Cross-chain Oracles with Jakub Wojciechowski of RedStone

Transcript

hi welcome to the interop this is the show which explores the decentralized economic networks that make up the inner chain my name is sebastian and today we've got a bit of a special treat for you i've got my buddy uh richard catano here who's gonna be guest coasting guests co-hosting this episode and we have our guest jakub wojciechowski who uh is ceo of redstone um so actually this is interesting because like we met last week at def connect in amsterdam and it was like a total chance meeting and we ended up having dinner and you know we got into like these deep conversations about um like you know crypto and data availability and modular blockchains and stuff like that and like

you guys are doing some interesting stuff also with our weave and like i was like okay i want to like dive into this and um and richard being someone who i know and trust and who's building on our weave i thought like this is the best uh combination to have here um on on the show so thanks thanks for the two thanks to you for joining me thanks for having us yeah our pleasure so uh yeah yeah um i mean maybe uh just richard uh perhaps just uh give us give our listeners a bit of background on yourself and um what you're doing uh right now uh yeah so um well

i've been been in the blockchain crypto um world for quite some time since uh nearly the beginning and that's how i kind of got to know youseb early on and uh currently we're building a comp our company accord we're building this uh privacy layer on top of our weave and trying to you know build a new set of uh storage tools based on web3 concepts where yeah where are weave as a permanent storage layer is providing a bulk of those services and we're also working with redstone quite a bit on on the development of our protocols and our contracts that run those protocols and that's i guess sort of how we all kind of converge together in amsterdam yeah

yeah i mean richard's a bitcoin og he's been around since like even before i have and um yeah we actually co-founded a company together a couple of years ago uh i forgot to mention that part those who listen to epicenter uh you know will remember uh stratum and so richard was the ceo of stratum and um you know we've uh we've we've come through that uh intact and so i'm happy to be doing this with you and uh yeah yeah coop thanks for coming on can you can you tell uh tell us a bit about your background and like how you got into this space sure so thanks for having me uh so i'm a computer scientist by trade another education that i used to work in

traditional finance um like a fintech insurance tech etc and in 2016 we decided to look for something new me and my wife we moved to london i was looking to to explore new technologies and that's how i discovered blockchain participated in the very first blockchain i got unlucky enough to be to be a winner that was the year when they team came out and then i co-founded the startup it was called alice was a philanthropic startup and then obviously it's not uh that the greatest business you can imagine in terms of the commercial potential so we we slowly started to win down the

startup when the copied extract and in the meantime i was smart contract auditor and participated in multiple hackathons and that's how i discovered more and more issues with the current implementation of the oracles and decided to to do something about it joining the r uh first batch incubator program hello yeah that's interesting so the so so you guys are both doing stuff in the r we've uh ecosystem um richard you're building a company in uh in every are we've ecosystem and and so are you a yeah um

so so this is an ecosystem that i i don't have a whole lot of visibility on um you know we we interviewed sam on epicenter and like i thought okay this is like a cool um data storage play and but but they have a really i think unique vision compared to some of the other uh platforms that that i know of like ipfs and scia skynet and these sorts of things like what's what's unique about the arweav ecosystem and about the r-weave technology that most people perhaps don't really grasp or or understand and you know maybe maybe both of you can can give a shout out that okay so i can start so what i really

like about the ecosystem is it's very technical like in a lot of crypto universe you got different kind of people some are there for quick quick money like trying to to shield some crazy coins and i've never came across anyone like that in our week everything is like very technical engineering focus super friendly cooperative so like for me it's an amazing place for builders to try out the ideas etc so yeah the cultural aspect is super important for me as well yeah i think that one of the

i think one of the important things that our weave brings is this uh you know this idea of permanence and and i think it's quite timely um of a solution you know so i think um i think when i discovered our weave shortly after it was like an ft craze and all these nfts and people were spending tens of thousands of dollars for these images and they were stored on uh temporary protocols such as ipfs or or even worse some centralized solutions and so um having the ability to write those bits out onto a chain and then being able to incentivize miners to um you know to to protect that data into

the future i i thought that was quite novel and it was really a zero to one that i haven't seen before um and kind of extending that observation outward more one of the things that i saw is just it's really the it's really the opening door for what we call web3 right which is complete ownership being able to bypass centralized forms of control and restricted access to certain types of media and things like that our weave kind of blows blows past all of that um so yeah and then and then there are also these like kind of like political and social kind of like reasons that are we've um exists which in terms like you know preventing censorship and so one of the interesting

projects that we see right now is someone or a team built a bot um that is recording like very sensitive documents in the ukraine uh a russian aggression and and and making sure that they're able to store those documents so that later on historians can kind of filter through it and like find different things about it so i i don't know i find on different um on different layers different levels like uh our weave is quite quite an interesting project and that that's what um like sold us that's what got us really interested in it yeah that's interesting um yeah i think one of the things also that and you know we'll talk about this here that that are we've uh sort of innovates on is the the the smart contract paradigm i

think in in our weave from from what you guys have told me is like totally different from uh from what we know in ethereum or solana or the cosmos ecosystem and you know one of the things that i want to talk about today is this shift towards more modular blockchains where there's like a there's a data layer uh and smart contract execution happens like in a separate layer and this is a trend that i think is happening more broadly in ecosystem with projects like uh like celestia and like even theorem two moving towards that model so um so yeah i mean you know i'd love to get get into that and like and get a deeper understanding of all this stuff but but first i mean maybe kuba for you like um what what made you want to

what made you want to spend your time on on oracle's because you know like if you look at the space like there are multiple companies and projects building uh oracle uh systems you know like chain link and there's a whole bunch of other ones and so um [Music] where where did you see in this space like a project a problem that really needed to be solved and how is redstone addressing that okay so it started out of my own frustration with the current level of oracle cells participating in multiple hackathons and often we got a really good idea and you need some some additional data like the data about

interest rates on chain or or the liquidity of different pools etc it's like perfectly easy to grab from the afternoon environment but if you try to consume that within a smart contracts then you can encounter some issues and i was really curious like why the current incumbents with a lot of experience and definitely very major technology cannot cover a weight our range of different feats and definitely the most popular uh

design pattern of providing data to the smart contract is still through the engine storage which solves a lot of issues with data availability with the security of data transmissions etc but on the other hand it's very expensive so i think there was a moment when this the cost side was actually really really heavy and it was like um preventing the the oracle ecosystem to cover more and more assets or any interesting data feeds and yeah so i was trying to do something about it because as you mentioned channing he's i

think his amazing company is getting a great reputation but the problem that it solves right now was a problem of the the blockchain ecosystem from like three or four years ago when there was only like a few few dozen of interesting tokens blue chip tokens and only a ethereum was a very dominant network while at the moment we got like like a cambrian explosion of not only layer one networks that are evm compatible but also in a plenty of l2 networks and obviously even thousands of interesting uh assets that you can query price of not to mention other interesting

indicators so so for me the the goal was to build a really scalable solution that can out of the box cater to any evian compatible chain and also to lower the the data access cost to minimum so so we got a crazy idea of you for utilizing for storage and network that was designed to be a storage network that's how we encounter our with instead of utilizing a storage of evm chains when when on that kind of an architecture the storage is heavily penalized because you focus more on the throughput and you want to discourage people from extensive usage

of input and output so use refresh storage and we got a really innovative technology of passing data from rv or from within our nodes to navm compatibility chains like i can go deeper into the technical aspect if you're interested but yeah yeah for sure um but uh but so so you guys brand yourself as a cross-chain oracle um what what does that really mean and like are you do you mean from that that you're pulling data from different chains or that you're able to provide data to different chains or both and how does that actually work compared to like this more uh traditional oracle design

that you were talking about earlier yes i'm happy to take so transfer it shortly it's both but it's not a pure market i think having a lot of oracles can say that they're crosstalking because if there is new chain then can spend like three four months to integrate with that new chain and be compatible in in our case if tomorrow there is a new evm compatible chain we can start serving data to that chain without any changes in our architecture so because we do not touch the storage of the target we do not need to to integrate so so we are really really focused on being interoperable and yeah it's it's actually a

a fact let's say deeply rooted in in our architecture okay um so what what kind of what kind of data uh sources do you pull from what what's the the kinds of data that we can um we can we can get from redstone so so at the moment we got 60 data sources but we are still like quite early and it's also very easy to add new data sources so we work with a few projects and we get open architecture code is fully open source that then even can implement their own fetchers conductors

adapters to external data sources and in terms of the the price feed that we currently cover that obviously like crypto tokens more than a thousand tokens and from other interesting price fees we get currencies a lot of currencies from for example the the lighting of south in america which we we get a lot of requests for that and stock prices commodities what we are looking to are obviously the nfts which will be coming soon we got the capacity with the low cost to cover a lot of the nft prices and also we'll

be rolling quite soon a very customizable oracle so anyone can can specify in a uh user interface set of data feeds and then start putting data so yeah our idea would be as broad and as possible to truly utilize the low operating cost so i had a quick question maybe a little bit to dive in a little bit deeper from your introduction um with the oracle service so kind of the two aspects that i think that are kind of um a relevant uh discussion have around oracle's is basically things like latency and uh gas costs and um you know how are your staking against that could you talk

a little bit more um about what you've seen different between like other chains and are weave and then also what were some of the challenges that you may have experienced in trying to solve those aspects and if you want to describe those aspects a little bit further that'd be great as well okay strength to start from the beginning the yeah the latency so so currently we we offer in a stable way that press fits refresh every 10 seconds we can experiment it with one second interval is is quite fine although some of the often data sources are not as fast so probably we need to to

better go locate our notes for that but i think it it should be possible in terms of gas costs we we got a hacky way of actually smuggling the the messages leveraging our user interaction so we appended the the price fit in a special format for this time so it could be cryptographically verified to a call data of a user transaction and then every user that would like to to execute the transaction within a context of that price carries the the price fit and then we extract it verify the the origin of a price feed

check at the protocol level where the protocol contrasts given source and if everything's fine if the time stamping find the kind of type of data is fine if the the creator of the data is also verified by a protocol a user may access that price information and all of that's written purely in the solidity assembly to to be as efficient as possible and yeah in total that the whole operation is about 10 000 gas costs which is super cheap so if you compare to the usual cost of the the referential price contract data access which is about like 20 or 30 000 gas so we are very competitive on that level yeah interesting i know one of the one

of the other topics i was speaking earlier with seb was um i think it was oh i can't remember how long ago it was but i think uh there was this article written by vitalik where he made the suggestion that uniswap could function as an oracle uh quite efficient oracle um what's your view on that um kind of like this dex versus oracle and where do you see advantages and disadvantages between the two and what what's your outlook on that subject yeah so i'm really impressed by the amazing job especially by the union swap v3 on the on the yeah how they optimize that the whole solution so it's yeah it's it's a great

step forward i'm a bit cautious always into trust a single source of truth so even if there's like a lot of liquidity in news web is still a single service so anything can go down wrong with a single service so always it's better to have some some backup some redundancy in terms of the data sources so that's that's my concern yeah we are actually as as you know all of the data has been persisted to our we've had at the moment we've got almost like a one full year of a data records that are publicly available for anyone and we started we recently hired person who is dedicated to to crunch those data and actually to answer the

questions that i got in my mind for a very long time like are we there yet are they the decks let's say the data quality from the access comparable to to the biggest off-team players and what are the differences etcetera to try to find the patterns yeah i'd love to share the insight yeah with you guys actually that that what you're discussing i find very interesting in that discussion which is so if you compare the nature of let's say like how a dex works there tends to be a lot of arbitrage to find you know what um and you know to find a spot price between but you're always kind of like arbitraging back and forth or counter balancing the trades

that are being made i should say whereas an order book when you place an order in you're buying something at a specific price all right so so the price mechanisms between the two types of exchanges um there seems to be a dependency going one way and that is saying that like the those who are arbitraging between a decks may be referencing centralized exchanges uh for spot price i i this is just kind of like loose thinking but what i mean what you're kind of talking about this what is your were your thoughts more about that like do you find that um that this phenomenon is probably expected early on but as dex has become

um more utilized and you know more efficient that maybe this um um you know this gap between the two may kind of go away yes it's definitely like having a larger on the decks and very more efficient yeah we have actually it's a very interesting topic like how the how the v3 actually like manage the the volatility of price so on one hand is it's like really great to utilize their ability on the other if you get out of this supported way and you really risk like a very sudden drop so that's also like super interesting

like how it will play out in the terms of the data stability and obviously this was like an answered question like how to actually utilize these two apps from from uni swap was the was the proper range because if it's like too short then yeah okay keeping in mind that it's a dex and it could be connected to a transactional risk into a really smart arbitrage but if it's too too long then you lose the the accuracy of price data especially important from the level so yeah i think yeah we've been discussing a few teams what's what's the like the best approach to that and i don't think that anyone yeah i think it's like um

because i i'm a big fan of v3 you know swap v3 um it seems to me to treat to treat trading on v3 sort of like a um limit order and on and rather where limit order would immediately sell or buy you know based on your position that you put in it's more along a curve right so you're selling along a curve until you're you're sold out of the range um i think what could make v3 better is um you know faster confirmation times being able to swap in and out i think one of the things that makes it unusable is not being able to get your order in or to shift the range if you need to and paying extra you know paying high

amounts of gas to do that um so i don't know maybe on a different platform that has a faster execution it might be more practical to use but i think one of the big drawbacks like you said is like you know if your range is not perfectly in some kind of action you're gonna you're gonna lose out so or from a global point if if the range is like too thin then definitely mark it yes sometimes you you see like a you know very long spike like somebody's just trying to like liquidate you know their entire position all right and you know it's like sometimes if if the range bounces in between that line they're just you know they're sweeping basically sweeping a large

amount of fees it's really interesting i agree so probably there's always like a like a trade-off between like super let's say efficient price discovery wins you should have that the same liquidity forever price range and profitability of liquidity providers so yeah i don't think it could be like perfect too much i'd like to talk a little bit about the the design uh the like redstone's uh design um and specifically like what the tech stack looks like um can you describe like

what what that is and so who are the data providers and how are they incentivized to provide good data um yeah like what does that look like from a sort of like design perspective of course so so a data provider is actually a node that um publish kind of we call it manifesto is kind of a publicly available sla like what kind of sources are going to be utilized what kind of uh aggregation methodology to be used whether it's like media and to a twab or anything like that and also what's like refresh interval and then it's being um accountable

according to that it's kept accountable according to that manifesto by by users who consume the data also the the data providers specify the the pricing policy and that at the moment obviously we use the easiest one as a subscription but we are going to to experiment with the pay as you go pricing modeling can be charged per single price fee that could be interesting for the space and yeah is so that's on the revenue side and also on the let's say guaranty site the the data provider in case if the price is

not prepared according to that manifesto maybe it wasn't being delivered or uh data provider cannot defend that it was taken from given sources and aggregated properly then a data provider could be slash and lose part of the the collateral which is strongly required for a provider so that's kind of an interest mechanism as i mentioned the the prices are this was like super interesting like how can you take a price from a data provider and and pass it to to a target blockchain as we do not push it to the storage so so we've been thinking a lot how to secure this

we layer part so it's very similar to like a xerox layer but we used a lot of red and dancing mechanism so we got our own light nodes like super dump nodes just taking data and copying data and making sure that the signatures are correct we also integrated with streamer which is a super interesting decentralized pops up mechanics having like few few thousand nodes so it's really not easy to deduce the service especially it's like in general it's not easy to deduce a pops up mechanism and here we also got our own uh caching servers which are much more heavy but but can handle tons of traffic so we got these different levels to the security

transfer data and obviously we got rv as a as a permanent archive we serve two purpose first you can have a full track of record of any data provider if you are doing due diligence and choosing the the best one so there is the highest level of transparency and also super important for us to have a regular heartbeat on the activity of oracles so to detect any issues as early as possible so if the oracles are forced to provide the information at every interval then then we can detect any issue sent quite quite early so on the guarantees side you said that the

the data providers provide a they they submit a bond and they can be slashed is it an opportunistic model where uh someone can come and provide a proof that that the um the data wasn't provided correctly or that there are discrepancies in the data or or is it uh or is it another sort of cons what's the what's the consensus model i guess around like providing guarantees yes so so currently uh and is it sorry is it one and one follow-up is this an ethereum contract or is it something like that okay it's actually the the smart with contracts based on rif

we call it dispute resolution mechanism and in the in the current form when we it's it's not live used but at the level that we are in uh we got the very democratic mechanism like you start a dispute then you can have a counter agreement then in case of the one of the parties not happy with the outcome it could be escalated and at the very extreme case the whole network needs to to vote who was right so it's like very still very many while we are experimenting with automatic proofs but it will probably came later um okay and so

and so the the data providers will will set their bond in in what currencies like are so that the bond is requested to to be set in our token but that the token is not down to it so okay yeah exactly the main purpose of the token okay so so the the data providers uh put up a bond then there's a dispute mechanism that allows a counterparty to dispute the ability of of data and then there's a process through which there's a decision that's made on the ability of that data and then the the bond i suppose is burned um

[Music] that it appears to me that there might be enough an attack mechanism here where someone could um you know if they want to sabotage uh a data provider um just create uh you know many disputes on um on a data feed is there like a also something at stake for uh one who is making a a dispute claim like a bond that they also have to put up yeah yeah so so being a whistleblower is not let's say obviously we don't want the protocol to be spent with the false alert so and we said though we need to

to provide a minimum stake on his side to prevent like raising falls around i guess i kind of want to understand a little bit more of this our re smart contract uh system so can i mean yeah i think i think you guys are maybe the two people that i know who understand the the most um and i i i some some people have tried to explain to me but it just hasn't i haven't quite understood so hopefully uh you guys can fix that um yeah like what how how does it work like i mean what how does it work and how is it different from say a contract on ethereum or

like a cosmosim contract on um on cosmos which i think most people most people are more familiar with those uh two designs which i don't want to start because i was like monopolizing it well i think well um yeah i guess to come in kind of a high level um so i think if you describe how the how uh san etherium smart contract works this you write code it's converted in some kind of byte code and then this is executed in what we call the ethereum virtual machine which runs on all the miners and all the miners execute this code and the execution of that code is

recorded on a blockchain and so then future miners can then pick up confirm and verify and continue right so so a lot of the so the execution is happening um within the miners and stored on chain and with rweave um since we have this permanent hard drive we could we could write quite a bit of you know of data to it and then deal with the verification and processing later and so the way that the smartweave contract works uh from my understanding and and jacob can probably go much detail that i can um basically we store the state and the contract on our weave

and the execution and verification of those contracts are done are pushed onto the client so i believe they call it um lazy execution uh jacob do you remember what the yeah lazy execution so basically you're only storing state and contracts on chain and then you execute and verify off off chain on the client and i think that's where we found at accord i mean we use redstone's smart contract system uh for what we're doing and um while that in theory works there there's quite a bit of infrastructure that has to be put in place to make it practical and you and you um being able uh to be able to use it in production and so i think that's

probably a good place for jacob to start his kind of go into detail about the red stone smart contracts and how how you've implemented and improved uh the smartweave contract system because basically you guys rewrote it from scratch right if i understand yeah so i think that the credit for that the idea has to go to to sam that the rv founder who who prototype the the first version yeah i was like it's super interesting so yeah so at the first glance the definitely adoptive thing that yeah is a scam it will never work etc because it's so different that the current uh design part of the smart contract space but if you look deeper it's actually

like super smart to to detach the two phases of the of the data storage and in the data execution so if you get a system when you can actually achieve a consensus on the the pure transactions without executing them so if you can order and persist transactions and with the assumption of the deterministic executions that the same set of transactions in the same order will always get you to the same result then you should be able to detach these two phases so so have a very cheap consensus on the ordering and storage of transaction and then delegate the execution to to other services i think it's also

getting popular in the the current roll up patterns when they try to to compress the transaction put put a proof on chain and then move execution to the layer to network so stuff like that is getting more and more popular and that's also the the reason why solutions like celestia are popping up which is a chain dedicated for data availability when you can actually have this transaction storage and then connect it to different execution environments and as as we just mentioned yeah in theory it looks really really cool but if you get to to this let's say implementation part then there could be

a lot of really interesting challenges like actually how to make it fast and efficient how to deal with the the data access the necessary indexing how to sequence the transaction put ordering how does to make it [Music] let's say efficient enough to be production ready because theoretically it's a fascinating concept but then if you delegate all of the execution you need to also think like who will be executing the code whether it could be fast if you need to re-execute like all of the historical transactions how to efficiently cash the the execution how can you take snapshots of

states not to rework the full history etc so that a lot of super interesting challenges and we try to solve with one after another still being able to offer the solution to to other projects so okay so maybe just uh to sum up here so with with a with a sort of evm design contract we have the execution that is performed by minders the state transitions themselves are stored on the chain um in in addition to the into the data and so

in order to reconstruct uh and to verify all of the all the transactions we have to go through the state transitions to make sure that all those are correct whereas with uh with an rv contract the contract and the and the and the state are stored on chain and what you do is you say okay this is the contract this is what was run this is the resulting state and so you have to rerun you don't have the state transitions on the chain but you have the uh the state and so you could technically just run a contract on your client locally and then store that on the ch and store the

the resulting state on the chain yeah so so in other words you got all of the the inputs on time and you can always re-evaluate the state to to get an output but it's like to tie the touch so so that the consensus mechanism doesn't require you to execute transaction and therefore it could be much more scalable so yeah so so the benefits that we get are that you don't have to store the state transition on chain which i guess is what takes up a lot of space you just have to put the contract and then the inputs and the outputs um is that right

yeah exactly so if you imagine a transaction that is like really resource heavy like like data processing transaction which is currently impossible in any vm compatible chain because it can slow down the whole let's say other transactions so it will get out of the block limit and and therefore if you detach the the execution and you simply store an input as an order to be executed later on you can fit a really heavy transactions and then try to execute that in let's say dedicated network whenever you actually need that state okay and who who like if if the if the execution is being done by a

single client who's verifying um who's verifying these state transitions um and yeah like how does that work in a practical sense yes sure so you got the code you got the inputs and if you are an honest user you always get to the correct result and so anyone can can let's say verify the end state anyone has all of the necessary ingredients to verify the end state and it's just quick to verify like can my client just look at the chain and

like verify an entire contract at the moment in the original form it's not quick to verify that's why and yeah it's let's say honestly it's nice approach to require every end user to recreate the site that's why we are building a dedicated like a delegated execution networks like a this is a complex structure that we are putting together that should help you to first get answer faster second level is the dedicated node to to evaluate the state and to verify the result much quicker so yeah like having like a snapshot roll

ups out of the networks that should be confirmed by by the major of rather like with a fraud proofs that should be refused by the majority so the client can trust snapshot and only re-evaluate the smaller part that should be possible for let's say end users okay so is this so and then what what are the what are the trade-offs in terms of like can you still achieve the same type of composability as you can on a on an edm chain or is that more does that make it more complicated it makes it much more complicated i mean

you you name that trade-off perfectly well so yeah we are working on cross-contract calls they are much harder to implement and they require some innovative thinking it's possible but definitely i'm not going to to refuse the statement that in terms of the compatibility we won't be as composible as as evm or networks like like solana when they get a very single state accessible to others but on the other hand we are much more scalable and we we allow like really heavy computation that won't fit in a block so so that's the spectrum you get a lot of flexibility you get a lot of scalability but then

the compass ability is much harder but but still it never be possible and we do cross-contract calls although they are much more expensive to evaluate also by its very nature it's parallel uh execution's in parallel right so you're pushing it to the client yeah and like jacob was saying you can have some contracts that could be very computationally expensive run that in parallel there's no no impact on the overall system so are there are there use cases that are kind of geared towards this type of uh design pattern or is it something that you all think are going to be is going to be generalized um

well if i may jacob i think from from a court's perspective it was a a perfect match for what we were doing so maybe just to talk two minutes about that so we originally built a cord with a particular block chain in mind but the blockchain didn't exist so we built this kind of protocol that we ran in our internal blockchain just waiting for the the blockchain to arrive and luckily it was our weave and so basically everything that we do in accord whether it be creating a new vault inviting people to the vault or members of the vault uploading files revisions of the files posting notes and things like that all of these kinds of activities which happen encrypted in a vault each each action is encapsulated as a

transaction on reeve and so what we're able to do is just map our protocol and transaction set directly over to smartweave's contract and that allowed us to just port things very quickly and essentially what we're trying to create is this like a kind of like audit trail of each of this um of the states in time for that vault right and those and pointing out that you guys are building a sort of decentralized dropbox right exactly with um versioning like document version and traceability yeah yeah and notes and editing and messaging and collaboration features all of that so it's very very interactive

and um where i was getting it was like so we're doing a lot of heavy encryption and decryption so that is one of the examples where we can offload that onto the client um it has to be on the client if you want into an encryption and then just send it over bits and then record those as uh state transactions in the protocol right in in our weave and so that worked out really well for us it we were able to uh deploy our protocol i mean within just a few weeks so it was pretty quick for us um where we ran into a hurdle was or we rented obstacles where um uh our we've started getting very saturated the gateways were saturated we couldn't commit

uh transactions and then um we were also trying to address some of the like remote calls and stuff like that which redstone was able to address and so um so yeah i think uh yeah i'll let jacob take it from here i think that's kind of a short story for us yes amazing example definitely something we've been thinking of and it's great that it's been validated by by richard and echo yeah and so so there is a the data processing logic when you need to have a a lot of access to the basic storage which probably is is impossible in interchangeable evm based chains but also

if you'd like to think of any application that actually makes some storage and logic like like native nfd processing when you do really some manipulation with the true nfts instead of like passing ownership of hashes or if you do any transformation like like atomic energies yeah like yeah a comp i was a company that uses us and that's processing the the digital recording songs pianity it's like super interesting so there's also like heavy data processing comes up probably from the metaverse uh ecosystem when

there is that the projects are forced to to process heavy loads of of data rendering information etc that could be quite an interesting angle for us yeah and any kind of like either heavy computation a lot of uh input output requests that's a good match and as i mentioned as a trade of the composibility could be limited yeah there's a couple other things i'd like to mention you mentioned the nfts um i think they're i believe they're called atomic nfts on our weave and so what's an atomic nft yeah okay so so crazy so earlier on we talked about how

with nfts you've got the contract essentially the uh token stored on one chain and then you have the data the graphic the jpeg or whatever it may be stored somewhere else and in this case our weave so you've got two you've got the the um nft split into two parts token and um data and so what our we've unlocked was okay we can not only store the data but we can also write the contract for the token and they're inseparable on the chain okay right so that's that's what it calls um atomic nft so it's just one complete unit and then there are some other projects in the ecosystem

um that allow you to bridge those tokens so you can have the token live on rweave and then trade it on solana or ethereum something like that right basically wrapping it and somehow so um so that's very interesting um there was a second um oh there was another use case that was really interesting um i'll come back to this later as i can remember yeah um okay i know that makes the i i i think i understand a little bit better now that like so how does this fit into the cross chain thesis because this is one of the things we were talking about uh when we were at dinner the other night and um

you know i really want to discuss here like so with regards to redstone and the the oracles that you provide the data you provide how does that uh fit into the improbability thesis that say we have on cosmos and um and i guess more broadly like where does this fit also in this uh modular blockchain idea like like this this idea that's pushed by project by like celestia um or we're hoping the cna ethereum two can you kind of unpack all of that and maybe make it a little bit clearer because it if it's still a bit nebulous for me sure so so

let's start with the latest the modular blockchain design so when we start building our smart contract uh infrastructure then definitely having to to work on the on the storage layer secure storage layer and then the secure consensus uh layer and the data availability proofs all of that from the beginning as a as a which is often called like a monolith if you'd like to create the protocol starting literally from scratch it will like take us probably like five to ten years to get this secure but actually there's like a pretty good and solid storage

network called riff also like celestia is entering this space which could be like a really good uh composible component because celestia focus on the short term very strong availability proofs when rwf is amazing in the long term data storage and having a full history of a chain and then obviously we'll be using cosmos for the [Music] interconnectivity of our execution networks so they can they're being kept accountable with their stake and that the correctness of the computation could be verified and roll up to

to that network so so we use of a major pieces of technology like like rwef cosmos probably celestia and try to glue them together adding our own magic sauce which is the the economic model and the ability to target the solution to a given use case which in our terms are data processing or let's say that the flexible smart contract execution so in this modular world you try to experiment using all of these amazing tools that were already developed and try to creatively join them together to to create a blockchain without unique

features so that's can you can you elaborate a little bit on this on this uses the cosmos sdk and [Music] because i'm not quite sure i understand like how you guys are using are you using ibc to for you know so so so we i mentioned briefly like it's not feasible to expect all of the end users to fully re-execute the transactions so there is a need for dedicated professional evaluators that confirm networks and to keep them accountable and incentivize them to provide correct answers there need to be a mechanism where they can

usually like leave a collateral provide a state that could be slash etc so there is a need for a network to manage that uh coordination and for that we'll be using cosmos so that we have a typical probably a b c i extension or the new plus plus version okay understood yeah all right so you'll be making use of the cosmos sdk uh to build a cosmos chain that will enforce um this uh sort of incentive mechanism to provide data okay so are you planning to launch this as a as a standalone cosmos chain with your own validator set or will you be using like interchange security or

so we are going to start from the standalone chain and we we see what kind of an extra value can we get from the interview t with the other cosmos chains okay interesting yeah because i mean at a minimum like i think like yeah being able to uh like have your token uh as like an ibc token like tradable on texas like like osmosis and stuff it's kind of interesting um and but i don't know like what other like what other uh yeah you know benefits would come from having um this chain be interoperable with the uh with the rest of the custom secret system i know maybe they can delegate like

you know data storage logic to us i think this is again amazing yeah let's see where to to explore the potential synergies yeah um and then so so okay so i think i think i'm getting a clearer picture here where celestia uh is really strong with short-term data availability and i like i haven't really dove deep into celestia yet like that's an episode that i want to do soon you guys are [Music] better suited for long-term data availability because you're using rweave as the data layer and so

is it something that can be used in um in coordination with celestia or is it for two different types of use cases is it like two different types of use cases or is it like a you execute on celestia and then like back up on on smart weave or are we looking at like different types of applications here yeah so i think they're very complimentary and we'll be thinking about the solution like like celestia to provide a really strong data validity guarantees especially between before the transaction could be properly seeded on the rwef network and yeah we are super super happy that the guys from celestia are implementing that we recently

visited the modular summit and definitely we are going to be like early adopters testing out the solution unfortunately it's not like production ready yet but i think it's amazing potential to to connect the two to side of a continuum of data storage to because celestia incentivized the the data availability in the short period of time why are we with some amazing of the the permanent storage guarantee but not as as quick as celestia to give the the data ability proofs and this is the amazing piece of technology with the erasure proofs really cool stuff richard you're muted

sorry about that yeah so i believe that the saying is like it's uh permanence over performance is what our weave is uh tuned for one one of the other things that just came back to my mind when we're talking about use cases with our weave i think there's something interesting to say um here so you know when we have data on chain when we have the application because application itself is data files that can be stored on chain as well and then you have a token with logic and and you know contracts and things like that so you basically have data application

and business model all running on the same chain and i think that's that is like where web 3 can take us like this whole idea that a business model and the data and app everything is together on chain and it's available for as long as that blockchain is available um that that that that starts to become something very interesting and for like for what we're doing at accord um there's this model that's i believe it was sam i may be mistaken but i believe it was sam he introduced what was called uh profit sharing tokens and so this is where you can build your application and then the business model that you know you you pay um so in this kind of model you only pay

for what you use whereas a web 2.0 model you have a subscription fee and you pay you know 20 bucks a month or something like this or 40 bucks a month with a profit sharing token model you only pay for what transactions you commit all right so if i create a vault i pay if i add people to that vault i pay and what that means is that each transaction the fee that you pay basically has three components one component is a transaction fee so you pay the miners to confirm the transaction the second component is the storage endowment so you depending on how much data is uh included with your transaction you pay up front for that storage on our weave

and that money go that the the tokens go to a crypto endowment that are projected to grow over time and then as storage costs decrease over time the difference between the two is the incentive for miners to store the data and then the third component of what you're paying for is you're paying a community fee and so if you have a protocol with token holders behind the protocol that community fee is paid back to the token holders so now you've got this like incredible potential built up in a flywheel right so as people start using the application they start using the protocol they're creating transactions those transactions are generating fees the fees are being paid back to the miners to the storage

endowment and to the token holders and then this just kind of like picks up momentum over time and that's an incredibly powerful model um not too many people know about it it's a very new kind of like interesting idea with what they are within the rv community but i think it's one of the ideas in that community that has the most potential yeah i know you've been telling me about this model for uh for a while and i know how uh i know how bullish you are it's the dream it's the dream right data app and business all on one chain like yeah and fully integrated um i think we'll we'll see a lot of that uh

develop into the future especially as web 3 becomes more adopted user ownership of data you know you're buying you're you're owning part of what you're already using and so if you look at the incentive structures of web 2.0 the incentive structures are set uh basically for the investors of the webtooth project money goes into the team team produces product product gets used by customers but the data starts to become very available so you look at facebook you look at a lot of social media becomes very controversial and then also you have like you know users are now locked in maybe they need this much of your product but

they still have to pay the same amount as everyone else right so everything's really stack incentivized towards those who hold data and those who are investing in services uh to build around data with the web 3 model where data application business model are all unchained people own tokens and that's their like stake in the protocol they're using the same protocol that they have a stake in suddenly the incentive structures are all lined up between the developers the investors and the users of a system and so that that's what i'm trying to describe in a flywheel kind of analogy that that once these incentive structures become all aligned um like really interesting things can

happen so i guess there's a lot to say about that we continue talking about that for another hour yeah we'll have to do another another uh episode about uh about the economics of our weave um because like this this is it's something it's something you can tell me about and i keep trying to wrap my head around and uh you know other friends of ours are also trying to wrap their head around and um you know you seem to be very very keen on this vision and you know i think you're probably right about it uh just where it's too early for us to even unders you know comprehend what's going on uh but yeah thanks uh thanks to both of you for uh for doing this with me and you know diving deep into

into all this stuff um uh thanks for having me yeah no problem where can people learn more about redstone and is there a call to action here yeah sure so so invite everyone to to our website redstone.find us or you can find us on twitter come there's a link to our discord come ask questions we are like super open we recently launched an initiative called to rest on academia and try to teach how to implement the contract starting from very beginning so even from the web 2.0 and not familiar with all of these crazy blockchain terms we try to take you by hand and introduce to all of that step by step showing the the tools how to

test your code how to deploy in the very safe environment how to enjoy step by step start building on on our with using our contracts and exploring the space so yeah we we are really open and try to to help you to implement even the most craziest ideas there cool and uh yeah rich maybe uh take this opportunity to to she'll accord a little bit uh and uh what you guys are doing well we're happy just to talk about you know i think on this show but the uh our partnership with redstone um we've published some articles on the blog talking about what we're doing together um

and a lot of this will be coming open source in the near future uh weeks if not next month or so so um yeah just uh if i were to show a chord yeah we're here to help protect the data that you care about the most um and we're going to help you take care of that data until the apocalypse so that's our goal the website is accords it's not core.com okay yeah i use the chord i like it i i used it for uh we're using it to like share legal documents and stuff like that and it's like this is this is like little vault sort of it feels like a signal chat but where you can also you know drop documents and things like that so i think it's like yeah has a lot of promise

well we hope to deliver the uh developer tools this summer so we'll be very interested one of the areas that we're interested most in taking it is to provide dow tooling so providing long-term storage for your dowels storing your votes your documents and things like that to help to help provide the dow members the information and data that they need um over you know a long period of time potentially multi-generational that is yeah that's cool it feels like it's your life's work i i've known you for long enough to know that like this is something you've been working for on for a long time in various iterations uh but yeah it feels very much like something that you like really want to achieve like being able to you know preserve preserve data

um you know have integrity of data and privacy around data yeah so cool thank you yeah thanks again and uh yeah have a great yes your day or evening all right thank you so much it's great to be there