Decentralized Storage for IBC Chains with Patrick Dunlop & Marston Connell of Jackal

Transcript

why is dpin so hot right now I think it's just kind of a question of development time and the differences in complexity of different applications that's kind of what's beautiful right that's that's really what's beautiful about this is that every blockchain if they're IBC connected have this functionality so every blockchain can have the functionality of Jackal we're obviously pretty excited from that side it's like self-hosting but the scalability of the cloud if you want something like that with Amazon just Geo distributed data in general for like say a terabyte you're looking at almost $200 if you want multi- region storage per month per terabyte Jackal is doing the same thing at $8 a month per [Music] terabyte welcome to the interop today my

guests are Patrick Dunlop and Marson Connell they're the co-founders of Jackal protocol Jackal is an interchange storage Network that delivers decentralized storage to blockchains and as far as I know it's the only live production decentralized storage Network in the IBC and Cosmos ecosystem so I'm excited to talk about them today because I think decentralized storage is one of those use cases that's really uh undervalued but is super cool and has a lot of interesting applications before we get started though be sure to subscribe to get notified when new episodes drop every week and remember none of what we discussed here on the interupt is investment advice and if you enjoy this content you should stick with us we're validating on EO Quicksilver osmosis and passage just look for interrupting the active set Patrick and

Marston thanks for joining me today no thanks for having us SE um it's just really good to be here it's always good to talk get a little bit of tech talk early in the morning up here in Canada but uh yeah know really nice having us just so everyone knows my name is Patrick I'm the co-founder of the jao protocol I'm CEO of jao labs and I'll kick over to Marson yeah thanks for having us um as Patrick mentioned I'm Maron Connell I'm CTO of jackal and co-founder of the jackup protocol nice so how did you guys get interested in this problem and uh you know how did you start Jackal to improve decentralized storage um because it is a it is an a market or sort of a subsection of uh blockchain

infrastructure where there are quite a few competitors so I'm just curious what the backstory is there yeah um so we kind of started in 2021 we were working in digital forensics and investigations for law firms um we were investigating like flash loan attacks on ethereum basically was our job uh back then you couldn't really go to the police cuz they didn't really know what this was or what was going on um so you do a lot of stuff civil that being said they would hire outside teams like us to investigate these things and track and Trace during that process of us working in digital friends ex investigations we got into a we wanted to build this internal tool for chain of custody of of digital evidence before the courts that being said said um when you build

something like this we were building it on polygon um it said who had access to what evidence at what time which is super important uh for kind of continuity of of evidence before the bench and we built it out and then we were like okay now we need like a clean forensic environment in the cloud and what does this mean it means only one person can have access to it it' be super scalable and it can inherently just plug into our application buildt on polygon so the no-brainer for us he said okay blockchains they like decentralization sovereignty ownership all of these things uh so we said we'll go and use a decentralized storage layer when we looked at what was currently on the market it was either Marketplace based so there was like no ownership and and there was no kind of self- custody

of the data that was being stored which was really important for us and the other thing is everything was using Cal work blockchain so they were really slow it was like cold storage archiving you take a file and say okay I want to put this to rest for a really long time you wouldn't want to build it as like a kind of a layer of the stack if you're building like a full stack web application so one thing led to another and we were building the solution for our problem and then the solution to our problem internally kind of became something that was a lot more interesting to a lot of people uh so that's kind of the story of why we ended up building something like this is we were looking for a solution to our problem we started on the secet network we were using file coin for Cold Storage

And archiving and then we started building middleware in between that was great and it worked until we started to hit scalability ceilings and we needed to alter the blockchain modules at their core um filec has taking a long time to give her files back it's up to 24 hours with the unsealing process that they have currently and um we said do you know what it might be ideal for us to go and build a cosmos L1 so we can control the modules have that sovereignty and purpose built an application for the use case we're looking to build so uh that's kind of the the Spiel of kind of how we got to the situation but uh that's pretty much it very cool okay I I didn't realize that you guys were doing forensic work before how did you get into that like what's the story there

like I feel there's there's an interesting story there as well yeah so I uh so my father ran a private investigation or intelligence firm here in Ottawa Canada and uh I was a student at University and he gives me a call like do you know anything about like Facebook I said yes Dad I know everything about Facebook and then he's like can you help me like scrape data on like a specific um situation that they were having with with their legal team there and they were trying to figure out uh just capturing digital evidence from like uh from a digital medium because they're kind of back then they were used to kind of doing more like human intelligence type of things um so that was my entry I I started to Learn Python I started learning how to web scrape a little bit and uh I entered there and

then I spent about four years at inquisitive Intel where we were doing all these forensic things and it turned more into a digital shop Marson came on board with us there and that's how me and Marson Met originally and I'll kick it over to Marson for his entry to to inquisitive Intel and the digital friends EX around yeah for sure yeah so I was a student doing computer science at the time and uh so when I was looking for a a internship position inquisitive intel was the spot for me and at first I was like so out of my depth because I'm like I'm a programmer what am I doing do an investigation but uh it it's funny the overlap that you find um being able to like Traverse the web and find little like loopholes and stuff in digital

evidence it's very weirdly related and then while doing that started work on the tool for internal use and that was kind of like my primary role was just getting this tool complete and uh I kind of Hit the bottleneck that we found and just kind of started this very cool uh well I mean uh it's it's great to have you guys your breath of of experience in in crypto now um so I'd like to maybe take a step back here and talk somewhat About decentralized Storage we we mentioned a couple of products out there but you know back in the day there was this this uh product called storage or store J that a lot of

uh early folks will remember U then siia I think they're still around of course ipfs um really I think uh created a brand new way of doing decentralized file storage and was quite Innovative at the time and um probably I think now captures most of the market of course then we have R weave which uh took it even a step further and offered a permanent storage solution um with this endowment that ensures that um the cost of storage is insured over time um so there's there's quite a few Solutions out there and I think I'm probably missing um some other ones that that I I'm just I'm just forgetting right now

but yeah what what is the what does the market look like in inter viw are these products comparable and do they serve the same use case or um are they sort of purpose built for specific types of applications and where does Jackal fit in all this yeah that's a great question so when you kind of look at the the competitive landscape let's call it of of data storage in web 3 um first thing first is ipfs what is ipfs um it's not really a storage layer but it's more of a discovery layer it's it's a way for you can you can index content by ID rather than location which is really interesting and protocol labs when they were looking to build a storage layer for ipfs they built filecoin and um a lot of them a lot of these storage

networks they were built back in kind of 2017 Ico era um and they kind of also had um the Tex St decisions where they were using the state-of-the-art back then and uh they haven't really been able to Pivot from that because it's really difficult once you kind of choose proof of work and then you get all these people to buy these really expensive servers and you kind of run down this this uh product path that was uh that was an informed decision back then um when you look at it everyone's kind of in a weird category of one in our mind so you have filecoin is primarily Cold Storage archiving put in data at AR rest for a really long time if you want your file back it can take up to 24 hours it's a Marketplace based so it's more of a uh like matching buyers and sell and

like service providers of storage rather than kind of being more of a kind of like a SAS product that we used to with like a Google cloud or an Amazon um AR weave is a little bit different uh they're also proof of work um they have like this block weed technology where it's a little bit different then again it's still kind of like this cold storage archiving putting data to rest and all of the players currently right now they built on these proof of work rails and they kind of go in the category of like vertical scalability right so if you want scale you have to like kind of build an L2 and an L3 or an L4 and that's the direction that they have to go based on the Texa decisions that they made at the time um interoperability is not really an option

for them um at the chain level if you want to kind of like interoperate with other blockchains they can't really do that and um that's kind of the state of the market of like the competitive landscape across the board what we wanted to do is kind of take a different approach we saw we have proof of stake speeds we have modular blockchains now we have IBC interoperability and we have the opportunity to build a product that isn't really a cold storage archiving product but actually a foundational kind of hot storage layer for blockchain applications so we can build full stack apps on blockchain Rails um that being said uh we had kind of almost want to say like a last M's advantage in a weird way where we were able to kind of see the landscape see where everyone makees their textback decisions and then make

our decisions based on all the information that we had at the time um and that's kind of a competitive landscape if you want to look at it but Marson do you have anything to add on that um just kind of like at a at a fundamental level with the way that these um existing blockchains were working is things like filecoin were really really great um you've got kind of like this data sealing process which made the speeds really exciting um or sorry made the data availability really exciting because it was proven over and over and over again and it was a really Advanced step forward and trying to minimize the amount of work that the blockchain was doing you had to seal all your data that was kind of like as Patrick mentioned

kind of the state-of-the-art at a time as 's gone on when you look at things like scoin they Limited did a lot of their bandwidth and how big the file chunks were that were being proven for this exact same reason we're at a point now where blockchains have gotten so much better um even just like at a core level that optimizing for very specific things is still important but we can get away with doing a lot more on a lot less because of the way that we're processing a block every six seconds instead of every five minutes so the amount of data we can funnel through those blocks are a lot higher um and what we're left with is just a really scalable really fast data storage layer pair that with a lot

of the advances in proof of stake where you've got uh like interchain security and the like what you can start doing is once you scale out of your own block space you can start creating these like subnets and those subnets aren't any different than the main chain that's running but they're all running off like interchain security from that main Hub so now you're left with this really horizontally scalable environment that just wouldn't be possible without IBC just to add one more thing to that it's kind of the modular versus monolithic thesis right let's say that like I'm I'm a builder on arbitrum I want to have the ability for me to use decentralized storage for whatever application I'm building let's say it's like a social

app um just because of the way that these model chains are built my options are I choose like a Amazon pinning service and I pin to ipfs myself or I use a centralized solution um to like run an API other option is that I leave and I build on top of these monolithic storage layers um so I actually have to leave my community leave my liquidity leave all the benefits of building on like a general purpose L1 and I have to move my entire application uh we think with the kind of development of IBC now we have the opportunity to deliver the functionality of Jack to all the chains and provide value rather than kind of fighting a zero sum game for devs and users which is a little bit interesting so it's a little bit different of a

approach but um at the end of the day it's really just the monolithic versus modular debate but on the storage scale I guess would be a good way to put it no okay understood we we'll talk a little bit more about inability and how this all fits in the modular thesis in a little bit but but first I wanted to get your thoughts on why people are so excited uh about you these sorts of products now you know deepin like what is I mean dein's been around since I mean if you if you consider you know Saia and storj and all these projects as as deepin why is deepin so hot right now H and uh and what do you guys think of this whole narrative yeah I guess if you build for

long enough The Narrative comes to you I guess that might be the thing but um I think there's kind of been a reason why deepen hasn't been the flavor of the month for a long time um number one is when you look at these deep in networks they're multi-dimensional that being said it takes like multip pull things to get right in sequence and working in harmony for you to even provide like a minimum viable let alone sellable product this is uh this is interesting because it takes a long time to build these things if it's like comparing uh like Snapchat to boss and Dynamics when it comes to the complexity of what you're building right doesn't mean they're going to they're necessarily like a better business model but it definitely means that it's a harder thing to build and

when you build a really hard thing it takes a long time to build it and deploy it and and have it really working at scale in general um so I think the timing is associated with where blockchain development has gone um the the teams that are kind of being the winners of the dpin have been building for a long time um we've been building for almost three years now um and we're just starting to find like a taste of product Market fit now so it's uh you have these multi-dimensional products you have like for us we need the storage layer we need the blockchain we need IBC to function we need all of our algorithms for redundancy detection of loss we need cross chain Outpost and all these things just working perfectly together before we can even get one thing right so um I think it's just kind

of a question of development time and the differences in complexity of different applications if it was a DEX or um something a little bit more simple simple to deploy um they're really easy to build find product Market fit really quickly and um just kind of have the ability to um just scale right out of the box rather than these things are a little bit more comp take a long time to build and that would kind of be my answer to dpin and why why now yeah no I think that makes sense um let's let's talk about the product so you know how does Jackal work and how do how do developers interact with it of course um so at a high level uh you can kind of look at jacka's two peer-to-peer

networks you have a blockchain and then you have a storage layer which is our second like storage peer-to-peer Network um at the end of the day the blockchain and the validators you can think of them as the police of the network is like who has what storage deals what storage provider have what files who has access to what files um maintaining redundancy and making sure that everything runs smoothly the storage providers as a separate peer-to-peer Network they uh they store files um they store encrypted data of users and then they submit attestations to the blockchain Via um we have some off offchain consensus with things called consensus spheres that Marsing can get into but um that's kind of at a high level how it functions so basically you show up to the blockchain you tell the valers hey I want to store this file update my file tree create

storage deals and uh just one click sign one transaction and the way you go now your files being stored Jack was a little bit different than other storage layers in the sense that most of them were marketplaces and what we look to do is we we see our blockchain more as a SAU application so you pay per terabyte per month of storage um it's kind of like a prepaid debit card at this current moment until we get like streaming subscription payments through wallets but uh let's say I want to store this file I can store for a month I can store for 200 years and uh I just pay that transaction and it's the validators take care of your files so you don't have to Fumble around encryption Keys manage your storage deals you can just set it and forget it so that's kind of a

really nice thing about working with Jacko but uh on the development front um I guess Maron if you want to get into the difference between developers on our Chain versus how we look to provide value to developers on other chains I think that would be great yeah um I think kind of one of the big things uh um that we kind of really prioritized early was the ease of use for frontend developers and Library developers to connect straight to Jackal um when you're looking at something like file coin for example you have to upload your file to a node on that Network that node then has to kind of like make that deal on your behalf so then you end up paying that node it's a very complicated process to go from starting a file

upload to getting your fil onchain um if you look at like stor a they use like Central nodes to kind of distribute that file out um our goal from the very beginning was to have the users talk straight to the storage providers we sacrificed some of that kind of like sealing time like filecoin was doing made it a little bit more heavy on the providers but what that allowed us to do is it allowed us to have the user's web browser talk straight to a storage provider and upload that file straight to them after making that deal on chain the storage goes okay I know you've already made that deal I'm good to accept this file but the upload is straight there which really makes it feel like a traditional

kind of Google Drive application where that user isn't going through a bunch of Hoops they're not watching their file get uploaded and then it's off their machine now but now they got to wait for it to upload somewhere else so that was a big hurdle for us to get over was just kind of that like ease of use on the user what that allows us to do in the IBC is even more special because we're kind of circumventing that whole middleman you just upload it straight to the providers if you go and say you're on like Archway and you make a storage deal over on Archway that storage deal through interchain accounts makes its way back to the Jackal protocol from there you know that deal is on the Jackal chain same thing your client just

uploads that file straight to one of the Jackal storage providers Archway doesn't need to have its own kind of middleware that you upload to that gets passed over to Jackal the files circumvent that whole chain layer which allows us to keep that user experience no matter if we're on Archway on our native chain or any other blockchain in the ecosystem all your file uploads will feel like exactly the same because it's all going to the same place with the same amount of hops which is really really exciting for us makes user experience far more deterministic um so that like that user knows that they're using Jackal exactly how they want to be using it every single time okay cool so let's maybe i' like to

propose an example here uh imagine we have okay something like uh like stargaze I think they're currently using pinata which uses ipfs but um you know in an ideal world they're using Jackal and you know you're uploading your nft collection and it's going to get stored on a decentralized storage because otherwise like what's the point uh then so you you create your your collection uh you you create that collection on on stargaze so there's transactions being uh executed there so you're interacting with the contracts on on stargaze uh then I guess there must be some kind of an SDK that you guys provide that stargaze can then sort of integrate within their front ends that allows users to just upload directly to Jackal

and so that that SDK then connects to Jackal nodes uh makes the deal there with the service with the with the U the storage providers and those files get uploaded directly s of circumventing um stargaze completely and then I suppose sends back that data to stargaze uh and sort of a call back so that it can be then kind of indexed in their blockchain you know that that's the reference to such such and such a file Etc is that that is exactly it yeah okay and how does like like the SDK uh what is it interacting with directly so when you're doing that upload that file upload so you know you've got a browser you're uploading a file right like it's getting sent to some API is that is that API kind of peer-to-peer with the uh

storage provider or is there a Gateway in between that's facilitating that file transfer yeah so it is directly straight to that storage provider so um if I'm on stargaze for example I submit uh basically like a request for a deal straight to the stargaze chain through that interchain accounts module that we're currently in the process of deploying to like archway and other networks so what we're doing is stargaze gets that deal request that gets kind of forwarded through interchain accounts to the Jackal protocol everything gets settled there once your client checks on that deal and says okay yep that deal has made it from stargaze to Jackal and

it's sealed it's all good to go then it uh it allows the user on their client whether that be a JavaScript frontend or whether that be like a goang server system they can then just send those files straight to the store providers of their choosing those storage providers then say hey I got that file this is the proof that I have it they send that to the Jack protocol and then because that reference is like saved on The stargaze Chain as this deal anytime stargaze wants or user can go to the stargaze chain and they can reference data that exists over on Jackal and then they can go and check it on those storage providers directly without even needing to query The Jackal chain directly that's kind of what's beautiful right that's that's really what's beautiful

about this is that every blockchain if they're IBC connected have this functionality so every blockchain can have the functionality of Jackal where simple smart contract call and then you have the ability to store data and you have scalable storage at the chain level which is just such an important just layer of the stack when you're building full stack web applications whether it's um whether it's stargaze and nfts or whether it's social or decentralized science or people starting to play with AI and ml use cases um at the end of the day it comes down to we need just such a core functionality um for scalable storage if it's over whatever you can fit in the blocks which is uh some data availability networks they let you do

like three Megs um a lot of other chains like it's like a few kilobytes so it's whatever you can fit in the Tweet um it's it's really important for us to have web scale applications on these blockchain rails if we think that this is going to be successful over the long run no that's very cool I can think of all kinds of applications here I mean you know if we look at something like uh like foraster for instance you know that is storing data somewhere and could leverage something like Jackal to save images videos uh you know text um you know I I I tried to build a Blog on R weave was unsuccessful like I'm my my dream is that my dream I'd love to have some kind of like a decentralized

blogging platform that I could use and like stop paying for some like WordPress thing or some you know ghost hosting and just be able to dog food this technology in a very simple you know application which is a Blog so if anyone's building a Blog for for Jackal like I'll use it and you know like we'll put our website up on there but yeah like these are the kinds of applications that I you know as a former web developer and guy who likes to Tinker at things uh you know I like to play around with and I think are just really easy sort of demonstrable use cases like I you know I use the cost for a while to set up a couple websites I was like this is super cool like I mean I don't have to go and create an Amazon account I just like send some tokens

here and I like here's my website right and um so that's that's I I love this kind of use case and I think they're they're very easy also for people to understand uh you know beyond the okay let's store some nft files or like let's do some crazy you know llm model storage and stuff like that like U the the simple use cases are easy to to create and to understand what's the role of the the cosmos SDK module so we describe here kind of the workflow uh you know we're leveraging inter accounts to be able to make cross contract calls do do do chains that want to leverage uh Jackal um storage have to integrate this Cosmos SDK module not necessarily um there's kind of a few things that we've

got going on right now where the core model is yeah as you say you do a chain upgrade you throw in that interchain account controller module into your blockchain and then you're good to go people on your network can start using Jackal over IBC okay so there is a network upgrade required in order to kind of enable Jackal on your chain only if you're going down that rout we have the other side of things where if you already have cosmism on your chain you can deploy a simple smart contract which uh we've been working on for a few months now with sar from the ICF and uh what that basically allows you to do is no chain upgrade required it just boots up it manages everything for you and it

is a full interchain account controller so any CM wasum Ena chain that exists can become an interchain account controller on Jackal without needing to do a chain upgrade which is really really exciting that distribution layer really opens up yeah yeah okay so the module exists for chains that don't have CMM I guess in the future you know a lot of chains probably have cosm wasm what about evm chains how can they leverage I mean I guess they could integrate the the cosmos module but there is there also a way to do this with an evm contract yeah so that's something that we're um it it it's theoretically possible right now um it's just managing to get those kind of like IBC calls over evm right now um I know like os does evm

so we've been chatting with them a little bit um to see how we could kind of integrate there as far as getting over to like traditional kind of existing evm chains like ethereum or arbitrum um that's kind of up in the air right now where we've been chatting with like teams like Union who are kind of pioneering that IBC connection to ethereum and and uh once we see kind of those things mature I I really guarantee that it would be as simple as posting a solidity contract onto ethereum to get those IBC calls going over to Jackal in almost the exact same way that we're doing it on kmom right now got it it's like IBC is kind of like our distribution in that way um everywhere IBC goes we go um and we're we we're pretty confident in ibc's ability to be

that standard for bridging whether it's tokens or getting arbitrary data between chains um obviously the we have backup plans if if in the event that IBC doesn't go everywhere which we think it will um we could also use like ax GMP or we can build our own solution as well there's also this proof of persistence algorithm that you guys have built and I guess this enables U users and and applications to verify that a file still exists how does that work and um how does the how do the so the storage providers ensure that files are like still being stored on their um on their servers and maybe also talking about that like what's the redundancy here and

how does that play into this yeah at a high level um basically we didn't want the end user to have to manage their storage deals locally and we wanted them to have ownership over those files so for our ability to do that um we needed essentially the chain to manage the redundancy and manage the encryption keys on the behalf of the user which is pretty fascinating um so essentially when you go and make a storage deal with Jackal The the validators will spit out three storage deals to three different storage providers um your file will be sent to one provider it will stay there um once it's downloaded um essentially the two other providers will start copying from redundancy to ensure that we always have 3x redundancy on the

network let's say that one provider says do you know what I don't want to be a storage provider on Jack anymore um or like say their electricity goes out or natural disaster happens or tsunami or something uh in the event that that happens and then that storage provider stops submitting attestations to the blockchain for a specific file the blockchain will assume that that provider has either lost that file or it's corrupted when that happens the validator set will break that storage deal with that provider and then send out a new one um to another provider to actually make that deal and then copy from the redundancy so it's kind of like storage whack-a-mole in a weird way um where you kind of um make the validators make sure that you'll always have redundancy on the network and by having this in place it giv us a lot more

flexibility uh when it talks about like the seiling and unsealing process but marison if you want to kind of do like a little bit of a deeper dive into like consensus spheres and um how that process works on like the verification of the files to make sure that we always have them on the network for sure yeah so at a little bit of a a deeper level every time a storage deal is made we build kind of a tree so we split all of the files into like 10 kilobyte chunks and then we build a hash tree so we take the hash of each chunk hash it with its neighbor and then go up and up and up and up until we reach that route that route is the only thing that's ever stored on chain which is really really exciting because it means that we only

have to store a little bit of data for an entire file no matter how big your file is that route will always be the same size which allows us to store like 10 gigabytes in the same kind of space that we'd store like one kilobyte so just want just stoping you there for a second why do you cut up the file in chunks why not just make a hash of the file so if we wanted to make a hash of the file anytime time we wanted to prove that file we basically have to create a proof that we know the entire contents of the file which can take a really really long time if that file is like 10 gigabytes what we do instead is we over the course of time randomly pick different chunks to prove and we kind of lean on statistics a little bit to say

that if you're proving a random piece of that file every 6 hours after you know 10 20 of those rounds it would be very economically un unviable for you to only keep the pieces that you want and then move the other pieces out kind of thing so that allows us to prove very very small pieces of data over and over and over again instead of proving very large pieces of data like once a month um so that's really so why why not use eraser coding here like I because like Celestia has uh Al also does this right where you are able to prove that data was posted by verifying just like small pieces of data on the Chain um is there something here

that prevents like does does file storage as a use case going to prevent uh that technology from being used or yeah no so um eraser coding is kind of like one of those things that we have kind of down our road map right now what we're trying to do is really build this like general purpose storage so if you want to store files directly as is you can do that if you want to store files that have eraser coding happening so you have like 30 different pieces of the same file and you want to store it that way that's totally something you can do as well what we're building right now is that kind of groundwork to allow users to choose what they want um and kind of embed that eraser coding technology into the storage providers instead of the

actual chain itself so that nothing's hardcoded into the blockchain you don't need to do chain upgrades to enable that eraser coding we have this really General platform where whatever data you give the network that's going to be stored that's going to be proven so if that is a chunk from an eraser coded file that's great it'll do that for you but it can also just store raw data okay interesting and all the files on Jackal are encrypted by default yeah so you can choose to just upload raw data to Jackal The protocol itself doesn't actually care what data is going up um whether that's encrypted or not but all of our client libraries um our like serers side libraries all that stuff by default it fully encrypts

your stuff we have a file tree module that manages all of your encrypted data um so that nothing encrypted is ever making its way on chain sorry no encryption keys are ever making their way on chain but through a series of like deterministic signatures and all that good stuff we're able to have a user unlock all of that encrypted data every single time they go to use the protocol just with their private key and that's really important because are all of our decisions being such a LEL the stack we want programmability for the different use cases right so let's say for example um a publishing application wants very different things than like a peer-to-peer file transfer application um for publishing you just

want public files you want anyone to be able to go and look at that file um through like let's say a web Gateway um that we have Shepherd for example so you can just publish data on the jackco protocol um but on the flip side some people want to use it like a private Dropbox where you don't really want people to look at your wallet and seal your files and having that programmability and having um the ability to be valuable for different use cases has always been really really important to us as an organization so I guess you you could build a decentralized Dropbox with this uh as you could do you know like nft storage as we've talked about before but could could you also use Jackal for uh like General data availability for

blockchains where chains could post data I mean I guess sort of similar to to how you would use Celestia yeah um I I really like to kind of make a an analogy that like a file is just a chunk of data in the same way that a block is just a chunk of data in the same way that like a proof for a file or a block is just a chunk of data um so what we can do is with Jackal everything is indexed on kind of the top layer of the chain which means that at any given moment if you request data you can tell very quickly whether it's available or not because it is sitting on the state of the chain rather than in historical data like a lot of data availability layers are so

what we can do is you know right now if you're looking at traditional data availability for like rollups you're posting the proof that that block is correct to that data availability Network rather than the entire block itself because posting that entire block gets very expensive on Jackal because that data is not stored directly on chain you can post entire blocks entire lists of transactions big snapshots of multiple blocks together all in these individual files and then you can take just kind of that hash of that file and uh you can store that directly into your rollup to say hey this is the block that we need and so now we're getting into a

state of the DAT is available on our chain because we have proofs running every like six hours but we can also determine that that data is ready to be pulled off those machines which is kind of a even an advantage over traditional data availability layers that we kind of see today is is you still need that data available somewhere even if that data is you know stored and verifiable you need to store it somewhere and that's where that kind of gets tricky and a lot of the the wording around data availability gets really muddy where Jackal you know it's there we can prove it's there and other than that there's really nothing more to a daily availability

layer for a rollup and it's kind of interesting the the Integrations that we're looking to build with uh for example Jack um from Strange Love and the RO chains application he was just on the podcast actually I know he's awesome guy um because we're looking to do a few things with them marrison do you want to because like the the earliest thing obviously is not necessarily using as availability layer for the the chain but also like no snapshots and things like that or or the the ability for us to integrate with them on that front and Mar like kind of like an MVP what do you see like that looking like with Jack for example yeah I mean as you mentioned there's kind of like a few things that we could do um the easiest step forward is just embedding kind of these Jackal

Outpost situations into exist like into new rollup chains so that you know day one they have the ability to store data to Jackal for their apps so that's really exciting um kind of as as a more data available type system having automated snapshots coming out of the chain where you know that block proposer every 200 blocks let's say they just take a snapshot of the last 200 blocks and then they load that onto the Jackal protocol so that other networks can just go and grab that data straight from there um that's really exciting and then obviously yeah like that last step would just be full rollups running on Jackal um and using it as a block store instead of the validator set at itself which is

you know a lot of R&D um making that experience feel good but there's there's a lot you can do with just raw data storage um it gets really really exciting if you if you think about like how even web 2 works you you have compute and then you have storage and everything happens in those two sections and uh if you provide a generic storage layer anybody doing compute whether that be a blockchain whether that be you know something like a c they can do whatever they want to that store layer now you have a full working computer in the cloud which is really really exciting I mean I've thought about this too uh at some point which is you know there's there's tons of compute in the blockchain space I mean execution is King there aren't that many like storage

has kind of been left to to the side and I I think that's why I think that's because the majority of use cases for blockchains rely on like ownership of of assets and those you know don't take a whole lot of data like I mean you don't need a whole lot of storage to track um uh the ownership of those things whereas you know storage is more complex because like with storage you have this well you have the problem of Al the size of the files and just the amount of data that's that needs to be kept and then like the redundancy and then on top of that you

know like the kind of Perpetual nature of block blockchains that that were like files need to be stored for a long time so that's a really hard problem like I think the reason why we haven't seen you know a very many use cases that leverage onchain storage or decentralized storage is it's because it's a hard problem to solve and for the most part I think um a lot of the applications that use lots of storage um can you know don't don't need to be of Perpetual I mean there are applications I think that um could be Perpetual could could benefit from Perpetual storage like um things like ancestry you know all sorts of

scientific data social social networks I mean who cares right like who cares if if you know our tweets disappear um but uh yeah I wonder yeah on the on the use case side what are the most exciting use cases that you guys have seen uh in the sort of Crypton native side of things but also just in the you know regular old data storage use cases yeah it's it's kind of interesting because a lot of people the value proposition of Jackal is different in a lot of people's minds right so let's say that uh for example we have a telecom company in Iowa that uses Jackal to back up all of their um user data um because they have legislation that they have to do that and what do they care about right they don't really care about decentralization or ownership or a few

things that just all like the core ethos of blockchain technology and our our communities They Don't Really align with that they want slas and they want all that stuff um but that being said what they do really like about jao is they get geod distributed data that's verifiable and auto heals if you want something like that with Amazon just Geo distributed data in general for like say a terabyte you're looking at almost $200 if you want multi- region storage per month per terabyte Jackal is doing the same thing at $8 a month per terabyte which is pretty fascinating so like they see it as a competitive Advantage for the verifiability um they also see the geod distributed nature as their value proposition web 3 storage um they care about different things

right until very recently we haven't had the technology to actually provide a good user experience in an active cloud environment um even on blockchain Rails so everything has kind of been more of like backup to file Quin or backup to AR weave and have an archive there um just so that they can hit the buzzword but not necessarily because they actually want to use that application very recently we've had the technology Advance we have IBC we have proof of stake we have speeds we have the boring cryptography work that's happened over the last let's say 10 years we're starting to get to the point where it's actually almost at Cloud parody we're not quite there yet but we're getting pretty close in that sense and if you're building an application over there what do they care about they care about well Integrations can I just call a contract

and have it work um do I have to manage my storage deals do I have to manage my encryption Keys locally do I have to do all these things um they care about the ownership the sovereignty the decentralization and and the ease of integration and it's a blockchain native for their application can I just embed it very simply into my application um so that's what they care about right and then that's kind of like the B2B web 3 front and then there's also like the end users in web 3 like all of our 930 people that use our Dropbox type solution right now um they just use it as like I just like privacy sovereignty security I want to own my data um in the cloud environment where I'm only the end user with my private key has access to this um to my file tree and my files and

and all that stuff so they kind of just see it as like a drawbox which is pretty interesting so it's kind of malleable and it kind of pivots into a lot of different use cases and there's a lot of different value propositions with the protocol in general yeah I think one of the coolest things that is coming out of this kind of decentralized storage um it it's it's combining a lot of Technologies um so if you're looking at something like um homomorphic encryption you know getting fully homomorphic encryption running on top of Jackal um kind of just as like a drop in replacement for the current like AES implementation we have for for certain use cases I know the the Z AI guys they've been doing a lot of really cool

stuff with homomorphic encryption and um kind of pairing that with AI model training and so what we're kind of looking at is these big data sets that you want to kind of train and AI I on what we're kind of looking at now is this you can homomorphic encrypt it send that off to a company that's running an AI training so that you don't have to have all the hardware yourself and they'll never know what that data is and they can return you a full AI model that they never even know they can't even use it because it's fully encrypted to your keys which is really exciting the thing that I it got me thinking immediately when I saw that is that data needs to be somewhere and having that dat stored

securely on a decentralized platform like Jackal you know you start pairing these different Technologies together where you have you know your data encrypted on Jackal and then you have this like secret AI training running in like in a cach compute module you've got a fully private endtoend system where the user can upload their data it's their data they train their model they own that model and nobody in the supply chain ever sees what that data is which makes it really really secure really really private and those types of like combinatorial Technologies working together create this really really nice use case and I think Jackal fits in there really nicely pretty cool pretty cool stuff at the end of the day we're

obviously pretty excited from that s just like it's like this it's it's like self-hosting but the scalability of the cloud it's like full ownership it's like no different than having your your files in in your desk droy but you still have like this beautiful scalability where only the end users access like we can't even access it and we build the thing which is pretty cool yeah I mean I think that's one of the one of the most important properties here is that you know everything is trustless encrypted with a protocol you can use the protocol in a full trustless way without the team having to having access to the files I'd like to talk a little bit about the tokenomics and and economics of Perpetual storage uh when I when I first heard this I think when we

interviewed Sam from R weave on on epicenter uh I I thought it was sort of crazy that you could you could pay once and have Perpetual storage it sort of breaks e you know the economics of most things that we're we're familiar with right where like you usually pay for something as long as you have it or as long as you're using it not like a onetime payment forever uh so yeah if if you could describe how uh the economics of Jacko works with regards to the uh the Perpetual storage solution but also like the drive storage solution where files uh can exist for like a a predetermined period of time or at least not in perpetuity yeah and before I hand it off to Marson there's

um forever is a long time right and it's it's making a lot of assumptions with with with a model like that it it's we we like to kind of have the authenticity of saying exactly what it is um so if you want storage on Jacko with our Perpetual storage option we say 200 years because that's what we can guarantee without any just models of just predicting what the cost of storage will be then and predicting that drives will get more efficient and having all those Associated risks with that um we like to kind of make sure that what we are saying is is is actually going to be delivered and with that I'll kind of

pass it over to Maron on how like 200e storage Works in general yeah so I mean as Patrick said um we we take a very very different approach to R weave um in regards to kind of our Perpetual storage where we price it out exactly as we kind of expect it to be priced so if we're charging $8 a month per terabyte if you want to upload a terabyte for 200 years it is $8 a month for 200 years and you just pay that upfront and then we drip feed that out to the storage providers exactly the same way that you would if you were paying monthly for an account um AR makes a lot of assumptions on the fact that like data will get cheaper over time and that's why they can price things out the way they do initially we

don't really like to make that assumption um there's you know it it is the case right now that data is getting cheaper but if data ever didn't get cheaper um all at once kind of a protocol that makes those assumptions kind of start being unprofitable for all the storage providers at the exact same time um and you know that's that's the thing with like economics of everything you cannot fully predict everything it's it nothing is perfect you know even Jackal if we're pricing everything out at $8 a month per terabyte you know if somehow data hit $20 a month per terabyte for the storage providers well then it's not profitable for them either we kind of just assume that you know it's not going to get cheaper it's not going to get more expensive it's just kind of going to stay exactly as it is

and if it gets cheaper that's great um if it gets a little more expensive the protocol will kind of take that hit um too much one way or the other wouldn't be great but uh that's kind of our philosophy on those 200 years where we just price it exactly the same as any other storage um already you know we're doing that 3x redundancy so that $8 a month is really covering eight like a terabyte times three for all that redundancy so you're already getting a a pretty good amount of storage for for that time but for like those 200 years you know if you're storing the same thing on R weave it is a little bit more expensive on Jackal but it's because of the fact that we're drip feeding it out almost the exact same way that we would if you were paying monthly um and we we

really try not to make any assumptions on speculating the price of data storage I mean I here's here's like something that I'd like to maybe double click on so in terms of data like so data storage costs I I guess you could say are you know like linearly declining but are somewhat of a fixed right like the the cost of data storage you you could probably create a model for predicting the price of data storage that would be mostly accurate however the price of the Jackal token can fluctuate based on all sorts of uh external factors and I yeah can you explain basically how you can

because I mean that you're s basically assuming that the that that the that that the price of the Jackal token The Jackal token U maintains itself in a way that data uh that validators want to continue running the chain I mean data providers need to want to continue storing the files but also the chain has to continue existing um how do you guarantee that like in five years from now or 50 years from now people will still want to validate this chain I think that's sort of like the first thing and then and that really I think creates a big dependency on the storage providers whether or not they want to accept the Jackal token as payment of course um the way that we do the storage payments is we we saw it we

we want users and businesses to make informed business decisions about how much the storage is going to cost so that's kind of why we' moved away from a Marketplace based and we don't really um the the fluctuating price of the jackco token shouldn't affect your storage um in the sense that we built an oracle into the chain to Peg the storage which is governed by governance to $8 a month per terabyte with three times redundancy um that being said it kind of creates this interesting economics around the the token in the sense that it it's kind of like a call option that makes an assumption that more storage will be stored on the protocol um if there's no storage being stored on the protocol then the value of the token goes down therefore um there's no use of running the protocol anyways which is

interesting that front right um so the whether it's a th000 tokens or whether it's 200 tokens for an $8 month payment for storage um you're still going to be paying $8 a month per terabyte for the storage user and the business's perspective regarding security we we see Security in a bunch of different ways and obviously there there's kind of this systemic issue across Cosmos where we have to have really high inflation to pay for security when you're in the Boost strapping phase and there's also um a I feel like a lot of chains stay in that high inflation period for a really really long time much longer than they need to be and we have made the decision in Q2 with our economics upgrade we're moving more towards a real yield slash like kind of like low inflation model

where it's a little bit of a hybrid where you can see the inflation as kind of like a universal basic income for the St providers and validators and then the real yield is is deterministic on how well is the protocol performing from a unique amount of data being stored um on the security guarantee front and all the things that we're looking at right now um meron I know you have a lot of thoughts about kind of um what's required for security in Cosmos and like obviously like we're talking with Babylon and things like that but I'd love to hear your opinion because uh you have a really interesting take when we talk internally about this yeah I mean so as far as like making sure that the storage providers are maintained Patrick said we're kind of going that real yield model where um we're moving away from just accepting

Jackal tokens as payment as well so we'll start taking like usdc so the storage providers will guarantee that their income is their income no matter what and uh they're probably going to end up getting a basket of different tokens whether that be like Adam whether that be Jackal whether that be usdc um so that kind of pulls away the dependence on the Jackal token for storage providers to keep their incentives going while still um kind of bringing value to that Jackal token by having users pay for that pay for the storage as well as a bunch of other things um like our name service and all that in that token on the validator front you know they're still making a little bit of inflation um in the new econ model but the main thing is that

they're also getting a portion of all of the storage deals that are being made so if you're buying month of storage the validator set and all the stakers get a little bit of that too so we're bringing that kind of recurring Revenue back to those the stakers and the uh validators from you know that that's in our mind um at least according to the econ model like enough to keep the storage providers and the validators up and as the chain gets more and more users everybody makes more and more money it's kind of an incentive model for those validators as well as the storage providers and even stakers to go out and try and get more people and businesses onto the network because that

means they make more um so you can kind of think of the whole network as like a software as a service company which is really really cool security wise for that validator set I'm really big on the the Bitcoin security and the time stamping through like Babylon um I think that's a really really great use case for Bitcoin it kind of creates this really big economic hurdle for people trying to attack the Jacko protocol because now they have to attack Bitcoin as well and so you you know pair that with kind of potentially trimming down the validator set making the validator set cheaper to run bringing that real yield revenue and then worrying less about security from a tokenomics perspective because you've got that Bitcoin security you create a really in my mind a very sustainable validator set

paired with the stor providers it creates a really kind of like harmonious ecosystem okay yeah I mean like there's I think there's lots of different Avenues here to um reduce the cost of the valador sets obviously like Babylon but then you also have you know rooll chains or uh ethos which uh which will will be live this year so I think the the options for running a chain more efficiently and more cheaply while maintaining a high degree of decentralization and security are are becoming more and more I mean there's more and more options available and I think you know we'll see by the end of this year probably a few of those go live and be used in in a number of

chains I know there's already protocols in Cosmos that are considering drastically reducing their valid a set and you know chains like noble for example that are already POA chains are going to become more decentralized by using roll rooll chains and and posting data to Celestia so it's all very exciting I think it's all positive for Cosmos chains that you know in all honesty like are very expensive to maintain and there's a lot a lot of inefficiencies there as well right if you look at that the you I think it was sunny who who posted this at some point but he had this this chart that showed the overlap between validators on different chains I mean there's no there's no reason why then chains need to kind of inflate um their tokens in order to pay for security when you know you could

have something some something canest security that uh would would really reduce the cost of security um across the entire ecosystem uh before I wrap up here I wanted to talk a little bit about interruptibility and IBC and sort of how that ties into all we mentioned it a little bit but you know what are the applications that IBC you combined with Jackal um unlock that you find the most exciting o that's a good question like right now we look at pretty much every application needs storage for something right and the question is is that where's like the best return on investment for the development time um is the decision that we have to make right now so um we see a lot of different applications whether it's like AI ml use cases which you're interest

thing with Morpheus and or chin like they want things like checkpoints they want their AI agents to not have dementia so they remember things and things along those lines um so those are really important uh nft use cases are cool but the thing that I'm most excited for associated with having a like a scalable and fast and actually usable storage layer in general um whether that's in the cosmos Eco system or whether that's using IBC to go beyond the cosmos seos system is what can we actually build now that's not already being built on blockchain reals because of this limitation um so I'm looking at things like peer-to-peer file transfer like sending files anywhere in the world peer-to-peer without middlemen you have functionalities where you can do dite

projects where we have advocus that is they're like storing human genomic data in a self- custodial way which is interesting for their science application the possibilities are endless social applications um other just web scale applications that we want to build on blockchain ra and people have tried and have failed um actually having like a good user experience and the speeds that we're used to in having Cloud parody obviously takes time for us to get there we're going to be a little bit slower than a gcloud because we have to encrypt the files if you want encryption so like it bloats in a little bit so it's a little bit slower but most of that is ux issues and kind of communicating with the U are on that front but Maron what are you thinking like what what do you are what are you excited about

overall yeah I mean we kind of we chatted about it a little bit but IBC opens up a lot of possibilities for a kind of just generic applications um whether that be an nft you know you're going live with 10K pfp all of those are you know you're able to upload those straight from whatever chain you're hosting them on that's really cool makes the ux a lot better for that um there's kind of the the other route of like more kind of web 2.5 layers like you know NF marketplaces or data market places where you know they're not actually running everything on chain because they have all these index xers and everything going in the background they can store a lot of like backups and a lot of that

user data that wouldn't necessarily need to go on chain anyways but they're still kind of in that web 3 space you can do all that straight through whatever chain again you're hosting everything on so um you know those kind of Big Data applications are really cool and then I think what's going to be really really exciting is it's it's not something that's currently really done anywhere um we kind of been talking about it with the data availability stuff is like if you can access storage on any platform and that storage layer is accessible on chain through like data agents you know no other platform can actually access that storage data directly on chain with

Jackal because it's so modular you're able to kind of use like data agents to go and fetch data from an actual storage provider and then post little chunks of it to the chain using those like Merkel tree proofs again um so it becomes very secure um very fast you can kind of you know look at chain data from any other chain what that allows us to do is it's kind of extends Beyond what's currently possible and I don't know if there's you know a really good use case for it but imagine you know you're running a rollup and you post all of the block data in individual files to this network imagine running a smart contract on like Juno and being able to access the block data

from a smart contract of an entire rollup and then be able to like run whether that be token transfers or anything based on the data that you're seeing in those blocks as they come in from another Network now you're communicating at a core level that's like never been done before in web 3 and so things like that really really excite me where IBC opens up all these possibilities that we normally wouldn't have before which is really really and that's crazy like I I also like like the most complex like obviously like data agent smart contract read and write to offchain data storage is pretty huge in our opinion and we're excited to like kind of aiming like Q4 for this functionality which um just kind of makes like decentralized apis in a weird

way um but the other cool thing is is sometimes when we look at Jackal some of the use cases that are killer are also like the simplest in a weird way as well um the one that we've been talking to a lot with organizations is with ledgers for example right um own my files own my tokens self- Sovereign but the functionality of just something as simple as like ransomware backups make sure that if your organization gets hit with rans schare you have a backup that's air gapped on on a multi signature wallet that controls the root folder a board of directors come into consensus to unlock their files that means that a hacker if they want to completely an organization and they need to get all the emission critical data if you're using like Jack as a backup solution to

protect yourself from rware they' have to physically like 60% of the board um to get their private keys for example right so functionalities like that where you have like this new level of like the highest digital privacy and cyber security posture possible in a cloud environment um so you still have that scalability and having a service like that where it's so secure we can't see it it's uh that's the beauty of it right it's such like a low level of the stack really valuable um I am sure that there's those kind of things that people are going to build that we don't know about either man I just I I just want some I just want something simple I want like a Blog and like decentralized notion that's it well kind of funny you

mention we we we do have an alpha of of Beacon which is our kind of blogging platform check that out yeah yeah it's uh as we say it's still early days but uh it really does you know there's no centralized third parties it's just you upload everything straight to jacel and it's a it's a fully kind of MV P of a of a Blog platform which is you know somebody out there could probably spend more time on something like that and make it a better ux but uh yeah looking forward to playing around with uh with these uh early uh products and tools being built on Jackal um work can people find out more about Jackal and you know potentially you know if someone's interested in stting building

on Jackal where can they go of course um so the the best way is to kind of come and follow us on Twitter come to the Discord uh we have telegram as well so that's kind of where our most of our community we're a little bit more active in Discord than we are on telegram so that's probably the best place for it um outside of that if you want to come build directly on ja that's awesome but what I'm excited for is US building our outpost on other blockchains so if you're a builder on Archway pretty soon you're going to have all the function functionality of Jack there so um we're excited to also providing Jack to other development communities and giving them a toolbox of all kinds of stuff that they haven't had previously which is super exciting guys thanks so much for coming

onks apprciate thank you sir