#195 Loi Luu: KyberNetwork – Towards Truly Decentralized Crypto-Asset Exchanges

Transcript

[Music] this episode of epicenter is brought to you by shapeshift.io the easiest fastest and most secure way to swap your digital assets don't run the risk of leaving your funds on a centralized exchange visit shapeshift.io to get started and by Jack Jax is the userfriendly wallet that works across all your devices and handles both Bitcoin and ether go to jx. and embrace the future of cryptocurrency wallets hi welcome to epicenter the show

which talks about the Technologies projects and startups driving decentralization and the global blockchain Revolution my name is Sebastian K and I'm Mah Roy today we have as guest ly Lou who is the co-founder of kyber network kyber network aims to be a decentralized exchange protocol that has features which would approach uh which would make decentralized exchange technology accessible to end users we're going to take a survey of the decentralized exchange field talk about the challenges of building such exchanges and zoom in on kyber network before we start L welcome to the show thanks for the introduction Y and thanks for having me

so so you have you've been a researcher at the National University of Singapore right yes tell us tell us how you got interested in the cryptocurrency space as a researcher right so I first heard about Bitcoin and blockchain back in uh late 2013 uh but not until like mid 2014 I started working on it um so the idea of payment system that a justess and you know operates in an open network attract me you know as a good research topic so since 2014 uh I have kept working on cryptocurrencies and and blockchain uh during my PhD and I was one of a few uh you know PhD student working on the

topic back then uh including many well-known ones like Andrew Miller Christian deer and other other jice um so so today I have written and published several um academic papers on you know top secur top security conferences um about cryptocurrency to to name a few you you might have heard about orente and about smart pool these are you know two of my work in my PhD can you talk a bit about uh uh these two projects and how they LED you to become involved in kyber network yeah so uh so orente is the uh first open source smart contract analyzers that allows you

like to check whether your smart contract has any security vulnerability so the idea is that uh you know you just need to to you know send your contract to oent and O will just you know try to explore all the other the possible execution part of the program or of the smart contract and then for every part it will check whether you know the contract has any security as defined in the tool and for smart pool is it's a first decentralized and efficient pool maning protocol for cryptocurrencies using smart contract uh so if you uh if you see in uh crypto currencies like Bitcoin and ethereum you have like several like two or three mining pools uh which account for more than 50% of

the mining power in the network and these you know two or three pools can just you know collude and you know do whatever to the network right so um the the goal of smart pool is to you know to to replace all the mining pools uh with with a smart contract so that people don't have to trust the the the poor operator and and everyone can be you know uh can can Monti can can can monitor the uh the money pool so there's no single point of trust anymore and and we have uh deploy and and maintain M uh smart pool in uh in the main net and so far we have mine uh more than 100 you know blocks on ethereum uh

Network so you you mentioned earlier that you worked uh uh in your PhD you worked on some of the early versions of uh trit uh can you talk about that and and how that work led into the work that you're doing now with kyber network back in 2015 I I wrote uh the verifier dilemma paper with Jason who is the the co-founder of trit now so the the the the high level idea of verifier dilemma is that you know uh if you ask a cryptocurrency network to to do a lot of computation on chain then there is a there's a dilemma that you know the verify they they don't know whether they should verify the competition or not

because either way they do they will be vulnerable to some attack right uh so that's why you know uh so that's why the truebit project exists to to address the the the problem so the idea is that you can still do do a lot of you know computation but you know only some of them will happen on chain and the rest of the competition will happen off chain right I have been you know focusing on on you know security and and scalability of the decentralized application and and you know uh from Smart pool orente and and trit and verify dilemma um and still you know cryptocurrency like Bitcoin and and

etherum Advocate a trustless and decentralized environment uh that changes the ways that people can judge and and manage their digital asset yes still you know exchanges uh the most critical critical part of the network are still uh heavily centralized because if if you see currently you know more than 90 more more than 99% of the daily trading volum happens on Central exchanges and you know big exchanges they keep getting ha so either due to you know external hikers or due to internal fraud and that you know put either those fund and personal information at briak uh so that's the main you know motivation to to motivate us to work on

kyber network so just just to recap the I think I remember the verifier dilemma is this problem where if I have a proof and I I I want you to be able to verify it I have to send so I send you the proof and then you have to have the proof that this is a valid proof and then I have to have the proof that you received the the valid proof I mean can you can you sort of walk us through what verify because I my uh my cryptography theory is a little uh my proof verify theory is a little rusty right now right uh so uh in the in the verifier dilemma what happen is that you have some computation that you want uh the network to to do right uh so what you do is you you upload the computation

to the to the cryptocurrency network and later on when you submit some input the entire network has like to do the computation and compute the output right and verify the output themselves right but you know uh so some people like you know get reward for doing that like you know the miners they who who who find the block but some people do not get any reward uh when when they verify the output right because because the the the the the generation fee is is sent to the only uh to the only minor not not to others so that's why there's a there's a dilemma there so whether should I verify the the the output or should I just skip it and

work on the next block so if I verify I I don't receive anything and I might be you know vulnerable to some attack because you know the the miners can just the minor who who find the block can just you know put a lot of competition there just to delay my uh the the the process of mining the next block from my miners right and if if I if I if I don't verify then maybe the block is invalid so who knows so I I may you know have the RIS of mining on top of invalid blocks and wasting my mining power so that's that's you know there dilemma there and how does that tie into the work that you're doing uh at at cab network but perhaps let's let's let's

use this to to Segway into decentralized exchanges um how does your your research and your work in in the verif with regards to the verified dilemma uh play into uh this idea of a decentralized exchange right so I guess it's it's not much uh related between uh you know verifier dma and and uh kyber network but there's the connection between smart appr and and kyber network right so can can can we like you know focus on that okay so so um I have been you know focusing on on developing like decentralized uh protocol which is efficient and and practical uh one of the example is is smart pool uh that that that you know allows miners you

know from all over over the world they they can just you know uh M together uh via the smart contract instead of via some centralized servers um so I I I continue the the the team of this you know work in kyber network by uh you know developing a decentralized exchange for cryptocurrencies which is scalable and practical I mean decentralized exchanges of course have been this now quite old dream right of the of the cryptocurrency space like like even when you go through like Bitcoin talk uh in 2012 or 2013 people were talking about and inventing decentralized the exchange protocols and

over the years like the designs have changed a lot but uh decentralized exchange technology hasn't really broken into the crypto mainstream right it Still Remains this French thing that people do I think the big change has been that uh back 3 or four years back there was just theoretical protocols now there are actual implementations where there are low trading volumes exactly so like like give us an overview of what makes this decentralized exchange problem hard why has it taken actually so much time to even build the first few implementations yeah I I would say uh

making a secure you know decentralized chain is easy but making it usable is hard because you know if you look at like Bitcoin back in 2014 you know not many people heard about cryptocurrency right but once you know ethereum started uh and you know uh like making it easier for people to understand and and to and to use cryptocurrency more and more people like you know understand about cryptocurrency and started using it so it's it's more about you know usability uh than you know uh Security in terms of adoption right so I think we can apply the same uh analogy to to uh decentralize exchange because now with ethereum we can you know easily build

like usable uh decentralized chains uh which is actually like easy for people to to use and doesn't require them to you know uh set up or you know to understand the entire uh concept behind it um yeah so I guess it it wasn't like you know practical and and feasible before so what are sort of the ideal features of a decentralized exchange right let's say let's say assume like 10 years down the line there'll be the perfect decentralized exchange what what features must that exchange have in order to be usable right um so I guess there are a few features so first of all uh it must be secure and and trustless right because otherwise we can just use

some centralized chains and secondly it's going to be the user friendliness like you know it's must be um easy for people to use uh even like you know non-technical users um the trading cost must be cheap because you know if it's if it's expensive to use then then no one want to use it um there must be like liquidity because um you know because when you want to trade there must be you know some someone else uh to to to trade with you right so uh I guess these are the few properties that you know any decentralized chain must must have in order to get you know uh Mass adoption so trustless here implying that

when I'm doing exchange there there is no point in the cycle of an exchange so when I'm exchanging say e ethereum like ether for ethereum classic yeah it's for ETC with you at no point in the cycle of this exchange is there um a single party that can um that can run away with our funds right that can choose not to honor our our funds not only that right so there's also the RIS of the you know some party which doesn't match you with the best you know counterparty so so for now you you still have to trust the centralized CH to match you with the best uh you know counter party so so somebody not being

able to run away with the funds is security yeah it's that is what security means here like and trustless here means that if there's like a matching algorithm or like there's an order book then when I enter something in the order book I am assured that it will get the there right and then when my order is matched with um which with the requirements of another person I get the best price and it it executes Faithfully so so those two those these things like embody the notion of uh trustless and security exactly and then the exchange must have uh liquidity liquidity meaning um a lot of people that are doing trades

concurrently right unless we use a design like bankor which is is a separate Topic in itself so there has to be liquidity and then there has to be low trading cost right so it it should be easy to trade right and and it should also be quick to trade so I I I enter my transaction It's Quickly accepted and I and I get credited my my my my S exactly so it's like all of these features have been been hard to put into one system until until now at least exactly yeah so if you look at uh other project uh like you know e Delta or or

bankor or even swap Tech and zerox so ether Delta has been around for like more than a year I guess but still the volume is still high because they they requires you know you know you know technical uh understanding from the users so so not many people can use it right this episode is brought to you by shapeshift the world's leading trustless digital asset exchange quickly swap between dozens of leading cryptocurrencies including Bitcoin ether zcash gnosis Monero Golem orer and so many more when you go to shapeshift.io you simply select your currency pair give them your receiving address send

the coins and boom shapes shift is not your traditional cryptocurrency ex change you don't need to create an account you don't need to give them your personal information and they don't hold your coins so you are never at risk from a hacker other malicious actor shapeshift has competitive rates and is even integrated in some of your favorite wallet apps like Jacks so you can swap your digital assets directly within your wallet just as easily as putting on your slippers whenever you see that good-look fox you know that's where shapeshift is so to get started visit shapeshift.io and start trading and we'd like to thank shapeshift for their support of EP Center let's like sort of walk through like all of these projects and the different approaches they are taking so

you said ether Delta swap Tech 0x and Bank call so ether Delta what how it works is that it maintains the entire aut book on chain so every time you want to trade something you have like to to you know send the transction with your order there and every time you want like to modify your your order you have like to send an the transation so it's it's going to cost you some transation fees for for every modification right so it's going to be firstable expensive um because you know now we have to pay transation fee for every uh modification and it's going to be uh uh it's going to be expensive for the

entire network as well because now everyone has like to store their Autobook on chain so it's going to blow up the the state of the entire blockchain and swap Tech and ZX uh are doing something similar except that they move the aut book offchain so it's like you know uh they maintain the the a book between the the the buyers and the and and and and the seller uh op chain and and there will be like some third party that that match that help them to match and when the settlement happens uh it happens on chain so the the the problem with that is that uh so first of all uh now we have like to trust the third party to to match uh between the the buyer and the

and the seller side and and there's no guarantee on settlement because because there is nothing on chain to guarantee that the set the settlement will be successful right um and Banker um Banker is is innovative in their in the in their formula uh that decide to sell and and and buy price but you know the problem with that formula is that it has not been tested and there's no you know guarantee that it's going to work in practice right uh and as you have seen there's a you know low adoption in in uh in banor so far uh I'm not sure whether

it's because of you know uh of the decentralize change in general or because of the approach that banko use so with ether Delta is like you have the whole order book on chain so in in in that case like who actually does the matching uh the smart contract will will uh you know do the matching okay but somebody has to trigger the smart contract to match right so it's like you know every time you send the transaction right uh like you want to sell something uh that transaction will you know check you know on the on on the buy side to to see you know if there is any match okay so this must be super resource intens intensive right because like all of these orders have to stay on chain like

if I put in an order and I cancel it it consume storage and transaction cost even though no trade really happened okay and and you cannot do any like State pruning or anything to reduce the state size of the blockchain because you know you must you know keep all the ERS but but ether Delta even though it has uh even though the the design might not be very scalable ether Delta does have users today so there's actually trades happening using using ether Delta exactly so it it is probably one of the first decentralized exchanges that does have some volume right right prior to this probably all the other decentralized exchange ideas didn't with like theoretical Concepts that didn't end up

getting any volume at all uh bankor does get some for the you know bankor to Ether uh exchange bankor to Ether exchange yeah all right so with with bankor the advantage is that there is no notion of an order book at all like there exchange technology doesn't have a notion of an order book but on the other side the design of that exchange is new and experimental and it could have have its own challenges that uh may not be apparent at First Sight yeah exactly and swap Tech and 0x uh what they essentially do is they end up it's like the order book is offchain

but the custodians the smart contracts that are holding the funds are on chain so they ensure um security against loss of funds exactly but they assure like security against loss of funds probably some kind of speed and liquidity as well but they are not able to assure the trustless in the sense that uh you need to rely on a third party to do the matching exactly and there there there might be um some in theory the third party could preferentially match and give you a bad deal and itself get a better deal so that so that um security element is REM is is fine but the trustless ele element is is not there

yeah let's talk about the design of the of cabber network uh as it Compares perhaps to these other approaches that we've discussed so far okay awesome so in in kyber network what we do is you know we bring uh so first of all we remove the Autobook entirely because we realize that it's expensive to to maintain the a book on chain and secondly what we what we do to uh guarantee the the uh liquidity is that we uh we bring you know some big market makers on Chain by introducing the the concept of Reserve so every time there's a trade comes in right uh we just go and ask the reserve what is the rate for this you know for this pair and

then we just you know get the best rate and then you know uh procet the trade and return the tokens to the users so so there's no aut book um the liquidity is guaranteed by all the market makers and we are we are going to be the first Market maker of the first Reserve in our system um in terms of CH L um we like everything happens on the contract so we we cannot like decide on how like who we want to to to match is purely based on you know the rate of you know or the reserve uh uh provid right so the the thing that I'm starting to imagine is that kyber Network would be something like shape shift right so I

don't know how many of our listeners have used shape shift but in shape shift like if you want to exchange like Bitcoin for ether you send Bitcoin for shapeshift and shapeshift the company sends ether to you right right so in in in a normal exchange you're interacting with another person or an entity that is not the company but like in this case there's the shape shift send money there and you get the what you want right so what you're trying to build is like shape shift but decentralized yeah so what we are trying to build is shape shift but which is decentralized and secure because you know user they do not need to to deposit to our to our uh you know uh to uh to our chain what they need to do is they just need to send the

The Ether or tokens to our to our contract and they will get back whatever they want to trade right and it is theen caliz uh in terms of using smart contract and in terms of you know having multiple Reserve to Pro to provide the liquidity so now everyone can be you know part like anyone can be a new you know shape shift in our system right because they can just register their reserve and you know process of the trade so of course with with shape shift the not to you know make too many comparisons with this particular uh uh product but in in the background what happens is that shape shift is actually connecting you to changes and making the trades on your behalf so there is an order book there there there are order books in in the back end that are uh

with which your trades are being matched and that's how they're able to uh to to to provide the liquidity for the for the for the trades um when you say that you removed the order books can you go a bit more into detail because it's not clear to me how you can have a trade without an order book are you just simply relying on liquidity provided by uh Reserve um providers that you know they they'll take any amount um for any type of trade right so uh in terms of like so first of all we we remove order order book on chain right and and we rely on the reserve to do the uh you know

liquidity providing right and what the reserve needs to do uh in in the back end is that they need to rebalance their portfolio uh by you know exchanging uh by by you know U converting their coins and token in some other Central exchanges which they have you know uh connection uh so that activity we uh we we you know we do not like um concern much because it's it's the uh Reserve uh owner's responsibility to do so the the reserve provider could be an exchange is is this accurate to say that it could be an Exchange in the back end it could be

an exchange it could be individuals with a lot of coins and tokens okay I see um so let's let's go deeper into this the design here we've talked about the the reserve uh owners can you describe sort of the the the high level overview of what the kyber network looks like who are the different participants uh and uh how they interact with each other you walk us through a trade like from from the user's perspective walk us through how he you when he sends the coins to the network what happens you know where is the liquidity coming from who interacts with a smart contract Etc okay so it it's a pretty simple so if you are a user uh you send your contract to our kyber contract H sorry you you you send your

tokens or eer to our kber contract and our contract result you know find the best Reserve uh that provide the best rate for for for you and we just you know send ether or token to that reserve and you know get back the the converted token tokens and and then you know forward the tokens to you everything happens in one transaction right so so here we have the users we have the kyber contract which uh you know accept the the trading request and and match to the corresponding reserve and we also have the reserve uh that provides the liquidity and provide the rate for you know the token pass right and so we also have the kyber network operators that

will uh you know do the you know listing of the tokens and do the listing of the reserve because we need to guarantee that you know the reserve will you provide enough liquidity for the system and we also need to do some sort of like kyc check on the reserve before we uh you know have them in our platform and the reserve will be maintained by you know Reserve uh operator uh that Reserve operator can be individual can be a dow a smart contract uh or can be a group of people uh um so basically we have like five uh five parties in in in our in our system the ERS the kyber contract the N uh kyber Network operator the reserve uh and the

reserve operator okay so for example if you were if you were to like sort of personify this us all of these people it could be that um so let's say uh lo you are the uh you're the operator you're the kyber network operator right uh I'm I'm let's say a a a reserve provider right so so my job is I don't know I have let's say a million ether from somewhere and I'm like putting that million ether on a on a smart contract and uh if if somebody wants to buy let's say melon tokens sell melon tokens and get ether then The Ether have put on on one of

these smart contracts will be used to fill that order right exactly so so that's the reserve and I'm the reserve operator uh law you can be the network operator and the other other party is sort of the smart contract right the uh the kyber network smart contract and then Sebastian can be another Reserve so uh so I I'm one company providing a reserve Sebastion is another company providing a reserve and the two of us are somehow competing in order to get as much volume through our reserves because I'm assuming we make some kind of transaction fees exactly so how does the reserve quota

price so how am how so let's say like so Brian initiates a transaction he has like th000 millon coins he wants 10 ether for those th000 millon coins and like he sends the transaction into the into the blockchain now how does Sebastian and I how do we indicate that we willing to accept melon coins and what price we will accept them right so what you need to do is you just need to send a transaction uh to kyber construct to to uh uh provide the transation rate as the the the trading rate between melon token and ether and you uh you essentially you have way to do that every like you know five or 10 minutes uh depending on you know how frequent

you want to update your rate yeah so based on the rate uh the kyber contract will just select the best you know rate for the users so if if if I'm Furnishing reserves and my reserves are going to be used in let's say 10 different trading pairs like melon to Ether I don't know dgd to melon let's say there's 10 pairs then I have to keep providing rates for all of these 10 pairs yeah so so Reserve management is a very active function right like yes but but you you you can like provide uh you know the update for the 10 pairs in one single transaction so you don't have like to send like 10 different

transaction to update you know 10 different pairs so in uh in in this in this great grander schema so there the reserve operator and the reserve manager are those two different parties uh oh they they they are the same they the same they are the same party I'm messing up with the word okay okay understood and then and then you we are also assuming that like if I'm a reserve so I sold my ether and got melon but but I might also interface as a reserve with another Exchange in the background get rid of the melon and get ether and like replenish my keep replenishing my my my my reserves exactly yeah right and and how do I make money as as being a

reserve as being a reserve you make money from you know all the trading spread right so you make that so every time you make a a a a uh you you process a trade you earn some you know trading spread there uh so that's how you make money by you know joining kyber Network as a reserve so in in some senses like kyber network is not exactly a trustless and secure exchange it is like it is removing the risk of trust from the user to the reserve provider right so the reserve provider now needs to go to a centralized exchange and uh needs to like do the opposite trade and like keep

the reserves replenished so it's the reserve provider that is taking on that risk exactly and the user is not taking so it is normally with with a user the user goes to a centralized exchange and is taking the the risk that the exchange will be hacked or something but out here you you say that the user doesn't need to take that risk but there are these professionals which are these Reserve providers they will take that kind of custodial risk on their behalf so so in in some sense you know all the reserve players or or the reserve manager or Reserve operators uh they they have a lot of coins and tokens right and and they actively you know uh you know participating in in centralite exchanges

already because you know they they they make money from being market makers there as well uh so we do not like increase you know any risk for them at all yeah I I I just hope all of these Reserve operators will have good security practices CU that this this could be a point of ha right so uh in the future we can think of some some model like you know we allow everyone to contribute to the reserve and and build up the reserve uh via some smart contract like the one that Mel Port is building so in in that scenario uh there's no trust on the Reserve at all right and people can just you know uh every time they they're running lower eer they can just add

people to you know contribute more eer to the reserve and they you know they have some incentive to incentivize people to do that uh but it's it's not in the early stage of of kyber network tell us then how how will this play out for the user um it what do you envision in terms of applications or user interfaces for the user to be able to trade coins using the kyber network do you have like a wallet or there like thirdparty applications that are meant to be built on top of apis that you provide that's that's a good question so we are we are building our own wallet uh that has that has all the features that we want to support in kyber network but that's not enough because uh you

know if those in cryptocurrencies they are really loyal um it's it's hard to like you know ask them to you know move away from their favorite wet so so what we're trying to do is we we trying to bring you know kyber Network to popular wallets like you know status uh MIT wallet and and others uh so so if they can just you know they can just convert their tokens instantly uh when they using their wallet they they don't have like to go to some centralized change or you know uh some other website to to trade and do do you envision uh what kind of like business models do you think can be built on top of this because I presume that then uh as a as

a as a d provider you know applications can be built on top of this and you're probably thinking of you know different types of business models that people can provide exactly so uh so we are thinking of so first of all um we can support like you know proxy payments right so we we can allow like people to to pay to anyone in any token like if you just you know receive if you just want to receive ether and you know I have only like you know uh melon pot token then I can what I can do is I can send via kyber Network and and kyber network with just you know forward ether to you and they do the you know conversion from uh melon token to Ether uh in the background so so so that payment service is one of our

application um there's another one that is um you know trading Advanced uh Financial uh instruments something like derivatives and uh you know Hing right so it's like you know when you invest in some Dow H fund and you are going to like receive your tokens in like 3 days uh but you but you but you see that the rate is is uh you know at best today and so what what you can do is you can you can do the you know you can do the trading with kyber network like today to receive the price the the the the the best price and three-day list the kyber network can you know just do the settlement with the Dow hash fund to to to get the to get your

toin okay so I I let's go through these different use cases so the first in the first use case you're what what essentially it comes down to is as a merchant I let's say I accept ether and you have melonport tokens and you want to pay me uh rather than you know trading those tokens on an exchange for ether uh in my payment platform you know I might have an application or you know some some sort of an API that would allow me to accept those Mel and poort tokens and receive ether right away um sort of like what people are doing now with uh with shapeshift can you describe this uh uh this uh the second us case case of

hedging is the kyber network providing those features to allow for settlement to happen later uh after a trade was initiated or is that something that you would be built on top as as an a third party application right so it's going to be a third party application because we need like to uh interact with the with the Dow uh contract as well uh so we need like to work with the the the the counterparty to uh to implement that Fe so this is not going to happen in the early stage of Cyber Network let's take a short break to talk about Jax Jax is your wallet your complete user interface to cover all your blockchain needs I've been using it

and I've been loving it now Jack supports a lot of different cryptocurrencies supp Bitcoin ether Litecoin ethereum classic zcash or rep and they're adding many more keep responding to users needs now with Jack the nice thing is that you can manage all of those coins within a single wallet and you are in control of your own private Keys they're not on their server and there's a single 12 words seed that you can use to back up your wallet all your coins and sync them across different devices talking about devices they're on pretty much any device that you can think of you can get it on PC Mac Linux you can get it on smartphones like Android and Apple and iPhone you can get it on tablets or you

can there are even browser extensions for Chrome and Firefox and on top of that in Jax you can actually exchange different cryptocurrencies for each other because they've integrated a shape shift and more Partnerships and Integrations are coming down the line in 2017 that are going to make Jax even better so Jax is really making blockchain and cryptocurrencies accessible for the masses easy to use for the masses make sure sure to get your own Jack wallet at ja.io or you can get it from any of the app stores you are using we would like to thank Jacks for their support of EP so let's talk about the network operator um can you talk a bit more about the role that the network operator plays in

you know governance of the Dao and some other roles that they may play that we're not seeing here so the the only um role of the network operator uh in in kyber network is to to guarantee that you know the platform happens uh correctly and there's no uh you know so for example the network operator has to uh list and the list the token pair so you know if some token is is not you know uh popular and like not not many people are trading it then we we we we will just delist it right and if there's a new token we need to list it uh to our platform and we also need like to to list and delist uh

Reserve um and in order to do that we need like to do background check uh to do kyc check to guarantee that you know the reserve is is um uh you know providing enough liquidity they hold enough you know portfolio uh to provide uh you know uh liquidity in our platform there's another important role of the network operator that is to to guarantee that there's no bad price for the users so so what what we can do is we can provide some sort of guiding price and if you know some Reserve uh offer some really like you know bad price uh we need like to you know stop them from doing that and you know do the investigation uh in the back crowd so you talked about listing and

delisting can you explain or go into detail as what is the what is the process there like is there a a set of criteria is it um uh are we talking about something here that also takes In Crowd input or is it solely uh the responsibility of network operators and and my second question my followup question would be like who are the people that who are the custodians of of the kyber network there right so uh in the in in the early stage uh we so we like as in kyber Network people have like to to to play that role so first of all we need like to do it manually because you know some some tokens they may have like different format they may not be like erc20 compatible so so

adding them may you know create some some failure to the network so we have like to do everything manually so we have like no criteria are in in you know list like list or the list you know tokens as long as you know they are they they w't you know create chole for the network and and they are you know compatible to to the network who will be the curators of the network like who will have the keys essentially that allows these I guess I guess there must be like a transaction that is sent to uh to the Dow uh when you want to list or D list like who who holds those keys initially right uh so in the beginning it's going to be like a few people from kyber Network like you know Alco Founders and and uh the lead

Engineers uh but later on we we can uh we can Envision the future that you know the the um you know our token holders will you know vote and decide you know which token to list and they list uh but it's going to be like in the future not not in the early stage of kyber network so in general I think we we are open like to list any tokens uh we we have like you know no um like not much concern about listing new tokens at uh as long as you know they have enough liquidity and and and you know uh truction in in in the community so one of the questions I I would have here is um so you have this network operator which is essentially

playing some form of administrating administrator role it is just ensuring that everything runs smoothly the right token pairs are there if there's some fault there's a way to recover from that fault and so on mhm um and I'm assuming like this this network operator can can charge some fees for its service for every transaction right um my fundamental question is why decentralizing Network operator why not keep it centralized why not why not just the small team of kyber network being the network operator from now until Infinity right uh so that's that's the good question because so first of all in the early stage we are not going like to

decentralize that process only a few of us will decide you know which token to list and which one is to theist right because we need first of all we need like to to guarantee the liquidity of of these tokens because if there's no Reserve like you know trading that tokens then we should not list it right um uh but so so the network operator they do not take any feat all uh or you know at least we do not take any fee at all uh the fee uh so let's talk about the fees later but um in the future I think it's it's good for the for the token holders uh in in the network to decide you know which token they should list or they should not list right because

essentially they are the main uh you know owner of the platform not not the you know uh kyber Network operators like not a few of us so that's that's that's the you know main reason why we should like decentralize uh the decision uh you know whether to list or not to list any tokens I I see a lot of projects and uh that that have this that that have this idea that they going to decentralize some function over over a lot of people yeah right so out here you have these decisions the network operation decisions and you're saying when you start out probably it's your startup there like three or four people that are going to make those decision decisions

and they're going to make the platform run well and then you're saying from three or four people you're going to go to at least 400 or 500 people when when you go to the Dow right now now let's assume that there is some economic value to this activity meaning let's say I don't know let's say there's like $1 million in in trades happening in your platform let's say $10 million in trades whatever and this network operator is making I don't know .5% 25% out of it so if it's $10 million um this becomes like $25,000 a

year now this $25,000 a year is being paid to the operator and right now with your startup there's only like four people that are playing the role of the network operator so each one of your employees that is is doing activity worth 6.25k right 6 $6,250 now when you decentralize it to like 400 or 500 people the same decision- making process when when it's decentralized over four people each one is doing activity wor 6250 and it makes sense for a human to learn this thing and do that for $6,250 if you decentralize it over 400

people now then each person is going to only make uh I don't know $62 and they have to learn how to make these decisions in order to uh be a good Network operator so I think like my my feeling is going from 4 to 400 is not going to make sense in terms of time like if I become a kyber network token holder I need to learn how to do all of this voting and stuff make all these decisions but what I'm being effectively paid for is just like $62 a year do you face this kind of problem in your in in your design right so um so voting is is always a a good question like how how do

you like incentivize people to vote um if you if you see from most of the election like even in the US you know like we have like 50% of people like you know appear and you know vote right uh because essentially they they they don't see you know much you know incentive for them to to go and vote um and and make the decision for the entire system because you know someone else will do that for them um so we we haven't like you know uh dive much into you know this this you know concern because our our our only concern now is to you know develop the platform and and you know make it run as soon as possible but I think that's that's a good question to to think of in

in in the long run like you know how to govern how to develop a governance model uh that you know can incentivize people to and and you know and like you know make it like practical one the governance issue is is is a is a really important issue and I I personally feel that a lot a lot of projects right now are focusing on new types of innovated use cases and are not addressing the the question of governance I'm I'm I'm not sure if it's because governance is hard or maybe it's just not very interesting uh you know if you compare it to some of these other uh interesting use cases that are being developed or maybe because you we're

waiting for uh for other uh projects like you know DED or tasos to solve those issues but it is something that at some point you we're going to have to address if if we're talking about you decentralization and at the core of it there's there is a team that is managing um the keys to a dow that will continue to remain a somewhat centralized system so one point or another we will have to face the issue of decentral of uh of decentralized governance no matter how complex or or uninteresting it may be from a you know user facing point you know use case kind of point of view okay um I think I think that's that's a that's a valid concern

uh but like so for so in kyber network um decentralized governance is is not uh something really important because you know what the most that the network operator can do they can just list or some token right they they cannot like you know uh hold userve fund or you know hold a reserve fund or they cannot like you know go away with others money so so it's it's not you know some security concern or you know even like just less um concern that that we need to address in in in the near future no I I think that's I think you're right there but one one thing that uh uh Network operators may be able to do is you up upgrades to the protocol

and once you have something that has a lot of value uh and you get into these uh decisions having to do with upgrades to the protocol and and how those play out then you fall into you know similar situations as we've seen in Bitcoin recently where you have a lot of value behind a certain Network and um uh there there are factions that you know may want to go in One Direction or another and that's where I think governance uh you know will play a major role not not just in the decentralized aspect of it you whether or not the the the the app the DAP can can can run away with your with your funds but how what is the governance around how the system continues to evolve and upgrade yeah so so so I think if if we are you know if

we uh ever go into that point uh that's that's a good problem to have because now the system like there's no much so much St in in in the system so people will have more incentive to vote and and you know to uh to participate in the the governance right um so I I guess that will address the incentive issue that that was raised like 5 minutes or 2 minutes ago yeah if if the system becomes very successful then there will actually be incentive for people to learn and participate in governance if you want like 400 people to participate in decision making you realistically need a trade volume of 100 to $500 million in order to have sufficient incentive for

that to happen yeah I agree right right so it is certainly that it is certainly like some some ways out before U that incentive structure is created otherwise you will you will only end up getting voters that really don't care because their effort is only worth I don't know $100 a year yeah which is like which is like tiny right like uh you could you could you could make more doing an epicenter episode yeah so right which isn't a lot by the way yeah okay so if if you take a look at at the trading volume at the daily trading volume of you know all the

cryptocurrency now is like more than $1 billion per day right so if if we like you know can take like 1% of the pie then it's going to be like you know a lot per year so with with a lot of with a lot of these decentralized exchange projects that are built on etherum you know the focus has been mostly on trading ether for other erc20 tokens and and and it appears that when you want to interface those decentralized exchanges with other blockchains say Bitcoin zcash or or other um decentralized internet platforms like Cosmos polka dot Etc uh it it it sort of becomes kind of hacky and you have intermediaries there or

some sort of Gateway or something like that so how will users be able to or is this part of the road map for users to be able to trade say for instance ether or erc20 tokens with uh Bitcoin zcash or through other uh platforms what do you envision there in the future so there are several like approaches to to do like you know cross cross cryptocurrency trading right uh so we can we can wait for Cosmos and and polka do uh to uh and then use these protocols to to support that features another approach course which is already practical and existing that is to use all the chain

relays and uh we just so just like two or three weeks ago we we made a a good progress in building the P relay between ethereum and ethereum classic so now you can basically chestless move your uh Etc to ethereum and allowing people you know to trade uh between eth and you know any erc20 tokens to Etc uh so that's that's really practical and you can you know do it today right and there's also BTC relay which is which was built like a year ago by Z zose Char um so but you know still you you need you know some hiking there in order to make BTC relay practical uh but

I think it's not hard uh we just released a blog post like uh a week ago to discuss about our approach in in you know shading uh different cryptocurrencies uh and after maor police I I think um building uh a relay for Z cast is going to be practical because uh you know etherum will support uh uh you know uh zero knowless operation and also they may uh you know support uh the the H function of of Zas proper work so that we can practically you know build a Zas freay on ethereum and allowing people to trade between Zas and you know any uh ESC 20 tokens or even ethereum so before we wrap up uh

tell us where you are in terms of developing this platform and what's your road map for the future right uh so we are releasing our uh our demo or our test net deployment in two days with two or three days from now uh so we are we are pretty close to the to the mainnet deployment uh so we are targeting but you know the end of this year uh to for you know midet release uh that will allow you to trade between uh you know ether and any erc20 tokens um and our our road map has several uh you know major releases um we are going to have a release that allows you to support to do TR to to to do all

the you know uh payment services like you know you can pay anyone in any tokens uh we also have another release that allows you to uh you know uh trade between like Advanced Financial uh uh instrument like you know hatching derivatives um and the the the another major release will allow you to trade between different cryptocurrencies and by then uh we we expect that you know Metropolis is already uh released and hopefully cosmod and poad do is will be launched by then as well otherwise we will just go with other chain relays approach that that we have and what's the business model here how do you plan

to make money with this right uh so we are going to to take a cut from you know all the transation or all the trade uh in in our platform so first of all the the company uh is not going to to you know take any fees or any share from that uh other transition Holder will uh you know you know enjoy that that generation fees uh and and we are still working on on on the on the model for that does uh your company plan to be for instance like on the all Al on The Reserve side you plan to also provide liquidity in in the network somehow or would you strictly be be on the technology side and as operators of the network right so when we first launch we

need like to to to provide and maintain the first reserve of the system before others can you know participate and guarantee enough liquidity for the platform so in the beginning we must you know play both the roles uh the platform provider and you know Reserve provider where will you get these reserves like what are are you partnering with exchanges or will you have the Reserves to so that's that's why we need the the the the crowd sale uh so part part of the money raise in the crowd sale is going to be spent on the reserve okay so there will be a crowd sale at one point okay yeah all right uh well thank you so much uh L for coming on today and telling us about

kyber Network we'll be looking forward to seeing how the project develops and uh how it will play out in in this ecosystem of the decentralized exchanges I think that uh uh this is a this is a space that will continue to evolve and uh this type of decentralized platform providing liquidity and providing uh the uh instantaneous uh trades is something that will be sorely needed once as as more and more cryptocurrencies enter enter the ecosystem thank you both for having me and thank you to our listeners for tuning in we are part of the let's talk Bitcoin Network you can find this show and lots of other great shows of let's talk Bitcoin of course if you like the show you can support us and there's multiple ways you can do that you can leave us a review on iTunes or any other platforms you listen

uh and you can also leave us a tip our tipping address will be in the show description so thanks so much and we look forward to being back next week [Music] [Music]