Back to videos

Video to Elon: How to Build Decentralized Twitter Using Doge and Cardano for Profit and Pleasure

Sunday, November 6, 202257:2958,860 viewsWatch on YouTube

Full Transcript

hi everybody this is Charles hoskinson broadcasting live from warm sunny Colorado always warm Always Sunny sometimes Colorado so today is November 6. I've eaten my Wheaties worked out meditated a little bit I got myself a ridiculously large bottle of water good old Atlantic water it's pretty good stuff so I'm ready to go you guys ready to go this is going to be a long video it is live I haven't really prepped too much for it just been thinking about it so it's going to be a big stream of Consciousness and we are going to talk about how a person would build basically decentralized Twitter using the cryptocurrency industry best available components and pieces that we have so I am going to do a whiteboard and I'll just share my screen and we'll just kind of start chipping away at it little by little and hopefully we get somewhere so it might be a few hours might be 30 minutes I don't know it's dream of Consciousness and this video is to of course the Twitter staff and Elon Musk I hope you guys watch it enjoy it and my hope is it gives you some inspiration as you think about what is required to actually build a decentralized social network okay so having given you a little Preamble let's share the screen there we go okay all right so first things first let's bring up the Whiteboard now when we talk about a decentralized social Network that's a kind of a difficult thing because we don't quite have a really good term for what exactly is decentralized technically a social network is composed of thousands or millions or even billions of people so when you look at the social network side it's already kind of really decentralized you have all of these people floating around doing things and they're connected in various different ways and information is Flowing between them but there's more to the story than just nodes with connections between them social networks tend to have certain nodes that have a high degree of clustering there tends to be a small diameter okay so diameter is basically the you pick any node can you get to another node in a certain amount of hops this is the Six Degrees of Kevin Bacon if you guys are curious to look that one up okay and you tend to have a situation where there needs to be some sort of overarching curation and that's where the centralization tends to come in so that curation usually is behind storage information flow computation there's AI assisted stuff and there's dozens of other factors that make this quite hard so in the cryptocurrency space we're pretty good at peer-to-peer we're pretty good at protocols we're pretty good at crypto cryptography pretty good at incentives engineering okay we're pretty good at a lot of these different types of things not necessarily super good at AI we're not necessarily good at decentralizing computation we're not necessarily good at decentralizing storage curation well we can think of that like maybe a Dao for organizational design you think about maybe a smart contract who knows and so it's a really interesting challenge actually at its core of how does one build a decentralized social network so let's think about Twitter as the case study and then I have some papers and we'll kind of work our way towards a straw man design so in Twitter you as a user you sign up all right you create a profile okay so you create that profile and let me tell them all about interesting stuff and then you select your preferences and then you've entered the great Network okay the Twitter Network at that point all right then you can view the Tweet space that's what I'll call the place where you can see everything so it's that box where all of your tweets are you have tweet one two three four five six seven okay so when we talk about curation we go up here and we say hey curation really curation and information flow that's the algorithm that Elin is talking about saying potentially is biased or there's an issue when people talk about Shadow Banning this is what they're talking about when people talk about bias this is what they talk about because what you see as a user is completely dependent upon some form of an algorithm that basically filters the flow of all information from the social graph so all of these different users these different nodes as they are sending Tweets they have to be ordered and some decisions have to be made of what you see when you see it and how you see it okay so this is kind of the Crux of the Twitter experience and each step here from how you sign up how you create a profile how you select your preferences or they are discovered thank you all right that's kind of like AI assisted the more information they have about you the more they can infer what you and what you don't the end goal is to give you the best possible tweet space view of all of these things then if you can view it you can interact with it okay so this is step four and then five is interact with the Tweet space okay and you don't actually have a lot of behaviors you can do there you can add a tweet so you can tweet okay you can comment so that's reply you can so I guess that's a little heart thing that they have there you go you can quote okay so that's just basically a reply plus retweet okay and then there's retweet which allows it to spread and that's usually what's called a Contagion Factor how quickly can things spread throughout a network and there's all kinds of models that you can use to study those types of things so algorithm algorithmism is built to maximize your level of Engagement and the time you spend in the Tweet space so it wants you to be scrolling forever it wants you to be interacting forever it wants you to be liking and quoting and commenting and retweeting forever okay so one of the first problems with that approach is that this has nothing to do with the quality of information in fact it's actually the opposite it's negatively correlated high quality very deep information tends to be nuanced and require quite a bit of time to digest and what that effectively does is it either takes you out of the Tweet space okay or it slows you down and you don't want the algorithm if you're maximizing for advertising Revenue to take a person out or slow them down take a person out means they've clicked a link and they've left your platform they've gone somewhere else they're reading an article they're doing something they're not in the Tweet space they're they're not doing stuff slow them down means that they're not adding commenting liking quoting retweeting okay because some of the things that are inside the Tweet space are actually advertisers all right so people pay to get your app eyeballs so at the end of the rainbow It's always important to understand that you are the product here when you create your profile and you select your preferences these things allow the network to micro Target you foreign and the advertisers pay Bookoo money to be able to do contextual ads and they want to make sure that you are always always either looking at their stuff commenting on their stuff liking their stuff quoting their stuff retweeting their stuff and the only time you're taken out of the moment is if you're going to an Advertiser so the algorithm is optimized with a disincentive to go to good information so non-paid information high quality information it's incentivized to maximize your level of participation with Advertiser content either to see it or directly interact with it but other than that just keep sharing just keep participating that is the current state of Twitter and that's why everything is so up in fact there's great books one is called surveillance capitalism and there's a Litany of other books that if you just go to Amazon and search you can find them and recently published that talk about how these types of algorithms work on Facebook Twitter and other places so if we're going to build a decentralized social network the first question you have to ask yourself is do you want more of the same or do you want something different just because you go put the D back in decentralized doesn't really guarantee that this algorithm is going to go away okay unfortunately this guy is has been around for quite a long time and unfortunately this person makes billions and billions and billions and billions and billions and billions and billions of dollars every year for Google for Facebook and for Twitter they're not exactly going to go ahead and fire them that's the last person to get laid off and so you not only have to figure out how do you handle the curation challenges of billions of messages how do you store not just terabytes but EXA scale exabytes of data how do you get the network resources to allow exascale data to flow every video shared every picture shared every tweet shared if just to give you a sense of it for Twitter's traffic if you sum up all of the cryptocurrency storage so every single blockchain storage okay all the storage of Bitcoin all the storage of ethereum all the storage of Dogecoin cardano so the size of every blockchain with a full node you sum them all up together for the entire 13-year history of the cryptocurrency space that would be less than an hour of traffic on Twitter okay so you can't use a blockchain structure for all that data flow and storage and moving it all around just will not work in any cryptocurrency and obviously we don't really have a lot of cool computation and AI assisted stuff all right lots of transactions lots of things lots of streaming lots of discretization so we need massive infrastructural Innovations okay and then we need over here a business model innovation because the algorithm is not built for what we would like to have which is that over time the quality of info increases that's what we'd like to have we wouldn't a situation that over time the quality decreases and unfortunately that's where we're at and the social media space all right well let's start chipping away at this a little bit so first off let's start in the very beginning the representation of people so we have a pretty good solution for this if you look up here we have this concept of a dead a decentralized identifier and this is an open standard from the w3c so the w3c is not just a small organization these are the guys that standardize a lot of the cool stuff inside the web like HTML and CSS and so forth and for a very long time many many many people my company and others Microsoft IBM have been working very hard at trying to create a standard for representation of identity not just of people but agents so smart contracts iot sensors autonomous vehicles and so forth so right off the bat you should represent your users as dids okay for your candidate solution so when we go here we go down here I'll change our colors up a little bit we'll put some orange in there all right so so when you sign up you get it did okay and that did has two parts there's a did address so did ID and then there's a did document the ID is basically just a string of characters kind of a web address and the document is the actual description the metadata and everything that's associated with it so they're interconnected to each other so when you sign up you get a dead and when you create your profile that's you filling out the document and you have two workflows here okay you have verified and unverified so your first major decision point you have to come up to when you create these types of things is what level of verification do you want your users to have and who is responsible for that level of verification that the users are going to have so there's some cool protocols like for example we hunchang over at Virginia Commonwealth University created something called proof of human and there's some follow-up work that could do it's a super captcha so you can certainly verify that a person is not a bot there's plenty of different ways to do that okay so you probably don't want somebody to be a machine so no robot so no no bot under the unverified you have to say well people are going to make a lot of claims they'll say I I'm a doctor I'm a lawyer I'm the president mama and usually what ends up happening is you have some form of an idv a provider like accurate or others that can help you sort out and the credentialing bodies that can help you sort out whether these things are true or false okay so you just kind of have to work your way through that so during account creation there's a question of granularity and how much you want in that profile and how much is going to be verified the profile most social networks because they want to maximize user acquisition they make it very easy to follow this flow so you can get as little friction as possible the problem with that flow is that you end up getting lots of false accounts or inaccurate accounts so basically accounts that are kind of bad okay so dids are great because it's now an identity ecosystem it's blockchain neutral it's standards and there's tons of infrastructure that's being built we have some called a teleprism that we designed for cardano but it's also cross blockchain so it'll work on bitcoin ethereum and all kinds of things and prism V2 is coming out in q1 of next year and it's based on everything that we did in Ethiopia dish and other places and it's a massive improvement over the existing framework that we have so we're pretty good at the issuance you can store these on a blockchain the cardano blockchain or Dogecoin or these types of things and then on the document side what you do is you create an API ecosystem [Music] where you have verification as a service okay so basically people come in and they offer the ability for people to verify things on their profile now right off the bat I think what we should do is augment some of these standards a little bit and there's actually a piece of crypto that the algorand guys have been working on that is quite good and I want to give a shout out to them and just point it out as an example of something we can add here so over here there's Falcon fast Fourier lattice based compact signatures over and true so this actually they've been working with nist and I believe it's becoming in this standard Falcon these are some of the smallest and best signatures it's a post Quantum signature scheme and they've done some great work cleaning it up and making it where it needs to go but this is basically something that seems to be resistant against quantum computers the underlying hard problem is a short integer solution problem sis over injury lattices for which no efficient solving algorithm is currently known in the general case even with the help of quantum computers so right off the bat not only do you have opportunity to represent identity in a new way which is universal and create a Marketplace for verification where third parties can come in and verify all kinds of things about people you can actually put things like Falcon in and Associate them with the did document so you can have next Generation cryptography yeah that is immune to quantum computers okay so why would you want to do that well when you enter the Tweet space here and you tweet or you quote you're actually adding content the problem Twitter has is that none of that is signed foreign so if somebody hacks your account they can impersonate you okay so if you have a signature system that sits with the creation of the Dead and you put a good thing like falcon or any of these other Next Generation things then then you actually have an unforgeable signature that you can use for every tweet you produce and you can associate that with the user so you can always prove who you are you can sign contracts with it and every tweet you have is verified now the other thing is that you can layer more sophisticated crypto on top of them that's what we're doing with the prism framework so our goal is eventually to have zero knowledge crypto right now a lot of people think about zero knowledge for scalability and privacy but you also can use zero knowledge crypto for basically Boolean proofs about things like for example did you pay your taxes or not are you us resident or not are you allowed to see this content are you over the age of 18 yes or no like for example there's a question of adding only fans Style content to Twitter elin's been talking about that so allow content creators to charge people to see their videos almost certainly some of that content will be adult content so you need an age verification okay and these types of things so the framework that allows you to extensively add crypto and add more crypto and more crypto eventually allows you to build up to things like one-click age verification and creating a verified Marketplace basically allows third parties to come and provide those services and do a profit share with the provider in this case it'll be Twitter alright then you select your preferences and that that's fine you can just copy the existing experience that Twitter has but then when you interact with the Tweet space this is where you start making some major changes to the guts of what you're trying to accomplish so if your goal is to never take you out or never slow you down and just have you continue consuming consuming consuming consuming so the robot up here can keep making money then you leave it as it is but I'd like to believe that we want to innovate we want to do something cool so let's kind of break this down first off when you when you take a look at a piece of content whatever that content is so this piece of content it has a bunch of claims that has some evidence it has an origin okay there's all kinds of stuff that you can put inside that bucket and the the question is do any of that stuff okay so if if somebody says something like Trump died of a heart attack breaking news in Florida all right that could be a tweet you'd probably see that tweet and so that's a huge tweet right it's a huge claim so wait what evidence is presented who is the person tweeting it there's all kinds of things that you're going to think about whenever you see something like that breaking news okay is it Reuters that just said this is it just some random account so there's a lot of referential stuff now the right now my options to interface with it is this heart okay I got that little thing here I I got my retweet and then I got a a comment that's all I have so if I heart what the does that mean I don't know if I retweet I'm retweeting without context I'm not making any assertation about it but what I am doing is spreading it so I'm increasing the contagion factor of it I can certainly reply to it I can comment on the tweet and that reply I could say I don't believe it or whatever but I'm engaging with it and what's happening is the algorithm is actually the more engagement it gets more likely than that it's going to enter more and more people's tweet space why because it increases the level of Engagement for everybody it's controversial news have we gained anything from this have we learned anything from this is there any consequence for people sharing it is there any consequence for people liking it these types of things especially if they know it's not true so what you'd ideally like to do is to improve the Nuance of these mechanics of how you interface with things that you see inside your Tweet space so for example have many more options than just alike so you could have things like suspicious you could have things like dislike a down vote so you can start curating these types of things you can have things like tell me more fact check etc etc you can create a whole catalog of things that people can request we are dealing with crypto the superpower of crypto is it's a payment system as well so the power of having a payment system with tokens is guess what you can do you can put bounties so you actually offer money if people do certain things so you already have a smart contract option and a catalog of requests that people can have and you can create a marketplace around desires for curation retweet you can have retweet with context so that's just a stronger form of clicking a button you can add all kinds of things commentary and so forth people would argue that the quote is this way but there's got to be some simpler mechanisms like for example something to slow the contagion down it's I think this is actually fake news as opposed to real news okay and then replies they tend to get lost pretty quickly so the problem is the system tends to lack memory unless you get a really Zinger of a reply that particular tweet in its entire thread that tends to deteriorate a little bit so you could actually do all kinds of things with the cryptocurrency in this respect you could turn each tweet into an nft for example and it can have a reputation score by the way at account creation you can have every user when they get a did also associate that did with an nft and this is called a progressive nft so a progressive nft is one that grows over time we're doing this with crypto bison and we're doing it for entertainment okay but you can do that if you really want to so if you treat this as something that is an object onto itself it can be taken moved around extracted analyzed played with old conversations can be archived pulled up they have a copious amount of metadata and then also you can start inferring a notion of reputation and that reputation can work its way to the people who interface with the information so if you have a habit of sharing things that turn out to be misinformation like every Thursday you kind of tweet out Trump died of a heart attack or these types of things and he didn't you're proven wrong then you can be lower on the totem pole and if you have a habit of sharing really nice information then suddenly you create an account that actually has more value behind it more weight behind it the other thing is that in these marketplaces these can become what are called prediction markets Robin Hansen is the big guy here and basically you can create a Marketplace for a person's belief of truth of the thing that's been reported and there's money behind it so it's like saying well who's going to win the election on Tuesday we have a big big election in the United States and on Tuesday November 8th well some people say well I think the Republicans are going to win and some people say I think the Democrats are going to hold the house well it's one thing to say it it's another thing to put money behind it are you willing to put a hundred dollars down that the Republicans are going to retake the house are you willing to put a thousand dollars ten thousand dollars hundred thousand dollars and so forth so you can create marketplaces for the truthiness of information and you create a bounty system for example the person who tweeted this they could also even put what's called a veracity Bond so a veracity bond is basically money that they put on the table that they stand behind that tweet and if it turns out that that tweet is false they lose the money so all all of a sudden your Tweet space looks very different you have much more nuanced tools that are extensible to curate the information and you have economic agency behind the information okay so you can put money on the table you can create prediction markets on the table for these types of things the people tweeting and sharing it have reputation so they have something to lose for veracity bonds can be put into place they can lose money in these types of things you also can have a situation where there's a realm for incompleteness so all kinds of news is shared it'd be nice to say well is this unverified is this verified under what circumstances is this verified and so forth this curation right now in the centralized social network is where most people are running into the big problems with Twitter Facebook and these other networks because they're sharing things that they think are true but then the fact Checkers are basically saying we don't believe that they are we think that they're false and not only are they de-platforming the content meaning removing the content from the Tweet space they're actually creating account level measures and I.E Banning people for sharing what they feel is bad stuff okay so we're just kind of working our way through this and so we've already discovered that we can sign tweets and by the way having digital credentials associated with your account means that your accounts are all super secure it also means that your users can move around and the reputation scores and the users themselves can actually live in a decentralized protocol when you follow somebody instead of just following them you should actually subscribe to them a pub sub okay and we've been working on that technology for quite some time so this was something in the incentivized testnet that we actually play around with it was called poltercast it came out in 2011 and one of the use cases they actually were talking about is a decentralized Twitter so the basic idea of poltercast was a decentralized pub sub protocol now we learned an enormous amount from poltercast and actually what we did is we contacted the author of the paper and cut them a check and now we're updating the paper to create a better Pub sub protocol based upon a lot of things we've learned and a lot of things that they learned but actually some of the people who participated in the decentralized incentivized testnet they they'll tell you stories about their experiences with it but basically a follow can be now much more meaningful because what you can do is now subscribe to a feed now the current Twitter algorithm just kind of mixes everything up so when you follow somebody it's not a guarantee that the things that they tweet will appear inside your Tweet space but what you could do is break it from one unified view to an extended View so let's go there this is kind of now going into the realm of what Mastodon does but you can have this idea foreign of these extended domains so your Tweet space would kind of live here but you can choose to move your eyeballs into any of these areas and these can be different curation rules these can be things like my friends or my follows so their exclusive information these can be things like trending and this is kind of a representation of that aggregation but you can have as many circles as you want now a protocol would treat this these circles and the Box as programmable foreign so that curation is something that you would be able to select so for example let's say that you're an Advertiser and you only want your ads to appear under a certain curation regime okay we'll we'll call you super woke Corp not to quote anybody but you only wanted to appear under a certain curation regime something that lives here okay well what you can do is say well I'm only going to have advertised content in this particular box because it conforms with my values so my brand knowledge reputation will not appear over here that's a different part of the world and then over here you could have gun Corp evil gun Corp here we go Gotta Be bipartisan in your criticism right so evil evil gun Corp and they they say I know we only want our ads to appear here and and we don't want anything to appear over here so these circles and these boxes are quite programmable and really you have to think a lot about how do people announce to the network who they want to follow that is basically a pub sub you subscribe to people and then you have to have the ability to have multiple circles some that are curated by the network some that are derived some that are curated by the user and then some that are are an effect of basically some safety Provisions that are put in and people either follow them or they don't like for example you could say something when you have these new tools to share information and you start talking about the curation of that information you could say I only want to see things that have a veracity Bond behind them meaning that people have actually put money on the table so if CNN or MSNBC or if Fox News they they claim hey we stand behind our journalism then you the user can say Okay CNN MSNBC Fox News all these other people Bloomberg are you willing for every tweet that you make about your stories to put a thousand dollars on it and if you turn out to be wrong proven wrong you lose the money that's a very different thing yes and by the way if you can nft these things they have a history behind them they could be traded they can be owned they can be referenced and recalled in their immutable time stamped and auditable they can never be de-platformed or taken down what does that effectively mean it means that you now can look at people's track record that reputation metric that you would apply to a user node when somebody signs up a person over here this guy would also apply to a corporation it would apply to a political party ubiquitous curation in that way oh suddenly the powers that be are held a little bit more liable aren't they and that's interesting so something like polarcast for the pub sub protocol okay you'd have to come up with a protocol for how you want to do your circles and your boxes so the boxes are multiple views so these the Tweet space right now you have a Consolidated view but you'd be able to switch between them it could almost be like Spotify playlists where you can adopt other people's views and these types of things and then you can also set your own and put custom logic and rules by how you want to build that particular View and that's going to be endemic on how the information is curated in that respect by the way there are people who try really hard to come up with some principles on proper curation of content and there's something called the Santa Clara principles on transparency accountability and content moderation and a lot of different people came together on this 12 big companies Apple Facebook Google Reddit Twitter and GitHub have endorsed it they don't really follow it but they endorse them and there's some content ideas and there's all kinds of people who kind of talk about it so this would be an example of a view so you could have a Santa Clara view so one of your Tweet boxes could be under that one of them could be Anarchy land and live here and you actually have both views and you can see both things and you can toggle between them and you can use prediction markets and other things to create basically marketplaces for information curation now how does an Advertiser interact with this because when you create this did document you're generating lots of personal data okay so you own your own did you potentially own your document well there's a great Venture I'm using the brave browser well the brave browser was built on this concept of the basic attention token bat you should expand that so what we should do is create an advertising Marketplace and what you do is you say that the user owns their own data okay and then you create a token this is a token to basically create a direct relationship between the ad and the owner they want to micro Target they want them to see things and if they want it to be placed there well these curation domains you can actually connect them and use a token to display ads bats has done a really great job constructing an ecosystem there don't reinvent the wheel just take that model refine it and prove it and then there's one thing that I'd like to add to it we have something called Kachina foundation's a private smart contracts so if you're going to do a lot of private data mining this is a paper that we came up with and we're working real hard on it then I think if you take that and you combine it with private data mining then you can have an entire ad ecosystem suddenly everything is starting to come together because this advertising token could also be used to give economic agency to the information advertisers could offer bounties for things for participation you could do tips you can do micropayments you can create prediction markets with it there's an actual natural use for this content for the creation of all these things also these are nfts if you want to make them which means that they also potentially have embedded intellectual properties you can create marketplaces to trade them so artists can just simply release the rights to their movies their videos other things as tweets and if people actually want to buy a license or not then they can pay with that same system that same advertising token effectively okay now how does Doge connect to this well Doge will not be able to provide any notion of any of this infrastructure all of this stuff is well beyond it because it's a fork of Bitcoin and I'm not being critical of the Doge ecosystem this is frankly true of most cryptocurrencies what this needs to be is this needs to be a side chain okay and I'll explain why in my thinking behind it this needs to be a side chain and it needs to be a web 2.5 application the goal here should not be to build some ubiquitous decentralized social network protocol because the problem is that you have to solve a decentralized network or decentralized storage and decentralized computation and actually completely resolve all of that and make it incentives compatible and profitable people are working on it the chias and the file coins and the heliums and the world mobiles and all these others to do these types of things but that's a lot of work and if Elon wants to do this next year or within a three to five year Horizon and he wants to use something in our industry to do it then it's just not going to happen in that respect so what you do is you do a side chain you say okay well the immutability the did space the censorship resistance the auditability payment system these are the kinds of things that you care a lot about and generally speaking what a side chain does is it takes a layer one okay so let's say cardano and we spent a lot of time effort money as an ecosystem to build a very solid route of trust there's these 3 000 stake pool operators so what you do is you sort them and then you create a side chain and then you can run it with a super fast bft protocol okay and here's an example of one narwhal and Tusk it was from George Genesis at UCL and a bunch of other people and it's a bft dag based mempool efficient consensus protocol 600 000 transactions per second without any latency issues okay and that's because basically they're able to bootstrap it from a trusted source so you could take cardano and you could do a super fast bft protocol and you get ultra high speed throughput and then what's nice about that is if you if you structure the security model in just the right way you get immutability you get censorship resistance because it has the same security properties of the main Ledger it's a blockchain so you get auditability and now you have an enormous TPS rate but what you can do is you can augment that side chain with service providers and they can provide the storage they can provide things like idv so for verification marketplaces and these types of things so this is actually really nice for the fabric ecosystem IBM fabric from hyperledger and the concept is that your service utilities you can express in chain code and you can write as many of them as you need to provide all of these things and then basically what you do is you take the distribution of Doge okay and then you do an airdrop they live here and Doge becomes the advertising token and the veracity token and the marketplace tooken for prediction markets and the token required to issue nfts and all these types of things and we can embed that with all the logic extended utxo so it can be a great YouTube system you can put Pluto smart contracts on it or other languages you can make an evm system if desired you do all kinds of things but I'd recommend doing cardano-like things because we understand that but yeah so you take something like fabric you put a super fast bft protocol on it you combine it with bespoke chain code to basically build a dogis ledger and then you add prism capabilities to this and that basically gives the identity layer so you have a strong crypto layer you have an identity layer and you have all this magic here and then you just basically do an airdrop of Doge over and users can redeem on this side chain the side chain would be able to handle all of the traffic and then you combine some of the things that dns's zendrose and others have been working on this this concept like proof of proof of stake these things this idea of chain free mining all right so he had log space mining these other things because you're going to have gargantuan amounts of data here and basically the people that maintain consensus would only have a small part of the overall State and then you'd have archive nodes that basically live behind and those archive nodes are an example of service providers and they provide that availability because that's the Twitter company that's their interface that's why it's a web 2.5 because there's going to be a lot of little stuff that is super hard to put into a blockchain space and really the core Innovation here is you're creating a Marketplace for information curation a Marketplace for advertising with various values behind it and a gigantic identity engine and that's what you want to focus on for your version too you don't want to figure out how do I build it a centralized storage system and all this other stuff you just basically have service utilities and you plug them in and these things would work quite well and chainless mining these types of things are the kinds of protocols that will get you where you need to go and there's already enough Innovation the bft space that I'd say within about 18 to 36 months you can have a bespoke protocol that would be able to meet the needs of Twitter's daily transaction volume in a sense and then of course all the data is here so AI can be added and marketplaces can be formed so this is basically kind of some free form thoughts but as you can see it's it's a very very complicated thing and we've only barely scratched the surface of it but we've already made ourselves post quantum we've already completely reconstructed identity around a new standard that's significantly better we've created a Marketplace for verification so Elin doesn't have to figure out how to verify everybody in every claim you can just basically put all an end we've created in the Tweet space instead of just an algorithm that's optimized to make money and then at the same time make you angry you've created a Marketplace for people to kind of derive the type of information they care to be you've created a reputation system and you can track not only people but organizations in that respect and you can now demand that media organizations be honest about the things that they fall on if they don't they lose money you can have different views for different philosophies and you can subscribe to people's views and so if you want to see what a left-wing person's Viewpoint and tweet space looks you can like a Spotify playlist subscribe to it you can do the same thing with the right wing person and so forth you have a token and that token is a basically Doge and you already have a fair distribution of it so you just do an airdrop and it's basically a bat token it's a web 2.

5 application you could take something like Fabric or any of these permission blockchains but a super fast bft protocol that's built bespoke for it alongside a pub sub protocol and some Network improvements and then you can create all these service utilities and you just have to get to a certain density of them and add chainless mining to it so that they don't have to have the state of the entire system technology like mithrol would mean that every single that's something we're building for card Auto but every single transaction that people see they're able to verify client side without having a full copy of the blockchain so given that these things grow so large you need to have that kind of technology for inclusive accountability what's nice is that advertising token as this network grows is likely to be worth tens if not hundreds of billions of dollars in a total market cap and as a web 2.

Found an error in the transcript?

Help improve this transcript by reporting an error.