes uh this is a problem because approximately that's going to cost you about 32 Grand just to save the game once now it's still feasible right this is crazy you probably wouldn't want to do this right just for one single game but it got me thinking and I started talking about this with some of my co-workers you can actually use a validium and actually if you you don't need this data to live on L1 which I don't know if you need your you know acarina of Time game save State all the way up to L1 it can just live offchain then these transactions go down to be pennies right that's the majority of this cost where it's coming from so theoretically you could have a fully onchain Nintendo 64 emulator experience all right let's go into idea number two the burner phone smart wallet so for those who are not familiar with what a burner phone is it's this uh prepaid phone that you can buy typically for cash at like a local gas station that people use and then like throw away once once it's done and it's you know has a lot of implications behind it but that's the general idea could you have that sort of experience is there I don't know what the need is you know I'm not going to speculate but could you do this for your smart wallets right and and as I start thinking about it it's really could you have prepaid gift cards for smart wallets where you know you can only put in money once and then from then on out the funds can only exit right technically it's possible with some things like account abstraction you can control with Logic the Ingress and egress of your transfers you can add business logic in there to say hey you can actually do this if you really wanted to idea number three what if your smart contract TR s changed ownership randomly now that's really really scary for some people because there is a lot of money typically associated with a smart contract and handing over an ownership of that contract means you are now exposing and handing over the ownership of that money but just leave it up to r& GES is and it's totally possible it's a terrible idea I would not recommend this at all but what if the ownership of it changed every single day a does that did you just create like a lottery system by doing this right and along the way as I was thinking about how to implement this you have to I like started learning about different ways for role-based Access right this concept of ownership can be defined in a few different ways and there are different ways of having role-based access also you would need to have some sort of scheduler right I had mentioned at the beginning like you can't just execute a uh smart contract function fully on chain based off of a schedule however after going through some of the research it turns out that there are schedulers that are available as a service it will exist offchain but they do exist and I started going down this Rabbit Hole of okay technically this is possible I don't want to do it but it's possible another idea what if there was an auto automated cat feeder based on eth tips so imagine you had a animal shelter right where there was a webcam stream pointed at a bunch of cat feeders and on this website you could donate or tip eth and that would automatically dispense some snacks and then the cats would interact this would sort of uh as you know I imagine me you know at the beginning where I wanted to spend that $100 I would love to have donated that money right it's going to a good cause it's going to keep that animal shelter open but it's also that instant gratification once that transaction goes through you're going to see oh the food gets dispensed and the cats are happy and they're all playing and like probably you know might be fighting each other for the food but it's you know it's rewarding and I I know where that money is going right and the money that you're Distributing for something like this it could be used for for keeping the website running the power the uh maintenance the uh anything else just goes to the animal shelter like I described before and who's to say this has to only be an animal shelter for those who might have been at Dev connect in November in Istanbul there are a bunch of just you know uh wild cats that just roam the streets they're totally friendly they'll walk up to you they're uh at most any park but who's to say the same setup can't be set at at a remote location where these cat hotels sort of exist in these Parks all you need is someone to constantly fill it up and you know it's the same experience and then I I I heard this whenever I pitched this to a few of my co-workers so if anyone wants to do this please F free another one is this idea of a timestamp Oracle so what if on chain you had your smart contract or your application had access to the current timestamp down to the 100th of like 100 milliseconds right now the only way you get access to a time stamp is what is the current block Tim stamp depending on what network or what protocol you're using that could be a few seconds apart right it's not an accurate time stamp uh to what the actual time is right now it's just when was the last block mined and what you'll really need here is a network that can handle hundreds if not thousands of transactions per second because again you're hitting this at 10 Herz so uh 10 times a second to be able to have that sort of frequency but what could your application do if it had the most latest up-to-date current timestamp on chain another thing that you might realize is this is going to get expensive I'm making 10 transactions a second like that's I can't keep this up forever even if I wanted to sponsor an application like this put in the funds uh that's going to drain my account real quick but that's when you know you start diving deeper into uh applications or networks that have state diff based fee models so with these sort of models you start learning that oh you only really get charged for the difference in the data at the beginning and at the end of your batch so even though you're updating it 10 times a second if there are if the batch is let's say two seconds apart you're only going to get charged for the beginning value to the end value another idea what if your biometric Telemetry your your health Telemetry is fully on chain right what if similar to that time sample orle it is constantly being updated and just available right what if uh well let's go into some examples so uh what if you could prove that you exercised at least three times a week right is that cheaper life insurance what are some applications behind that is there any way to anonymize this data right that was the basic uh as soon as I asked this you know crazy idea to some of my co-workers it's like I didn't want to do that I don't want anyone knowing this information that's you know very private that's pii uh but maybe if we anonymized it that could work what if also you use this as a reward system what if you built an ecosystem where if you proved your heart rate was above uh normal for 30 minutes a day then for after a full week you get you know a small prize pull that gets distributed between all your friends right and it incentivizes all of you to make sure that you're exercising at a constant pace and you're keeping each other credible as you're going through this so here are some final thoughts here are some takeaways that that I want to make sure that I I sort of cover here which is one you know get creative like these don't constrain yourself to the idea of just the limitations of the system right don't be afraid to share these crazy ideas like something like a hackathon is perfect for making sure that you feel like you can just you know try something for the sake of trying something and ethereum's ecosystem and the feature set is just expanding rapidly so take advantage of it right State uh diff based fee models or validium or you know whatever all these examples that I went through you know those are becoming a reality now so understand what they are and try and use them another big one for me is we're better than just another Finance dap right these all none of these ideas were like you know Finance based but I don't I really cringe at seeing another just nft shill you know sort of app that's calling itself a game and and lastly like asking these crazy questions or or crazy ideas is going to help you learn something new and think differently whenever I bring this up this presentation up with my co-workers I left every single time knowing something different whether it was hey that actually existed about two years ago with this one side project or hey uh actually that might not be that might be possible now because we're going to have something like validium so one final note is ZK quest for those who haven't signed up it is a web app that we built just for E Denver where you can go and complete uh coding challenges and puzzles to earn eth this is something that we sort of built together as a different approach to hackathons so if you want to sign up uh the booth is going to be here all week it's going to be down that way you probably have already seen it um and thank [Applause] you yep good he's getting pressure here TK safety and we are back with the next panel thank you thank you very much for your patience everyone let's get a round of applause for everyone who joined us [Applause] [Music] today this panel is called z z snarks rule everything around me that vitalic presented at atorn in montro and we were very inspired by that that's why we decided to bring thought leaders of the space starting with Jason Zach Ben Jordy and Benedict Porter will be the one who is hosting the panel so let's just welcome them again and let's hear what they say all right welcome everyone I'm super excited because basically the way this panel was created was I got to ask all of my favorite guests and they all said yes and we're going to talk about how ZK will rule the world uh but first we will do the the slightly boring intros so if everyone could just say your name introduce yourself and your project yeah I'm Jason Morton from e he is a project to make AI uh verifiable and private with ZK cool hi I'm I'm oh hello yeah I'm Zach uh does this need to be a boring intro extra boring who hears out of Aztec yes so yeah that's that's that's that's my jam so um yeah I'm a CEO co-founder and we our mission is to add strong privacy guarantees to this wonderful distributed well computer that we're all building and using uh Ben Liz VP of research at ZK sing I'm Jord Vina I'm the technical lead at polygon zbm we're trying to just skill ethereum that's the main goal and yeah and very much a developer engineer and ethereum fund I would Define myself yeah my name is Ben buns I am one of the co-founders of espresso systems uh we're building a shared zerve for ZK rollups but also rollups in general and uh you know I have some takes on where ZK actually doesn't rule everything where you need other tools such as consensus to solve some issues um but I also have a second head I'm also a professor at New York University where I work on Zer knowledge proof so I also do believe that ZK does rule a lot of things nice welcome everyone so uh we will have some spicy takes uh I know a lot of these people have been on a lot of panels before and I don't want this to just be another boring panel uh but starting off we're going to all those panels were very exciting right right sorry very exciting panels and we swap seats around and then we go speak the same stuff at the next conference um but anyways let's zoom out and let's go 20 30 years in the future and I want you guys to imagine all of your dreams come true ZK is taken over the world or or however much Benedict wants it to and so like your engineering goals are hit like all all the things that you want to happen what does the world look like at this point in time sure I'll start um so one of the things we think about is not just AI but kind of General analytics that can be done in ZK um I think one kind of final scenario for ZK is that uh completely replaces all the kind of Erp databases everything that lives inside and between companies and so every time you look at a financial statement on Edgar uh you know that that number underneath it is a proof of every single transaction that went into that number um and I think that would be an interesting New Foundation for the economy um it would allow a lot more uh trust uh as well as confidentiality but trust in the numbers that were presented in Investments and in other decisions that we make yeah good answer what's what is blockchain really for um I would say that it's it's a coordination engine and it's trust infrastructure um and it's at the moment it's a very primitive limited coordination and it provides limited trust because of lack of strong privacy guaranties and limitations with scaling and um data and ZK can solve all of these so where are we going to be in 20 20 30 years we're going to be in a world where it's much harder to create a monolithic company that leverages its profit through information asymmetries uh if we succeed I hope we create a world where users are much more in control of their own data once that becomes practical and possible I suspect that there's going to be a strong push to make it inevitable um people are going to be much closer to the institutions and structures that they are um governed by um and will ideally um if not themselves actively participate will have an expectation of being able to do so um which means having a more pluralistic um like forms of governance both politically and socially um I see it as a world where it's much harder to create monopolies a world where it's much easier to start competing in um in Industries because everything will be much more composable and to injury much lower because of the composable nature of what we're doing um but ultimately it's going to be a society where there's more trust it'll be a strong trust Society not a weak trust society which is where we're heading at the moment where people don't have trust in their institutions and they don't have trust in the governance structures that they're subordinate to I mean look so fundamentally this is about uh building um an affordable uh high integrity computational layer with optional privacy in a way that cannot be exploited right it's a bit of a mouthful but that's sort of at the end of the day that's what these systems become and uh certainly we uh already see uh enough of sort of traditional Finance functions being move to that but I think it's going to be super interesting uh to watch as we have sort of users other than like humans actually starting to step into the space that is to say you have agent autonomous agents things of that sort uh programs that react to data I think that's where you know basically this high integrity layer will start shining and where the demand for higher capacity for higher throughput potentially lower latency where all of these things will become even more important than they seem now yeah I I it's more philosophical maybe uh answer here but uh I believe in uh free world when I say free world is uh free people living in there people that just live uh happy that has some happiness in there people organized in communities not in states or not in these so committies that you can choose to opt in or to to opt out and uh for that uh you need some system of trust that you don't need to trust in a specific persons you just trust in in numbers I think you were mentioning very much before that and uh yeah to create this uh community base where mean people just thinking having fun and uh working for the other people's to have fun and uh enjoying uh where we are living in this world uh yeah uh we need to solve uh conflicts that we the humans uh have today and if you want do a deep analyze of what are the big problems that humanity is suffering right now most of them you will discover that end up uh trust problems if we can solve that or we can add a piece I mean if we can add or or or or or or peace uh so that the the trust of humans is less needed and we can trust in in in math in science I think this is uh we will get a better world and that's my my goal for the for my life I mean but I don't know if it's going to be 20 30 50 years whatever it is uh but this is what I'm working for yeah I'm not sure I've met Chad I would love to live in all of these worlds but yeah I think one very concrete thing that I don't think will take 20 30 years is that there's a lot of generative data you know with the Advent of AI and I think that sort of necessitates more uh like sort of verifiability of where that data is coming from and I think that's a very immediate problem that we're facing but that I think is very solvable uh using ledgers and using zero knowledge yeah nice I think it's actually a great summary all these Visions are really cool and they actually do seem all really solvable like you guys are all working on projects that directly do this so is there any reason why these Futures wouldn't work out or any roadblocks or bad things that could happen along the way yeah I mean I think one of the roadblocks that we face every day um maybe not so much here in you know magical uh ethereum land but if we try to sell this to someone outside you know who's not here in Denver with us is that they have a trouble understanding what exactly it is that we're doing what the guarantee that we're providing is um and whether the trust assumptions that we've now made are actually better than the trust assumptions that are provided by like a trusted Enclave or just by a legal contract um and so I think the biggest barrier in my mind is not really technical I think we've shown that uh the technical Innovation can happen at an extraordinarily fast pace so the real barrier is is more social and educational and kind of marketing can we figure out a way to explain the benefit of this technology um such that everyone can understand it right we we've done a tremendous job for example with uh htps right with secure socket uh type technology and certain kinds of very limited encryption um but you know outside of the crypto space uh people don't even really feel comfortable with the digital signature and we're asking them to understand a generalized dig digital signature so I think we have a lot of work to do in in making it seem useful to the average person yeah I'm not I'm not sure I agree about it being really about education I think the fundamental issue is the tech sucks like it's so bad you know who has tried to write ZK snuck circuits come on raise your hands right who's deployed ziki snar circuits um in any kind of production setting yeah yeah it's hard sorry for swearing um so I think yeah people don't understand it and it's unfamiliar blah blah blah that's because it's it's it's weak and it doesn't reach into their lives in a meaningful way and we have a lot of work to do there but that's not why we aren't going to succeed because technology is inevitable there are so many reasons why we'd fail let's just look back to the first way generation Cipher punct in the '90s right the battles that they fought were around ensuring strong encryption was available to protect user data um and so they had to have a fight to to to legalize encryption software and they won all these battles and you know but but the war that they were fighting was a war for an open permissionless internet where information was free and humans would be their own autonomous agents surfing the information Super Highway and and my god did that not happen um you know like we're all we're all bloody commodities you're a commodity you know like what's the what's the sum total of the of your commoditized information versus the amount of money that that that um black market Smugglers could get for your organs I reckon the information might be more valuable at this point you know you're not the customer you're the product like for every single goddamn application that you use online so the cipher punks failed um and now look at us you know we're talking about oh we're going to change the world you know we're going to we're going to create this wonderful world of free agency and and democracy and openness this is A1 trillion Casino you know like that none of what we're doing is reaching into the lives of the average people Ordinary People nobody buys their daily goods and services with cryptocurrencies unless they're very technologically well connected Bitcoin is not the kind of threats to Sovereign currencies that its creators intended and um there are so many difficulties with pulling out like any kind of positive utopian Vision I think out of the the relatively corrupted flawed uh foundations that that were that that we've had to build off of um so what are the problems like privacy does not get accepted um you know we we as in we the the entire Community just can't make it work in a way that users are willing to uh willing to tolerate it um I'm wrong with my thesis that privacies the missing link to make blockchain reach into the real world and become mainstream without privacy we are building the tools for the next surveillance State we're building a a networks where every single transaction is happening on a global permission permissionless so for now Ledger where everything is visible where the sum total of all of your transactions and all of your payments what you own what you have is is viewable by all like what the hell like everyone's oh yeah privacy yeah we'll add that on later you know it's just an easy fix it's TurnKey like are you mad you know like you're you're like this we're building dangerous technology masquerad as as as as a dumb casino at the moment and um we're just repeating the same mistakes as the wave one Cipher punks the way I see it but like what do you think are the mistakes because like I don't know like uh a product that I think about sometimes is is pgp um like just encrypted email right which is not very technically hard like what we're trying to do with zero knowledge proofs is significantly harder it is absolutely awful right like no one has ever used it like seriously um you know it could be the default like you your emails could be by default encrypted in some way I feel like that wouldn't be that hard to do and yet it doesn't exist and I'm not entirely sure what the reason is is it like hard to like is it to deploy or is it you know like I don't think it's like you know evil Uncle Sam you know trying to prevent it like I I like genuinely wonder why not and like how can we learn from these failed attempts in the past I can haard a guess is that the problem is society benefits as a whole when people use these technologies that that are backed by strong encryption for the individual it's a a meaningless difference you know you take some random person so just some Average Joe right going about their lives do they care about their emails being able to be like read by by by the government like no and they don't even know that that's that that can be the case um not entirely I mean so every just about every survey you do in privacy you'll basically have this sort of uh 108 10 kind of thing which is 10% don't care about much of anything then 10% are sort of the tin foil demographic and then 80% you know sort of negotiate with me right so basically depends on the circumstances and depends on the value prop and then you know there is a bunch of products out there that are quite addictive where the value propers that people are willing to give up give up on their privacy to get you know free usage right or seemingly pre usage I think that all the I mean all the Technologies technology is I would say quite neutral by itself I mean all the Technologies can be used for good and for bad and and but the technology I mean if we think that the technology is a Bad Thing uh then we would not Advance we would not do technology and if we think that the technology is a good thing uh I mean then uh maybe somebody is going to take over so we need to we need to understand that techn the techn the Technologies can be used for good and for bad and is a responsibility and here I fully agree with with us Zac that it's our responsibility that this technology as it grows it's also used for good things it's used for improving the the good of the humanities of course there are risks and we I agree that we need to be aware uh of the risk the risk that you mentioned they are facts and they are reals okay so we need to be aware of that and that's why I mean supporting privacy is is is is is an important thing but the technology is not only I mean this technology is not is not only privacy I mean scalability is important usability is important I mean there are other things that are important and needs to be we improved and when you join all together all the pieces of course including privacy including uh censorship resistance I mean including all the values that uh are important if we put it in the right way we will get a better Society if we put it in the wrong way we will have a worse Society I agree with that I just I'm saying that without privacy it is a worse society as in you like transparent distributed ledgers I would argue if that's the end State they are net negative on on this planet net negative because of the amount of social control that can be exerted by having perfect information transparency around around individuals um resources how they what they own and how they spend them is far too damaging I think one uh okay sorry to take to take a slightly different take I think part of the problem with privacy as we're conceiving of it right now is if you try to sell privacy to an individual you have this problem they don't care it's a lot of work and the government would actually rather that they not have it but there are plenty of other actors in society that have an actual obligation right so they have an obligation under regulations to keep user data private or not to collect user data or not to keep it and so I think if we for to focus our energy on kind of companies that have to maintain confidentiality companies that have to use their users data in some specific way we might have an easier time actually delivering privacy to the end user let me be a little bit controversial here with you Zach uh so I agree with you in the Privacy you know that but I would prefer a world where all the transactions are public that a world where very few people and people that you don't control have access to your transactions and not the rest of the world yeah but that's not what where we're at at the moment at the moment everyone's like transactions it's it's a mange of chaos right um we have cash we have an enormous number of differentiated payment platforms where it's actually quite difficult for at least in in a in a nominal democracy for for a um uh for for a government institution to actually have perfect visibility over an individual and their and their assets and their finances um es especially not in a very algorithmic deterministic way where you can just like run a query uh and on anyone and get that information please don't don't misunderstand me uh uh I truly believe that we need PR pracy and privacy is a value it's a fundamental value uh in the space and we need to solve privacy okay but said that I mean I prefer the transparency which is another property that this that we are adding here is also important and it's also value that's interesting to have uh I was just going to say look let's not forget that the public uh blockchain use case is not the only case that's out there right situation is very different if you start speaking to Banks corporates uh consortia of I don't know car manufacturers uh people who deal with uh Logistics and all these other use cases that in some cases you know go back to like 2018 2019 I mean those people yeah to them like privacy is uh very much needed for all sorts of reasons that have nothing to do with like personal freedoms that just have to do like they make good business s right and then you don't have to sell them on that they understand why it's needed yeah I agree I mean these are all able problems um I wouldn't be here if I didn't think that um uh but it's it's it's a challenge that needs to be considered as in the way that we're planning on going about it is not like privacy is not the um it's not the product because not enough people care um it's privacy is something that is an a essential requirement for certain products and services where whose value prop is complete completely unrelated um and those I think are the um are the targets that that need to be need to be chased in the early days but Zach the the other point is that uh we are the starting point right now and uh regarding privacy currently we have zero privacy the starting point that's not true that's that's absolutely not true um look maybe we maybe we had privacy more maybe we have more privacy 20 years ago before the mobile phones before we had the mobile phones today all the governments have access all the governments have access to all your data that you have on your iPhone this is where we the starting point right now I think that's to H I think that's a bit reductive because what they do not have is coordinated accesses and they do not have a single source of truth that's fully legible that they can easily run algorithmic analysis on you know I'm talking I'm talking about the blockchain I'm talking just real phone yeah yeah that's I'm saying that they don't have what what I just said they don't have you have kyc you know who you are you know the phone and they have I mean there is Services right now there is services that all the all the states come by and they I mean they just say I want to spy this we're talking P each can what I want to say is there is friction governments can acquire information about an individual but it takes effort they cannot acquire all information about all individuals at low f with with a very low cost basis and perform a lot of analysis on it exact they're using AI to organize all this information okay I think I'll I'll take I'll bring it back and take us to a new question but that's great from all the panelists um let's let's go to Benedict and talk more about the technical side of things so you've published tons of papers you're working in both industry and Academia um could you talk some about like the balance you see between those like industry at least in cryptography actually seems to be doing in general a lot of work that doesn't start from Academia and how do the two work together yeah I mean I think that um you know I think it's a it's a complicated space and I I you know sometimes I have the feeling oh like we haven't reached anything yet and you know we still have so much to do um but then I'm also there's things that make me incredibly optimistic and one of the things that makes me the most optimistic is is there is a beautiful Synergy um right now and I think there's inherent reasons why that is between Academia and Industry we have really like cool ideas that come from industry that then lead to sort of you know more academic research which then leads back into industry and so on and so forth and like hard problems that like like people literally are expecting like oh we have this hard problems but we believe in signs you know we believe that science can solve this problem and smart people from wherever they are whatever background they have they can tackle these problems and that's just so cool to see and there's fundamental reasons why that and it's way better than in AI where like you know which is also an emerging field where Academia and Industry have a thing to play but you know their like data is private right and then you know access to data is really important in order to make you know even just new Innovations and like certainly products and one of the cool things about blockchains is that by its nature it's it's open source and it's open accessible and it's you need to actually you need to actually show that the product that you're building has value and is trustable right and that works so well with this academic model that we have and like yeah so there's there's which uh is I think you know also like uh they used to say that like in the best case it takes seven years for some cryptographic Innovation to be deployed in in practice right and now I don't know like I've had uh yeah less than one year right it's like or I think I I wrote this paper and then within like seven months it was deployed in Monero and and you know sort of powering Monero and yeah it's it's a it's a really really fun space to be in and to work in and it uh makes me like when you have both uh companies that are profit oriented which is totally fine and smart people in Academia working together and tackling these hard problems that is what makes me optimistic that you know sort of at least the technical challenges we can solve and and it's been like yeah it's exploded and uh it's it's really you know I cannot be more positive and optimistic about it let me let me add to that so I've been an academic uh for about 20 years right mostly within industry band but like I've been publishing papers for about 20 years and um frankly uh I have not seen a space or have not experienced personally like firsthand a space that moves quite as fast as this one and of course AI sort of competition that we have but the level of uh sort of intermingling the level of uh kind of exchange of ideas and the speed with which it happens here in ZK specifically is quite astounding right and again I lived through a previous sort of wave of zik innovation back around 2013 2014 right but some interesting results back then too but I mean nothing compared to what we're seeing now right just like the uh you know the speed at which uh results are coming out in the form of papers right I mean like often times we don't even bother submitting things they just go on ePrint to Archive and people read them within days right um or or like the speed with which you know experimentation is happening uh again quite astounding yeah I mean I I agree completely I think that the the kind of the intrinsic incentives of blockchain are are extremely positive sum in many ways and um it's it's an area where good ideas R to the top because they're they're not controlled by like individual factions um that are trying to monetize them uh they they just become another piece of a very composable framework where they add value to a lot of different um uh different projects different people um and yeah it's it's really positive to see I think the quality of opportunities available to competent people in the space is so much greater than any other industry I've ever either been in myself or or or observed from the signed lightens yeah I think we we talked a little bit uh earlier about some of the political problems and uh sort of Human Rights issues that we hope to address with the technology I was also an academic I was a professor of mathematics um before uh starting this company and you sort of see saw I would say four waves of turning math into money right like what are the options if you want to make a career outside of Academia and sort of wave one was defense and wave two was maybe trading and wave three was maybe advertising and this is wave four and this is the wave that I felt okay leaving that academic job because I felt like I could wake up and do something about problems that I saw in the world and that the actions and technology that I was building actually could have a positive impact of course there's a lot of froth of of course there's a lot of issues but I think it's easier for me to sleep at night I think than if I had gone one of those other routes of turning math into Money Jim Simons did very well in wave number two right in the trading wave Jim Simons yeah I think if we have one one one value that probably right now is in danger and is this uh collaboration this collaboration environment that we are in the space of course this is why acad acemia likes because Academia has been also about always about collaboration on that this is an important value here for me it's a call especially to the tech people to the people that's interested in Engineers people mathematicians people that's interested in in the technology to have to keep this value in mind uh the pressure from the I mean from the society I mean from the competitive from the marketing teams from the uh uh I mean from the business teams I mean money money is money and and and the pressure is H but here is a call I mean for uh technical people for free technical people for people that's interesting to keep this collaboration and this is if we keep this this is much better for the industry it's even better for the for the for the I mean for the organizations for the corporated organization itself okay because the they been a revolution okay so we get uh and we are in a blue ocean in this space so this is even good from the business perspective but we one of the biggest dangers that we have in the space is that at some point we stop collaborating so even if sometimes goes against your hurt we need to impose ourself let's act as free persons and continue collaborating do not hide information do not hide technical information for patents for uh business reasons for whatever just share what your knowledge and if we all technicals we keep as that we are going to advance much much much faster and this value is important and and and here I think that from the academics that's if we keep this the academics they already been doing that for years uh uh on that we should be an example for other Industries and and and and keep and and keep this value as we have been doing until now I I think sorry like there's one uh like wow on this topic there's uh one like sometimes there's some small frictions uh between sort of Academia and Industry uh and also by the way to your point keep funding research um but it's a one small friction is that you know academics like to write their papers you know with you know like very formal and they like to get their citations and brownie points and the greatest trick we ever pulled is that you know as Society is that we you know get to underpay academics but they get their brownie points so keep giving them brownie points it's very important but then you know technical people like to bring things out in in blog post very fast you know just put the idea down and I think both are great and we need both and I don't think one sort of like like obliterates the needs for another only because you know an idea has been written as a blog post doesn't mean that you know someone else can you know formalize it and write it in a in a much more you know like formal way with with mathematical proofs and all of that and only because you know something uh like you know only because you have an idea it doesn't mean you need formal training in order to publish it if it's a good idea just write it up in whatever way you can or care about uh but put it out there just exactly as Jordi said we need to keep this culture of of open source of open publishing uh I think it's one of the strongest assets we have as an industry it should not be incompatible uh to be formal and to be uh clear and uh that accessible uh for the teams I think it's I mean it should not be in compatible I know that sometimes you need to do an extra effort for compatibilizing both things but uh the I think it's I mean uh if crypto went so uh fast and here I'm looking some here people that wrote some some papers is because the people that wrote the paper they they not only did the formal part they also did a a huge effort to go a step further in the paper and explain the the protocol explain the the the whatever they was explaining in a easier way for the people that are less uh used to read papers to uh fully I mean to understand uh the basics Maybe going behind going after the the the formality which again I respect and it's important it's essential in the in the crypto space yeah and and it doesn't have to be the same person right I I mean I think ideally you know you do everything you write you know formal academic paper and informal blog post you know in implementation and everything but not everybody is suited to do all that right like someone might not have the formal training to write a perfect paper that's okay you know like there's I think we we can again use the power of collaboration uh to have you know you know someone else can explain a paper I think that's totally okay and and and I've seen and this is happening which is great I think there's tons of willingness to go up and down the Spectrum and again like not everybody has to do everything right I think there's a lot of willingness to sort of do handoffs of course people need to learn how to do it better and this takes some time but frankly look I mean for years I worked in areas where like you know you have a paper and then it takes at least 5 years to see any glimpse of that in production right any glimpse of that seriously and so compared to that I mean things are happening like blazingly fast so and uh so that's one thing the other thing is that I see a lot of willingness even in the rollup space right for people to do things together and you know we are on the research team starting some collaboration with other rollups and certainly we would love to see more of that and perhaps eventually standards around some of these things but at this point like let's just you know let's just see more of that and see where that takes us okay all right speaking of collaboration we have all of these super smart people on stage who are publishing papers and creating zkm and zki and privacy and crazy things and we have all these wonderful people in the audience here who are just dying to know how they can end up like you guys at some point so what advice can you give to the audience that they can start working on to end up helping with this future that we're looking for I can oh do you want to go first or go ahead I can give the cynical answer and then I can give the real answer the cynical answer is go to www.az tec. network and and download our sandbox and start developing today we've got a great grants program and and in all seriousness you know like it's amazing Tech hack on it um we've already had like several teams that have actually created startups out of what they can do with our Tech uh but but more seriously like okay how do we how the thing is if I look at how I came into the space I'm not sure it's a model that can easily be replicated because there was just a lot of serendipity and Circumstance right like I I came into the ZK space when it was still relatively new and um and I had basically I I didn't know ZK I learned it from scratch in six months and was able to make meaningful contributions and publish papers and that's not as easy anymore but I can I can give more general advice which is basically um find an existential problem that speaks to you you know find find a sense of of belonging like if you're working on a project and you and you're getting as like immen that intrinsic satisfaction from working on it and you feel like you belong you're in the right place keep going uh and if you're not there yet then maybe maybe you need to keep to to to explore other Alternatives um but like you you need that existential need to drive you uh because if you don't have that you're not going to be prepared to grind the like the what the like the 20 hour days or whatever it is um for for multiple years and and um change roles wear different hats learn different skills um to in order to in order to um to succeed yeah I can listen I think what Zach said is very correct I think it really depends on who you are um you know there's great uh academic like learning resources out there corsera has a great and free cryptography course that teaches you the basics of cryptography there's tons of YouTube videos on sort of learning about zero knowledge proves uh there's um obviously amazing libraries you know that you can hack around with if that's what you do you know just starting to play around with I think one thing that really like one advice that I would give is that it's really helpful to be comfortable with abstraction so for example not knowing exactly sometimes it's good to dive deep and know exactly how something works but sometimes it's also very useful to just know what it does so for example if you just understand what a zero knowledge proof does like it allows you to prove that something is true without revealing why it's true that's sometimes okay to understand everything you need you don't need to understand how they work or like sometimes I get people that come to me it's like oh I started to learn about Zer knowledge proof and then like I got stuck at elliptic curves I'm like I don't really know how lipic curves work either that's like way deep down the line you don't need to know that you can sort of like just be comfortable with some abstraction of it and then uh you know just knowing like how some things work and then you can build on top of these things but yeah it's you know just I think hacking around with a library is is is one of the best ways to start and yeah get get get your hands dirty I mean that's a problem that's common to everything uh I would say the vision maybe from the developer for people that loves coding I mean this is very much my passion uh just take some of the protocols I mean just like just try to create by hand an ethereum transaction and uh send it to ethereum transaction and and and see that mined and it's accepted if you implement the litic curves if you implement the cdsa if you implement uh uh numbers uh uh I mean the finite field uh some finite field uh libraries coding I mean just trying to understand by coding when you code something means that you probably have a good understanding well not always but but but in general in crypto if you don't understand what you are doing is is very very much you probably will will end up giving up so for me the way to get in the space is was coding it's coding that's what I like to do I think you can get a good level maybe if you want to write papers then there is other path uh we need more math more abstra action on that but uh I mean I just want to share here my experience that uh coding and trying to understand uh what's going on maybe starting a little bit of math uh to understand what's going on uh this is the best way to to to to to get in um so I I can also do a plug for for us of course one of the things we built was a system that lets you just write in Python um so you can just write in a python Jupiter notebook you can write ordinary python torch and compile it to an AI model which is a zero knowledge circuit that's a great way to get started and play with it um of course I agree the right way to do it is kind of play with the thing start to think about what you want to happen very soon you will reach the limits of the technology because it really is a young technology although it's moving very fast and then you start to dig in onto what you know how far down the rabbit hole you'd like to go and understanding you know the kind of math you need to add and that can go all the way to very deep algebraic geometry right so I started with algebra geometry from Robin har right and then somehow way down the line I find myself doing extremely applied algebraic geometry um so the math that it underlies is extremely beautiful and I would encourage you to dig into it but don't think you need to learn any of it to to start so look my take is that uh there already you know even on this stage there are people with fairly different backgrounds my background is in systems right let's say as opposed to cryptography but as the space expands I think there'll be even more diversity that you're likely to see I think what's important is that given how fast the space is Shifting is finding a large longterm Pro long-term problem that really appeals to you right on a deep sort of psychological even emotional level this is the kind of thing that you're happy to work on for the foreseeable future it really helps because this is how you you know this is what forces you to learn this is what forces you to overcome obstacles as they come this is what forces you to sort of reconfigure your forces as circumstances change and you know things change around you right I mean I don't know the Innovation improv systems we see like a new one being announced like every I don't know three weeks or something these days um so yeah that's that's that's my take cool all right I hope everyone is very inspired about the future of ZK now I think that's all we have time for so let's thank our panelists yeah that was fun got some controversial all right awesome my height oh wow okay yeah you want helping me get this up a little every listen to me am my loud and clear oh that's perfect thank you oh hello there we go we have a mic check I need to eat the microphone thank you I think I'm good yeah front and center yeah yeah yeah okay uh do you have a microphone for questions feel like I'm going to lick this you'll be a to all right cool cool cool if we got it we'll see if we have questions all right you're a pro I a little no I need I need good posture oh nice okay I might just hold on to mine I like fidgeting with my hands welcome everybody uh let's get this uh show on the road we'll stick right on schedule my name is Braden or BX paana uh from matter Labs the team behind ZK sync and you're joining us for The Good the Bad and the Ugly to take uh the next few minutes to take a look at the the good and the technology and achievements from different ZK rollup projects uh the bad some of the technical things that have they've run into and then we're going to take a look at the ugly and take a little bit uh turn the stone of community and take a look of what's going on underneath it instead of just the shininess on the outside so we're going to start off the segment uh just a little introduction from each of our guests and let them just give a little one one minute intro about who they are where they come from go ahead Alex hey my name is Alex and I am Community manager at Tao so I handle all things Community Discord Twitter uh I'm a little more on the ecosystem side so trying to find really fun interesting things for people to do on Tao and that's about it hey guys I'm sherar I'm on the Partnerships team at scroll so if you're not familiar scroll is an open- Source bico compatible Community First dkm we launched on mainnet in October and in my role at scroll I work across a bunch of different initiatives like defi staking reaking stable coins a bunch of stuff yeah hi I'm Emily I do develop a relations at Linea that means a lot of different things so I can't tell about like my day-to-day because I feel like it changes like crypto but anyways um lenia yeah we are an evm equivalent ZK rollup uh moving towards a type two eventually type one um and yeah where Dives Thrive hello everyone I'm CC fos I work at met Labs the team behind ZK sync I'm a community manager uh were focus mostly on the developer Community right now and been focusing on the users in the last two years and uh zync is a type four Z KVM we've been on mainnet since March last year last e Denver and and I'm happy to discuss with all of you um Good the Bad and the Ugly of Z kvms sweet thank you guys and we really appreciate you being here uh and like I was saying we just wanted to give an opportunity Community to just have a little bit more of a say and show to the world to the world of developers especially at a bit a week here uh what goes on behind the scenes but before we get into that uh I just want to make sure we co cover you know the technical bases so we're going to start with the good and just give you guys a uh a minute to talk about like from the technology standpoint what are the good things that you guys feel have have been accomplished by your project or by your protocol uh and like the future achievements that you guys are trying to achieve as or achieve achieve achieve what you guys are trying to do as well so uh let's start off we'll go switch up the Ord we'll go backwards so estan all right um so wow um it's been super exciting to launch the first KVM one year ago on Main net and that has been a great technical achievement I think we were uh on Main net before the W ethereum Community was expecting to see an EAS KVM uh on Main so that was like something that was amazing to go through more most more recently I think um one of the technical advancement that I'm super excited about has been our new prover um we um published and is live on production the new prover called Buju which um decrease substantially the fees on ZK sync and I think since then we've been consist consistently able to manage I believe the cheapest fees as a layer two and I think that's an exciting um technical achievement um I think buum is also a great system because it has a very low Hardware requirement so when it comes to decentralization one day and having a decentralized pro Network having low Hardware requirement is U I think something that would enable to have you know a proper decentralized PR Network so these are two uh yeah the two technical achievements that I'm most proud and happy about over the last two years nice excellent Emily yeah um definitely a lot of really cool things about Linea I think I want to speak more from like an ecosystem of Technology standpoint uh so if you guys don't know Linea is powered our circuits are powered by Gar so gar is an open-source ZK circuit Library that is the fastest snark on the market um and it's it's a goang and there's a lot of things to say about it but if you've ever interacted with like zkb andb um sellers ZK Bridge everything like granar Powers those and it's like that's an in-house team with the like that at consensus right so that's really cool I I kind of want to take the other view of like again leveraging ecosystem and uh open source projects uh so Linea and any of you guys call me out if this is untrue is the only L2 that is multic client right so we support nethermind uh bezu gu out of the box um which is really like I think a huge technical achievement I can't explain how they do it uh they're just intelligent people um but yeah I think that's kind of uh something that people don't really think about at all um especially when we keep talking about oh decentralization multi client security BL blah like oh ZK yeah yeah I don't know but anyways that that's really cool uh that I'm really proud of our team um and I I think a lot of it also is due to to bzu right hyperledger bzu is an open source team that sits within consensus as well so there's a lot of synergies that we've been able to leverage um from like the technical point of view yeah that's great thank you yeah so there's there's a bunch of different things I'm really excited and proud about for scroll so I'll try to go through them quickly one thing is like the cross-pollination of contributors between us and the ethereum foundation's Privacy scaling and Explorations team so that means like we're our contributors are directly working on what we think is going to be the future of scaling ethereum that's a really exciting thing and those people are also helping us out too uh the second thing I think is that we have been pioneering the multi-prover approach so just like you said with client diversity you need to have pro security for you know fault tolerance so over the next few months we're working on basically bringing in addition to our ZK evm prover we're going to have a trusted execution um environment prover that's going to be based on stuff like Intel sgx on a threshold signature scheme so that's going to help build another layer of security on proving and I'd say the last thing is that we've been working really hard on being really as compatible as possible and making it as easy for Developers in terms of like op codes pre- compiles and like a little known thing is as far as I know right now we're the only ZK rollup that has the pre- compiles necessary to deploy uh Noir contracts on Scrolls so that allows you to build stuff with privacy um so yeah those are all really exciting things all over the place that I'm really happy about fits in perfectly for our last uh panel though up here with the Privacy talks yeah thank you Alex well where do I even start I think Tao's uh it's a little bit different cuz we're still in testnet so a lot of the interesting things that have been happening have been like a long Evolution over like the test net phase so I think a big thing is is ethereum alignment so everything about like being a type 1 ZK evm so we can really uh you know work with the existing Dev tooling uh we can work with existing smart contracts you know it's it's it's ethereum extended um and I think it's been a really long interesting road with a lot of like open-ended research problems that we've had had to tackle from test net phase to test net phase we've tried out different um like tokenomics designs we've experimented with l3s uh you know we've taken a look at uh like like scroll like a multi-prover approach so we we're we have a a based contestable rollup that we're testing out right now in our u in our test net and so that's again similar along the lines where you can basically plug and play and use different uh proving systems in in the rollup so you can use like trusted environments like sgx all the way to you know the new up and cominging Cutting Edge proving systems and you know that's that's also been a thing is like the adaptability of like Tao as a protocol and how how things have changed you know the ZK world is very new and and up and cominging so there's lots of changes you know 6 months from now there might be a new novel proving system and so I think a big key has been keeping the protocol adaptable and able to to sort of be modular in a sense so we're not relying solely on like one proving type you know we're not but you know when things change we can keep up with the entire landscape you know and it's been it's been really interesting to see us kind of like adapt to that to that really crazy landscape this is awesome I I just reflecting on when starting in the industry here like where ZK was to now hearing about multi clients multipro being a type one like having different proving systems is like it's it's mind-blowing what we've done in the last what maybe four years so uh just wanted to highlight that really quick and of course uh we talked about the good in the the technical so we've covered the good now thinking about each of these great things that you've had I I want us to also explore like the bad all of these while we're trying to achieve like technical greatness uh there's a trade-off with some of these things and it's the challenges you run into or the trade-offs that you make or or whatever it is so just thinking about that perspective uh what has been like the challenges from the technical side that you've seen for your guys' projects or for your protocols uh and let's go ahead let's mix up the order make things spicy uh sh you want to start us off yeah I mean where do I start I think so scalability and costs are probably something that we all have dealt with the growing pains uh with running a ZK evm I mean like if you go to any like Twitter announcement probably from any of us and look at the first few comments it's like please reduce transaction cost please reduce bridging cost you know you just got to deal with it um I think with Zs I think that people to understand obviously is like initially the the prover costs are going to be uh advertised over fewer people but over time it reduces as you change the pr scheme as you change and adapt the circuits as you optimize those uh contracts and everything it gets a lot cheaper but at the beginning you can expect the same experience you're getting from optimistic rollups and so getting over that bridge of understanding for developers and users that it's not the same user and developer experience at the beginning has been the biggest challenge for us um and I would say just like like you said continually adapting to this like this ecosystem is is is really difficult right like these circuits that we build to prove these particular op codes and pre compiles we have to continually adapt and upgrade them if you ever want to change the prover and verifier scheme we have to adapt that we have to also make them simple enough that they're auditable and if we want to add any of this custom stuff that's part of these new narratives right like the modular narrative if you get an alternative da ler you have to prove that too so there's just like constant push and pull of trying to keep ahead of the narrative and keep the same ux and devx of incumbent alt l1's or l2s that have had time to catch up and get that existing user base um but I think you know I'm I'm very positive that we're getting over that inflection point now nice all right Emily yeah um I guess there's kind of two ways to look at it like I think generically bad is just trying to build a zkm right there's a reason we came out later um I think like at least the Linea team has taken a really interesting approach to it in the sense that it used to be two-step now it's four-step proving system where kind of the translation part of the proofs we actually just didn't do a snar it's something called Lattis Bas hashing where it's really good for recursion um and then at the very end it's kind of too bad too big and then we like shrink it down to a snark and whatever that's probably not the important part I think I want to do talk about like the comments on Twitter and like what that means in terms of the bad and I want to speak specifically to like I think the user like trying to explain gas prices to users especially since at the end of the day we are still affected by L1 prices and so like yeah if everything is pumping like ethereum's expensive whatever and like that's going to affect us right like as l2s I'm sorry yes the fee will get higher and it's like trying to manage the you know the community of like explaining how Gas Works to like a very technical thing to a non-technical audience many of them just aping into the the chain and things like that um um but I think probably the solution to that is just making everything cheaper right like whatever okay which is exciting right cuz I think from the linear perspective we did introduce data compression two weeks ago which reduced our prices by 66% which is pretty like like Hefty kind of yeah and I I heard Scrolls kind of would be doing something similar soon right yeah yeah which is great very cool yeah the the yoloing for some people who just Yeet right in you're like uh let's take a step back that's one um so I'd say that the explicit trade-off that we're making with ZK Singh is a performance equivalent one so we are not evm equivalent uh zkm so I'd say I would classify as the bad as being like there is an additional barrier to entering the ecosystem um you have specific tooling you have to adapt a little bit there is a little bit additional work for you to unor on Z casing and that's one of the main I'd say bad aspect that we se but we actively working on it to bring better tooling people I mean tooling on which normal devs are familiar with so um yeah I'd say that has been the main challenge is that we had great performance at the protocol level but at the expense of a bit of equivalency but we're working actively on on on that and making the developer experience as as possible yeah that one's yeah that's definitely a big challenge for us uh Tao Alex well it's been you know a long road I think being ethereum equivalent and being like a base rollup basically just comes with its own field of like future research questions so again this whole landscape is really experimental a lot of these like protocol designs in general like design spaces are really unexplored so it's just it's a challenge of seeing which direction is the right way to go you know where to put all of your energy and what to focus on uh because these are all questions that people haven't answered before this is the first time you know we're taking this on so it it gets it just takes some time and it gets a little difficult trying to figure out again like the best way forward because there's 10 million and one ways to like figure out a way to design this protocol um but I mean those are those are happy and open challenges you know there's been lots of like collaboration between like the ethereum foundation and researchers like like he mentioned with scroll uh Tao was also you know working on that like open space collaboration with like the Privacy and scaling exploration team on like the ZK evm circuits um um you know so there's research being done on like based pre-confirmation Justin Drake is talking about um uh shared sequencing and all that and there's multiple ways to go around that there's you know there's a million and one questions to answer and I think the the biggest challenge is is getting around to all of those and and and another less technical but more like prominent like challenge is just expectations you know you have this like really opinionated like you know like Twitter Community for example that you always feel like you're on stage like everybody's like waiting for you to make the wrong moves so they can you know make their Twitter comments and you know so you always you always kind of feel like you know whatever step you take you know you're always thinking about like what could people say about this what could go wrong you know so it's those those expectations of of always feeling like no matter what you do you're always going to like let somebody down you know whether it's with an airdrop or you know the direction that you're taking the protocol whatever you know have really really big expectations so trying to to supersede that and and to to make the most people happy I think is is a really big challenge yes and you're right about that it's like all of us uh in a in a sense are all like building in a glass house and it's everybody from the outside who has something to say about what's going on in the glass house but like we're the ones in there but everything you do is always on stage and like that's that's definitely something that's different about this uh about crypto that I enjoy and you know is also a bad so good and the bad okay well thank you guys and now let's talk about so let's look into the ugly side like we all know that Community is is a wonderful part an extension of crypto that makes really why most of us are here it gives the the sole aspect to web 2 uh it gives the ownership it gives this like camaraderie of place of belonging now we all we all know those are the good parts but what are like what's the the ugly side that like you're just like makes you like why am I doing this and like then you get back and you're like all yeah this is what I'm here but what what just like drives you nuts what's the the ugly side um of your role cuz not I know not everybody is community leads here but in different like facets of the community but what is the ugly side that drives you nuts and this one let's go with whoever wants to start I won't I won't Force this on anybody I couldn't start off with that so being a community manager I'm like you know all in on the community uh so that's my dayto day and it is a really like crazy space you it's almost like the elephant in the room you know you have like you know we're here in a builder Hub everybody's focused on protocol technicals you know ideas building stuff you know and the community is just kind of like there off to the side where you know a lot of people don't even want to touch it they don't want to look at it and I think you know it's you're referring to it as the ugly because you know I I've even noticed in my space there's like this this like drop farmer type of like you know Vibe there's people that don't really care about the fundamentals that you know they don't care about being part of a community they're just in out try to get whatever they want you know lots of back and forth about like oh this is better than this this is better than this this project is going to win you know you have to be a maxi on one project yeah and so I guess I have like a unique perspective so like with Tao I started out as just like a general Community member like I was just interested in Tao like I didn't know anything like super technical uh you know so I was just in there asking questions uh you know our Engineers took time to like talk to me answer my questions gave me enough knowledge so I could personally go out there you know in our Discord server answer questions to other people you know really inform people about like you know how to run a node you know what's going on with the protocol and then from there you know they really they really cared you know they they gave me a moderator position and then from there Community manager and so I guess I have this really like personal you know interest in like stake in in in building like a really robust and interesting community so I think like a a big Challenge and a big key has been trying to stray away from like the financialization of like a community and so it's kind of banking off of that ugly so it's like you have people that you know join your Discord server join you guys on Twitter you know for whatever reasons that they have you know potential airdrop this this and that but people are just there they're in your community and it's how you Foster that so what we've really tried to do and what I've really tried to do is Foster that environment that I was given of like education so you know people have joined you know our Discord server they've never opened a terminal before they've never installed Docker you know and then within a few days you know they're there like running a node running a proposer you know answering questions for other people and people get really excited so we've seen lots of like Community initiatives and like projects where people start you know we've had our own Community built like Prov Network where somebody was just like in our community asking questions and they were like oh hey I could build like a distributed approver Network and that was all Community run you know so there's been some really beautiful like projects and like initiative started just from our community because it's really been all about like cultivating you got people there for whatever reason that they're there it's all about you know treating everybody with like respect you know giving them time and really showing them that like you can learn it's all about education you know really giving people the time of day to like Foster that like environment nice way to way to take the ugly and then make it beautiful again like see it's like it's so hard to make Community ugly uh thank all right so let's I'll leave it up for the next person who wants to take it uh yeah I can take it so I think there's two ways I can say like one um and I think this is like a one- sentence thing I absolutely hate how everyone calls it an L2 War which is like the dumbest thing ever it's not a zero some game like multi-prover multi I guess Len is the only multi Cent but no I'm just kidding um yeah so stop that that's ugly um and I I think I want to touch a little bit about the financialization of community um and I think actually what I think is really ugly is I don't know if you guys feel this but I feel like there is an element of developers like pitted against users sibles Farmers right because you know like I I think there's an extent of like oh we worship the developer we worship the Builder which is true like they are the ones who build the future but I think what Linea has done and I I really like from like a L2 perspective of the opportunity that we have is we have a lot of eyeballs right and also I mean like where do those eyeballs come from so if we look at the uni swap uh the uni drop right that's like average $2,200 which will feed what 44 people in Vietnam for a month this is like life-changing money right which is insane and so you have a lot of attention and so like l2s we have this opportunity to direct the attention back to the builders and say like let's like let's Dam your network your load balancing needs work or like your ux is really bad and stuff like that right and like also from like an L2 ecosystem perspective right we are connected to everybody we are connected to users Builders from the developer level like developer tooling Builders from like I don't even know anything like you mentioned as well and there's so much like Capital that we can do and interact with people that I think is ugly the way that we kind of just kind of I call it like don't villainize the farmer cuz cuz they do a lot for us at the end of the day um and I guess the the the thing that is ugly about them I'm very fortunate I'm not in your position of community manager I'm going to be honest but I talk to them and it's ugly how nasty internet and nons can get yeah so please be kind um I'm sure none of you here are mean in in in DMS hopefully not yeah cool so that's where I'm at that's yeah they can mean visible wall behind yeah I I'll say that I agree with all of what you said whether it's you know managing the expectations within the community and channeling the attention that is not necessarily uh focused on the protocol to towards the protocol and the mission that we are trying to accomplish with the zvm um I will always remember the first time that we try to just innocently run a survey through the community and we had like 15,000 people just sibling the the form all the infrastructure going down in between and we were like yeah I mean this is not I mean essentially these additional burden from you know the community trying to G protocols and the protocol trying to leverage and farm the community um these um adds a lot of additional operational burden for us and you know if you want to run a meme competition if you want to run um anything in which there is a minimum at stake you need to think twice about how you're going to have you know on board white PIP plan you're going to have because you know that it's going to be seil and you have to yeah take the necessary steps to to account for that so yeah um high expectations from the community creates also High operational burden on our side so um yeah that's I would say that that was a bit a bit the ugly side that we faced over the last few years but of course the um sometimes how mean the community can be uh but also how good and you know uh altruistic they can be you know I think that comes with the the job like we have this broad range of community members and sometimes there is a very minority frange that could be very vocal and mean sometimes always seems like a a constant fight sometimes like some they change something so then you change something and then it adds like then they change it this way and you're like I thought I already stopped that so it's like this never- ending cat and mouse I feel like sometimes too sh yeah I mean I agree with all the points you guys have made I think there's a lot of difficulty in in finding the fine balance between respecting and listening to your community and then kind of focusing on this vocal minority and like bending your will to them sometimes um and there's been a lot of discourse obviously about the toxicity of some of the community members and a lot of different reactions across the board I think my take here is that it's it's a no upsides move to disrespect or look down on the members of your community right there's ways to work around it and there's ways to focus on the people that deliver the most value but looking down on them on people like you said who for this amount of money is is is life-changing it it doesn't help anything and it just makes you seem a lot more highbrow and thinking you're above them when you know we're trying to all be very down toe people um and then one other point about developers I think is also really interesting is that we like you said we sort of like worship the developers in a sense and we think a lot about them of course for us like developer education and thinking about them is the most important thing but sometimes we Silo in on thinking about their needs only and when you talk to them they're like wow you guys are so compatible it's so easy to build these contracts but you don't ever listen to like users like I want something fun to do I want to just like be able to gamble with my money and that might sound trivial to us but this is a large proportion of what the crypto Community is and I think all this posturing about trying to be super developer or research heavy that has to be balanced with a little bit of Degen and a little bit of understanding where the community members are at and what they need um yeah that's my take on it awesome well so we get we got through the ugly and when we were talking about these things it made me think of something else like not a lot of people or when people think about Community they kind of just automatically put it into this subcategory of marketing uh or of growth and and that's what it is and has been in the traditional web 2 but web 3 it's I see it as completely different as what we were talking about earlier um how do you guys uh manage like the internal expectations your own companies of like we see I think from all of us all of our our mission statements or our Visions are like Community First Community built community community community but it can feel like sometimes in inside like you were saying like it's something off to the side like all right we we've got this Vision we got this we got this and like oh yeah and Community is doing something so like regardless of your role because I know you guys are are in different extensions but like how do you convey how important like either Community or what your role is is to inter internally into your Comm uh your your company so they know that resources need to also be put there like product and everything else is very important but yeah Community devil product or uh partnership like all those need attention to so how do you how do you say hey look at me like down here any order no any order any order I'm down I'm happy to go first um I so I like to think of I mean a community we always say that Community web 3 is Community First um essentially I think the reason why is that there is a supposed to be owned by no one else but D distributed community and uh so you know the same way that a protocol requires a certain level of engineering and at some point there is open sourcing and there is a community that gets involved I think the community also needs a certain form of engineering uh you need to design it to be able to have contributors I think um have yeah have contributors have specifical specific opportunities to shop as well and um yeah having the community being engineered um and talk the engineering language to Engineers I think and see the community as a product that needs to evolve in few Cycles I think this is a way a bit to to speak the M language with the engineering team uh I think it's that's that's a barrier that I've seen that you know um speaking different language the marketing language versus the engineering language and turning that into a common vocabulary is a is is a good way to advocate for what you want to try to achieve well I think like the way I look at it is you could build the best protocol out there that has all the best fundamentals you know you could have the fastest proving times you know cheapest fees whatever you launch on mainnet and if you don't have a community you don't have any applications on there or anything it's a you know it's just a ghost town what's the point you know what's the whole point of building a protocol that doesn't have anybody a part of it you know without the community the protocol is really nothing you know and there's there's a real utility to having like a really robust and like excited Community it's like you can like blend the community and like the developer side together so like for example like from my personal experience with like you know community building and all that we've gotten to a point where where we have a pretty large and excited Community behind our test net and right now I'm I'm basically utilizing that Community to to be a Launchpad for all of these new and exciting uh applications that are being built on Tao you know so it's actually enabling all of these smalltime projects that usually wouldn't get enough traction or have enough users to go through and QA test things and get feedback and all that uh you know having our community there and excited and like ready to to learn about stuff and like mess around with stuff on chain and all that has has really like fostered this like really incredible line lunch pad for like our our ecosystem projects you know so our developers on that side and our ecosystem you know they all come together and they figure out like cool ways to like you know be composable and build on top of each other and then you know they learn from stuff and then they get valuable feedback from our community and they have you know this this really great like developer experience all from our community you know and again it's it's it's when you focus on the non-financial aspects of the community when you when you get people excited it doesn't even have to be like really complex like you know appliations that are deployed you know it's like people love to play games you know people want to like play games on chain and all that so you'll have all these projects that come on and they have these new fun novel ideas and you're able to you know again bring community and developers together to create this really cool like like experience it's it's it's really interesting yeah the modular system all over again right put things together yeah um I think the point you made at the beginning was exactly right where you said these are like all extensions of each other it's really question of internal structure and culture like we see and I'm sure all you guys feel the same way Partnerships is an extension of dero and Deo is an extension of community and Community is an extension of marketing all of them flow into each other and especially in web 3 they all have to work together very tightly in these like Fast interative Loops I think that's how you get attention on community feedback like comms will highlight one thing and then that comes to Deo and we say okay is this a technical issue or is this an issue with a particular project does Partnerships need to step in uh if Deo has an issue they bring the rest of the team in so it's it's really about highlighting the issues and making sure that every single team member who might be relevant is able to see it and basically create a culture of like okay Community feedback is critically important just as important as an engineer's feedback might be in a Sprint that's kind of how you create a community-led or Community kind of feedback focused culture nice that's really good yeah yeah so I feel like you started off the question saying that I guess people don't care about Community or it's off on the side which I feel like is categorically false right am I am I putting words in your mouth cuz that's like kind of the first impression I got I I it comes in both it's like a a duel sometimes like people come in and like yeah I'm here for the community but I think you sometimes get to maybe more some projects who are like internally outside they're like yeah all about the community but internally it's like like Community will get we'll get to it or whatnot but I think a lot of people might like what I run into is some people like oh so you do marketing and it's like no I don't do marketing and maybe it's not just internally but it's also like from the external people's like Viewpoint when they think of of a community thing they're like you're doing uh ads you're just you're like a brand ambassador like doing those types of things so I guess something in in that realm that I mean okay cool I I guess maybe because we just have like different maybe opinions on what community is like I don't think like at least for us our community team does not sit on the marketing team right um but I think what I will say about community that isn't brought up that is really difficult to navigate in terms of maybe like resources from like a Time perspective is how diverse the community is in with regards to their needs right because user Community is completely different from what a developer Community needs and when like all of our teams are really small right like how do you dedicate time to like I need to take care of my user Community while at the same time taking care of my developer Community somehow collating all that feedback and then letting like the product managers prioritize that and then like even within the developer Community it's split into two groups too right because you have the daps who are wanting to deploy on Main net and that's like partnership side and I think in general for a chain that's usually your first priority because when you launch you want people to actually be able to do things right but when you start focusing on that then you start to neglect individual developers people just exploring the space trying to figure out what's going on and then you're like okay how do I figure out all this like mind share I don't know I think the answer is always just hire more people to be honest I there's not really a good answer there either overwork people and burn them out until they leave or like hire people and care about their work life balance um um but I I will say I think from like the linean perspective Community has been like critical I think to especially how we've developed the idea of how we interact with users specifically like Farmers or people like that expecting incentives and how important like transparency is um about what's happening at the time how like every action or time resource thing that you contribute will actually like manifest later on cuz I think when you pull the rug underneath people who are spending their time like building with you um like transacting on you or whatever like I don't know where I started the sentence but the point is like it it's it's something that you get a lot of feedback on from the community and like addressing that and making sure they're feeling taken care of and heard and like understood that they are the core of your protocol y um is super important and I I don't think that's a marketing function and I don't think that is ignored at least like across the board I feel like we're all here exactly pretty nice that's yeah that was exactly what I was trying to get across so that you leted into it perfectly so thank you um we have just a few more minutes left I wanted to just leave it open for a few questions from the audience if anybody had any of the more project Focus questions or Community Focus questions uh if you got a question give that hand a nice high raise and our lovely ABI codes will come give you the microphone give it a little a waight questions for the communities I know it's been a long day oh we got one right there the guy in the beanie hey um uh question uh how can uh the these unique communities all kind of connect and work together to kind of solve some of the the Bad and the Ugly that was discussed on stage ooh who wants to take that one I mean I guess I'll say I don't know if it's like there is an element of we are all suffering the same problems and I want to say just from like a perspective of working at Linea it's actually a very collaborative environment from like the L2 side like we're all working together and thinking about how like we're solving problems um and I think in the same way like each of us are figuring out what to do with our communities when they are ugly and when someone fixes it or does something cool one of us immediately copies it which is fantastic I think so and that's what you're supposed to do right like uh regard like copy paste other people's success I think is the way to kind of bridge the ugliness away um yeah yeah I'll I'll had to that that and um we are working also with different slightly different Tex Tack and so at some point at the engineering level I think there there should be a certain level of standard standardization that will also bring these communities together uh they will we maybe make the protocol more interoperable we know already that our communities are I'd say overlapping uh we all within ethereum but it's true that there is technical differences that at some point over the long term I think we going to naturally converge through a specific design architecture that will also bring the communities together yeah I I would say like for the community members like I think there's this like idea of this like L2 War it's like complete BS like all our teams know each other a lot of people probably could have or might have worked at other people's teams like there's a lot of overlap there's like no war happening over here so I wouldn't think about like pitting l2s against each other it's like l2s against the world we're trying to grow the pie like just try out everything if you have ideas for interesting applications try them out everywhere get engaged in each community I think yeah the pie that we're trying to grow out is a lot larger than what we're fighting for right now um so just be actively engaged in what we're building right now and I think it's going to be really exciting in the next like 5 to 10 years yeah and I think I I'll go a little bit of a different way I think it's just like what I try to do with like my community is just remind people that it's not a big deal like it's it's chill it's like you know it's it's not like life or death you know like there's no L2 War there's not this crazy like you know endgame or whatever it's like a really open collaborative like research space you know and so I guess when things get a little ugly or a little bit crazy it's just you know we we have like our weekly like Community calls and we talk with our community and there you know it's just it's really like getting that message across that like you know these things aren't serious it's not life or death you don't have to pick one over another it's just like come check out what we're doing we're doing cool stuff you know if you like this better you know that's cool but you know we're always around we're always here to help you guys teach you guys you know be be a a Launchpad for your project or whatever so yeah again we all have overlapping communi so I think it's just reminding people that it's not so serious all the time yeah that's great well thank you guys all again for your time today we really appreciate it uh and I just hope that there was something great takeaway for people in the audience as well just wanted to Showcase that the community as much as the rainbow and the fun uh and the the unicorns that are out there uh there are different sides to it uh and there's also this deep complexity level of knowledge that I'm I'm super impressed that a lot of people think that Community can be a little bit higher level but when you're in it you're you're cross functional you're going between different departments so you're absorbing the different knowledge along the way because you are the first like part of the funnel of people coming into the community so you have to have that that layer of knowledge to help them along so thank you everybody for participating The Good the Bad and the Ugly and uh thank you for everything you guys are doing for the ethereum ecosystem we really appreciate it thanks for having us than thank you very much thank you very much all right uh so yeah I'm giving a presentation on 4844 uh so who is familiar with blob transactions okay perfect I saw someone pointing at me thank you uh be kind of awkward if I wasn't so yeah basically we we're going to go through an overview of what 4844 is what are blobs how does a The Blob transaction life cycle differ from the normal transaction life cycle uh introduction of the new fee Market what changes uh we had to make to ZK sync and then what is going to follow so the tldr is blobs are cheaper uh storage on ethereum so before blobs we mainly stored Pub data in the form of call data and then with blobs we're now able to uh store it in an area that's uh relatively cheaper uh the back of the napkin math says it's about 93% cheaper when it comes to the amount of gas that's being used so as many of you may or may not know ethereum road map is moving to be rollup Centric and so that puts an emphasis on data availability uh so 444 sets the stage for this with uh full sharding while allowing rollups to take advantage of Cheaper data sooner so in the blockchain trilemma uh vitalic basically put it as a blockchain can only accomplish two out of the three uh kind of like sectors scalability decentralization and security we have uh those targeting decentralization and security like Bitcoin and ethereum um multi-chain ecosystems going for decentralization and scalability but with rollups and becoming a rollup Centric road map we can actually offload the scalability down to the rollups so ethereum would only have to focus on decentralization and security while allowing rollups to really uh put an emphasis on the computation and now so how do blobs kind of roll into this whole thing so with blobs we have the new storage type that's cheaper uh but it's only the reason it's cheaper is it's only being kept for a short amount of time around 18 days and now you might be thinking this becomes a problem but for many rollups 18 days is more than enough zero knowledge rollups will really utilize this data uh within an hour or so as we prove blocks uh prove batches to be valid uh about on ZK sync and an hour after they're committed and then on optimistic rollups that data is only needed for 7 days as uh that's the time for the fraud proof challenges and now each blob is able to uh handle the amount of data that we have in call data and given the new um the new targets we can actually fit three uh there's a target of three blobs per block with a maximum of six blobs it's not a ton right now but but this is really just setting the stage for uh full DK sharting that will introduce up to 64 blob slots and it basically allows them allowed ethereum to move the necessary changes for the consensus and execution layer to now while allowing full DK sharding to uh be focused on later and that covers data availability sampling so each of the consensus layer nodes doesn't have to hold each individual blob which would add terabytes of storage for them and in addition uh there's the introduction of the multi-dimensional fee market so it is really just like a complex way of thinking uh or of saying the way that blobs are being charged for changes based uh changes versus uh how normal transactions are being charged for so if you're used to gas going high for an nft mment or an airdrop or anything along those lines those won't affect the uh amount of cost that the layer 2os will have will incur because the the market for blobs changes differently than those for regular transactions so we can go into the blob life cycle so when a user submits a layer two transaction those get uh basically sent to the rollup operator batched together with a bunch of different different transactions for everyone and then sent to the layer one uh men pool once they're in the layer one men pool they get split off um into the execution layer payload and the consensus layer payload uh blobs only being available on the consensus layer payload and on the execution layer we're able to get a commitment to The Blob and that's how we can ensure that we know what data is in there um it's the same uh data that we assume and um it'll help us uh reason about it in the future and so this is another diagram kind of going through the life cycle uh I think it actually does a little bit better job explaining uh how things get split up so as you can see on the top right uh once the execution payload and the blobs reach the beacon proposer they get split up into the sidecar which is the kcg commitment the block blob and the proof that the kcg commitment references The Blob and then the execution payload which is everything uh including the hash of these commitments which we'll go into a little bit later so now if we don't have access to the blobs on the execution layer how can we reason about what's inside the blobs and that kind of uh sets the stage for kzg so kzg uh which stands for Kate and then I should remember the other two but it's the last names of the three people who came up with uh this zero knowledge uh proof and basically it is a way to form a polinomial commitment of the data and so if we wanted to know what's in the data we could just hash it and then on the execution layer we could just have a hash to that data but once we want to uh reason about what's inside that data we would have to then provide the full pre-image which would then be inefficient and basically um removes the benefits that we get from the kcg commitment instead we can take that data form a polom and then uh basically make the commitment to that polom what that allows us to do is we can uh we can prove what the data behind the polinomial commitment is by making a proof of by evaluating a point uh in that polom and basically saying I know that the value at this point is this value and over a certain number of points you then become more and more uh sure that the data in the blob is actually the pub data that uh you're committing to and so these the benefit of these commitments is they're relative they're significantly smaller than blobs so for instance a kcg commitment is about 48 bytes whereas uh Pub data which is the main reason we seal a batch on ZK sync goes up to 125 kilobytes so that is significantly cheaper and the people who ends up seeing the benefit of the reduction in this amount of data are the end users and so uh I spoke about kind of how we take the kcg commitment and how we re and how we uh are able to prove what data it is and this is done through a proof of equivalence and with the proof of equivalence uh we basically uh onchain we're able to take the version hash from the execution payload uh we use that along with the uh opening point and the opening value and the proof that that o that the version hash corresponds to a kcg commitment that when opens at the opening point will yield that opening value and with that we're able to then uh prove everything we need to know about the blob uh this is for those who are a little bit uh more mathematically uh inclined this is uh vitalic post on how uh the proof of equivalence uh can work and so we started to talk about this multi-dimensional fee market and how blobs are being priced differently than normal transactions and how this helps when eth gas price surges and so with that uh they operate kind of like the EIP 1559 where once we hit the target number of blobs that we want in a particular block uh the gas price will increase by 12.5% and when we re when a block has less than three blobs will actually decrease it by 12 a half% and this makes it so that uh we can always be sure that we know when uh our transactions will get included and uh when there's a surge in the normal um when there's a surge from demand for either again nft mins aird drops the rollups don't have to pay for this burst and data nor do they have to delay sending uh uh their L1 or their L2 batch commitment uh waiting for the burst price to go down um so this is talking about the block size so as we can see with Proto dang sharting we have an average case block size of 375 kilobytes and the real reason why Proto sharting was introduced wasn't necessarily for the average case but you can see how in the worst case the block size is actually significantly less than what we see in the status quo which is uh using call data or uh versus a different EIP which was introduced 4488 again not to be confused with 4844 uh which would essentially rather than introducing this new cheaper storage uh actually would reduce the amount of cost that uh call data would have so now the most interesting part how does this translate to ZK sync so in ZK sync basically the way it works is we have uh our security rooted in proofs and you can think about our L1 batches as going through three major uh steps we commit the batch we prove the batch and then we execute the batch and so when we commit the batch we make changes to the batch commitment now to include both the pre-image of The Blob that we're using as well as uh we add in the blob commitment The Blob commitment has the version hash the opening point and the opening value as parameters and so these values will then get included in the overall batch commitment and proven later on within uh the proed batches step and so with uh going from call data to uh blobs as I mentioned before we actually bring it down uh the potential 120 kilobytes to 144 bytes per blob and our uh system will handle two blobs so that'll be about 288 uh bytes which uh allows us to have double the number of uh PP data than we could previously have uh so that means uh with more B uh with more transactions and more statei within the batches our overall uh cost will go down per transaction and uh another important thing that we wanted to include within our contract changes was not just this ability to utilize blobs but in the event that the uh blob fee Market decides to go Haywire we want it to be flexible enough to uh switch back and forth between uh blobs and call data and so this allows us to basically retain uh the finality that we currently have while uh giving us the flexibility to either go opt for cheaper transactions or for um yeah and so with the batch commitment we also have to make Chang ches to our prover um which basically re creates the changes that we've done in the uh the commit batches step and So within our proof system we'll basically take the blob we'll repackage it uh in order we'll take the pub data package it into our blobs compute the kcg commitment uh along with that the proof that of uh the blobs so basic uh so the fact that the kzg commitment references The Blob and that using these uh blob commitments that we've introduced in the commit batches step that these are actually the values of that and proving that the and proving that all these values together will essentially uh give us what we've committed to on uh layer one and so going into what's next uh for ZK sync uh right now as I mentioned we support two blobs but in the future we plan on supporting any number of blobs as the number of uh blobs overall supported per block increases from 6 to 64 but also with tank sharting we have more blobs coming uh the implementation of data availability sampling so each individual node at the consensus layer won't have to um won't have to store all the blobs but can just store a certain part uh we'll have proposer Builder separation which is needed for uh increased security around uh different things like me as well as uh a new 2D uh kcg scheme and so I think we have a a little bit of extra time left if anyone has any questions about blobs or anything um about that all right oh I see uh hands up there I'm sorry oh so the question was who is going to be blob ready on day one um so I can't necessarily speak to the other chains uh because um we're not 100% sure uh what their progress is there but I can confidently say that ZK s will be blob ready on day one nice thank you all right thank you everyone and yeah