Avalanche: What are Subnets and What's Wrong With Appchains | Kevin Sekniqi & Patrick O'Grady

Transcript

all right everyone so on Empire you obviously know that we talk a lot about the institutions coming into crypto and that is why we are super excited to share that we are hosting the digital assets Summit we hosted this since 2019 it's coming up in London March 18th to 20th don't miss your chance to get ahead of the curve you can get 20% off with code Empire 20 we'll see you in London this episode is brought to you by toku the first comprehensive Global solution for both token compensation and tax compliance toku makes implementing Global token compensation and incentive Awards simple with toku you get unmatched legal and tax support to both Grant and administer your Global team's

tokens across the entire token life cycle make your token grants easy today with token all right everyone welcome back to Empire uh very excited to be joined by Kevin uh co-founder of avalanche and Patrick ogrady VP of engineering uh at Avalanche so Jen welcome to the show thank you for having us yeah us I was uh joking to some of the guys before this that um I hadn't spent too much time in in avalanche world in the last couple of months and this uh this podcast forced me to go deep and I've I I I will say I'm impressed by everything that's going on there's a couple interesting acps hyp SDK uh some cool plans like future plans for for consensus uh fee isolation firewood um warp messaging a lot of cool

stuff I don't think we can have that that conversation though until we kind of just remind folks like what in the world you guys are doing so Kevin I think I'll pick on you for this almost a too high level of a question but like there's this big landscape of blockchains today l1s l2s um wh why does why does Avalanche exist like where do you guys fit into this whole landscape um yeah so at the end of the day it comes down to a simple thesis uh that I think I mean the market still has to play out from a adoption perspective but I think it'll become at least you know internally on the engineering side we hold this opinion I think lots of other people actually ultimately would hold this opinion as well even

externally um and I think it will materialize in the next three to five years uh but the future of blockchain ultimately blockchain infrastructure is extremely fast highly optimized chains and a lot of them um the simple reason for why a lot of them is because there's just far too much potential data that could be produced by users I mean just back of the napkin calculation uh if you were to have let's say I think uh goon had tweeted this recently as well um just back of the napkin if you were to have four billion users uh and each one of those users on average produces about 100 megabytes of data uh per year not

even more which is a trivial amount of data but just you know back of the napkin um and this is not actually hard to to imagine uh especially if you're considering nft applications SK gaming skins uh all kinds of defi applications on chain 100 megabytes is nothing uh but still supposing back of the napkin we're still talking about a amount of data that is greater than the entirety of the Facebook data L right now uh per year um it's just INF feasible completely INF feasible under any infr under any uh you know thing that we currently have right now where we can possibly envision to have everything on a single chain um so the simple thesis is okay there's just

all this data that's going to be generated from in you know gaming to music nfts and regular JPEG nfts and uh uh you know social fly coming there is real world tokenization of things and all the applications built on top uh as as stxs um how do we manage all this data well the simple answer is we shove as much hardcore engineering ing to make each chain as fast as possible under even gigabytes or terabytes of load and yet at a certain point everything will break so we distribute it horizontally uh when you distribute things horizontally into their own environments where each environment has isolation for storage has isolation for execution

meaning its own potential you know virtual machine execution uh isolation for security as well uh you still need a way to make things interoperate so seamlessly that at the end of the day if you make a transaction on chain a and it has to get routed to chain B and chain C for liquidity the end user will feel like this just happened on the same chain there will be no visibility to the user to the end user of of what just happened so you know imagine for example you have on on chain a you have a you know token that you want to swap to token uh X you want to swap for token y you have an aggregator application it

notices that there is actually a better pricing on chains B and C it will automatically route your asset X directly into chains B and C it will swap there and then will reroute them back in a single transaction lasting less than one second um it's the gold standard but it will make basically uh you know this Global I mean this this universe of many super fast blockchain chains look like there is actually just one single chain it's the it's the ultimate instantiation of sharding even though I hate using that word because this is not what sharding is but it's the ultimate Nirvana instantiation of sharding basically um so I think this will become so obvious in five years uh maybe a little bit longer but let's

suppose five years it become so obvious and uh you know Avalanche has built all this infrastructure precisely to go towards that direction so I think that's a very high level overview okay so go one level deeper we've got the pchain we've got the C chain the pchain the xchain and then all these subnets so maybe walk us through how you guys have actually set this whole thing up yeah so Patrick I will let you dive as deep as you want on this but very quick high level on this just for a little bit of um Clarity there's a bunch of for historical reasons there is some uh name nomenclature that is still stuck around that is probably not very accurate um and and it gets confusing uh but xpc chain and all the other chains

that currently are whether on mayet or on testnet um they're just chains um and at the end of the day none of them are special the only one that's special is the pchain and I'll just give you a bit of a uh architectural overview so um you know X and P and C chain totally not special they're really just subnets themselves they're just happen to be subnets that everybody body else also replicates right now so they're Global subnets they're basically replicated across every single validator on on the on the Avalanche Network whether you're running maybe like the I don't know singling out like the gonzilla subnet or whatever um you know other subnets don't have to run the gonzilla network but they have to run the xpn C chain

currently but the current ACP is to remove that completely uh and only ultimately everybody would have to run the pchain and pchain really just is a metadata chain it's a platform that's what it stands for and it's the basic most minimal requirement necessary for us to be able to run awm which is the interoperability uh infrastructure that we have um it effectively says that if you want to interoperate all these chains seamlessly we need to know at any given point what the latest set of stakers is that's precisely what the pchain would give you it would also give you some ability for simple payments and transfers and so on and of course taking but the most important thing is to

respond to the question okay I'm I'm subnet X I want to interoperate with subnet Y which are or what are the current set of stakers for subnet why and I need to know this because this is the only way that I can do a trustless transaction between the two trust is being defined here as nothing more than the uh the the set of validators for any particular chain like no Reliance on any third parties um so that's really the architecture here they're all just horizontal completely flat um you know all separate set of validators some interlaw is because we need all this metadata um right now the C chain is also replicated and that's more of just

like you know you need smart contract permissionless evm based so you can use the C chain but there is this ACP right now to not require the C chain to be replicated across the network and you know make it make it more you know Niche uh but that's basically the the architecture here we'll get into that ACP in a little bit because I I read that is really interesting um Patrick maybe I'd throw this to you when I think of what you guys are trying to build the the the like word that comes to mind is app chains and when I think I think when a lot of folks think about app chains it was originally was it was actually polka dot which I think a lot of people have forgotten about now and and and Cosmos and what's happened is a lot of actually

app chains have been built not in Cosmos land but in um uh on on rollups actually and I'd be curious how you think about like how subnets like compare subnets maybe to Cosmos and to uh uh building an app chain on a rollup yeah the first thing I'll say there is um I think app chains has been a term used for a while I prefer like I think that the industry should move towards more like using the term optimized chain I think for that term because I think it is a better description of really like what the like actual qualitative experience of using them is for many of them now um like it's not typically like a toy but like some you know specific set of trade-offs that's optimized for a use case so uh

appin s like kind of like a toy uh a bit to me but uh I think on the uh on that front that's the question we always get asked right is you know you got subnets right but there's already l2s there's already zones there's already par chains there's already you know to use to bring back sharding like Kevin mentioned shards I guess and other ecosystems um and uh I would say like for people that aren't familiar with Avalanche it's much closer and probably closest to the cosmos World which is it's optimized from the get-go for Sovereign security so like Sovereign sovereignty all the way through so there is no like an L2 an attempt to try and share security from like a single base chain or origin chain where data somehow needs to be

incorporated or even in some cases included in that chain um the difference being primarily that you know as Kevin mentioned with having really fast but a lot of chains the first question that comes up is like well great you've like now made the problem worse right because you have a bunch of chains instead of like one like we can't even get people to use one chain safely or easily like now we got like a ton of them so the the first thing that comes up right is well how could you connect them or like how could you make it feel seamless and so the big difference there between Avalanche and Cosmos is Cosmos there is no like Mothership in some sense where like some of the registry and information is held so OS zones need to constantly keep each other up to date

with like the channel state to actually send messages across each other and not to say it's good or bad like if you're not sending a lot of messages it's great right because you have a very low fixed cost you really only send like Channel synchronization information uh whenever a new message needs to be processed or there's some time out but I'm not going to get into that um avalanche on the other hand says you know we're willing to accept a fixed cost AKA syncing the pchain but then have a very low variable cost so on Avalanche between any two subnets there is no um like it's just as expensive to send a message from subnet a to subnet b as it is to send a message from subnet a to subnet C but that's not quite the case in Cosmos where you have a smaller set of channels because

there's this overhead of maintaining Channel synchronization information now all that being said is to say if you have a lot of chains it should you should be able to communicate with them but I think in terms of the L2 World um it really depends on your view on how security comes to be and what is required for a community to succeed so in a world where you uh have no interest in creating like some sort of like Community source security budget uh and L2 is a great choice for you like I won't debate that right like you're you just want to spin up something and use e security uh rely on fraud proofs or you know some sort of ZK mechanism to secure

the chain like I totally understand that uh but there are sets of communities that have very different security standards maybe like let's say some of the the partners that have you know started to use Avalanche for their subnets like a JP Morgan you know they understand that eth is willing to provide security for an L2 but they have a couple thousand people working on infrastructure like in their own company or tens of thousands and would much prefer to provide their own security similarly with a new community or ecosystem the in inve to provide security can be quite a powerful flywheel to like get ecosystems and communities off the ground right if you only have your native unit of account and that's only within your community

there's clear incentive amongst everyone there where maybe it maybe is more expensive to get started but there's a different type of flywheel than using the assets or security of a different ecosystem where you're you know maybe getting off the ground faster but have to answer questions on a long-term sustainability basis of how you actually if you want to incorporate some sort of token what the utility of that is over time not going to get into that any deeper I'm not a lawyer I don't want to be a lawyer but I think there's like interesting conversations to be had there um but my main takeaway is there's quite a spread of different options I think time will tell in terms of like what actually ends up being the most meaningful um but the two theories that I think adland is oriented most around is transactions filled the space that's

allocated to them and within that people want to communicate between these different communities and that's our architecture in a nutshells based around those two core ideas just to say one more thing that's uh very very important it's all about optimizing speed and efficiency between these chains uh rollups have this problem well besides you know the centralized sequence or I'm not even going to get into that stuff uh putting that aside the one thing that's that's really really important that that's almost fatal at this point um you cannot do the one second transaction cross rollup right now you just will you will

never be able to do that uh under any circumstances you just don't have the settlement speed that you would have from something like uh the the Avalanche Network because the trust assumptions assumptions are a little bit different uh that poses new challenges uh for the rollup operators that are really really hard to overcome um but and then on the cosmos side you know the app chain thesis was probably right the way that Cosmos designed it uh earlier on and I think it will in fact play out precisely at a high level in the next five years I have actually no doubts of that it was just that the implementation was done incorrectly so the vision was there but

implementation was wrong IBC is extremely heavy for example it's deeply deeply heavy I mean you have to carry so much information that it becomes like almost it becomes so hard to use and so cumbersome and just the performance of the system uh is significantly um outd it at this point I want that's the best word I can use um and then furthermore interoperability I mean the the the SDK while Cosmos has a really fantastic SDK I still think it's deeply outdated at this point from like a VM perspective and so on so the vision I have no doubts is going to be correct the implementation was uh was maybe maybe one of you guys could actually

expand on the economics of subnets as they relate to fixed costs and variable costs spec uh with this spectrum of like there's Cosmos there's Avalanche and there's rollups can you walk me through like the three those three as they relate to like some have low fix some have high fix some some the variable scales quickly some the variable is minimized like how how do you guys see that what is what what does that look like yeah so um I think the this is another one of the bigger trade-offs between like Zone subnets and l2s um which is like when you're using the L2 the your ability to scale there is based on how much data you end up replicating right because the security comes from elsewhere so if you want to like pose a

fraud proof or like prove that that data was actually correct you somehow need to store pay to store that somewhere so inherently that data is variable cost right like and I think people are currently using eth for that like obviously like the arbitrum and optimism side of it where they're putting it on chain and you know dank shardy and everything that's coming there should lower the cost of that other people are trying to do an entirely different da thing like some of the IG da stuff where it should be a factor lower but share some of the security properties not exactly as an L2 but like pretty close to it for good enough for most people um the difference with subnets is like the the subnet itself solely replicates the data so there is no external fee rent charged by Avalanche the platform um so

for people that are getting started right now all you need is one validator to create a subnet so right now and as we talked about ACP 13 we can go into that deeper later but requires you to be a validator which is requires 2,000 minimum of ox but then once you have a subnet and like some set of validators there there is no additional variable cost charge to Avalanche so in the sense and I think the very fair critique is well isn't that kind of high for like a new project getting started yeah and that's what the whole idea behind ACP 13 is is to try and lower that thresold to make it more competitive with other uh projects that are trying to introduce this sort of like Craton blockchain scaling um you know

property fully understand that so roll you could say rollups on eth are very are actually low fixed to start but High variable they SC this the cost scale very quickly obviously someone like a maybe a Celestia is trying to solve this um but that's kind of the case right now Avalanche is actually has inverse inverse uh economics to Roll-Ups where they're actually maybe high at the beginning um but the but the variable is decreased is that fair to say yeah well it just has to do with like where the security is is sourced right like the if the security is external naturally there has to be some exchange of value to compensate someone external for supporting your effort right and again instinctively they're probably not going

to do that for however much you want to do right if you want to shove like 100 megabits per second in they're probably not going to charge the same amount as one megabit per second so naturally there has to be some sort of scaling there now on Avalanche with subnets like there is a variable cost to the infrastructure provider right like you if you are processing more data you need better Hardware but from a charge to the network itself it's entirely controlled by the subnet on how it wants to do that yeah that's the Nuance yeah said maybe slightly with a few different examples um a rollup operator would have to um fully pay for the hardware costs the users would pay for the fee costs uh

when settling everything so they would they're the one carrying the actual like you know um transactional cost but Hardware cost is fully shared by the rollup operator um it could get very big so you have to run a lot of Machinery there uh different for a subnet uh you run the subnet you if the subnet has value associated with it and the rewards that are paid in the native token are valuable there's going to be an economy of people running this chain immediately in the same way that there's an economy of people running eth and Bitcoin and you know everything else there's a value to running it because you're accumulating the rewards um that's basically what it is on the subnet side so I would probably say it's after the ACP if it's passed it's low uh cost to start a low cost to maintain as well uh

versus anything else so that's uh at least hopefully if the ACP passes if it doesn't pass then right now it's probably a little bit High upfront cost and then lower afterwards uh but you know you can easily imagine in a situation where you run a subnet um and uh you know it has this native staken token the Yano token and then all of a sudden everybody stting validating it you don't need to run a single piece of infrastructure yourself it's fully permissionless people other people run it for you and it just takes off I think one of the things that folks maybe didn't like who wanted to run their own subnets was that you had to I think also become a validator on the C chain I'm going to completely

botch this but I think you had to become a validator on the C chain I think Kevin I think you wrote this piece path to a thousand subnets and I think you guys were thinking about relaxing the require I'm mixing up acps I'm sure already but starting to relax requirements to actually validate the main chain what's the decision- making behind this yeah so the path to 10,000 subnets um was a piece I wrote kind of originally a blog post and then turned it into ACP um but that feedback is exactly the feedback that prompted it right people want to have their own network their own Sovereign chain and there was a lot of confusion first of all like why I have to validate the C chain right that dramatically increases the amount of like Hardware I have to allocate to the

primary network but also from like a regulatory perspective some of the people that were launching subnets could not or were not able to like actually validate like a permissionless smart contracts chain so these are the people you know like more of the publicly listed regulated companies that are like okay I get the premise of a subnet but I don't understand how I can validate this thing and you know continue to actually operate the way I'm supposed to as well as um like great I want like super subnet scale but that seems to like be conflicting with the need to validate these other chains that I don't need really to run my Subnet um and so that was the culmination of you know a ton of community feedback on like uh on the way that people felt the avalan network

should evolve um and the in a nutshell idea and then I'll turn it over to you Kevin to elaborate on any aspect of it is uh really the only thing that's needed from a uh ability to run a subnet or operate a subnet is as Kevin mentioned earlier this knowledge of the staking sets um so like what is your staking set as well as what is the staking set of other people to verify warp messages and all that because of the original design is contained on the pchain um so really all you need to stake is like uh ver verify or synchronize as the pchain you can't fully validate it because you can't there are like transitions of transactions between the X C and P and

if you don't verify stuff on the C and xchain you can't properly fully verify the pchain so you still have to trust the consensus of the primary Network to sync the pchain um but the idea is it will dramatically lower the the actual like cost to to run a subnet um from an infrastructure perspective and then from there uh because you're not a validator you don't have to put the 2,000 of Vox up front and that ACP goes into some ideas of uh different ways that subnets could work with the primary Network such that there's still like this deep connection between the two even though subnet validators are not primary Network validators gotcha so you're basic introducing this new type of Staker I think you called them subnet

only validators where you can validate subnet and participate in all the new thing like Avalanche warp messaging awm but you don't have to sync uh back to the primary network is that right that's correct yeah you don't have to validate it and so you're basically saying I am I don't want staking rewards from the primary Network I don't want to run the X and the C chain I just want the benefits that the pchain offers which is securing my stake and uh letting me uh basically operate warp messaging um and in return you know I will uh use this new Staker type so not mandatory you could still do both if you wanted to but we expect uh many folks to do this but

the yeah the outline from there goes into like how things could even be better with pays you go uh staking which basically sets a market for how people join and become subnet participants based on the capacity of the primary Network and then uh avox augmented security which is well now like there's this disconnect between Avalanche the primary Network and what it can offer subnets and so the idea with with that is that people can use Avalanche to bolster the security of a subnet um in the first probably well uh well elaborated intent to have some way to share or provide help with security as the network is small we can go into that later if you want to but there's a few different direction before we're going into that

give yeah before so actually something Kevin said made me think about so I actually didn't this is gonna be a really show show how uh left Bell cve am here on on Avalanche but uh I didn't even realize the pnx chain were were just like these Global subnets it but it makes me think about what the end vision for subnets are um and I think there's probably two worlds here and I'd love to hear maybe Kevin your vision for this one world is there's millions or tens of millions or hundreds of thousands of sub subnets eventually uh another version is you basically have these like big Global subnets that are uh maybe there's the gaming subnet or the defi subnet um or the you know that the nft subnet what what vision is uh which one of those versions is like more correct

with your your vision of the future uh right so um maybe differently said is um whether or not there's going to be um one only a couple of big subnets or a lot of like smaller subnets uh is not the question to ask it's more of like regardless of what happens how can we minimize as much as possible overhead of launching new blockchains that interoperate basically you know as as minimally as possible like reduce those requirements as minimally as possible and it comes down to basically this ACP that Patrick is talking about which is you know get rid of the C chain

get rid of all these or reduce the amount of stake for validation uh and at that point just let the market decide what he wants to do but if we get to a position where we are having billions of users and if we're getting into a position where you know these users are producing you know data that is comparable in the dozens of megabytes or even hundreds of megabytes per year there's just no way a single chain can ever handle this full stop so it's there's no doubt in my mind that the only proper architecture for the next five years or so is multiple of these and so the question is what is the absolute bare minimum required architecturally to to make all this stuff work as easily as possible and that's effectively what is being

designed here with awm with the new acps that are coming in uh the only bare minimum here is the the pchain everything else is optional and um you know optimizing the living crap out of these chains in terms of performance in terms of cross-chain interoperability in terms of flexibility for developers um and just let people go wild with with old this tool link we can switch gears a little bit and talk trade-offs so there's no there's no perfect blockchain uh everyone's got a tradeoff right you could uh you could have higher performance but you're probably trading off some flexibility of smart contracts right you could increase throughput but you're going to probably raise the requirements for Hardware as we've seen

other other folks make that decision what is the trade-off that you guys have decided to make how dare you say a tradeoff there's no tradeoffs there are no trade-offs you guys have Sol the no no look I we'll be very straightforward with everything you can increase the blockchain I mean the the hardware size you know probably there's going to be acps to increase the blockchain the the hardware size as well for for Avalanche um at the end of the day just very quick back of the napkin the data that I'm selling you at that particular volume you can't have it even under the most hardcore block Hardware out there you do need to replicate things or you do need to separate things horizontally across multiple blockchains fault isolated performance isolated um so you know what

are the trade-offs for for Avalanche I mean the trade-off is fundamental in the building of a distributed system I don't think there is necessarily uh a tradeoff of avalanche versus other blockchains it's more of just like is there a trade-off to using a blockchain at all versus not using one and you're just using a database um if you want to use a blockchain and you want to replicate things and you want things to be distributed and you want other people to replicate this data and make it persistent for a very long time you're going to fundamentally have an immediate trade-off in performance in scaling and so on that just running on a data Lake on AWS will not give you that but then again uh you know that's that's fundamentally what it comes down to uh

from a tradeoff between Avalanche versus the other ones I you know I'm like just totally being honest right now let's think it through it algorithmically probably the fastest one right now in terms of time to finality there is some like trade-offs in terms of latency that we're working through and optimizing but one of the fastest is not if not the fastest in terms of time to finality um in terms of um uh uh storage uh and Hardware requirements you can increase this as much as you want you know it's just really a question of like uh you know will the community accept this uh insanely High Hardware requirement maybe they do that's great so that's not really a trade-off uh in terms of like database stuff and all that you know hyper optimizing the living crap out of

those as well um and in terms of uh the the interoperability protocol um almost certainly between IBC ccip and awm probably by far the most efficient across all of these um in terms of uh you know maybe like the trade-offs might come in this in the form of like oh I I don't want to necessarily bootstrap my own security I want to rely on E sec security and but but then again why is eth secure it's really ultimately comes down to private Key Management uh so you know maybe that's the trade-off between a subnet versus an L2 but I'm not exactly sure that's necessarily a good

one so the question to be asking here is uh if I'm devel if I'm a gaming developer and I'm building um you know secondary market for my gaming skins do I do it with a blockchain or do I not do it with a blockchain at all that's the really the the trade-off question but then otherwise if you're if you've already been sold on the benefits of of tokenizing everything and making them available on an open network of of blockchains then I don't think you have any trade-offs here with other blockchains versus Avalanche um at least not that I can see if that are obvious to me at least so that's the tldr I see three if you want to talk about those I think I think the three like first big one is like I think the

benefit of like that we get for messaging comes with having this pchain right like so some people would argue that they don't want to have like this kind of registry train like a cosmos world right so they only have point-to-point connections so there's like something something that uh you know it's again a preference Choice uh the other one is I think the security topology so none of avalanche's like virtual machine designs are meant to be provable on the primary Network at this point you could certainly go down that direction if you wanted to but it's a very different goal than maybe some of the other groups that are that are trying to balance Stu uh and then the last one is flexibility um kind of an extension there is like we are trying to

achieve like this multi-m world on Avalanche where you you may spin up a wide collection of different uh things and inherently when you have that communicating between like really flexible VMS can be uh challenging right so some of the L2 designs for example like I think there's the super chain pack Manifesto that optimism is pushing is like to have a single like kind of Base same thing with ZK sync uh they want to have everyone have the same core ZK module uh and what that allows is like easier integration between the SE like the different chains that are being sequenced um and so we have felt that the more important aspect there is to like really optimize how broad you can

go in the crypto design space and accept maybe more flexibility at the cost of maybe having some of the interoperability between those virtual machines be a little bit more difficult uh to manage so I think from a technical perspective at least those are three clear tradeoffs that I think we've made all right everyone so we talk a lot about the institutions coming into crypto on Empire Santi and I are both headed out to London March 18th to 20th for Block works' 8th ever digital asset Summit Das this is an Institutional buttoned up conference that we've hosted since 2019 I like to joke that it is probably the last remaining kind of suit and tie event in crypto people are still wearing suit and tie it's pretty funny

you'll actually hear from a lot of the largest institutions in the world coming from Standard Charter FIS JP Morgan framework folks coming out winner mute vanc Goldman Sachs there are a couple big themes of this conference one Bitcoin Catalyst the having and the spot ETF two A View From the buy side three rw's tokenization and stable coins four Global regulatory Frameworks five institutional infrastructure including banking and payments and six the macro case for crypto if you have anything to do with the institutional side of crypto you have to be there Santi and I got you back we hooked you up with a 20% off code it is Empire 20 there is a little competition running internally at blockworks to see who can drive the most

number of tickets so help Santi and I out register with our code and you get 20% off that is Empire 20 this episode is brought to you by toku the first comprehensive Global solution for or token compensation and tax compliance if you say yes to any of the following four questions toku is a no-brainer solution for you number one are you planning to launch a token number two is your token already live number three are you currently granting your employees or contractors vesting token Awards and number four are you trying to figure out how to take care of taxable token events for your team if yes you have to get in touch with the toku team toku to high level makes implementing Global token

compensation and incentive Awards simple you get unmatched legal and tax support to both Grant and administer your Global team's tokens toku navigates this across the entire token life cycle from easy to use token Grant award templates through tracking vesting to managing tax withholdings toku makes it simple for leading companies in the space including protocol Labs dydx Foundation Min Foundation hadera nosis safe getcoin and many more reach out to toku that is tok.com Empire T ku.com ire click the link in the description or DM me on Twitter and I'll get you connected to the

team what do you guys think about when you see these new as like the industry has branded them NextGen chains like the you know whether it's a manad or a sui or apos they you know they they've made some they're very interested interested in like I guess Max throughput is what I'd call it what are the mistakes that you guys have probably gone through and made that they maybe don't realize that they're gonna have to make or that they might fall into soon oh let me let me get into this more real quick so um so the Kevin just like cracked his finger stretched his back out no no I'm not I'm not gonna talk negatively at all because they're all actually doing uh uh

doing lots of great stuff um parallelizing the evm is good work although there is arguments on whether or not the evm is sustainable long term honestly in five 10 years time frame are there maybe better uh virtual machines arguably yes so maybe that's a different conversation but at least for the evm parallelizing it is a is a good Endeavor um the thing is we actually have an enormous amount of data on where a lot of the bottlenecks are like real world data uh from networking to uh execution to blog propagation and then storage and uh memory um and so on um the the parallel execution is not even remotely

a bottleneck um it is a bottleneck when you have an empty state so if I launch a new blockchain and there is no State and the database is empty then I start bombarding with transactions indeed you could run into the problem where parallelization is actually really nice um as you start scaling up the amount of storage that has been uh used up uh by the chain you will very fast like so quickly find that it's not even like it doesn't even come in the same zip code as the overhead of actually just writing and not reading but predominantly writing the data to disk and then managing it easily so it's still a huge problem it is a completely unsolved problem even algorithmically let alone

implementation wise how to actually manage active State longterm uh the one answer that is that is obvious is to to separate it logically between multiple different kinds of applications that's exactly what Subs are doing it's a form of sharding but I want I do not want to use that word so do not confuse that with you guys really hate that word huh it's just not the it's not the right word but uh but yeah so um you know it's just not the bottleneck right now it's ultimately it all comes down to um at some point it will grow like the state will grow very big uh how do we manage that so that's a problem that salana

will run into very soon um eth probably will not run into that problem anytime soon purely because they have parameterized their blockchain to currently not have storage as the bottleneck but to actually have the consensus be a bottleneck meaning that like they literally just tuned it all the way down to whatever the TPS is uh at at Max Capacity um and that's because they're just trying to throttle and avoid going into like the state blowup issue too fast um but but if you don't have that uh uh otherwise you know if you remove the the artificial bottleneck that has been implemented in ethereum um then ultimately over a long enough period of time storage always becomes the the

bottom leg so all these other things being optimize great and we can probably use a lot of This research directly into Avalanche frankly um but it's just not the bottom neck right now yeah we can get into um some of the things that you guys are working on going into 2024 there's uh there's hyp SDK there's some fee isolation stuff I saw goon had this post about um Avalanche go which seems pretty interesting uh though a little over my head technically uh there's firewood uh there's warp messaging um where do you guys want what's what what's like the most interesting in your mind a lot of it is really interesting Patrick what are you thinking where do

we want to start here so I think it makes sense to start with the stuff that is uh closest to use I think for the average person in the Avalanche Community um and I think that that probably starts with warp so for those that aren't familiar with it and we keep saying warp warp warp warp right like it's this cross subnet messaging protocol um it actually has been out since uh last year actually like winter of of 2022 um but the issue is that we uh like adopting it to work well with the evm uh has taken time and so finally after what has been you know about a year of development um we have a protocol uh ad adaptation of warp that is part of the

evm and then wrapped in something called teleporter which is like a smart contract interface that will let people take assets from the primary Network like avalan or any erc20 and bring it into a subnet using this like warp protocol and so people have heard about warp but they actually haven't used it yet so what this will allow people to do is you know easly easily bring assets between the two without having to rely on a you know a third party Bridge operator so from like a what can I soon use on Avalanche that will totally change how everything works like that is going to be a big one uh for for folks and then from there you know once we we have like a credible path to actually having these things communicate with

each other a lot of the work just goes into performance so like you know no one's happy with like 10 or 100 TPS on the C chain or on any of the subnets and so like how can we entirely change how that's done um and then which is obviously a quite an engineering feat as a lot of other teams have also seen um and then the other one being um yeah I'll pause there yeah I just have some questions on um on on WP message so like what what sets awm uh and teleporter apart from someone asked this on Twitter and I thought I thought it was a good question apart from the likes of IBC and axel when it comes to trust minimized uh bridging and liquidity

access yeah so um first I'll kind of describe a bit about warp and so the the mechanism of what it is and so as we mentioned before there's this P chain this registry chain validators on any subnet will register uh you know a special key a BLS key in this case not important um and what will happen is when they message with each other uh the validators in a particular subnet will sign uh that message and then once you have so many of those signatures you can aggregate them into a multi signature this is similar to like what eth2 does with validator attestations and you can take that multi signature and bring that to a destination subnet and it can verify it now the key there is in that verification step on the destination

subnet is where a lot of these things differ so on Avalanche the verification works as the subnet says hey who are the validators on the subnet that's messaging me I'll go ask the pchain the pain tells me what the stake weight was and their keys and I can see did the message that they send me was that valid from like my parameterization of is it a sub that I want to listen to messages from and is there like a substantial amount of stake that has signed it that's aving the difference with Cosmos is an IBC generic messaging format THS is pretty geared towards like Cosmos SDK interoperability but like it doesn't need to be um and the way they do it is they're constantly synchronizing the signers between two zones and so that

when you verify a message on a cosmos Zone you're looking at like that Chain's Channel and record of the validators and then looking to say okay is this message valid so they have like a local record on their Zone but they're not like asking some external chain in a sense like they have records from that external chain that they internalize Axel is just a multi- signature similar to thorchain which is that they have some set of validators that produce like typically some sort of threshold signature where as long as that threshold signature exists you're all you're doing on chain is checking the cryptographic validity of it but you're not like looking externally to say like oh some amount of Stak signed it or something now they all have you know different properties as

I've mentioned U but that that's the big difference between the three and again is there a perfect solution well it depends on your preferences right like the cryptographic solution axar has is super generic right you don't need to be within Avalanche or anything like that you just have a signature that I can verify on bitcoin or I can verify on ethereum ethereum as long as it can verify the signature doesn't care where it came from right so that would be one example of something that's interesting tradeoff like warp works really well within Avalanche but not out of avalan so okay so warp solves the need for cross-chain communication between subnets will you guys yes will you guys launch like native solutions to communicate from Avalanche and from

subnets to other chains so this is one that um will be possible in the future but what it requires is synchronizing like the uh stake that's on the pchain with other chains right because the validity of warp relies on checking signatures against like that staking set so what would have to happen is like the pchain would somehow have to produce like some sort of signature that could then be imported by eth or imported by something else but like there's inherent limits on some of the chains capability to do that right like we couldn't import that data into Bitcoin for example so like warp is probably never going to be compatible with Bitcoin similarly eth at least on the execution layer doesn't support

quick verification of BLS signatures so it probably be really expensive actually to integrate it within ethereum the idea I think we've gone for so far is like sure like as I mentioned there's this like broad spectrum of different cryptography could be used for this we really wanted to pick the one that we felt like was best within Avalanche and make sure that that was really really good and optimal for that and then in terms of external bridging discussed that once we felt like within Avalanche there was a great bridging story which hasn't been the case yet because teleporter and some of like the evm adaptation to warp hasn't really rolled out yet but I think following that if if we really see great adoption of that I think there'll certainly be a push to

understand how you could bring that technology out of avalanche but to me that's really like running before you walk in in a sense yeah when does this go live Patrick um so we have actually a demo website you can go visit I'm called om myw war.com I don't come with the name it's hilarious um where you can go move assets between different subnets um but we're targeting uh you know it's up for ACP review right now but the hope would be that at least on Fuji uh which is our test net that it would be activated by the end of the year so people could actually play around with it so very soon is the answer uh to that and and then if Fuji goes well as a test net then main net could be a few weeks later if the community decides to go with that

nice okay so that okay so warp solves subnet subnet communication um what about hyp SDK my understanding of H hyp SDK is it's basically like the first of many future Frameworks that will help make it easier and faster to basically launch your own BL blockchain on a subnet I think or launch your own subnet you do you need a job that's better description I've probably been giving people uh yeah um like so the hyp SDK yeah I mean long story short you want to create your own blockchain and you like most people these days you mentioned these like nextg chains that are targeting High throughput you know you want high throughput but you want it to be your own chain you want to use your own token

or whatever and um the most common subnet if not the only subnet people have deployed on avalan so far is evm based and there are just certain decisions made there that are not competitive with something that's like the super high throughput and so at that point what do we tell people either like H Avalanche isn't the one for you or like we don't believe high throughput is possible or whatever right which is none of those are true and so it's just clear that there needs to be some sort of avalanche oriented and avalan optimized framework for building chains and that's the hyp SDK so uh that's what I spend most of my time on is is getting that to a place where we think it's really competitive with with other of these so if your decks or a game or something like that instead of focusing on like uh

you know all this all the complexities of basically doing this you can get you know 500 lines of code set up your own set up your own blockchain basically so the idea that we also yeah exactly so the idea that we also are trying to really drive home is um and we've learned this now that we like have like three chains on the on the main Network or the primary Network we call it is the maintenance of like having like of these virtual machines is massive and you can talk to anybody about like just look at the version spread of people that have like zones or chains or whatever like there are some people that like once it's up like don't fix it if it ain't broke sort of situation right like and so the hyp SDK wants to give you a lot out of the box but make it easy to you

know keep up to date with changes and so only exposes things that it feels like you really need to manage and that's the whole point of the multi framework which is to say hyp SDK is an opinionated way to do this on adlan is it the only way no like we've seen other people building on top of subnets like movement for example where they have like a move dialect as part of the framework for launching a chain and like some people just want that they just want to use move right but like it it makes move and aptose derivative specific ways to scale and the hypers SDK takes a very different approach so but it's the idea is to give people a framework for scaling to me it's like uh it's not just the the complexity of maintaining and

building your own stuff uh like for example building your own rollup and you know whatever happens in the future and M maintaining it and so on it's also just uh you also need like to update with the latest in optimizations and everything which can be a nightmare it's a really it's not something that most people will even want to do instead you just kind of borrow what the best engine is out there let the development cycle make it faster and faster over time and otherwise just focus on building your own thing it's the equivalent of you know if you have a web uh if you're building a web website some web app um but instead of just all just worrying about building the web app you also worry about building your own web browser it's impossible it's just not something that anybody would would ever

want to be in a position of the web browser is Avalanche you worry on the web app which is what everybody else would want to worry about yeah uh zo zooming up from the technicals for a second Kevin I have a question for you about just like applications and like working with other Founders and Avalanche what's been the um like hardest part about getting people to build on Avalanche that you maybe didn't expect so I actually uh probably not the the best person to ask this I I don't deal a lot with this kind of stuff I Mo most of these days it's just technical things um I think frankly a lot of the you know a a lot of the difficulty right now is just on these last few things that are still left in order for the full Avalanche framework to be done we

still don't have for example C chain based um uh uh tokens available as native tokens for other subnets which is something that is that is an important one uh there's a couple of things for awm that are still not done uh ACP 13 has not been pushed yet um so this Vision that I mentioned at the very beginning of the of the podcast is something that is probably going to materialize within about a year uh but it's still not there yet and it's a big pain Point um but otherwise you know there's also a lot of noise around what others are building and you have to differentiate yourself yeah but I think you know that's kind of ultimately pretty easy I think once you know the the next set of performance tests come

in and we show that like you know you can get 250 millisecond on the C chain type of thing with Avalanche and blow the second best competitor out the water as well uh I think it will just speak for itself so uh yeah it's just a lot of Hardcore engineering work this one what um good good infra is usually informed by good applications like what have what uh yeah how how's that impacted like the Avalanche Vision uh really good question so the very biggest uh the very first thing is you know I want to run an application or whatever and I want to be making sure that like know nobody else is competing with me um I just don't want this to be

on an open permissionless chain in the same way that for example the C chain is that informs a lot of institutional applications as well this does not mean they're necessarily permissioned they just don't want the the uh C chain style permissionless they want more control over things uh it does not mean that it's closed either it's fully open you can work with it you can play with it uh but that's a huge one uh and and so that's precisely actually the one solution that subnets provided no nobody else provides um there is a lot of of course uh feedback around fees uh and fee isolation being a very important one salana does actually not do fee isolation despite them saying they do

fee isolation they just divide things between four cores but they're not actually dividing any uh smart contracts on their own little environment that's something that we're working on um that's very important with hyperdimensional fees we're designing sdks with hyp SDK to literally allow you to provide or to specify uh uh uh fees per operation code so like wres for example much more expensive than than reads so very fine-tuning and that's feedback that we've gotten so uh the ability to manage the environments better the ability to to uh uh specify fees more dynamically and more precisely um speed always comes down to speed at the end of the day so hyper optimize on being the number one in terms of speed time to finality um the number one in

terms of uh uh you know overall TPS although TPS is largely a function of um Hardware as well um but I don't see a reason why Avalanche can just also boost its Hardware requirements up but uh yeah that's just a hardware thing so first off I think the community's been great in G like anytime you like have a whole new community or like a new project right like there's definitely some people that are you know the the Pioneers right exploring exploring it for the first time and I think like dfk and Kata for example were two that really explored subnets for the first time and made them a lot better for you know everybody I think my main takea away from this has been a lot of people are targeting a very specific niche of the crypto design space which is massive

and largely untouched because I think a lot of people are just familiar with that so like you know bundled sequencing or bundled replication sequence sequencing execution like you know very specific paradigms for like running code on chain um and I think like the feedback we've gotten from Builders and I think salana and some of these other ecosystems would equ the same sentiment is some of the builders coming in are like we have no idea what you're talking about and could literally care less how everyone else has done it up to this point like the things that we care about are this you know I want to be on a blockchain because it provides sovereignty to the people in the community that I'm with and I can do things that I couldn't do by having this like shared State and shared computer and when you look at like the design of

infrastructure from that higher level perspective it really makes you question like some of the core ideas that you know Bitcoin brought on the table but like May no longer be important to some of the builders especially if you can launch your own chain or launch your own Community where you can really play around with some of the TR tradeoffs there um as you mentioned earlier so I think that's that's thing that at least excites me most about the space is like a lot of the community people aren't saying like I just want to do the same thing as everybody else like give me another one right like it's it's a lot of you know you're telling me all these things are constraints but like you're basing it off of this assumption that you know I really want to mirror what a lot of other people are

doing and frankly this is what my application is especially large Enterprises and I know you're saying that this is how it's always been done but like that's just not good enough or just not something I'm interested in like I'm willing to do this instead if if that means that I can have my shiny like higher throughput or you know more flexible Computing surface and so U yeah it's been it's been great to engage with a lot of really great great engineers in the space uh it feels like Avalanche is I'm Switching gears here Avalanche is winning the institutional game and I think part of that is due to I mean you guys have a really good team like Morgan and John and obviously all the way to the top like you know John John two John's over there and Morgan obviously are doing a great job um but what about

on the technical side why do you guys think that folks like whether it's a KKR or JP Morgan like why are people choosing Avalanche um just goes back right to what Patrick said that you know if a KKR and so on comes you know with their developer team and you know where can I develop and then somebody says well you should develop on ethereum because of this and that they're just going to blink blink they're going to be like what are you talking about like I'm GNA need much more customizability on all this kind of stuff I don't care about what the narratives are I want customizability I want ability to manage these things a lot more you know more a lot more fine-tuned way sness provide you basically exactly that uh that's what it effectively comes down to you can do basically almost any of the

applications that are possible or that are uh U that come along with the institutional side with anything but subnets so um I mean I don't know if if Cosmos could do that as well I don't know if it's mature enough at this point but yeah that's what it comes down to I said in I said in like round tables with some of the technical contributors from like some of the larger like Enterprises and institutions right they they obviously want to understand the technology they're playing with and they bring up conversations about like you know what are the different options here um and when it comes down to it like whether or not the crypto Community wants to accept it or not like some of their needs are very different than what I think we hope their needs are um so like from the outside looking in

it's like oh they're adopting crypto and like really leading in because they want to be more decentralized that's actually very much for them not a consideration or even something that they're looking to do um they're very interested in the way that it allows them to better communicate and interoperate with partners and like other people with even within their own company right like you have massive lines of businesses in many of these multinationals that like really operate as separate companies and and they want to find better ways to interact or offer more value to their customers but they often feel like their customers aren't asking like hey can we like put you know Bank large Bank a like and just tear down all the walls and like it like

that's not their intent and so I think a lot of avalanche's architecture imposes like very little con like belief or constraint on like what you're doing like if you really want to run it with like a subnet with 10 noes in your own data center but then you have interoperability with other subnets you can do that you can even run it privately right where like you other people can't even connect to your subnet they don't even know what virtual machine you're running like that's all possible and for a lot of these Enterprises they're interested in moving on chain but at a speed they're comfortable with and they're interested in like this public private tradeoff which subat again let you do via this warp thing where there is no proving on

the warp message for example of the state of another chain or something like that so they don't actually have to communicate that state externally because they very much may not want to but it they can still issue messages externally and if you trust that that subnet with what they're doing um it's totally fine so my my ex my all I'm saying is I think the needs of some of these Enterprises and financial institutions don't map to our expectation of what they are maybe someday they will but right now they don't and so they see value in the technology and Avalanche I think because of its flexibility allows you to use the technology in the way that they want to whether that's the way everyone wants to probably not but in some cases for

Enterprises they just want very different things yeah this is super key um and and maybe it'll be clear in a couple years but maybe it seems uncontr controversial right now but you gota that the you just have to design for things that are as unopinionated as possible and that's effectively what avalan is trying to optimize that was the word I was looking for unopinionated yeah it does not try to opinionate on no no no JP Morgan you should do it on an L2 because it listen it shares the security of of the the what no like they know what they need they know what they want they want to

you know tokenize they want to be a part of the network they want to provide liquidity for their secondary markets whatever but they want to do it on their own terms so if you're building a blockchain network you got to do it in a very unopinionated way you just got to make technology that is extremely fast Best in Class sdks that are extremely unopinionated but very powerful and just lets you spin up things and then otherwise you know do not get in the way you know I think Avalanche becomes with this OS which doesn't just just does not get in the way in any way like this blockchain OS basically where you just deploy things you run things and they just kind of spin up and work and otherwise developers get their full flexibility and they don't need to worry about future optimizations that just

kind of come for free by just being part of the network uh and they otherwise focus on on their restrictions and their applications that's really like why why this design is so compelling ultimately yeah how does everything that we've talked about in the last hour impact the value acral of avax like for for one like this something that comes to mind is we're talking about removing the necessity to validate back to the C chain um uh so right right now I think the way that it works is subnets that you have means more avax is locked uh because every validator needs to stake to validate uh this ties down Supply and I think I read that all fees on the primary Network get burned or at least some of them uh what the if if this new ACP gets put in place this would remove

this would actually probably hurt the value acral to avac but it actually makes it more attractive devel to developers which I'm assuming is a conversation you guys had internally um like how do you think about this um but Patrick I'll let you add by I'll just give it my quick opinion because I think this is one is a very subjective one and everybody has their own different opinion unlike you know engineering which is just make things efficient this as much about the narrative as it is about the technical yeah it's very subjective so um to me at least I actually um sounds maybe controversial but I don't necessarily think at all about that I just want to build infrastructure that just feels like like basically you don't even have to worry ever about blockchain uh scaling it just

kind of works um and you know yes you do as you remove things because you don't want to get in the way for developers you do remove uh requirements for aox I don't necessarily think you know we're just trying to build technology here that will scale up so easily for the future I don't think any of these other uh uh things are necessarily to me at least um worth like concerning yourself with um you know there's still use cases for a Vox that are obviously important it is uh a central hub for the pchain it is a central hub for the C chain it's a central Hub probably for a lot of uh when I say h i mean like the token itself is a central liquidity provider

for a lot of different applications but so will usdc so will usdt potentially um I just think that if you are an engineer and you're thinking about value acral it's probably almost certainly a recipe to design a system that will hurt developers and will hurt usage down the line I as Engineers it's probably our number one priority priority to just optimize the blockchains make them really fast don't get in a way build things that are just incredibly easy to adopt and then let the things afterwards uh uh kind of play out so to me I don't care about that kind of stuff sounds very controversial but I honestly don't I care just about building really fast

blockchains Patrick I don't know what what you think about that yeah I mean again this is this is my opinion I think on on this one but uh you know Avalanche is subnets like I don't think the intent with Avalanche from the community or anyone is to have like one chain or like one monolithic chain and Avalanche I think buil grows in adoption the more powerful subnets can be now people may say like oh you know the staking ratio could go down if you don't have to stake U you know to launch a subnet um I look at it very differently which is to say that the primary Network for many people is the uh jumping in point to Avalanche from externally as well as like the

point of Providence or liquidity Hub that everyone shares now that liquidity Hub and point of providence is only interesting and only interesting to secure if there are like really novel and cool use cases actually interacting with it right like why would you bridge anything to Avalanche if it's the same thing that you could just get elsewhere like it doesn't it has never really made sense to me that that's like the primary motivation and so what I think provides a lot of interest and like adoption to the primary network is this like really expressive world around it that is subnets um and so I think in the path to 10,000 subnets or that article I wrote I talked about uh that last item which is

the augmented security which is also saying you know great you know there's been this historic relationship with subnets and the primary network is it the best one I don't know people have a lot of different opinions but still makes sense to have some deep relationship between the two if we can um and that's where this approach comes in which is to say if you're a new subnet like a new permissionless subnet and you know I'm I don't know maybe it's Patrick coin right like probably going to be a small Network at first right but like if I have a lot of value on it hopefully I can drive the security budget of that Network right we're designing again for these Sovereign ecosystems where you're not borrowing security from the from the from the primary Network so what can you do well one option instead is to to almost flip

the the uh the story which is to say instead of subnet validators actually doing something with security you can have anyone in the community augment the security of a particular validator and say hey I'm going to pledge Security in this case Avalanche tokens to the subnet validator and if they like equivocate or have some issue um some of that like my Avalanche will be slashed or taken away um but in the case that they don't I'm actually rewarded with the subnet uh token as the reward rather than Avalanche and so in this world you know you have this really nice relationship between subnets in the primary Network that seem to be really like a a productive

relationship that you couldn't get on your own which is to say I want to be a subnet that can hold substantial value but I can't do that if my security budget isn't high enough and hey I'm on the primary Network maybe I'm staky maybe I'm not but I want to use avalan differently on the primary Network while I can support support the growth of subnets which then you know increases the value of the liquidity Hub that is the C chain so um I think that my takeaway here is to say the 2,000 ofox that's locked to become a subnet validator there's probably a better way to use that that has the similar properties of like locky and if people are interested in that that really makes subnets better and I think that's probably the way I

look at it yeah if I had to if I had to push you guys on the avax avax I still don't know don't know how to pronounce it um like value a cruel question like if you look at uh you know I think a lot of people compare kind of avalanche and salana and look at those two and they're kind of make deciding between those and which one to you know make a trade on obviously none of this is financial advice but you you're you're seeing something like salana run up from you know 10 or 15 bucks to 60 and I think you avax has run run up from like 10 to 20 like how do you think about that internally or how should other people be thinking about that um I I personally don't honestly at all um it sounds very boring of an answer

but I I really don't I think uh nobody can talk about any of these kinds of token prices like predictive in any way they're kind of silly so I don't concern myself with that kind of stuff personally cool um at the end of the day you know I just want to build technology that does not get in the way uh and and you know I would I would any day of the week uh take a blockchain that is extremely popular and technology that we've built being extremely popular over anything else whether or not it has any value of cruel so that's my that's my very boring answer Patrick you can add yours have a feeling the same I'm not gonna comment too deeply on

it I mean I think that the you I've been in crypto for quite some time and I think I'd go mad a few times over if I tried to like EXP explain all of that I mean the way I look at it is we're trying I think everyone in the space is just trying to get people on Shane um you know I think that's a great cause I think there's a lot of benefits to society from that um you know what comes from the value of cruel of those mechanisms I think is just a reflection more so of you know when you're coming on chain you know where is the interesting stuff happening yeah it's pretty much it and I think um salana uh you know because of their like feet topology like I think makes them very nuanced in the space right now relative

to a lot of the evm chains because a lot of the narrative there is you know we want decentralized blockchains that uh in their in their terminology decentralized in terms of like the average compute cost or resources that any validator has to use and then um as a result we're constrained to some sort of throughput and there's someone now that is pitching a very different story which is like let's have many more resources and just do it differently and it turns out people are interested in that you so to me it's more a reflection generally speaking of I'll just talk about popularity not value of anything and to say like there are many areas of interest in this crypto design space and over the last 10 years we've really

stuck to one and I love the explosion of like looking at different parts of it and people just seem to get really excited when we've discovered a new part of the crypto design space I do too right like it's cool right like oh I didn't know you could do that on chain like people freaked out about friendtech when they turned out when you realized you could connect social to like some of this social trading right people like get euphoric when they're like wow like that's something you could never do before same thing with flash loans and so I think when it gets down to the layer one layer of wow like this is a totally different tradeoff and it lets me do things faster cheaper like interact in different ways it just gets people excited and so I think with teleporter and some of the subnet stuff

coming um you know I think that will tap into different areas of the crypto design space as well and I think that developers will find that interesting that's probably all I guess as we wrap this up Kevin and Patrick anything that you guys think is really important to think about about aval laabs and Avalanche going into 2024 that that we didn't hit on um I mean I think again the most we're in a very exciting time I will say in blockchain engineering um I think the narrative ative that which was wrong completely wrong narrative SL assumption whatever that um uh you know the blockchains you

know that are very high performant are centralized is completely wrong we're moving away from that thankfully salana is and a lot of that narrative counter attack uh in a lot of ways that's good uh so that means now that there is a really exciting uh new kind of like wave of hyper optim ization while keeping everything actually very decentralized which was always possible and always the case so I think over the next year I'm personally just so incredibly excited about like actually getting to frish all of the stuff on the engineering side um you know whatever happens with the crypto markets honestly you know you'd

go mad to think about any of these things it does not matter I excited about just the applications and the engineering I'm excited about social f excited about uh everything that's coming on the institutional side I'm excited about gaming I'm excited about all these different applications I'm excited to see these things play out in really interesting new ways we've never seen before um that's not a comment on necessarily Avalanche it's just more of a comment in General on the cryptospace um really exciting to see let's see what happens there um obviously I've always had the same thesis that you know when you allow the programming of value uh in the same way that you allow the programming of information the sharing of information like the internet did

that thesis has still to play out at incredibly large scale but when it does it's going to be really exciting uh I think we are at the edge of some really interesting applications at at Mass um and so I'm really excited about that just honestly really really excited about adoption so I know Patrick if you want to add anything yeah I mean I'll say that a number of people have asked uh particularly about the PR DK like what makes it so different and special when we're talking about scaling chains and so I've been working on a long form uh piece about more of that uh because that's very fair question is like you know you have all these other like high throughput chains like what are you guys doing Special that makes it even that stand out in this really

competitive space so just working on that we didn't talk much about that but that's I'd say more on the super more on the technical side and so we'll uh once I share that maybe we'll chat again but uh yeah I think that's been our the the the one thing that I would like to uh to the space to uh to view uh as like more of an Apples to Apples comparison is which one is the fastest to confirmation per core of Hardware or like per I don't know gigabyte of of Hardware whatever it is like per the same kind of Hardware which one uh uh which implementation will actually be the fastest I want to get to number I would love to see AV get to number one

um and uh I'm really really excited about that it's going to be a really exciting time because I think we're just seeing now that like you know it's time for some serious optimizations in the in these blockchains yeah Kevin Patrick thank you guys great conversation really appreciate thank you really appreciate it thank you yeah thanks for having us yeah hey everyone thank you so much for watching today's episode really hope you enjoyed it we wanted to take a second to just remind you about our upcoming digital asset Summit in London March 18th to 20th Santi and I got your back seats are limited and we hooked you up with a 20% off discount code it is Empire 20 if you heard it earlier in the podcast there's a little competition running at blockworks to see who can drive the most number of tickets so when

you register for the digital asset Summit make sure you use our code Empire 20 see you in [Music] London