Alex Shevchenko : The Rainbow Bridge: the First Fully Trustless Bridge to Ethereum

Transcript

doing the um you know the lectures it's super unusual to see first three rows without any people you can distribute a little bit if you want okay it's like like everybody's sitting on on the back and cheating you know just reading some stuff yeah okay so I'm ready you can start that's a nice move okay hello everybody thanks for joining this talk my name is Alex shuchenko I'm the CEO of Aurora Labs the company that

is developing Aurora uh solution but my talk today is not going to be about Aurora I'm not I'm not here to sell you Aurora as a project I'm not here to sell you a part of our project which is which is Randall bridge I'm here to share with you the knowledge about how to build Bridges properly and what problems are there right so I'm going to begin with a simple thing a schedule like the agenda here right I'm going to talk a little bit about the bridges itself what are the problems they have then we are going to transition into the architecture of the Rainbow Bridge how we build it then we are going to talk a little bit

about the governance and what hurdles would have and and then go into practical aspects and and then obviously to open problems uh there are lots of things to cover so yeah I believe you will have lots of questions right so let's start with the first piece uh with introduction and the bridges itself can can you raise the hand your hands if you if you ever have tried any Bridges right and you understand what it is okay okay quite a lot of this so maybe for for you this this slide will not be actually very very important or informative right but just for for everything for everybody else here yeah

for you just to know uh what is the bridge and this is the slide right so you know that the blockchain is is a Sandbox solution this is a solution inside of itself it cannot communicate with anything outside right you cannot just just put in the smart contract there is no such function in the in the ethereum SDK to call the API that is outside of the outside of the solidity sandbox right you you cannot do it and there is a strict reason for this right you need to have the consistency all of the data all of the execution needs to be consistent right so you cannot overcome this this is the you know the core of the blockchain

so the bridge is a solution to connect this sandbox to something that is located outside of it right in case you would like to query some apis you have Oracle Solutions right and if you're talking about the bridges this is the solution that connects blockchains in between itself right so in general when we're talking about bridges people people are thinking about token Bridges so reasonability to send tokens from one blockchain to another blockchain and yeah this is the token Bridge quite recently there are additional bridges that get developed they usually called as generic Bridges which allowed

to transfer any data and by transforming data I mean that some kind of execution of some kind of transaction right so you're executing the transaction you've got the execution results it can be and some kind of log from the transaction right it can be an event that is emitted and then you're able to move the data from that event to the other blockchain there are lots of existing Bridges and the Rainbow Bridge is not far not the first one right you probably know many of them and and as as we also you you have been trying it right

the ordinary architecture of this bridge works the following way and I'm going to talk about the talking Bridges on the source chain on the chain from which the transfer is happening there is some kind of escrow account so what you need to do is to move your funds into the escrow account and then there is some kind of entity that is responsible for Bridging the tokens and this entity it actually monitors the balance of this escrow account and tries to understand who was sending the transaction who was sending tokens to this escrow account and obviously this entity may be just a

single entity or it may be multiple entities together combined together into some kind of a set of the participant participants some of the networks are calling this set of the participants the bridge validators right and as soon as the entity is seen that tokens were transferred to the escrow account they agree on issuing the IOU tokens right or like a digital certificate on the other chain and for this they are calling so-called Mentor contract right so they're able to Mint additional tokens on the other chain

and obviously the problem of this approach and this is like the ordinary architecture all most probably if you haven't ever used Rainbow Bridge you have been experiencing absolutely this architecture and you have been using it and the problem is obviously in the middle the problem is these Bridge validators because what you are doing literally with these transfers is actually you're transferring your tokens to a multi-sig wallet right so that's it that's it your trusting all of your funds to this entity or entities and well in some cases uh in the in the

market right now there are bridges that have millions of bridge liquidity more than like tents and I believe even hundreds probably and the and the the thing in the middle the entity in the middle is just two of three multisig and all of the three keys are kept inside of the same company right obviously there might be not a very big incentives for this company to cheat but in general your protection as a user as a owner of this assets is not that high and well the core problem is just the centralization of the solution right and we are here you know all all of us

sharing the the vision of the ethereum community right of the decentralization trustlessness so that's the question how we can do it better is it even possible to do it and the answer here is yes you can do it but it is damn hard to do it it's just super complicated thing it is absolutely on the other level in comparison to the multi-seek approach that is that is implemented in in many other Solutions and yeah we did it like it's not even me it's the team and I was actually joining the project um like in the in the last phase of it yeah I just needed to to wrap everything

up so all of the credits to to the world and one additional thing I want to to talk a little bit um because like Roll-Ups uh right now also is a pretty hot topic and and bridges are also you know connected with Roll-Ups you need to move your funds or you need to move your tokens to the roll-up in order to be able to to use that in inside of the roll up and uh the difference here is that uh these Roll-Ups may or may not use the blockchain underneath it but in the end what they have is the experience and the security level of the

source chain this is what they they are building on top of so that's why the transaction finality or the you know the the finalization of the transaction in rollups is depending on the on the connection of the Roll-Ups to the source chain and in this case the finalization may take some long period of time it can be like naturally for many Roll-Ups it is a week something like that so still you you can move talk and like and this is obviously this is obviously true for the transfers of the token from the Roll-Ups back right so in case you are doing the transfer to roll up again

you're depositing your fund somewhere this can be as fast as ethereum can give you so as fast as the finality of ethereum transactions which is also an empiric thing right you can have like 10 20 confirmations what is okay for you or what is okay for for the roll up itself but then after it you need to bring money back or bring your tokens back and it can take up to one week so long story short there is a problem let's try to solve it and Rainbow Bridge is is the answer or the or the architecture of the Rainbow Bridge is the answer to this problem now this is a pretty complicated slide this is if only you know the the tip of

an iceberg of the architecture of the Rainbow Bridge and we are going to to go a little bit deeper into it in just one minute but for those of you who are deep enough in the in the blockchain uh on the text side of the blockchain uh just one minute pitch about the architecture of the Rainbow Bridge is that it literally implements the like client of the source and Target networks as a smart contract on the opposite Networks and because of these light clients we can get an ability to prove the inclusion of the transactions on the other chains just absolutely as as your wallets are

doing and yeah and because of this we are able to understand that something was deposited in this scroll contract and that's why we need to Mint something on the other side and this is the high level architecture but let's go a little bit deeper so first thing that we have in the Rainbow Bridge and I was talking before about the source chain and Target chain just for the generic Bridge the Rainbow Bridge right now is implemented in between ethereum and near blockchain and just a 30 second speech of near this is a gen 3 blockchain with a proof of stake and and horizontal scale and implemented through the dynamic sharding so so we have two blockchains and what we do

first is implemented like clients of ethereum and nearest Smart contracts on the other chains so literally on near blockchain there is a smart contract that does validation validation of each hash right and on ethereum blockchain we have a like client of near that does the validation of ethereum of near consensus um and because of the way how these these like clients are implemented and because of the some restrictions that that developer SDK is imposing on us ethereum like clients on near is implemented as realistic like client which means that as soon the block committed there it means that it is

available it is final right obviously depending on the all of the rollbacks that are there in the ethereum blockchain right so if you have I don't know like 10 confirmations of the block in the ethereum like client on near then it means that more more or less this block is final while near like client is implemented as an optimistic like client and the reason for this is that near blockchain has a proof of stake consensus so everything ends up into checking the signatures of the and the signatures of the validators since near is using

ed25519 which is advert's curve cryptography it is hard for for ethereum to validate because ethereum native cryptography is is sepkeys right and there is no the eip665 it is already four years on the considerations whether whether this needs to be included and unfortunately it is not yet there that's why there is no host function for validating this adverse curve signature which means that there is no simple way to validate the the signature and unfortunately this needs to be implemented as a smart contract and we are these crazy guys who

implemented cryptography by ourselves and not even the the cryptography itself but also is a smart contract right it is 1600 lines of code generated absolutely unreadable by ordinary normal human uh it is generated by by the program right so specifically for this we needed to implement it as a smart contract and obviously since it is a smart contract it is super expensive we cannot run it every time with at the moment 60 validators that near blockchain is having so we just cannot check it all the time that's why when the block is committed

to the near like client on ethereum and again ethereum fees you know so uh so every time the block is is committed it is stored in the so-called untrusted State and only after some period of time in case the challenge was not issued then this block is treated as final and who is able to to issue the challenge well we have for these real layers and and watch dogs right so these are the services that are running outside of the blockchain and doing pretty simple stuff they need to keep the networks synchronized right so there is a real layer that is sent in ethereum blocks to near blockchain

directly the headers obviously right and then the like client is very fine the the if hash and all of this stuff and there's the other the other relayer that is sending near blocks to ethereum right and the challenges issued by the Watchdog the service that actually checks whether the submitted client is okay or not and in case there is some kind of and everything is verified in the near like client except for the signatures it takes about 1.5 million ethereum gas for the validation of one block but even a single signature costs some hundreds of thousands and we just cannot check everything so in case the Watchdog is Asian the

challenge then this the particular signature gets verified but what is the what is the economics behind this well we need to introduce here we need to incentivize watchdog's existence right we need somebody to run a server and constantly check and what are the blocks that were committed to to near like client on ethereum right and we need to understand and like issue the challenge which is also an ethereum transaction that costs some money right so how we would do this well literally we are we are saying that if you want to become a real layer you need to stake a little bit of eth on the ethereum or on the ethereum side right

and then in case you are committing run information then the Watchdog will issue a challenge and you are going to get slashed so you're not you will not be able to retrieve your deposit okay so a little bit more complicated stuff right but let but now let's go let's go deeper so we have the setup of the synchronized blockchains through the like lines how we're going to use it for this you need to to convert it into a service we have prover contracts this prover contracts they are simply answering the following question in case this is my transaction was it included in ethereum

blockchain or was it included in near um so this is just literally a service or like an interface to the like clients and things that you are able to prove with provers is actually anything that you are able to prove with the ordinary like client right because it has the the information uh the the information from headers so for example for ethereum prover on near you can prove things like the inclusion in try and you know that there are three tries right the the try for transactions the try for logs and the trifor uh storage right so anything that

is located in these three places you can prove you can prove the uh you can prove potentially the the height of the blockchain you can prove some kind of time stamp of a block you can prove stuff like that so anything that is provable can be proved and on the uh with the near prover obviously similar things but uh because of the specifics of how the blocks are constructed you know they are a little bit different but in general anything again just bottom line here anything that can be approved uh in general in the blockchain it can be proved through the provers right and a good thing that obviously that we have not built all of the potential

provers many of the provers are just not yet there we just do not have you know the capacity to prove everything that can be proved right and the good thing is that well like clients are there Watch Dogs and relayers are can be set up and and they set up by us but anyone is able to connect there right but then Proverbs can be developed by anyone this is just just the smart contracts that are sitting there in in the blockchain you can access them you can access the information that is there right so absolutely no problem in doing this and even more what's the reason for you to have approver well actually you need to do some kind of logic on top of this information

and this logic is is incorporated inside of what we call connectors right so the connectors are the smart contracts that are actually implementing the the messages that are sent in between two blockchains so until now we've got only the state of the sync blockchains and ability to check stuff and now we we are implementing the actual messages that I that are going from one blockchain to another so for example some some examples of the connectors these can be erc20s that are bridge from ethereum to near this can be near token Bridge from near to ethereum right well and obviously back this can be F this can be erc721 this can be Dow resolutions that

are working on top of near blockchain and you would like to pass these votes this can't be cross blockchain calls this can be anything that you can you can imagine right and again everything before are the smart contracts on blockchains that's why anyone is able to build their own connectors right if you have some specific information that you would like to pass from one blockchain to another you're able to do it for free because you just just can call other smart contracts you would need just to pay the gas fees that's it that's it that was the hard part I believe some of you you know go in a

little bit asleep here but now we're going to go into into some interesting problems on top of this maybe you actually have heard a little bit about the the Rainbow Bridge before I'm not not sure whether whether you're following the developments there uh but yeah but let's go to something new or something that was not widely uh you know communicated it was accessible but I just would like to put some emphasis on this stuff right and first question is the governance the problem with this bridge setup is that on the smart contract level what we are doing is that we are very fine the uh the headers right

and as we know from London hard fork and you know there's this madness that is happening with Robson right now we know that uh hard Forks can change the the block structure hard Forks may change the way how the consensus algorithm is working right and what does it mean it means that the smart contracts must be upgradable because in case not well then you will need to force all of the bridge users to withdraw their funds and deploy to the new bridge right this is this is just hell um it will be it will cost a lot of money and it will be just super inconvenient

so well from my points of view okay the maybe must is is too much there right but it is strongly recommended for the smart contracts to be upgradable right and in this case since the header verification algorithm is not the part of this Mark it cannot be accessed on the smart contract level it means that somebody needs to develop this solution it needs to be uploaded there and then somebody needs to vote for the upgrading of the smart contracts you cannot access this data on the smart contract level you cannot unders your smart contracts cannot understand what

is the change in the um in the in the in the consensus in order to implement it right away and to be able to check the new implementation whether it is good or not you just cannot do it because because of this you need to have a body that is going to make the governance of the bridge so we started from the from the thing from the problem that there is somebody in between who is who is doing the the you know the bridging and now we are ending up in the same situation right we need we need to have a governance body but you can implement this governance body uh in a much much better way right

and obviously the most the most close governance approach to the decentralized nature of the blockchain is obviously a Dao right in fact we have quite a long roadmap for the governance upgrade for the for the Rainbow Bridge there are multiple stages there how we are going to this to this now now is not implemented for for the Rainbow Bridge right now but we are going through it and we we've checked with our community whether whether this approach is okay I'm not going to cover all of this there are lots of things there are lots of reads um it's not it it the the link to the to the discussion is here but you can check

out the near Community forum and you will you will be able to Google their Rainbow Bridge governance and upgradability plan so the the ultimate setup that you can get here with the convenient infrastructure for the users is actually you need to have a dial on the target chain because from obviously you are bridging things from from The Source chain and the reason why you're abrasion because you need to you know lower the cost or give this interoperability features right so probably on the target chain it is much cheaper to run it down so so the ultimate setup is to have a tower on the target chain you need to this doubt to vote for the

upgrades the votes of the upgrades are applied to the bridge smart contracts on the target blockchain immediately or like straight from like straightforward in a straightforward way but the updates on the source chain they need to be bridged through the bridge through the existing Bridge protocol so the Dow votes about the new implementation on the source blockchain and then the vote is bridged to the uh source blockchain and then this votes choose well causes the change of the implementation

this is the ultimate setup none of the bridges at the moment is having it including Rainbow Bridge but this is the way how you need to do stuff this is the most decentralized way how you can build a bridge that is convenient to users and they do not need to constantly think about moving their funds from one implementation to another one and now some practical aspects so right uh Rainbow Bridge exists right now uh with a little bit simpler governance not like that but let's let's figure out how how do you uh if you if you're using

it what you can expect from it so actually Rainbow Bridge it was in development starting uh actually by the previous speaker Anton from one Lynch was contributing a lot into the development of the Rainbow Bridge unfortunately he left um and the development started in 2019 late autumn and then everything was put in production just this uh this Sprint and actually this is the wrong information it's not 15th of May it is 15th of March so two two months earlier there is no intrinsic fees inside of

inside of the Rainbow Bridge we do not charge anything the only thing that user is paying for is the guest fees in case you are transferring 100 usdt tokens through the Rainbow Bridge from ethereum to near you will get 100 near usdt tokens right the only thing that you need to pay for is the gas fees on the both networks right uh the total transacted volume is around 5 million a little bit more I believe because I was I was composing this deck around a week ago and and even then it was 5.1 so the bridge code was audited back in 2020 but since then it there are some changes applied obviously and we are now

undergoing the next security audit of the bridge because we obviously understand that this is a super crucial piece of technology for for Neo Network in terms of costs Rainbow Bridge is quite an efficient bridge and it does not incur lots of lots of you know cost associated with the fees for the users and for example ethereum to near transfer costs only 40K as yes it's only twice worse than just send in F from one account to another it is almost almost the same as just a plane transfer

of an erc20 token super cool Super near yeah and this is the cost on the ethereum blockchain because costs on near blockchains they are they are negligible the cost of near to ethereum transfer where we need to you know check the proof and because of this this transfer is is actually not that simple it costs more it costs 330 000 uh so it's like two swaps on uniswap or something like that um and I was not putting here any numbers and used dollars because nowadays you you know the the gas price and also the exchange rate uh it goes crazy so like okay so you can estimate it a little bit I'm sure you you're

pretty used to to recalculating gas into dollars right now what about the transfer time and remember I said that the like client from of of the ethereum on near blockchain it is realistic that's why as soon as we can treat ethereum block as final it means that we can validate the transfer right we can prove stuff and usually it takes five minutes right and this is like with extra precautions and all of this stuff you can have you know 10 confirmations 15 confirmations that is enough right so this is this is what we're saying five minutes is absolutely the most uh well an order of five minutes is some is absolutely the lowest

that you can get because of the design of ethereum blockchain if somebody is saying that they have built a trustless bridge with the transactions over the bridge happening within seconds they aligned to your face it is not a trustless bridge okay so five minutes uh near to ethereum is an optimistic right that's why we need to wait for some time for the Watchdog to be able to issue a challenge and because of this like this this time can be it should be long enough we were doing some math some some modeling on the on how the gas price can

may change there and we figured out that four hours is the lowest amount of time that we need to wait and right now Rainbow Bridge is operating with the transfer taken from four to eight hours so four is the absolute minimum eight is at Absolute maximum that you can expect from it so in the mid six hours uh what else we also have ERC 721 connector under review lots of people from the community are developing their own connectors they're they're trying to to do stuff with it um yeah and also one additional things one additional thing is that we are

extending the the prover uh the theory improver as a smart contract on your blockchain to support not only logs but also State and transaction inclusion proverbs and we have a little bit of a little bit of sugar on top of it right so because of the way how the fungible tokens are working on the near uh because of the the great nap 141 standard developed by evgeny who is sitting here in the second row we have an ability not to transfer the funds from ethereum to near wallet but directly deposited into the smart

contract and actually do some operations on top of it with a single transaction you can think of this as a composability of the transactions while there is not you know the district analogy on in your blockchain for the compatibility but something like that can happen so literally literally a user can deposit his funds to some kind of smart contract even without having a near account and this is something that we are heavily using in Aurora as a project right so when Aurora just again I was I was promising not to pitch you because

I'm just so just for your information right so Aurora is the is the project that enables an ability to to run solidities more contracts on top of near and users are able to deposit funds in debt in that runtime right in the evm runtime the second thing here is a fast finalization options right so we potentially can speed up the the bridging transactions with some people providing liquidity for the fast finalization right so in case you see imagine the foreign situation you see that there is a there is a transaction happening from from near to ethereum because this is this is the transfer

that you would like to speed up there is a transaction happening and it is going to be proved at some point in time in eight hours right but you would like to get some premium for transferring for finalizing this transfer up front with your own funds so you can send to the recipient of the funds everything with except for the fee and then you can prove to the bridge that you already did it and the fee will be returned to you this is something that can be implemented and with this approach we can we can reduce the hurdles of the optimistic region right we can make it work

like the fast way and the last but absolutely not least is that because we have implemented near like client as a smart contract on ethereum blockchain it means that the whole ethereum blockchain becomes a light node of the near blockchain what does it mean it means that near blockchain right now only because of the bridge is as secure as ethereum with regards to the retrospective attacks retrospective reordering stuff like that so this design is super beneficial for any Target blockchains right so I'm

showing you the design not not the Rainbow Bridge itself but the way how to build Bridges right okay lots of good stuff but there are lots of open problems here and this is one of the my main goals here to raise these open problems and and to gather people to discuss this and first and I believe one of the biggest one that that is in front of the uh in front of the multi-chain community that we have is the following imagine that we have the die token on ethereum and this this die token is breached to multiple chains so to near beats in polygon we have ndi B di PDI right the

the bridged token the bridge IOU of a die on the other blockchain but what will happen if the B die is going to be bridged to near right then we will have the near B die right the near P die because in case we have near BSC and near polygon Bridge we will have these tokens and actually moreover you can Bridge the other tokens to the other blockchains and then you can Bridge them again in case you have four networks you will not be able to limit the amount of the tokens that you are getting from one specific Source token

are these different clones of die on near blockchain actually different well probably yes because the technologies that that like people are relying on when they're using these tokens are also different so for and die user need to trust ethereum consensus it need to trust near consensus and it need to trust the implementation of the Rainbow Bridge but what about NP die right now user needs to trust the bridge in between ethereum and polygon and then from polygon to near right so there is different set of well a stack of technologies that the user is using when they are using these tokens and that's why they're different and

probably the right tokens are should cost a little bit less than the left tokens right just because higher risks involved there is it convenient to user well absolutely no I have I have a hundred bucks in my in my wallet I absolutely do not care in which countries these hundred bucks have been settled right I do not care whether it was coming through Japan or directly from the US right and and all of you are not taking care of it this is super inconvenient to have different tokens there and what about metadata so some of the some of the tokens are actually updating the metadata and this

is this is pretty common and inspect specifically to to some of the projects the updates of the metadata is in the core of the logic of the token like ample Force for example so what what was with the metadata how you're going to bridge it what will be the meta Dev for npb die right is it going to be different from the metadata for the other token I name it the consistency of the tokens yeah or the token consistency problem you can do it you can try to solve it with something like a curve approach right you have similar tokens you can introduce curve uh curve markets there but as soon as you have you know at least four networks as I said the amount

of combinations is unlimited and you will just not have enough liquidity to cover all of the trades so this is solvable as a first step this is not the ultimate solution to introduce the the amm solution there something needs to be done here it should be something like a cross chain standard for the token bridging and this is this something I encourage all of the people from the community from other networks who are building bridges to come together with us and let's discuss it let's make a cross-chain standard for the benefit for the benefit of the whole industry right

now let's go let's go even deeper how to remove the need for the governance entity right because we started with the multi-seek approach not that good but still you need to have a governance entity but how to remove it well we probably somehow move the header description and the verification algorithm to the blockchain or do we need to be able on the smart contract level to check whether the consensus algorithm is correct or not for this specific thing right this is the first time in the history of ethereum

when this type of question is actually Horizon and nobody was never ever thinking about this stuff and uh well who knows we we we with this setup we cannot with this like with the current stage of ethereum we cannot develop a secure trustless decentralized Bridge without any governance entity it is just impossible so maybe we need to change something in the blockchain maybe we need to do one additional hard Fork of ethereum to help solve this issue right again open open question so I'm not giving you here the solution right I'm just inviting you to discuss

and what about the security incidents right a bridge may be broken right uh there might be some kind of vulnerability discovered how we are going to deal with it how do we implement the the Escape hatches how the users can withdraw their do we need to restrict bridged assets uh from moving in between accounts when when the governance entity introduced the freeze period right this is also something usual for uh for you know Advanced governance do we need to all of that stuff or maybe we can skip something I don't know and I was telling you how to make the fast token transactions right yeah if in case you have liquidity providers but

how we how the heck we are going to do the fast transactions of an arbitrary data like people cannot provide can people cannot be the providers of arbitrary data right so you cannot speed up the data right just like that so but it will be super neat to do it right so these are all of the open questions I'm leaving it I'm leaving it for you let's gather together let's discuss we have some ideas how how we can organize this um as a some kind of discussion club or but at this stage uh I would like to to end

and thanks for being with us and I believe you have lots of questions so I'm here for the next uh several minutes to answer them thank you guys foreign yeah uh thanks for the really good talk um it's kind of mind-blowing thinking about the ethereum blockchain validating the near blockchain it's kind of crazy to me um really cool but yeah um I was kind of under the impression with X die for

instance that their plan is to sort of slowly introduce like more validators and kind of come up with some sort of incentive scheme for them to like validate it properly um could you talk about why that's sort of inadequate and why their approach kind of can't work long term I guess well from my point of view as I said you know introducing the breach validators is is not like or at least in my head he's not he's not aligned with the spirit of the blockchain and the spirit of the decentralization and many projects are going that way right and then they obviously introducing the incentives there and so on but like it's still you need to trust you're not trusting the

blockchain that only the consensus is of the Target and the source blockchain you need to trust something more some something additional entity and all of the all of the ideas behind the the Rainbow Bridge that we cannot sacrifice this decentralization we cannot we cannot do this stuff right um in order to be aligned with with the uh with the ideals that we have so so that was that is the story behind uh Rainbow Bridge that is the story actually behind Aurora but this is for absolutely another talk uh yeah why xdi is doing like that and whether it is going to be sustainable um from my point of view it's just not following the like it is not fully aligned with the decentralization principles that's it

so I'm going to ask some maybe like subtle questions just to have an interesting conversation um so in this in all optimistic systems you need to have a large number of Watch Dogs because just to give a simple example if you have a town and you want to steal money from a bank or do something bad and it in the town there are 10 cups if you bribe all of the 10 cups that's true so it's really important for you guys to have an economic system where the number of these Watch Dogs is large and they have to be paid so how and I think other people also

like optimism PBC they never talk about this but they have pretty much the same problem uh so what's your economic model and how many of them do you think you will have on average like 10 100 right you know right the fair answer here would be that right now we are hosting just a single watchdog that that that that is that is hosted by us the code for the Watchdog is available there anyone can deploy it and run it it's just quite a simple process it is going to cost you something like 10 cents per month or something like that so with the with the AWS uh setup so it is super simple it's just every minute checks what is there and it has

it needs to have an access to the near blockchain right so super super simple super cheap uh right now it is what it is but again anyone can can do this and I'm with you I'm with you with the with the premise that that we need to have lots of Watchdogs so if you have a single Watchdog I mean it's fine you guys are like you're developing your thing but currently you have a single Watchdog that means that you are controlling everything no one else is watching and if you want to steal the money and go to the islands no because in order in order to steal the money you need to submit the incorrect block right and then after after the submission of the incorrect blocked you need to you know which dog which is

controlled by you no no no no no so so look okay so we are going to go pretty pretty deep into the into the into the tech stuff right in order for you to submit the in order for you to be able to steal the money you need to prove to the bridge smart contract the fact that there is a transaction happening on near blockchain that is uh causing into into like into the withdrawal of the and in order to prove the fact that there is such a transaction happened in

the in the near blockchain you need to create a block that is signed by the validators of two-thirds having two-thirds of the stake of near tokens and then you submit this right this block and what you can do there is just you can propose the the incorrect block there that is not signed by by the two-thirds of the validators right yeah or like there's some kind of one signature or something like that happening there in case in case the in case the signatures are um so so like you can with the with the absence of the

uh of the transaction you are not able to withdraw the funds you are able to to create some kind of other block with a different order of the transactions or something like that right you can do that stuff but you cannot create a transaction that actually was executed correctly and was introducing the invalid State change without having the a big chunk of near tokens in your wallet and be able to to work with it so it is a little bit more complicated thing but what's the purpose of the Watchdog then why do you need the Watchdog then you can remove it or I mean I'm just maybe asking no no questions but yeah so well I propose we

we start the discussion a little bit outside because I believe we're we're running a little bit out of the out of the time but uh but yeah still Watchdog needs to he needs to make sure that people are not submitting the blocks uh with the incorrect set of the set of the uh set of the signatures from the validators remember because this is the consensus algorithm of the near blockchain yeah but the but again coming back to you to to what you have you what you have seen I'm completely with you the centralized set of set of the Watchdogs making the Watchdog as simple as possible without any big hurdles for it uh this is the way to go and yeah if anyone would like

to run the Watchdog here well first of all the incentive is in case someone behaving that you are going to receive the slashed amount of f right uh but also I believe Rainbow Bridge team and Aurora team as a whole can can incentivize you with some a little bit of money so you would be able to set up an AWS instance for you yeah Okay cool so uh thanks everybody [Applause]