Dieter Fishbein & Joe Petrowski: Polkadot – Claiming Digital Sovereignty on Blockchain #347

Transcript

I and welcome to AAP's enter my name is Frank rain my name is maja Roy today we did I think a third episode on polka dot you know we've done some over the years but of course like as with many blockchain projects it takes a long time for those to come about and now finally polka dot his life so we had dieter Fishbein on from the web three foundation and Joe Petrovsky from parody to dive into various things about polka dot especially about economics and the pair

of chain leasing and you know some of these interesting interesting features that pair that polka dot has now you don't hear so disclaimer that you know with course one we are currently running polka dot validator so just be aware of that and of course if you do have dots on a stake then you can do some of us so if you want to learn more about that just go to course 1 so that's ch or us dot one and then there's gonna be information there and anything actually where we have been working kind of on a polka dot related thing with course one which is that we've been working on this bridge between a substrate chain and the

cosmos SDK chain so we talked a little bit about both substrate and cosmos SDK in the project those are kind of the leading frameworks at this point for building your own blockchain and we actually have a grant from the web 3 foundation to build that bridge and so we've kind of just are in the process of finishing that first part and ok with that let's go to the interview so we're with Joe Petrovsky and Tito Fishbein to talk about polka dot today we've done a few episodes about polka dot already I think that was one with Gavin there was one of Peter Saban but now of course polka dot we spinning and making for a long time is actually life or you know sort of in the first stages of liveness

which is become kind of the norm for blockchain projects right where launches become this drawn-out multistage thing but it is life right and kind of with a proper proof of stake network now so I'm excited we can revisit polka dot again and thanks so much for joining us today yeah thanks for having us yeah thanks so much maybe you guys can each give a little bit of introduction about yourselves like kind of you journey in the larger crypt of space and then like how you've ended up working on polka dot so I mean I guess like I've been interested in the crypto space seems like 2011 or 12 and then got more into it around like 2015 2016 I was a professional cyclist and just kind of

like looking for my next thing to do after I retired just a funny word to say when you're like 29 and I'd been under Krypto for a few years already just out of interest and then I started my background was in like time series analysis and stuff so I got into like trading algorithms and all this sort of thing and worked on that for a couple years and then ultimately decided to join Paradis and start working on polka dot and I've been here like about a year and a half now I actually discovered crypto around the same time as Joe so I was graduate student studying cryptography and it was about 2013 and I remember like seeing the Bitcoin white paper at the time and

I actually think this journey sounds probably the same as so many other people but I saw the Bitcoin white paper and read it and was like okay that's super interesting put it to one side continued on with I was doing finished my graduate degree and ended up working in finance for about four years doing stuff completely unrelated but still following the space in about 2017 when the space was starting to get pretty pretty large I decided that at that point I actually wanted to work in it so I ended up joining a small blockchain focused venture capital fund in Toronto and Hong Kong and really just learning all about layer 1 protocols and about the ecosystem in a much more deep way

than I had been which was really great it was super fun and about a year and a bit ago decided that I actually wanted to work in the space doing something a bit more meaningful than just investing so joined polka-dot and yeah have been here ever since let's just start from a from a big picture probably most people have you know some familiarity with polka dots but can you guys give a little bit of a background like what is the polka dot vision from a very high-level polka dot fits in with like the rest of the web provision of like creating fundamentally new ways for people to organize themselves outside of like what we would call traditional organizations or societies and a way to organize without trust and

so I think I'm more concretely how polka dot expresses or that contributes to that vision is that it is a suite of protocols that allows like specialized box chains to communicate and interact and it provides the context for this interaction so for developers like what this means is like they have a platform that gives them like secure interoperability and allows them to write a block chaining that specializes and it kind of lies somewhere on the spectrum between like a smart contract and building your own blockchain from scratch so a lot of people find that like smart contracts are either to Bilotti in their execution do so like gas fees or like unpredictability or other smart contracts on a shared platform or they just they're not able

to upgrade or express the complex logic that a developer wants to be able to express and building a whole box chain from scratch is like a huge undertaking that can take a year or several years to do and a lot of its kind of like repeated work of like the networking and RPC in database type of work polka dot really is trying to create like a new class of developer where you can target this like intermediate zone of having a specialized blockchain with a lot of customization but also without the full work of building out like the underlying infrastructure for that blockchain so yeah I like to kind of wrap it all up like it gives you this platform to deploy this specialized blockchain without worrying about like the security

or like the underlying mechanics of it that's exactly right I think what Joe just describes is really the vision of poccadot for developers for users polka-dot can actually give us new forms of organization so I mean fundamentally polka-dot exists to give us the ability to coordinate in much more sophisticated ways than we've ever been able to and ultimately this will help us solve like much more complex problems than we ever thought possible these are large problems requiring the pooling of resources around multiple stakeholders and I think we'll see kind of the launching of groups that wield similar power to nation-states and own

properties can enter into business agreements derive income from sources and you know I think this is more of the end-user vision than the actual developer vision let's dive into what poker actually looks like so what are the key components and technologies that make up poker on so we can kind of break it down like technically there's this real a chain and then pair of chains that connect to it and so the real a chain it really provides like finality guarantees for the pair of chains and the context for them to interact with through a protocol that we call cross chain message passing or X EMP and then kind of like underneath the relay chain we have a

consensus mechanism it's like a hybrid consensus called babe and Grandpa and it's secured by a proof of stake algorithm called nominated proof of stake where people can select like this is a set of values that I deem acceptable or like that I want to nominate to be a validator so then like if we go out until the pair chain space you like pair of chains and pair of threads which we could go into a little bit more but the way that these interact with the relay chain is really interesting so polka dot is a sharded platform so you can think of these pair of threads or pair of chains as individual shards in the network and where it gets really interesting is that from the outside in it looks heterogeneous Lee sharded so each of

these pair of chains is its own individual block chain it's going to have its own API it could be like an application it could be like a decentralized exchange or a stable coin or a smart contracting pair of chain and these all look different so if you're from the outside of polka dot you know it kind of depends like which parity nu goes you it's going to provide completely different functionality but where this gets interesting is that the way that the relay chain interacts with them is through web assembly and web assemblies like really the glue that holds this whole thing together and through our libraries like what we've provided is that all of these power chains compiled down to web assembly which gives them an execution interface for validators on the relay chain so these pair of jeans kind of all compiled down to like pair chain validation

function to guarantee a proper state transition so from the perspective of the relay chain this is actually like a homogeneous lis sharded system all of these pair of chains look the same it just they all expose an API that has a validation function that says like validate my pair of chain and that's it this interaction it provides like a lot of flexibility and then if we kind of go like beyond the technical stuff we can talk like the ecosystem also like Kusama and our testing at West End and how those play into poke it out and like I don't know dieter can probably talk more about Kusama if you want to talk about like how that fits into polka dot as a component Kusama has launched about a year ago and originally when it was launched this was

kind of early version of polka-dots and I think we launched in about August of 2019 the idea here was that it allowed us to go through this process of launching the network before we got to polka dot and allowed us to you know actually see how these things were deployed in the wild but you know the use case has actually really really developed since then and when you actually look at what polka dot is compared to kasama is Kusama is this less robust less secure version of polka dot and this ultimately means that we can see development take place in Kusama in a couple of ways you have this canary network use case where we've used it to

actually test out kind of new pieces of technology before deploying them to polka dot and we actually see this canary network use case evolved into like a canary network use case for others and so we're seeing a lot of pair of chains that intend to deploy on polka dot actually want to deploy pair of chains on Kusama as well and this is so that they can test upgrades that they're planning on doing and test things they're planning on doing in an environment with real economic consequences before eventually deploying to polka dot like a good thing to add to that just like talking about like upgrades and probably what I should have mentioned when I was talking about like the architecture and components is this

concept of like being able to upgrade your chain and this is also part of how we architected polka-dot where a client doesn't like Express the polka dot logic itself it it's actually kind of split into two parts it's a host environment and for this web assembly executable it's the same like compilation target for the pair of chains as for the relay chain so it's actually hosting its own logic and so this allows the relay chain to upgrade because we can just place a new webassembly executable on chain and without anybody upgrading their clients the system can upgrade itself and so polka dot like it hosts itself which i think is first and block chains like being able to have for clothes upgrades and upgrade the chain without having

people update their clients and this provides like the same functionality for pair of chains because they're just exposing this like parity and validation function for the relay chain validators and well they can change that so if they have if they want to upgrade their pair of chain and come up with a new one and that's also hosted in the pair of chain storage and they can say hey I want to upgrade my pair of chain here's a new one and so it's somewhat of a risky thing to like because we've seen mistakes on Kusama of like upgrading the relay chain and some problems there so that's obviously like something you want to try out on a test net or on Kusama or something before moving to polka dot because these are new things that we're trying to do and trying to express and so having these testing grounds that are

more than a a testing ground like Kusama is not a test net but a lower barrier to entry lower risk don't working on polka dot is really important you can think of it as the minor leagues basically where early stage teams can fine-tune their product and business model before potentially moving that to polkadot if they're able to to build their community of users and actually construct a business model that makes sense and it's getting traction I mean one thing to note too is polka-dots always going to be a hundred percent focused on reliability and security however there are going to be teams out there that don't need Bank like security and robustness that are willing to trade you know this higher throughput for

lower security and robustness that they can get on through sama so I mean some use case verticals that could fit in here are like gaming social networking content distribution things where there aren't really kind of high amounts of necessary value transferred and so they could actually deploy to Kusama and this could be a platform for their production deployment and then you know things that kind of need this higher security or maybe for enterprise use cases or things like that could eventually choose to deploy a polka dot as well maybe like one way to think of this spectrum is if you think of AWS I can go to AWS and I can give an image for a machine that I want AWS to boot up and

AWS will boot me a machine corresponding to the image I I give it so this image is essentially like this this binary an AWS will boot the machine and it will run it for me and I can and of course when I give it the image that image can include my application in it and in some senses polka dot and kasama are similar I can come to polka dot and Kusama and I can give them a validation function that compiles down to web assembly and I can somehow give it to the network and the network will boot up and run blockchain for me yeah that's actually an excellent analogy except that instead of trusting AWS to secure data center you have this open permissionless system of relay

chain validators the difference between Kusama and polka-dot is that in some sense when I ask polka-dot to run my validation function when I run my block chain and I ask to someone on my blockchain I am more sure that the particular blockchain I launched in polka-dot can't be attacked as easily as kasama so maybe there's a failure rate associated with this these block chains and Kusama failure rate is let's say like once in 10 years polka rot failure it is once in hundred years so that's the qualitative difference for me but in order to have the block chain with a lower failure rate in some sense I will also need to I or my users in the end will also need to pay more through some

mechanism which we will discuss exactly and I think one way to think about this is really like from the perspective about how a deployment process should look and I think teams will originally start kind of deploying their blockchain to West End which is Kusama x' test net and there they can kind of flesh out their tech and make sure everything works and then after that successful deploy to Kusama and then I mean I explained that process before like maybe at some point after that once they build the community of users fine-tune their business model then deployed a polka dot if it makes sense on some level that we can trace polka dot back to the challenges of aetherium and challenges of a theorem and scaling

and of course that has led to you know so much different stuff right whether that's now eath 2.0 whether that's maybe something like a cosmos vision we're all there will be many blockchain sentain operate whether that's visions like the polka dot one where you have this one tightly interwoven system of many block chains that then can be separate you know then you have like things like near solana like so many other things when you guys kind of look at this landscape and how this has evolved what do you see as the most important dimensions in which to differentiate these different approaches and like where do you sort of situate polka dot in this landscape a lot of early motivations for these

kind of next-generation protocols came out of scalability several years ago people knew that like aetherium wasn't going to be scalable for building production applications if it was going to gain mass adoption and so we talked to start talking about like different scalability approaches and we saw a few so like you look at like what year is doing and there's a well we're gonna still do smart contracts but we're gonna do sharding on a single state aetherium to said well we're also gonna still do smart contracts but we're gonna use eve azam and we're gonna have these sharded chains that are all the same they all just have like an e wasum host environment and then we saw cosmos say like what we wanna do like interoperability so we're gonna have like specialize box chains and we're gonna let them communicate with each

other I think there's another dimension in here which is security and so I wouldn't really say like one of these is more important than another one I would say like we try to look at all three security interoperability and scalability as like three dimensions of the same like fundamentally the same problem so the interoperability kind of naturally gives you scalability because you can if you can send messages between different systems then all of a sudden these systems can specialize and you just get like horizontal scalability right that's kind of obvious but what are those messages mean and without security you're kind of relying well if I receive a message how do I know that that chain hasn't been forked or

attacked since it sent that message and so all of a sudden you're kind of stuck with well how do I interpret these messages that I'm receiving like I have this scalability and interoperability like that's great but like what do I do with these right the angle we're trying to take is like providing the context in the form of security to give these messages meaning so that if you receive a message you have some guarantees about the provenance of that message like who sent it how it was generated if the message says like hey I burned 100 tokens so therefore like you can mince 100 tokens on your chain I want to actually know that like you burned 200 tokens on your chain right and that you didn't just send me this message and likewise like for the sender you know I

want to know how this message is going to be interpreted is this person going to 200 and then come back to me later and say like yeah I actually 200 cool no you don't I would say like the other dimension that we go to is just like pushing scalability or pushing this work to the edges of the system right like a lot of shards systems have some sort of like central gatekeeper that's kind of like three layer between all these different chains that are in the system and with polka dot like one of our core principles is like we really want the relay chains do you like as little as possible the relay change to really provide the context for these messages and fourth sister operability but it shouldn't actually send them

itself so we let the chains actually just like talk to each other like chain to chain doesn't even need to be through valador like if they just share a normal full node they can actually exchange messages but those messages are still guaranteed by the security of the relay chain to kind of like go back to like your original question like I don't know if I'd particularly like pick out a single dimension of this that's the most important I think you need like more of a holistic solution about how you actually scale web 3 and make these different systems interact so the way I think of it is like you mentioned there's an interplay between scalability and messaging and the idea that system like cosmos and system like

polka are ultimately is playing with is you can have many block chains and they can message with each other the difference is that in a cosmos like system if you imagine a blockchain and it has a post box and other blockchain scans of deposit messages letters into this post box and then this blockchain will process each of these letters in its own logic in cosmos when the letters let's say stack up in the post box is a stack of ten letters from ten different block chains each of these letters has different risk right so for example the first letter it might be that hey there's a once in ten year risk that this letter is actually a forgery of

some kind meaning that it's not really valid but it looks like a valid letter the second letter is that also runs that risk but maybe it's a once in 100 years risk there's a third letter that turns a once in two years risk and fourth letter that runs a once in 50 years risk and so each of these letters essentially has a heterogeneous risk and this blockchain in some way we need to deal with the heterogeneity of these risks and the risk comes from you receive a letter from particular blockchain you're trusting that blockchain right so if that blockchain ends up being corrupted then that letter could end up being I don't know invalid or due to some hack

or what's happening with polka dot is when a blockchain runs on polka dot and it has a postbox it's also receiving letters now these letters are of two kinds like one kind is it's receiving letters from inside the polka rot ecosystem and there's a there might be other messages it's receiving from outside the polka ecosystem but let's forget the once it receives from outside the polka ecosystem and zoom into the once it receives from inside the polka ecosystem so these are other block chains yeah and I think outside the Polka Dot ecosystem you could kind of consider like a chains API right like how does it let the user's untrusted people interact with it so when the messages for come from outside it will have the same problem cosmos chains have

which is it will need to deal with the fact that all these letters have intrinsically different risks but if you look at like the sheaf of letters coming from inside the pull cord ecosystem what polka rot has essentially done is even though these letters are from different block chains all of them have exactly the same risk and it's trying to make that risk be below it is trying to make that risk B once in 50 years or once in 100 years it's not just that they have equal risk it's that they share common execution and state logic with yourself and so if you compare this like smart contracts interacting on a theory I'm like like smart contracts on aetherium can

interoperate with each other and the reason they can do that is because they're sharing common execution logic the fact that they're on the same chain and that they're charted that's what allows them to interact in a synchronous way instead of asynchronous but what allows them to interoperate in trustless way is that they share common state and execution logic so contract a knows that if it receives a message from contract B there's no possible condition where contracts B gets like 51 percent attacked and gets rolled back to a state before it sent that message and contract a doesn't they either both get rolled back and attacked or they both don't so you know that like if you receive a message and there are some attack you would also go back to a point before you

receive the message and we'd get to reap regress through and maybe that message gets sent again or maybe it doesn't polka dot provides that similar guarantee where if some other pair of chain in the system sends you a message there's no way that pair of chain can be reverted to a state before it sent that message without you also being reverted to a state before you receive that message and so that's one of the things that I mean by contexts that the relay chain provides is that you can't attack a single pair of chain without attacking the entire system if you look at compare cosmos and polka dot like this of course you know many pros and cons on each side right because they're just different approaches but a

big pro in if you run a pair of chain right is you don't have to worry about you know the validators you have to find the validators and you have this kind of like shared you know the shared security the colonies i think that you have more complexity right the system is just very and then especially these in that outside event that you maybe it's never gonna happen but where you know there is this corruption and having to roll back all of the the chains in in polka dot that's gonna be a challenge actually hasn't happened in test nets I mean we've rolled back Kusama before by a few blocks but was that just a single chain with I was having to roll back also we like pair of chains that

are connected with it exactly like we haven't done a system of like rolling back a hard work with multiple pair of chains now that hasn't been done yet hopefully it doesn't have to be done well I think that kind of ties actually into something that would love to dive in here because right now you know to be they chain is live which is still restricted right so there's no token transfers yet there is staking so can you talk a little bit about what are the stages of this polka dot launch and you know what do the next two or three years look like or what has to happen for kind of the full polka dot to be in operation yeah so like we launched about the end of May and like a proof of authority

phase and the purpose of this phase is really to like allow people to claim their dot tokens and set up their staking infrastructure so I know that's like you kind of mentioned at the beginning of the show this is like a thing with with Bakhtin's now of like having a bigger rollout process and if you think about like when Bitcoin and aetherium launched there wasn't really like a threat of like launching and having like a really low security and then getting attacked but if you look at like modern ball chains when there's like a lot of attention on them there's much higher risk so you could imagine like a scenario of say you want to block chain as a hundred million tokens in it and on day one there's only a hundred thousand of them have been claimed some

person could come along and say like claim a million of them and easily like this individual could attack the chain in that moment so they could prove Authority fate is really just to allow like enough stake to be claimed and put behind validators that we can like securely transition to a permissionless proof stake which we did after I think about two weeks so I switched to nominated a proof of stake and then relatively quickly scaled up the valley to account to about two hundred holiday ders so that's where we are now is we've been running for two or three weeks now with about two hundred validators on polka-dot with still quite restricted functionality like limited to like the staking and proxying accounts and like kind of getting your infrastructure set

up our next phase which will hopefully be in the next two or three weeks we'll be rolling out governance and removing the pseudo authority you know I poke it out as this module it's called pseudo that it takes the place of governance before governance is like fully in place so we should very quickly see the removal of that and the implementation of like our full full governance protocol suite and then after that it would be enabling balance transfers and hopefully not too long after that so I think I think we'll probably see that around like late August early September and then hopefully not too much longer enabling pair of chains and then cross chain message passing by the end of the year I would hope so I think like that's kind of what the next six months look

like when we talk about like getting to the core relay chain functionality and like opening up all these features when we talk about like getting 2 full polka-dot vision I was saying the next 2 or 3 years will still be a lot more infrastructure so I wouldn't expect like a super exciting application to be out there like in a year or something I think it's like polkadot is like fundamentally like infrastructure for infrastructure the first wave of pair of chains are actually going to be like not so much an application in themselves so there might be stable coin chain decentralized exchange chain an identity chain and then I know you guys want to talk about later but like some more system level pair of chains moving some work off of the real a chain and

putting that into para chains and that really provides like infrastructure and composability for other applications so the next wave of applications becomes and after that well they have an identity suite they have a stable coin they have a decentralized exchange they have a smart contracting chain all of this stuff and so that's what I'd expect in like the say like four to ten year frame as these applications can come in and they have all of these primitives that exist within the polka dot network that they can compose and build higher level logic and I mean I would say like when we talk about like the full vision of polka dot I don't think we can anybody can predict it like if polka is successful it won't be something that I could tell you today my imagination is kind of like restricted to like things

that I know and like maybe like a little bit more I would hope you know I don't want polka dot so like just recreate stuff that exists today and like redeploy like I want to see new stuff new things that don't exist yet and I think that just takes like people coming from different backgrounds and seeing things like in a new way it's not necessarily possible to see like from your one angle so that's what I would hope to see like in the longer term is more of these applications coming in that I'm just like oh that that doesn't even exist in an on blockchain world but it's being deployed on polka dot and I think one really powerful thing which Joe alluded to is you're gonna see like a very different notion of application like now when we think of applications

we think of them living on on one layer one and you know that provides some limitations but with polka dot with the really intimate notion of interoperability and with kind of this environment of all these you know applications specific and also more general pair of chains you're going to basically see this notion and application that lives across pair of chains and that's going to make things much more scalable and much more powerful in a lot of ways yeah I can't wait to see like an application that's not a blockchain itself but it uses like all these blockchain primitives so like instead of PayPal it uses like a stable client arm and you know instead of some like email login it uses some like web 3 log and type of thing and the application itself isn't a blockchain is

just using blockchain is like as its back-end yeah that's I mean first of all the first thing comes in I think that sounds very you know in the timeframe that is probably realistic but it's interesting kind of like pointing out just a long time scale be talking about and of course this is not just polka dot right this is like same with f2 and like all of these other networks but we're just seeing this is taking such a long time you know to be in the place where actually you know block chains are able to support you know end-user applications with you know millions of users that yeah it's it's a long it's a long road that is ahead of polka dot and and the entire block chain space I mean one one really nice thing

for polka dot is is even though we're just getting around to launching polka dot now substrate has been around for I don't know how long like like well over a year right right Jo a year and a half two years and so I mean something we haven't really touched on but I mean to build one of these pair of chains like substrate is this very versatile framework for building kind of either your own independent blockchain or your own independent blockchain that can natively connect to polka dot as a pair of chain so we've actually been very lucky to see a ton of development in you know all of teams building on substrate and I mean we actually we actively track

the number of teams building on substrate there's something like 70 teams doing this and it's it's really really special to see you know this ecosystem already forming and so we think I mean when polka dot is launched I mean this gives us this kind of massive bench yeah I mean correspond right I mean running around it is on polka dot but also we've been doing that for centrifuge now for for quite a few months which of course is built on on substrate I'm kind of curious like because you're running Valladares on on centrifuge like I mean like I'm mostly in this like polka dot and Kusama ecosystem like how different is it to run a validator on a like an arbitrary substrate based blockchain versus on

polka dot the validator experience is very similar today the same tool said yeah yeah yeah things everything's the same yeah I mean that's awesome I'm like that really helps like scalability and like hopefully that eventually transitions over to like collator node yeah I mean we absolutely I mean I think we are seeing that also from you know there's now we are probably running like six blockchain spilled on cosmos SDK and the the ease of onboarding these networks it's just 20 times easier or something then like some totally independent blockchain and I mean the same is gonna be true with substrate right and also as you pointed out many many chains also deciding to

build on substrate and I think that's going to be a great advantage right because you're gonna have to laying around that framework and people are gonna be familiar with it and that's gonna so I think actually when it comes to this sort of blockchain frameworks I mean at least the way it looks like to me at the moment you know I think we're gonna have like substrate in cosmos SDK both having a big market share but I think it's gonna be difficult actually if somebody else to come in without doing something like dramatically better yeah and even even from my vantage point while it's like a very technical point the substrates designed that like pieces of substrate compiled down to this

poison and then they can be upgraded independent of each other is such a massive strength of of substrate it's a strength that's really hard to appreciate until the point you run a blockchain and you need to upgrade something yeah I mean it's like we saw this kind of recently and in Kusama it gives us a lot of flexibility to push like optimizations in the underlying client to like we didn't upgrade I think like version like 2012 on Kusama and the client actually like for validators was throwing some error about like having difficulty finding the deciding key for the session but those validators were just able to downgrade their client to

like the previous version but without affecting the actual blockchains upgrade so like we just had all the vouchers running like like the client version of like one earlier than latest but the actual runtime the blockchain was on the latest version and that didn't affect anything so it's like pretty cool that it gives us the flexibility to like fix bugs both on the runtime side and the client side so talking a little bit about you know kind of what's in development I'm in at the moment so there are a system pair chains that you guys mentioned right that will let's say stable coin or identity that will then be available to other pair of chains so what are some system pair chains that are like you know currently

being developed and yeah maybe start there I wouldn't call those system level pair of chains I guess like the vocabularies not completely flushed out here you know I mean those are still like infrastructure broke maybe coming from like some other other party and like somebody who's actually like bidding for those pair of chains lot because like they want to launch their identity solution versus somebody else's eye view system level pair of chains as offloading work from the relay chain so you know like the relay chains primary responsibility is valid in the state transitions of pair of chains but it also has all these other things like it has dots oaken's and so a stood to you with like balance transfers and staking and governance and all all of this stuff

and that could all be offloaded into pair of chains so you could have like a balances pair of chain a governance pair of chain a staking pair of chain and all of the transactions that are related to that type of activity take place on those pair of chains which just leaves room for more pair of chains in the system to exist because you're pulling all of that stuff out of the relay chain really the only stuff that should go into the relay chain is the like the updates to the pair of chain head so we kind of take like I don't think it's like fully decided but like the like a hash of the header of the latest pair chain block or the State Route or something and that gets updated in the relay chain block and then some like other critical stuff so like if somebody

observes some like misbehavior from a validator like flashable behavior that you want to report you don't really want that like going on on a pair of chain you want that to go like right to the middle so that everybody knows about it right away but the real a change is really just be doing this like super like low-level core stuff and so like I've used system level pair of chains there's like anything that just takes functionality away from the real a chain and puts it more on the edges of the system the way I think about this breakdown is you have pair of chains you have the collection of pair of chains and then you have a category of kind of governance allocated pair of chains so I mean these are pair of chains that are some sort of public good where I mean these slots weren't

bid on by kind of individual projects I want to build their business and then you have system level pair of chains I mean these will definitely almost surely be assigned by governance but then you probably actually have this kind of other category of governance assigned pair of chains that aren't actually taking functionality off the real a chain but are still some sort of public good that shouldn't be launched as a business necessarily or at least I mean this is very subjective but at least at this time we don't we don't think that that should be true yeah I would see like system level pair of chains or like anything that's awarded by governance I was like kind of like a good for the whole polka ecosystem probably is like a token less pair of chain the fact though that even possible kind of just like

speaks to the security model of poccadot as a whole that like you don't really need to depend on having your own token for like block rewards or like having value at stake for your validators in order to be secure because you're kind of like outsourcing your security to the real a chain and so you can really have these pair of chains that don't have any token or like economic model with them at all they just provide some sort of service and and that's primarily what I would see is being allocated by governance a good a good example would be bridges probably I mean they cannot have their own token but I think I mean our view is that they shouldn't and these should be launched on kind of governance allocated pair of chain slots

and any transaction fees should be paid and kind of some other tokens basically and probably go to like some pool of capital that serves to keep these pieces of infrastructure updated and and maintained yeah I mean if you think about like a polka dot aetherium bridge or something if you introduce like a third token for the bridge the bridge is kind of just like acting as a middleman and like I'm gonna like take my cut touken or whatever so yeah I think it would be ideal that we can just say like hey this is a useful service to the whole poke out Network to have a bridge to aetherium it doesn't need its own token so we're just going to like provide this functionality and it's not only as benign as this token just being

like a middleman it actually brings in real security concerns depending on how you actually use the token so I mean in some designs if you know certain kind of participants in the bridge have to actually use these tokens as collateral to stake then you're now introducing this this concern that what if kind of there's you know a market dump of that token and it has no value and then these kind of stakes participants no longer are subject to say slashing and then there's potentially like security concerns that that introduces whereas if you kind of keep the types of collateral to kind of more liquid more you know assets that have other use cases and

other purposes and I think you mitigate that risk yeah that makes sense okay that brings sort of a one of the one of the really cool aspects of poker the things that I'm I find it just sort of like is elegant and interesting which is this idea that okay you want to create a pair of chain and then you basically can lease a pair a chain slot right which which will mean with the exception of the things you mentioned now right if there's some sort of like you know public good pair of chain but of course the other thing can be that you know I have some idea for the centralized application you know I could go and build it as a smart contract on aetherium somewhere else I could like

create my completely own blockchain that you know I have to get my own validator set etc or there is this you know hybrid approach where I can kind of get my own pair of chain so it's kind of your own blockchain and you can decide a lot of things about what that looks like but at the same time you know you delivering all of the polka dot thing and you basically putting up this Leasing you know the starts to lease that slot and of course it also means I can still have my own token on that chain but you know today we move most of stake if you created a mostly for sake networks you're launching a network and then that token is used you know to pay validators to put up collateral etc

whereas in the polka dot example you know that's decoupled right you have dots that are used to her the collateral but afterwards you can kind of do whatever you want with the token so I think that's really interesting it's it's a novel thing right that we don't have today so I'm curious what do you think are some of the your business mall or interesting use cases that are gonna kind of arise from this new paradigm yes so I mean you bring up a good point again the the spectrum of things that you can build on polka dot basically I mean there's two dimensions to this I think the things that you can build on polka dot can look anywhere like applications built on etherium to

platforms like etherium basically and then you know I think you can kind of split business models in in roughly two ways you know one you you can still have these token models and I think I think actually when you know we did a survey of several dozen kind of core teams that are planning to build pair of chains and we asked him a ton of questions and one question was do you plan to have a native token and I think almost all teams said yes basically which which was quite interesting and and I think you know generally like the use of a native token is one of the uses will be to roar'd co-leaders basically and co-leaders on pair of chains are they they basically take pair of chain

transactions package them into blocks that are then you know reviewed by validators on the relay chain essentially so there's a much different security model for co-leaders that then validators you just need one honest collator to make sure that actually the blocks produced by pair of chains go to the relay chain successfully so they can compromise availability but they can't compromise safety essentially all this to say that incentivization of co-leaders matters quite a bit less than incentivization that validators and there's much less that can go wrong so when you actually think about having an inflationary model that rewards co-leaders this doesn't need to be of

the same magnitude of having an efficient airy model that rewards validators so you can get away with a lot less inflation I mean this is a it took me a lot to get back to this point but but basically I think you will see token models for platforms that are very similar to kind of token models and proof estate platforms but have way less inflation you know making it kind of cheaper for participants in the long run I think that's definitely one thing yeah I mean it's it's great you bring that up because you can of course look at at this problem or this situation that you have to put up dots to Lisa pair change slot and you know in some level it there's also something problematic about

it right because you have let's say a new team coming up and you know maybe they race maybe there's just a startup right and they don't have any capital or let's say they raise capital from from some VC and investors I think though they will be very very hesitant or wary of the team then going to like purchase thoughts to lock them up for a long time but I think the natural thing to happen here is that you know you'll have on the one hand people with thoughts and then on the other hand you have people who like have a business idea for a pair of chain and then I think so from my perspective like your obvious thing to do would be to create some sort of like a market place right where I can say oh

I'm creating my own pair of chain it's gonna have its own native token I need somebody to put up the collateral for that pair of chain but I'm gonna pay some inflation you know on that token to basically compensate people for putting up the collateral I think that's sort of the natural thing in my eyes to like happen there and it could be very cool to have to sort of you know crowdfunding pair chain leasing marketplace yeah and I mean this is something that we've arrived at as well and in fact as there's a substrate palette called the crowdfunding module and this is actually deployed on the relay chain and it basically allows for exactly what

you're saying here that the crowdfunding of pair of chain slots so how this would work in practice is a project that is seeking to raise enough dots to acquire a perishing slot would essentially use this module to accept contributions from dot holders into an account on the relay chain and you know if the account obtains enough dots to win on the next pair of chain action then the dots you know would be used kind of as the bond for the duration of that auction and then they'd be returned to dot holders the dot holders who contributed them and then you know of course if the auction fails you know these dots would be immediately returned to contributors and because this account is on the relay

chain and it's not like contributors are sending these directly to the pair of chains this in and in a way this is very secure like dot holders you know as long as they trust and the security of polka dot know they will receive these back and there's no risk say of like ICO risk where you can kind of send your eath or Bitcoin to to a project and you know they fizzle out and you lose everything in a sense it's a secured debt investment into a pair of chain and so you know this brings in to the question which you're saying Brian like how are these dot holders being compensated for that so you know this is left open and this is not you know role built-in

rewards is not something that the crowd module the crowdfunding module handles but we fully expect projects to you know incentivize contributors here by setting up you know some sort of offer and I think the most obvious offer and I think this is what we're gonna see early on and potentially for a long time as as you said you get like some sort of share of that pair of changed native tokens and and the way you set this share is you probably have to make it more attractive than stinking in a sense taking asks as the risk free rate here it's actually much more complicated in that because there's there's different risks associated with sending your dots to the crowdfunding module actually

probably lower risks than staking so it's a bit different so you can see dot holders are contributors rewarded in a pair of chains native token but you can also see other stuff right like you don't have to have a native token to use the crowdfunding module so you could see projects that either don't have a native token or really really believe in what they're doing and want to retain as many tokens as possible I guess in the short term probably use say stable coins as rewards or use dots as rewards and this actually gives a much different risk profile to what basically to the contributors contribution you can also see choice here you could give people

different options and yeah I think this will be super super interesting so we'll definitely see that and actually one one interesting thing to add when we asked projects if they plan to use the crowdfunding module to raise dots I think all but maybe one or two of the projects we served were yes so I think this will definitely be probably one of the most predominant sources to actually raise this bond there's a very interesting differentiation of where inflationary tokens go in for example cosmos as decay chains which are independent and in in the para chain so in a in a traditional cosmos as decay

chain let's start with 100 million tokens and generally these chains need to have high inflation so they might be seven percent or ten percent inflation so let's say 10 million tokens your printed and it might be the case that nine million of these tokens already go to the existing token holders and a million of these tokens go to the evaluators to run the infrastructure so in some sense in these networks the validators are the beneficiaries of new token supply now in in the in the polka dot environment the validators are less clearly the beneficiaries of of new tokens

because when you when you create your own token and launch a pair a chain you don't need to give extra supply to the validators it's the para chain that figures out how to remit the validators so you don't need to give the new supply to the validators but you may know the validators don't even know anything about like the internals of your pair of chain or even like whether as a token or not exactly exactly but you might may slash probably need to give some of your new token supply to the people who put up dots so that you could lose your parity in slot so in a in a sovereign cosmos SDK a sovereign substrate model the validators are the beneficiaries of

new token supply in the polka dot model the dot holders are the beneficiaries of the newly minted token supply which is very interesting because it's almost like the capital allocation role is shifting here so in many times in sovereign change it's like when you earn as a validator you go and validate these chains you are doing some form of capital allocation because I you want to go and validate those chains in which the token value will go up or validate those chains very successful but our pins in pole cut are that kind of role of picking who will be successfully shifted onto the dot holders yeah so I think like the word sovereign carries

some like decision-making connotation to it and I would say in the sense that pair of chains are actually sovereign box chains they have their own implementations of governance and and the way that they handle the logic of their chain when it comes to validators I mean people might be getting confused that must be like break down the whole like validators versus co-leaders thing here but when it comes to validators they really have no choice about like which pair of chain they validate and it actually changes every block so like using like a random of vrf which that can go in another direction but like using our vrf we randomly assigned Vala there's two pair of chains every block and so there's actually every single block this just gets reshuffled and so

volunteers don't really have any choice about para trainer valide just jumping in here like if you shuffle it every single block I mean don't you have as validated like sync the blockchain or like I know so this is like work it's really cool about like webassembly and like some of these like really early decisions say like two years ago about the way that like the host environment is architected and isolated and how we use webassembly and like why a light client was built into substrate from the beginning so you can think of the relay chain as a light client to all of the pair of chains and you can think of every pair of chain as a light clients to the relay chain and

so when avow deter shows up at the at a pair of chain to validate it already has the previous state root of that pair of chain and what what the co-leader from the parish chain actually hands over to the the validator we call it a proof of validity so it it has the block so you have like a list of lists of state transitions and like the header information about like what is parent block is and what the new State Route is and then it comes with this like pruned state tree so anything like a Merkle tree where all of the leaves are just hashes only the values that actually got updated would be and what the handover the rest would just be like the hashes and so is to like quite a large chunk of

information and that goes over but they basically have like the previous state root and I have like the different like basically like the diff over the miracle tree of the state of that and then they have the list the block which is the list of state transitions and so they can actually and then they have the actual runtime like the wasum executable of that pair of shane and so they can like they can validate this list of state transitions against the difference in the state and they can show that they validly get to the new state route and that's like their process so you have this kind of like cluster evaluators and they're all going to kind of like gossip this around like their little like micro cluster say there's like 10 voters on a

pair of chain and they're going to attest that this thing is valid before what actually goes into the relay chain is just like the header information or like a hash of the header of this pair of chain and that's what actually goes in and there's like a whole bunch of like other availability and validity checks that go on before this gets finalized and that stuff pretty interesting too but yeah it's just like to briefly answer your question and that's how they're able to do it without any kind of sync issues yeah that's actually super cool one of the concerns I have as a putting on a hat as somebody looking to Lisa Peres chain one of the concerns I have is that this model appears to be one in which pricing is

potentially volatile meaning if I think of an AWS experience I want to be able to go to AWS and have a machine for like as much time as I want with a fixed cost that maybe goes up with the rate of inflation that's the arrangement I want but in the in the polka dot model I have to get some people to put in dots that might be volatile themselves and because they're putting in something that is volatile I will have to give them something that is volatile as well and there has to be a financial contract balances volatility it feels like it's not the ideal customer experience for

somebody money to launch parity so I mean with that like I don't think you necessarily have to reward people giving dots with another volatile asset I mean I think some some people will want that because they'll be risk seeking and want to take that but I think they'll be probably an element of contributors out there that will appreciate a stable return to their volatile asset which which they will get back so you know I don't think that's everyone but I mean to your point on this is an ideal user experience I mean I think at the end of the day this will really actually help select the good projects versus the bad projects and you know by by basically having this process where you allow the

community and dot holders to actually drill into what each of these projects are doing I mean I think value propositions will be challenged enough that we'll actually see some some really interesting models here and really see good projects actually get the funding required to to participate in this well I think I also the parroting leases for two years which you know in crypto space is kind of an eternity for a project to get going but there's also it's not like you lose this lease and then you're all of a sudden like kicked out or whatever there's a lot of other options so like you can convert to a pair of thread and have like more of a pay-as-you-go model and you're not going to get like the same execution throughput but you still

get the same security guarantees the same API the same experience it just might be like a little bit slower and so I mean that's kind of like an an on-ramp and an off-ramp to being a pair of chain right is um that like maybe you don't have the ability to to raise all of the tops that you would need for a pair of chain slot but you can get started as a pair of thread and kind of like prove your application and develop interest in it to get like the next one there will be some stability even though like early on like some of these auctions could have wide variations but the actual like swapping doesn't only take place like during an auction pair of threads and pair of chains can actually just kind of

like agree to swap so like if somebody decides hey I don't want my pair of jeans thought anymore but I still have another year on my lease if you find a pair of thread that's willing to take over your at least you can just swap places and so I think with that kind of like slot liquidity if you will that allows us like intermediate swapping I would be surprised to see in the long term like too high of a variation there because like people will always know like oh if this if this law auction goes for some obscene amount you know I could like kind of offer to take somebody else's like off their hand in the mid term for something cheaper we also have this idea that there's going to be kind of one substrate based chain look to a

pair of chain slot which which is true it's technically to think that has to happen but I think what you'll actually see potentially is like one chain that wins pair of chain slot maybe they don't actually need all that throughput and you know maybe they could you know it depending on what they're building this this will vary quite a lot but you know if they're kind of building the stable coin platform maybe it makes sense to have you know a substrate based chain that's building a defy application actually deploy on their chain as well and you know maybe they can actually say okay defy application do you want a portion of you know do you want a portion of my runtime and I mean with substrate sophisticated governance I

think there's actually probably quite good ways to manage the sharing of these pair of chain slots like that cool those are some very very cool features I'm amazed that it's just possible to kind of like swap out the pair but you guys brought up pair of threads maybe it's worth spending like a few minutes talking about what pair of threads are yeah I mean like it couldn't they we're kind of born out of this concern about like well what happens if I lose my parents slot you know like what am I just out or something and so I mean you can think of like an application on a computer that you know you can kind of like pull in some application execute it on a thread put this back in memory and

then come back to it later rather than keeping it like in the couch and executing all the time there are actually quite similar pair of chains I mean they have this exact same API they have the exact same security they have the inter chain message passing or cross chain message passing there's going to be more latency because like the messages aren't going to be delivered on like every single relay chain block so the big difference is really like how you allocate these slots so like we predict that for every 10 voters on the relay chain we can have one pair of change slot and so you know if you have a thousand dollars of 100 slots but this is kind of like fundamentally limited it can only execute so many pair of chains

in one relay chain block and so pair of changed you know they put up this bond for like one or two years and by bonding these tokens they basically have the right to execute a block of their own every time that there's a relay chain block so they have very high throughput they're just always able to a buck pair of threads they actually don't have like a bond or it might be like a very small bonds to become a pair of thread but then they have like more of a pay-as-you-go model so you might have thirty of these hundred pair of chain slots might be allocated two pair of threads and you might have a thousand pair of threads out there that are all bidding for these 30 slots and so they're just going to like bid hey I'm

willing to give five dots to have out it or if they'll take my block and execute it somebody else would say well I'll give seven whatever and this is on a block basis not on a transaction right this is on a block by block basis and so it'll take like the top 30 bids and say like okay well these are like in next block I'm going to take these 30 pair of threads and execute those but who pays like who makes I mean let's say there's this pair of thread which sort of like a little bit like the block chain right and various uses sending various transactions conveys sort of like pay a little fee with each transaction and those get aggregated up and if it's enough they get the block or is this

that payment happen in some other way right I mean it depends on the pair thread I mean like the payment would be like in dock tokens and so like however they wanted to acquire dock tokens like whether they have an account or I mean like they will have an account on the relay chain but like however they wanted to acquire them and like decide how to allocate them or charge their transaction fees and their native token and whatever logic they have for converting that the dots opens to make the bid and it could be inflationary so they could say you know like my block reward is going to be like one native token if it's been say ten blocks since the last time we got a block in the relay chain but if it's been a hundred blocks since we've gotten a block in the

regular chain our our block reward will be say 50 tokens or something that will give that collator more incentive to say like hey I really want to get this block into the relay chains well bid more dots I think a lot of applications actually make sense as a pair of thread and not as a pair of chain so like if we think about like higher throughput applications of course that makes more sense to be a pair of chain and they might start out as a thread to gain some traction and then like raise the capital that they need to become a pair of chain but applications that are say like read heavy versus write heavy those would make perfect sense as a as a pair of thread so if you think of like a Bitcoin Oracle that just has like hey this is the latest like

block header that I have from Bitcoin or whatever that's only like every 10 minutes you don't need this to have a a block every six seconds for that or so you're like a domain name service you could update your registry every hour and still let people read from it you don't need to submit a block every six seconds so you could just kind of like collect this change to your DNS and then submit like a batch of this every hour or two hours or whatever you want there's no need for this like superfast throughput so I think for some applications actually makes more sense to just stay a pair of thread so in some sense it is like this like on-ramp to convert to a pair of chain or back to a pair of thread but in some other sense it's just like a different economic

model that is like fundamentally like more opposite to certain applications than others so in some sense the the unifying principle appears to be that put in polka dot you're really auctioning off Peres chain block production rights right so this auctions at two times case one is a per block time scale and then there's a one is a two years worth of block production time scale right and Holkar is a system that just believes in auctioning off these these assets and people can bid on those assets at these two different time scales and then and then they can create financial engineering on top of it to

auction of those resources to other users yes like the primary resource is state transition validation right do you have any examples of teams that are building like this financial engineering on on top of Perak chain on top of para chains of para threads like they're they're taking these production slots somehow they are acquiring these production slots somehow and then they're selling them on further or something like that and that is itself decentralized application I don't know about selling them I think the most interesting one that we've seen there's a kala I'm really like I'm excited in general like I said earlier

about like things that don't exist before I don't I'm very I don't know the right word like skeptical or maybe just like negative about defy in general where I see just like kind of these like same financial primitives that a lot of like the early blockchain movement was kind of like opposed to like this is the problem and now we're like well let's just recreate these exact same instruments on blockchain and like I think it's kind of silly and I think it's like it's more interesting to see like something new that's never been done before and so like what I call is doing his um they're calling it a decentralized sovereign wealth fund where they have some like dock tokens outside of their parish reinstall that they're staking and these are actually

owned by the chain so they have a pair of chain with its own like Native token and Treasury but this para Jane has an account on the relay chain and they can also interact with other but like other pair of chains in the ecosystem and so it actually owned some dock tokens that it will stake and use those taking rewards to start contributing to its perishing slot and steadily return whatever crowd fund and dot Hogan's I got to third users who crowd funded them and then in that sense they just become like a self-sustaining pair of chain and they can they can take stake in like other pair of chains or like when we have a bridge to a theory on Bitcoin like they could take they could take a second those and I think that's like

it's interesting because we're not talking about like an organization that's decided to take a stake in like some blockchain token or an individual or something we're talking about one blockchain owning a stake in another blockchain it's something that like we can't really express in like the etherium our Bitcoin world like we can talk about like interoperability that like we have Bitcoin or not the area or something but you can't really have like aetherium take a stake in Bitcoin like that's a new thing like that doesn't exist and I think that's like one of the really cool things that's being on polka-dot I don't know if you have more dieter yeah I mean just one thing about that is basically like a

portion of transaction fees from what a Cal is like from a calles stable coin protocol which is essentially a maker like protocol so a portion of stability fees and and others basically go into this decentralized soft sovereign wealth fund and initially the goal is to just acquire dots until they get to some like critical mass of dots where kind of they think and obviously this is somewhat subjective because of the auction process they think this will be enough dots through staking returns that they'll be able to you know afford this pair of change slot in perpetuity and then they'll be kind of making more you know strategic investments and you know other protocols probably mostly on polka

dot at least to begin with where they think it could make sense for some reason I mean when you think of like parallels here one is obviously like a nation's sovereign wealth fund where nations you have excess wealth instead of like putting that directly into the economy we'll put that in some fund designed to grow returns and use that in the future and then the second parallel would be how corporates deal with strategic investing I mean sometimes corporates will invest in other corporates you know if it's for some sort of vertically integrated reason if they expect to use the services of those corporates down the line you know these are kind of strategic investments and so

this is a really really good example of where you actually see blockchains and poke it on and what we're building actually kind of cross the barriers from just like decentralized applications to some mix of a corporation and nation-state that's able to actually have its own agency which i think is something that we've never seen before so some some computation can be sovereign and that it has no human owners and you think that it needs to keep running are these Pera chain block production slots right in in some sense

the acquisition of these parity in blocks production slots is again subject to some kind of payment that these this sovereign computation can bid for and so so it's almost like you can have sort of a sovereign agent that can pay for its own hosting that can own pieces of other sovereign agents and as long as it's logic is such that it keeps getting on more and more assets so that it continues to host itself it can it can go on indefinitely that's really the future that you know people in the kindy cryptocurrency space

have imagined for a long long time it's like computation that is self owned can pay for the resources to have it executed but whence is executed it on small resources and then can pay more to have it executed and it just keeps on going forever but it says in polka dot it's given like a more elegant economic substrate to to do its operation oh yeah and I fully agree a lot of the frictions in polka dot that come up around like para chain particularly around pair of chain bonds and the auction process here really come down to and they can all be corrected by is this prop do people

believe this project will add value and has this is that is this project do they have a likelihood of succeeding and does kind of do dot holders recognize that and so I mean from that you basically get the like hopefully like the best projects out there that continue to win these slots and by extension continue to be able to succeed on the platform yeah cool well guys thanks so much for coming on we said it was a pleasure to kind of talk about polka dot again it's exciting to see this project you know live and now unfolding and I think especially all of these interesting economic experiments and experiments around like

coordination and that's I think that's the part that I'm especially excited about when it comes to polkadot and I think in the next in the next few years as this actually goes live I think they'll be incredibly interesting experiments happening on polkadot yeah thank you back I'm super excited to to see how it plays out yeah thanks so much for having us it doesn't end here there's much more of this conversation and you can hear it by signing up for epicenter premium as a premium subscriber you'll get access to a private RSS feed where you can hear the interview debrief which goes on for an extra 20 minutes you'll also get exclusive access to round table

conversations with epicenter hosts and bonus content you won't hear anywhere else go to epicenter dot Brock's slash premium to join the community and support the podcast [Music]