EthCC 2: Igor Barinov

okay guys I'm going to talk about text day so nothing about security documents and regulations so it's more technical talk and exit is a what is X - X that is a first table stable chain with Bridget dies in a Jif token and it means that it's a it's a new block chain based on aphelion protocol with with a new token created from existing token from another network right so this concept is known for many years as a hard spoon when you start a new block chain and you use existing circulation of another token without creating additional emission you move part of the submission to a new side chain and this etching gets this token as the token of the same nature over the token of another nature for example this idea was thought about let's say a theorem when we can take ether token and we can move it to another chain and create negative ether token another chain and use it for example for scalability right and uh next day one token or form of LC 20 has a different nature of native token on the side chain so why it's important because how we think about this it's irresistible to use after you first time try to use X day it's it's very hard to use networks with volatile tokens especially if you are developing depth and you need the scalable infrastructure with predictable platform usage fees right so the one you develop on X day usually you have stable user balances so like if the user has one tie today it will have one tie tomorrow if you will not spend it so platform usage fees are so stable it means that you can calculate how much money your contract or your user will spend on on using your tab it also means that the the platform can use existing liquidity of die without creating liquidity with traditional methods like mining or ICO or bonding curves a recently and the platform can just use existing liquidity if this liquidity will be moved from from network to network right the the network itself can be set up with and set up with different parameters optimized for transaction speed and when the network has started it started with sometimes people called empty but we call it with record with a free capacity it means that network is always at least at the time when it has empty capacity accepting your transactions and execute them if they are correct right it means that when a user submit transaction to the network as net the transaction is accepted and that's for for many users it's a blazingly fast when the network has this free capacity capability right and also the the the problem of full capacity is what we think is overestimated in our in our industry and futures with full capacity and most chains are with empty capacity and it means that platforms of these chains and let's say foundations or entities which supporting them can subsidize platform usage right when you think about this some cheese offers three transactions or very inexpensive transactions some cheese offers meta transactions where someone is paying for for for someone else execution and at the moment there are much more resources free sources and infrastructures and plant infrastructures then the app developers and actual depths which can use all the all these infrastructures right so it means that this gene is is always ready to accept new new developers the consensus of this chain is is environmental friendly so it's not the proof of work we're going to talk about this on the next slide but this consensus is proven with with a scalability and security assumptions which is working in production since 2017 and the the idea of having this breached chain with different types of negative tokens in in this last scenario type of bridge token is scalable and secure from 2017 till night till now right the the main think about the consensus is of this exact network is like who is who are validators and how this validate herself well data set is managed basically well data set as a list of validators who are participating in creating blocks of the chain so this consensus is based on ideas of decentralized autonomous organizations and it's a self governed Dao so this network started with first initial validator which we call meister of ceremonies with a trusted set up so on the network with one body there is a heavy centralized network with centralized power and after it achieves next step which were called minimal viable decentralization this is when the the network can can be managed by this governance system where any listing validator can propose a new validated for this network and at the moment there are four validators in this network which is enough to keep a decentralized nature of governance consensus so trivial data's are required to add remove or validator and with for validators there is at least one valid data which can be kicked from the network and more validators can be added and the initial validators who started this network at POA make make a double proto fire in givethe there are five more applications from burner world alpha wallet Porteous it's interestingly many words applying to be holidays and my Krypton hurlan Network also discussed to be of Auditors so the technological advantage of this chain is that the old legend is based on the theorem 1.0 protocol so it's compatible with existing EVM and it has a dynamic validator set but by managed by this what we call exclusive self-governed decentralized autonomous organization the native token of the chain is hard spooned so it's it can be moved by by any user from if you network with and the network started without initial emission Island no emission at all for for this native token and with zero prime I'll write otherwise this network will inflate die on this X dot inside and it's it's this scenario is possible but but will be differences in in the imbalances and in case of mass exit from the sidechain it will not be possible to exit from the sidechain right so this native token is breached one-to-one so it's it's a form of atomic swab when the user when the user moves his die into X die this day is locked on the inside on the my net side in a smart contract of the bridge and a new X die is created on the X - inside so this is how the bridge is working I think this is the main innovation of the of the this type of chain so think about this like a user can buy a token die token on the on exchange on the theorem site and after he is locking this token in bridge contract on the theorem side and and after a group of Auditors observing this event and each of them submitting his own vision of of this transaction independently from from validators right now the security model is three or four of our data's are required to relate an event so one example for example when these guys and dye into inter contract and this day is locked the contract emitted an event and within this event there is a information about this transaction transaction hash addressed route which blocked funds and amount of coins locked in each validator observing this event on a theorem side wait till probabilistic phonology of the network in our case for eight blocks and after submit a transaction on X date inside saying that I observed this event on the film side smart contracts on inside after receiving after receiving three confirmations from three independent auditors regarding one token transfer center command to consensus of the of the chain and and the consensus omits block reward for the address which which locked died for for this event right so the it means that bridge directly included into consensus of the of the side chain and native talking in native tokens can be produced only by external events from the from the bridge the one the user when the user wants to exit the bridge he sends transactions to the smart contracts on X dash inside and with his ex die coins the burned on on X decide and an event is produced on by smart contract on X decide and validators observed events on the X dash inside and after delay this event to do with helium right and after three hundred three affirmations three affirmations combined on totally burnt X die the lid is unlocked the corresponding amount of die unlocked on my net side it's a bit complex when you think about it for the first time but this this idea of analyzing events is very flexible so on this bridge it's used in a form of lock/unlock of ESC 2013 and mint and burden of native token the same model can be used to lock/unlock FC Twente and lock unlock FC Twente on the other side or lock unlock of ESC 20 and mint and burn of ESC 20 so for example if you would like to let's say have died on Tron right you can say lock unlock die on the film side and mint and burn representation of die on Tron side right using the same britain same concept so the bridge itself I think we launched the first production bridge in the D theorem ecosystem as far as I know and on our bridge for for the for the pure network exists in some made 2018 and it holds millions of tokens and more than million dollar worth of tokens but this exact bridge forest is quite it's quite small nowadays the market couple the whole network is around twelve twelve thousand dollars and this is interesting because this actually shows how the network is used and that the token is not speculative the chain itself is not competing with with platform of a video so it's I think about this as a like temporary scalability solution before we have you know if you 2.0 or or other extensions to the platform right ok so let's let's compare a little bit so on on my net the the the big problem with with platform usage what I want to stress here is that platform usage is two dimensionally volatile right so you have token price which is volatile and also guest price is volatile right and when I feel our prices up and they will have cryptic edges then the platform is very expensive to use usually side chains and another platforms can say ok now on our platform we can subsidize gas usage but well for for unions of accounts you still need to use a volatile tokens right and that's that's also not convenient for many use cases and what what is a very attractive to developer so next day that platform usage is table because token price is non-volatile and and gas price is subsidized the both platforms are what we call like X I chain and our first side chain which is pure a what we call with minimum viable decentralization and we plan to increase level of decentralization by introducing new economic mechanisms to achieve security on side chains so the slide we made today I'm very proud this is a scale in scale out idea of say cheese now basically what people are asked what people are asking now you have frico prices here but what if many a depths will migrate or you know start to use this platform how are you going to achieve the same parameters that you have now right and in traditional system engineering like cloud computing we have usually two different approaches or combinations of them so one is vertical scalability right you purchased I don't know m3 nano instance on AWS and launched our website but after you have users what oh right you can you can upgrade your instance to another instance and basically get more storage faster storage more CPU right and more RAM and so that's usually three things which we can modify to beef up our node the same can be here if we think about parameters of off site change we can decrease to certain amount block time we can increase block size and there are side chains of a theorem with relatively big blocks and small block time for example and the centralized chain called goichi they have 136 million gas per block and four second block times and you can achieve like six hundred transactions per second with a single token transfers and and it's interesting that you know there are some public networks where you can actually see that if Hume can be scalable on using this vertical scalability but this vertical scalability has limits right because you know the rules of physics and and implementation details and also we need to think that this type of systems are distributed so we have network latency and you know data propagation and all network effects right so but it is can subsidize essential dubs that's what we implemented on on XD chain for example when we started this network we as like first group of Auditors decided that the bridge is the most important part for the for the system and without bridge the system may have like one additional risk for users which is called exit problem when users cannot exit their ex die to die because validators don't have money to relay the transactions so we whitelisted the the bridge on the sidechain side and which is basically free to use for validators and the same way some essential applications on the platform can be subsidized and also some essential like new op codes can be implemented on on new side chains of the same type think about privacy-preserving XD chain on this type of chain you may you may have you know bulletproof source Iggy snacks with less expensive than on the field right because as died now use the same guessed price for for for elliptic curve cryptography which is from my perspective too expensive and it's easy to understand because it's it's made to protect my net from from types of attacks like you know Shanghai attacks so second scalability is more interesting it's horizontal scalability and when we with out especially without shared security layer so differentiator of our approach from from approaches of other cheese with shared security is that we think that we can provide a set of tools and people can take this set of tools and launch their own chain if they want right and we don't force them to use our security model but the security model of Opus Dei and bridges are proven and audited and we think that okay each component is is proven but if you set it up not the right way you may you may have less security than a reference system and we think that can be different types of side chains and that's a beauty of this model where people can experiment with their own implementations right so first of all people can use different native tokens after we launched the side chain with xi1 project from from Israel Colo they are they launched their own network network based on the same idea of bridged EAC 22 native token but they used their own ear set origin instead of die right they don't need stable token they more interested in having like dub chain where the chain is running on their own you see 20 token right so it's possible to run the same chain with type of for consensus as we run or with our delegated row for stage you can run this type of chain with like one body there is a centralized system one one group of people asked is it possible to whitelist only one application and have extra chain with in public space but only like for one company it's possible to run exactly without without a fee room itself right so it can be a different platform so there are many variations of the same idea and we think that this idea of double token model where native token and the application usage token is a stable token and all other tokens are complementary tokens it can be used for very important aspects of the network like providing security or providing utility of this network but this model of stable is negative token astable and other tokens are supplementary will be used by by my major networks because it's very attractive in terms of platform usage and also it's it's becoming more and more important nowadays to to like decide in which echo system you are working right and not compete with in the second system because it's relatively easy to be like to switch to another echo system and you know kinda it will be hard to get you know everything within within this existing data system well and let's take talk about like a system the the X die is a project created on each building hackathon with but but we didn't hack it we just created like the ID on sponsors dinner and and it was like a half a year ago but now several partners already support support this this chain I'll start from curated registry of depth think about like Yahoo for for dubs and where you can submit your application and it will be with with and you can compete with with other depth of other platforms it's kinda hard to compete with us and steam but it's interesting that many that many applications on on the new types of chains can be discovered through this aggregators right it's also important to have wallet support the most well-known application of exodus burner world which was used developed by Austin Griffith and it was used in at the largest pop up economy on earth Denver hackathon and thousands of people participated you know buying food and drinks and also vendors participated in of ramping from X date time from die to USDA this effort created by by the community without any it's a central project management team so it's very interesting that many partners combine their effort together to bring this into this to this event and we see that this type of wallet is something that popped up on DEFCON 4 and now it's you know many people having tried tried this experience right and I personally would like to see this burner world mode in in other wallets where you can open an existing port and have this mode of partner wallet where you know if you want to use this type of world you can use it like within any world like incognito mode and browser right so now before the world supported by multiple wallets and and some some wallets like which is interesting trust world is planning to support it and unbalanced world and one project built built on if the americathon called zi is also planning to release and bring those transactions to X die and to burn our world the important part is analytics our friends from Black Sea launched the analytics of the of the if Denver event and after support the the analytics and you can see here you can clearly find where is easier right it's somewhere in the middle many transactions and large amounts of transactions and transfers and amount of tight transferred through this day days but the the system is its amount of centers per day right so during the Denver it was like 100 centers per day but but now it's it's not drop significantly right so that people continue to use it and centex die to each other which is quite good for the adoption right we we @p we developed open-source bludgeon Explorer or black scout these this project helping to solve a problem of centralized service in the fear America system called Hitler's can wear it so kind of hard to get the the same experience for for side chains force and something that is not supported by if you don't my net or test net so we decided to build full-featured let's say copy of either scan and now we're supporting this for 4x dye and eight more networks and and even support for some networks not supported by many other services like we support the freedom classic and and whoever wants to use it can can install and and get the full-featured blockchain Explorer dubs started to move to the next day which is exciting tomorrow Haledon network which is a prediction market based built on Geonosis is migrating from from mine from my net to exile the and this is type of applications which attracted by platform usage fees and empty capacity of the network and it's it's very exciting that the application started to move from from network to network right good thing about this model is that we we're not trying to make vendor lock and like some other projects which are trying to migrate their apps from from a theorem to another platforms we are saying yeah this is a kind of temporary solution if like everything is temporary right in our life and this temporary solution is is available now and when people migrate from from my nature to this to this network they can at any time they can migrate another network because when usually when they migrate their dab they already used some form of a bridge and they already experienced interoperability on if they have effectors for for governance of their depth like upgrade ability and they can gracefully stop there's a pond on one network and run it on another network then it's it's it's easier to manage this migration and there so we we expect that this prediction market will be happier next day next slide about resolution is not very good so it's about future of exile and it's about future for horizontal scalability and vertical scalability right so we are working on new consensus algorithm which we called POS down or post down so process now is a delegated proof of stake implemented in a form of dowel when we analyzed many consensus algorithms we kinda had problem that implementation and reference can be different right and if you want to look into implementation and understand the the consensus algorithm you have to look into many files and most likely unlike native in native code so it's very hard for or let's say for people who want to for this network actually deep dive into consensus algorithm change some parameters and launch their own side chain without understanding the whole underlying code right so we decided to build two layered consensus one layer can be used any DFT algorithm and we are using we're using order for now with with added rental and Aranda we're using for producing randomness and this is the central part for for delegated propose take and on top of this BFG algorithm a reward distribution schema or about it or set layer can be implemented and this part is actually is managing economic incentive and also security model right so to implement roles of delegated proof of steak like validators and candidates to vault eaters and delegate us who delegate their stake on candidates or on body eaters you need some some form of logic right so this form of logic implemented within solid it is my own chart and it's it's quite easy to read to understand and to modify for the for the applications who are interested in launching their own side-chains based on tricky that profiles take because we like the idea of double token model where this the native hand application token is a stable token and utility token for example 444 the network security utility token is a staking token right because taking provides security for the network for for this model the delegated proof of stake will make lot of sense and the depending on how important network is people can stake more and more right we will launch a reference implementation with with one a gift with dies a native token and separate staking token but anyone can take this implementation and run their own digital proof of stake with difference taking token right so you can run a side chain with die as a native token and die is taking token or you can take either in a wrapped form and move it across chain and then use it as a staking token so the possibilities are like lot of possibilities to play with the with us taking algorithm and we think that this is a missing piece in the video market system to have a delegated proof of stake for the video in 1.0 right eventually we're going to have it for a film 2.0 but we're waiting for years to have it on a version 1.

0 to have a the the same the same platform the but with with different economic incentive or validators and all sort of have this you know green consensus with yeah with with more decentralization than proof of autonomy which is our existing form of governance and eventually we plan to upgrade to honeybrush 50 which is interesting algorithm which we already implemented and audited and now we combining it with this incentivization layer so this algorithm provides censorship resistance both like external and internal because in this algorithm validators are operating on encrypted data by threshold encryption and before they create a block they don't have oversee of all transactions in the network so for them it's it's hard to Center some type of transactions and when when we think about networks like X dye which can be used for for payments its its regulatory wise can be a tricky area right if the network is centralized when you think about centralized payment network you have to have kyc and Emma right when you have decentralized Network it's kinda different but in decentralized network the regulators may ask do you have a way to censor some type of transactions maybe we don't like some individuals or some countries can we block them and honeybadger provides this ability to to resist censorship attacks even but even though what data's want to do this they will not be able because you know there is no possibility with with some assumptions depending on number of our data's right the protocol itself is leaderless so there is no because all validators are in honeybadger are working on the block at the same time it's good that if there is no leader it protects them from from some forms of attack on BFD consensuses also the the protocol produces unseen randomness because of a threshold encryption used in the in the algorithm and it has instant finality right with some trade-offs like more numbers of more number of validators required to get this protocol alive but yeah that's that's exciting think about honeybadger we implemented that extended the original specification to support dynamics of all data set that while the in a honey badger can be unboard and removed by the governess decisions and with with delegated proof of stake governance decisions are my designs within the protocol so for example when the news taking epoch starts the validators on board that by amount of tokens they staked and I greater staked on them and also by randomness which also participate in in selection of validators we with when we design this POS our consensus algorithm we designed it this way that all pools and pools by possibly we think that pose our delegate us and validators and like all the leaders who do I get a dress taken about it we made equally word between them so the the protocol will prevent forming mega pools so that's that's interesting in to run and and also it's interesting to run in this environment where like security assumptions when we when we start this type of network and bound the list of risk and reward very very small right because we're gonna start this XD POS as additional side chain and we're not calling a test net it will be more like the staging because native token will be paged line and it's taking token or obedience taking token and we'll see how this will be interesting to the community so it's almost the last light I have some free day so yeah let's play who is the fastest one right yeah you can you can you know you can make a camera so the idea is like I generated some feelings and it's a QR code when you scan it the partner world will will open pop up and whoever take it first will take this day turn 5x day well but if you're not the first you will see like loading so usually like you know 5% of people anyone yeah if it's not working that someone else claimed it no no I didn't well but but my designer changed some slides came here are you guys okay really but someone claims it how do they get it really okay 10x day [Music] yeah more people so when you get damaged I share some with someone else anyone because now will be a super price next one okay 1,000 where 1,000 that's next presentation but yeah good guess 1000 I think 1000 is over like regulation well at least in the u.s. we can only like giveaway I don't know $6,600 something or less anyone claimed okay tell down 15 it's the last one yeah 50x that $50 basically well you have advantage like half people that's good and next time I need to put thousand but yeah what do you received as X die is a wrapped form of die on on XD chain and if you want to exit there is a button exchange and you can you can convert it to today in a custodian free mode so there is no central exchange between you and and die and after you have died you you can convert it to whatever you want on feeding my net right using unit swap first for example or any other platforms okay so I don't have more money to give away so some projects are not giving away eat but I'm giving away die so that's my last slide I hope you enjoyed it we have like three minutes for questions question hello great talk Thanks can you detail a little bit more how the gas model works on next day because I read it's it's fixed to one Giga wafer guys yeah it's it's fixed on the validators nodes and and validators can increase it right so I can I cannot say that they will not increase it it's possibly that they will increase it and it's also it possible that someone will think here you know what I will create a script and they'll just flood this network with transactions and fill all bars all right it's possible and there are some ways felt how how we can prevent it or steal the network horizontally right and vertically so the the increase of gas price is obvious and there are some other models we exploring the model of having some thresholds and like you know hundred transactions per user per day without without any restrictions on speed and after validators will put these transactions into priority queue under node and well they will feed them based on priority and it's on validators how they decide when they should put transactions into Q right right now it's like it's a parameter of equal price but they can be increased thank you question question from the top microphone question yes thank you very much what is the incentive for the validators they get the transaction fees ORS are also a mining reward yeah as Ethan to be right as is most datum they don't have much economic incentive they get transaction fees but they're very low so they spent more on hosting compare notes now it's it's like a community project so pro-ball data's participating think about like why about data's are participating and test Network like worthy or Kovan right in the same incentive structure but in the future when the type of networks will grow we think that there are multiple possible incentive models so first model is a exit fee from side chain so when and when there is a ramp and users go into the side chain they get conversion for free basically subsidized by by by validators and when they exit used it's different type of users let's say merchants who accumulate as Amex die when they exit they'll pay exit filled this model is working in in many payment systems think about credit card processing or you know payments and messenger so this models working also with delegated profile stick staking token will be inflationary token so well deters will get this staking token and what we will be able to you know to sell it and get speculative aspects of the reward here so this basically to economic incentive for them they cannot inflate die so there is no block reward for them in die okay you I'll show you guys bye-bye [Applause]