Commonware: Building the Anti-Framework | Patrick O’Grady

Transcript

and what we talk about instead is the meaning of like what your application is bringing to people and like why it's a great product we don't talk about like Netflix's CDN architecture we just don't we just use Netflix because it's great the same sort of thing will happen with crypto and we need to move past like the TPS era and like all this fighting about infrastructure things and instead just as soon as it's readily available by anybody I think that will help us move past that because like the differentiator of an ecosystem being like some sort of particular primitive or like some sort of particular scaling technology no longer be becomes the cell the cell is like what that ecosystem together can do together hey everyone this episode is

brought to you by uniswap Labs who is launching the largest bug Bounty program in the entire history of crypto it's a $15.5 million program this is aiming to be the most rigorously reviewed codebase ever deployed to ethereum so if you want more eligibility on the scope rewards payouts Etc just click the link in the bottom of the show notes tell them that I sent you you'll be hearing all about Unis swap Labs later in the program hey everyone quick disclaimer before we get into today's episode nothing said on bell curve is a recommendation to buy or sell Securities or tokens this podcast is for informational purposes only and the views expressed by anyone on the show are solely our opinions not Financial advice our guest and I may hold positions in the companies funds or projects

discussed all right everyone welcome back to another episode of bell curve today I'm joining joined by Patrick ogrady the founder at commor Patrick welcome thanks for having me long time fan of the show excited excited to make an appearance here oh thanks man also a longtime fan uh lurker on many of your videos so excited to uh chat live with you here about your new project which we're gonna dig very deeply into and I'm super pumped about that um before we get into the weeds I just got to do a quick shout out to our sponsor a um many of you know them they are the largest Borland protocol in uh in crypto they've got 20 billion in tbll extremely trusted team just so you use a you know the deal click the link in the show notes um Patrick uh we got a lot of ground to

cover here and I'm sure many folks know you in the crypto ecosystem from your time at aval laabs um where you where the CTO and ran engineering and built it up into a tech Powerhouse um and now you have a really exciting new Venture uh in the form of common Weare here so you know before we get into uh and you know the intention of this episode is to explore you know the vision and some of the uh early Tech Tech architecture of what you're building but can you just kind of give us the backstory here you know what have you been working on in crypto you know what was the initial Insight or spark that led to this new Endeavor yeah I uh um you know I've been fortunate to be in crypto for for some time now like uh you know I don't I've

still still young but uh my entire career so far has been focused on integrating with blockchains like building blockchains or helping people build on blockchains um I'm originally from Wisconsin and unlike Olaf off I I didn't hear about Bitcoin in the midwest as much as I wish I would have I guess I was in you know the Wisconsin Echo chamber there um but I heard about it when I first came out to school in California and and have really been focused on it uh ever since so uh after school I I uh was I didn't have a strong opinion yet on like exactly what blockchains would look like or how they would really uh change the internet and so I decided where about where better to go than coinbase so uh worked at coinbase on the crypto engineering team there um and our job was really like how

do we figure out how to as quick as we can and reliably as we can integrate different ecosystems um it was a great role to be in because we got to meet a bunch of like great teams working on different stuff so I worked on like the ALR Cosmos polka dots salana Integrations and so we got to meet with their teams talk about the things that they were worried about things that they were excited about and figure out how co coinbase could really support them and that's when I really started to think uh more about like uh blockchain distractions and so while I was there I created a project called Rosetta uh which is what coinbase uses now to integrate different chains um and so the idea was can we create like a uh a universal like read write layer uh that coinbase could just uh basically

automatically test and automatically integrate with and um for a lot of people like coinbase integration was always like having the coinbase team do something and so the idea if we had an open spec was that we could turn that problem back to the um actual like teams themselves and they could integrate everything that they wanted coinbase to do um so through all that um started to get really excited about a few ideas that uh coincidentally led to going to aval Labs so uh around that time like any good conversations start Kevin reached out on Twitter to see if I had heard of avalanche um and I had there were a few people at coinbase that were super super excited about it and spoke super highly about uh goon and so after

a few conversations there you know really made sense to me and so I jumped in and so super thankful for uh the chance goon and Kevin gave me to build the engineering org there they hired some initial folks before I got there that were absolutely incredible and it was a a great uh a great opportunity to to work with all of them um and so yeah for the last three and a half years or so I was the head of engineering at aval laabs so uh what that meant was if something uh went wrong I got the page I guess and so uh it was a uh really captivating experience we'll call it but uh uh and fortunately we we uh we had a lot of fun and I think uh I think the record speaks for itself in

terms of some of the cool stuff that we were able to do there um so yeah but uh earlier this year uh in July or so I left uh aval laabs and decided to start a new project uh comp so that's where uh we are today and uh been a little bit of a you know like the proverbial coding hole for the last six months or so and so excited to hop on the on the podcast to tell people a little bit about what we've been up to so that's the brief backstory I'm super excited as well and you know we were chatting about this yeah before the podcast I I would actually love for you to I I think you know lots of people might not be as familiar with avalanche's tech which is extremely extremely strong um in no small part thanks to you and you know

I've always bucketed uh Cosmos and Avalanche as being relatively similar just sort of having this vision of different change that need some sort of common connective tissue there's an interop component there um there's like a standards component there and so could you just tell us a little bit about what was sort of the core stack and what what what are the components that make that up at avalan that you spent your time working on and then we can segue into how some of those Concepts um you know led to Common Weare yes so I mean I think of um I always thought about Cosmos and polka dot at least at this time so this was in like uh 2021 when I originally went over there um on basically like uh different ends

of the spectrum so you have Cosmos which is like everyone's on their own and then you have this like really elegant way for them to like different zones to interact with each other which is IBC um the whole idea is that zones are not like validators of other zones instead what they do is they're like clients so they there's no like o there's no overlapping validator sets instead you just verify headers and from that like the state proof in there you can just derive like what trans trans actions were coming to you super cool on the polka dot side um it was a totally different approach which is that you have like a very tightly integrated parach chain ecosystem uh that was orchestrated entirely by the relay chain and it was really like one Mega chain that gave you like people the ability to deploy these um like chain modules or

what they call like pair chains like I said uh and that was very tightly orchestrated by polka dot and like you had to uh at least at that time like lease a par chain slot to be a part of it like uh I think at that time at the beginning it was like a couple million dollars I think it's come down since then but um different approaches Avalanche I felt like occupied a really interesting space in the middle which was that you had a lot of chains um and the idea was that there would be many chains for different purposes doing all sorts of different stuff um still having the ability to like deploy as you saw fit so like deploying the subnet was not something that took like a big investment it just you could launch it um but still there was a instead of a um

point-to-point um like Channel created instead you had like an overarching interoperability function provided by the L1 itself and that there was like an l0o or something like that kind of making the interupt cheaper um so the way I've defined it or described it before is um Avalanche optimized for uh low variable or like low marginal message cost so that like you could have pointto Point messaging between all the different subnets or what they call Now call l1s uh without having to have the overhead of maintaining point-to-point connections so the idea is that with everything on the pchain is long as you validated the pchain you could verify messages from any of the other subnets um but because the uh like the validation and

everything was andse security came directly from the L1 rather than Avalanche uh it was a lot more scalable to have like many many more subnets so I thought it it served like an interesting middle ground and then Avalanche consensus if you're going to have like some sort of interupt thing that like underpins all this you want a lot of validators not a little bit of validators and so aval Anders consensus I think really opened up the door to having like a multi uh validator set to like uh SEC like basically facilitate the peachin operation between all these different uh subnets so yeah they have a huge upgrade coming out I think next week so super excited for them and a lot of the stuff we've been doing the last few years is finally uh coming out the door so I think massive shout out particularly to Michael Stephen I think

at that team who have been really pushing it so yeah so talk to me about you know you were very Hands-On not only in terms of the design um and are pretty opinionated about the design principles uh but also you work with a lot of folks as you said that were deploying you know onto either their own subnet or you know maybe the C chain at some point or whatever it was so talk to us about some of the the design principles that started to stick out to you and I feel like you know there was a niche that you saw in the market or you got pretty opinionated about this new group of builders that were building in a different way so talk to me about that yes I mean I think um anyone on like coming into the space will tell you like uh the crypto space

is very ecosystem driven unlike many other you're kidding unlike many other uh different technical vertical right like you don't have a religion around postgres or MySQL although some of the people may tell you that they are um and so I think uh early on in like the cycles of development of like an ecosystem system it's good to have like people that are super super passionate and the idea that like value is now like directly attached to that ecosystem I think clearly explains why that happens um that sort of connection between like basically what was product and like open software and ecosystem is what uh bothered me so um when I started common wear or like the the the basically the

the motivation originally is that I just wanted to commoditize state-of-the-art Performance for people building blockchains like you shouldn't have to join an ecosystem to get access to great stuff so like you want like to do a 50,000 TPS whatever um I don't want to have to come into your world and say like powered by Patrick you know whatever like it's it's just software like you're running software and I don't think we as an industry had taken that step yet so like if you're for people this is before my time so I can't say I remember it U but the early internet like there were all these branded internet like it was at net and like SP SPC you know whatever um and the idea was that you could capture all of it and

you could capture all of it by providing like all this great stuff that was just for you and because our stuff was so better like everyone would come to you um I think this approach just like with the early internet is holding back like a lot more developers from coming into the space and so the approach with common Weare which is why I put it in the name uh basically to make sure that we did we never deviated from that original goal was that we would it was all about democratizing just great performance uh to anyone that wanted it and so what we often saw or at least what I saw uh at different points in time was a lot of teams making ecosystem decisions before technical decisions um and I think that uh when you have

technology more readily accessible to people without any direct affiliation to like a particular ecosystem um you let people really push the envelope and new way and then from there join ecosystems that are intellectually aligned with them rather than just have the the best tech um and so I think that that really has been the the driving goal for cinare and why when I wanted to start working on stuff like this it didn't make sense to do it somewhere else or join somewhere else that would like sponsor it but instead uh create my own thing and so uh to answer your question directly what I saw was people wanted more and so any any time there was a chance of more they would switch

ecosystems or like move somewhere else to get that more um usually conveniently associated with like some sort of Grant or something like that but generally speaking um as soon as someone had something better why wouldn't you use it uh the problem is that it was directly connected to the ecosystem and I think it's it's it's a much more like sustainable option to delineate those two things where you have open Tech that spans ecosystems it has no particular goals for a particular ecosystem and can be used by like you know commware could be used by people in avalanche just as as much as it could be used by people building abss um and I think that that's important because it it lets the communities really form around their principles that matter to them rather

than just having access to some particular piece of piece of code um and so I think uh I think this is what really helps to push us forward as a space with like getting better developers in the space building better things um and so you know I thought about actually like originally starting common Weare as a nonprofit like to just like write software that was open or whatever like that um but I found that or I felt that it would be a lot more sustainable as a healthy business I think um in the early days people uh with Linux at least sometimes forget that Linux was greatly accelerated by Red Hat um helping to like ship Linux and bring it to the

Enterprise and like polish it up a bit um I think common wear serves conserve a similar function really helping solve direct uh developer problems with uh a lot of different products we'll offer um over time so that's the that was the original gist I guess that like really took me from listening to builders saying like I want to build my application and then having to pick an ecosystem and then picking a technology rather than thinking about what technology made the most sense for them and then from there figuring out how to deploy that within a particular security configuration provided by some ecosystem yeah so Patrick maybe this is a good time to you know we've alluded to it a little bit and maybe given listeners a

sense of what the shape of common Weare is but yeah could you just give us a you know give us a formal definition here like what is common wear actually maybe we can introduce this this concept that you're debuting of an exost stack yeah yeah so great transition there keeping me ey um yeah we um I the the idea here is that like how do you actually instantiate that like so could we just ship like a bunch of random things and give it to people and say like yeah it's like state-ofthe-art we're just going to keep shipping more um and they're not really connected to each other or whatever uh you could go that approach but we wanted to take a deeper look um at how people actually build an instantiate like blockchain applications to try and create um some sort of Symphony of parts I think is a good way

to put it uh that brings that's more than just like random parts that uh these things are meant to play well together but can still be tuned based on your application and optimized based on the needs of your users so the leading theory is that through this all we want to do more on chain I think a lot of people want to do more on chain um and we think that the way that that's going to happen and like the way dominant applications of the next decade are going to be created are going to be born from obsessive specialization so um you can think about it like you know when people first built shopping carts online like it was a pretty you know primitive Endeavor like they launch stuff um everyone's shopping cart looked the same uh now you fast

forward and like Amazon shopping cart is a very different construction than like Facebooks than whatever and the idea that people can meticulously optimize what they're building just for you is not quite something that today's Stacks uh give you um and so where does that come from well let's just talk about a few basic examples of things you might want to build today um and we can talk about how that might be possible or not so let's say that you wanted to just create some applications secured by an NS well you have to deal with this problem of having like a validator set that is external to the chain itself right the whole point is that the chain like the validators are defined

somewhere else now the stacks today that was not what they were designed for right the like Cosmos was designed to have everyone have their own chain have their own validator Set uh at different points of in time there was this ideas where like you kind of secure each other uh but the idea that like the validators that actually drove the chain like you stake on the chain from that you derive who the validators are and those are the people that run consensus so what people have started to do to solve this problem is Fork away from like Comet bft and introduce like mechanisms to basically pull validators from outside of the chain in and then uh export faults and like safety issues back to the ABS to

slash those people if that didn't sound like a complicated change uh it should like that is a very different way to build uh next like let's say you wanted to develop like some sort of like social network on chain um and you wanted to have huge blocks you know we're talking gigabytes of data or like however you wanted to distribute that through the network well most chains or like blockchain Stacks are meant are based on like the original approach like Bitcoin and ethereum where you have block is built block is sent block is verified next block is built block it doesn't really work to give like the performance users expect in that model like you

probably want to like pre- distribute data connect it like via some interesting like signature scheme into and like include artifacts from that like that have been distributed into the blocks rather than Distributing like a multi- gigabyte block outright I wrote a lot about this in a project called Vicks which is talking about decoupled State machine replication those changes are not just like oh slap on like a another config and enable like decoupled State machine replication it is a deep rethink in how these pieces fit together you know let's say you're building an exchange and you want to offer like dedicated meol access to market makers seems like a reasonable thing uh most time blockchain Stacks do not give you access to the M Pool so you

want to like change how the M Pool Works you want to do whatever not an option unless you Fork it um so like for example let's say that uh you were building foraster you were building like a Dux and then you were building uh ethereum and now I'm going to tell you that like yeah you you all need the same Med pool just not the same like the right construction and then like let's say you're not trying to build a payment app um well most times people that are building payment apps like really focus on both throughput as well as easy verification so you also have like want to use cryptography that's on your phone uh not like some uh cryptography scheme that was originally described in Bitcoin or something like that and so what

people end up doing or what you want to do is like change the block format or like change the transaction format again not allowed so like what you end up having to do is just pull together out all like the little things you want to do take what is the best in class or like the best available thing and then ship a bunch of your stuff into it now why this isn't great is just like this is a huge burden for people that are building these apps first to learn exactly how the thing that they're forking works and then second to maintain this across changes to that underline thing um it takes a huge amount of time and is an extremely expensive thing and then you also often

need to hire someone that knows how that original thing works to make sure you don't like break it right because like the whole point in the terrifying thing with crypto is that it crashes or like there's a bug that lets you steal funds or whatever um and so you can't just like ship a website and say oh like that sign in button's broken like we're going to go ship another version you end up takes forever because you need to get it like security audited you need to like test it extensively to make sure it doesn't break and um what this all leads to is a question of like what are we doing here yeah so what happened I guess and I think this is the line of reasoning I've at least followed into like where things went so

Bitcoin first real blockchain that really obviously came on the scene here I mean it did one thing it is a like peer-to-peer Cash System it is for transfers it has become more since but the original idea was transfers that's it now people got so excited about Bitcoin they were like how can we reduce the cost and complexity of letting PE other people create their own like Bitcoin thing right it's not that they wanted to just clone Bitcoin although a lot of people originally did that um the idea was like can we decentralize more applications more than just transfers and this was like all the excitement that originally came in like the 2015 era of like well what if we have tokens

what if we have a computer what if we have whatever and so uh reasonably so looking at what was out there a lot of people said well like in modern Computing today we always build on a virtual machine we always build on a kernel you know when you run Excel on your computer you don't run Excel OS you run Mac OS that supports Excel and the same idea was well we don't need an operating system for every particular thing we can expose like a clear set of Primitives or like system calls that people can use and then um as long as that stays ahead of where application needs are like it has in operating systems you know we're good we don't have to worry about this it becomes a

lot easier for people over the last 18 months or so uh it has become clearer that like that abstraction is not keeping ahead of the needs of developers and what has become the case is that increasingly like these the best Builders are spending their time working around the abstractions rather than using uh those abstractions so to be specific here so I think most people will agree uh like common bft and Cosmos I think really like set the tone for like the entire idea of like application chains and like how you could actually have like your own chains and then not actually have to rebuild everything from scratch so according to like their docs the original ABCI had 11

methods over all of networking mempool consensus State transaction verification sync ABCI Plus+ introduce five new ones to make it possible to like intercept some consensus stuff now when you're getting started having just 11 to 16 methods to like build on that's great right like I don't even know how this [ __ ] works like the whole Magic of it was I could just take an application and it would work right like I I could just like Define how transactions like change whatever the whatever I want to do and that's it um however once you actually deploy something the problem changes entirely you go from like how can I get

this thing up to how can I make sure or like how can I turn my application into something that is leading like either winning this vertical or like just absolutely delivering an experience you wouldn't believe is possible going from that model to that ends up forking or like you end up having to build like different sort of like Contraptions next to the chain to augment what you want to do um so I think the learning here is that what takes you from zero to one is not what takes you from 1 to 100 and what common Weare is doing is really dedicated for the people that are going from 1 to 100 like I'm not suggesting that what we're doing you want to pick it up as your first project here it's really for people that have like seen what it looks

like maybe deployed a few things and now want to turn it into something new and the last thing I'll say on this point and then I'll get into a little bit more about like exactly what we're doing here the root problem in my view is that this abstraction at least in the world of operating systems was based on the idea that you had more or less commodity pieces so like CPU RAM and network and then dis usually are like the core things now unlike in crypto those Primitives like mempool consensus you know Tree store like all these different things they're not as commodity like you want to change them in tune like they're very different based on what you want to do

and so the abstraction itself is just an abstraction and I think that treating it as law I think leaves a lot on the table and so um the transition for us is we're building a new type of Stack so what you what we call the exost stack um and we think this is an architecture it's not like a branded product from us or something like that um but the core idea is that we build a bunch of state-of-the-art Primitives for specific things so whether that be like P2P whether that be storage networking cryptography you know broadcast uh some sort of execution thing you name it um the what we try to do is make sure that they play nice

together which is the whole idea of like where the stack comes from and then the EXO is the idea that you can take external things so like stuff that matters for your application and bring them in without having to Fork everything else that you're not touching and so the analogy would be instead of giving you a house and telling you like yeah you can put on the shutters you can like change the color of the paint you can like do a new roof like you know build a playet outside uh instead we're saying like here's a bunch of wood nails you know bricks build your house uh we have some blueprints that you can use like maybe you want you don't want to build your own sinks or like you don't

want to build whatever um will help you uh get off the ground but the idea is that you're pulling together your application from a bunch of pieces rather than taking something that is monolithically defined as like a single like virtual machine or kernel interface and then building where it where they say to build um so it's a different way to go about it that um that really the whole idea is to break away from the framework idea when a framework stops serving you maybe it's time to think about a different way to to pull it together so that was a lot I'll pause here and we can go into any any part of it sorry all right Patrick let me see if

I can let me see if if I can try to summarize uh try to summarize here so basically you know I really like this analogy that you gave me of uh bu Builders bumping up into the limits of abstractions and you know how I would Define as a non-technical person and abstraction is there's this assumption that Builders are going to want to use these very different parts even of something that's supposed to be a very modular stack like the cosmos stack um in specific ways and now that we've matured as an industry to the point where applications are relatively differential and especially when we need to push up against performance boundaries of what we've historically needed to do those abstractions no longer necessarily work and so you have to create you have to either do

something like you are forking uh something like the cosmos s ke which happens an enormous amount of time and in some cases I love the ABS example that you gave of actually the core Assumption of the validators being native to the chain and that's how you even identify the validators is completely 180 degrees different from something like an ABS where you're reh tooca or leveraging external validators and so you just end up with something that looks very rickety right and I I also like this idea of the virtual machine or the kernel being the core assumption that modern Computing is built on you know that leverages these sort of commoditized components and that's just very different in crypto can you help me actually understand you know before we get into um you know some of

the more specifics of common wear what exists today in this library of you know nails and Hammers and wood that you're going to end up giving people can you use toine uh because this is always a tricky concept for Less technical people what a virtual machine is what were the you know just just give us a quick explanation of like uh RAM dis Etc and then what are the crypto equivalents of those that common Weare is going to allow more flexibility around yeah yeah great question so most people have heard like VM as part of like the evm or like the svm or the movm or whatever uh it's become like a branding tool to describe a very particular way to build uh at least so the basic version of it

actually I'll start off with like what where it comes from which is that you um have a bunch of stuff like different privileged resources uh in your computer simple enough um you don't want every application to have to know how to interact with every possible combination of resources so for example um when I ship some sort of product or application uh let's say that I'm building I don't know like a uh an IOS app we even make it so simple as that um you know every iPhone is different like you have all sorts of different pieces all sorts of different stuff and so the idea is like what if whoever what if you could build your

application for like an abstract idea of what you might want to actually like use on a low level so I don't need to know what kind of hard drive or like Hardware is in my iPhone I know that there's a camera I know that there's a way to like store stuff and then leave it up to the operating system to figure out how to orchestrate store that thing into like the all the particular device drivers that are actually provided to you at that time so um people that like run Linux know how complicated this is because Linux uh you know it's so widely used that like sometimes you hook up some camera or like the the joke is always that like your speakers never work um because they have different

drivers or microphones and so um the abstraction is to say you are an application you should need to know how every possible computer is configured to develop for it and so it's a great idea it is a huge reason why computers are so successful is that we were able to successfully abstract the low-level parts from the concerns of the application developer now in the case of blockchain what that looks like in EV is that you have op codes you have op codes that like add stuff multiply stuff subtract you know whatever you also have ones that store stuff and read stuff I don't know how many there are I think it's like 30 or 40 or something like that different op codes that do different things and so when you're a developer in solidity you just write your program and then the evm actually

orchestrates those different things and that can be implemented all sorts of different ways across different things now there's that level of development then there's the level of development where you are creating your own blockchain using something like a cosmos SDK and the features that are abstracted are a little bit lower level right like you define your own execution stuff and then the things that are provided to you are things like um like uh sto storage access and like state so like put it in a tree you have like a me poool um you have consensus you have U uh like RPC stuff like these really what are like sub protocols or like sub things that live inside and the idea and why the thinking was that the virtual machine

and kernel World mapped was that yeah consensus is just agreement right so like just like dis is just like storing stuff we should be able to abtract this so you as an application don't need to know what exactly the consensus looks like just tell me what you want to agree on and we'll go do that similar with a men poool it's like well like do you really need to tell me what the transaction looks like if not I can just make sure the transaction gets around and I think I mean the idea is makes sense like I don't think I would push back on that I think what has happened though in practice is that um a lot can be pulled out by like putting very specific

application guided optimizations in place of some of those General things whereas in a computer you know you just put in a faster hard drive and it just all works right like you're not going to change the way you interact with it in in blockchain it's so quickly developing that you almost frequently are changing exactly how you interact with it and so the idea that we can encapsulate it um I think we have seen in practice just has run up against that edge now it's unclear if there's like some magical um construction that like will come out in a few years that will just mean that we actually can go back to a kernel abstraction um my belief is that blockchain development is going to look

a lot more like uh distributed systems development backend development than like running a local machine um and that instead there's always going to be different types of subservices and different ways to do things that are more optimal for your application and because people are so sensitive to fees and latency that we will always be pushing the envelope of like the little pieces that make up your blockchain rather than um trying to accept or like fit into some abstraction now again it's not to say you start there this is the prom mode approach which is you have got something up or you know how this works and now you know that whole state engine that like manages you know a particular

tree format for you you're like nah I want to do something else and that's the people that uh I think are a small really motivated really talented group of people that today don't really have much that makes their life easier it's a lot of taking what's out there deeply changing it and then hoping what comes out on the other side is coherent or like piecing these things together uh to make it work so like for example in the case of pbra which is a fantastic team like super great at cryptography and have done I always joke around on Twitter that's like they've come from the future and they're bringing back um like crazy ideas that you some of the

stuff that they can do is just wild and what they have to do is they run Comet bft as like a stand process and then they've reimplemented a bunch of things uh they don't use the cosmos SDK they've like implemented their own version of it in Rust that sits on top of the com bft construction and then interacts with it over grpc um I wouldn't say that that is the going to give you the best performance right like there's a lot going on there and so the idea that you can remove that connection remove the commet thing directly integrate like low L consensus things and then mod them with like all the different things that Puma's doing to me that's where we're going as an

industry we're not going back to here's this box build your little thing in it we'll take care of the rest uh I think there's too many smart people to believe that that's the the way forward I mean you've uh you know smart people I've worked with some some crazy folks and the idea that it's like this is it like don't go past this line like users want more but like it's just not how it works and they're like oh okay well I'll stop here like it's just not just not how they think um and so you're going to have those people pushing up against the boundaries and working around them doing all this stuff anyways so you might as well keep that in mind when you're actually designing some like the actual

architecture that they're using for building and give that to them directly rather than having them end up there was something that was never meant to do that all right everyone listen up like I said at the top of the episode Unis swap Labs is rolling out the red carpet and launching a $15.5 million bug Bounty this is the largest bug bounty in the entire history of crypto to anyone that can find a critical bug with the uniswap B4 protocol this bug Bounty aims to make it the most rigorously reviewed code base ever deployed to ethereum now the Unis swap name is basically synonymous with security $2.4 trillion dollar in cumulative volume has passed through Unis swap V2 and V3 without a single hack so Unis swap V4 is dedicated to

bringing that same level of security and transparency to the V4 protocol so if you want more information or eligibility or scope or Rewards or payout or any of that good stuff there is a link in the bottom of the show notes I highly recommend that you stop listening right now go to the show notes click the link head over and tell them that I sent you start earing money today so Patrick I think you know what what I want to get to here with you as well is how should listeners sort of categorize this project in their minds I think a lot of people have you know there are all these Buzz words in uh crypto and you know people like to just okay I I think I roughly get this right there there are these stacks and there are various levels of different abstraction so maybe

you have less flexibility if you use if you're building on something like the evm versus something which is a little bit more modularized and customizable than something like the cosmo stack but even the different elements of the cosmo stack and ABCI Plus+ how those different components interact with one another there are a lot of assumptions and almost training wheels that are baked into that as well so this feel like this very comprehensive library of components that you can just use to build your entire house application whatever you want just directly from scratch um but I think people are still going to be wondering okay I I feel like I vaguely got a notion of that is this an ecosystem should I be thinking of common wear like an L1 or is this a some sort of SDK maybe the business model here is

a little bit more of a um cloud services or or uh Services type model how people bucket what you're building in their minds yeah this so this is why I uh decide to keep the stack part in it is that people I think already have at least some of the builders have a predisposition of like okay stack is like a bunch of stuff I use to build something yeah um and so uh the idea is that you would use this you could use it to build your own integrated just to use some of the par LS people have used like your own integrated blockchain you could use it to like Stitch together other blockchains as part of your blockchains and more of like a modular thing like I said with an AVS like you could use it externally now this is not about um oh

like we're going to do everything for you like you know everyone told you like you know you have to build this way and now like with this you can do whatever you want the idea is that the pieces are broken up in a way where you can actually swap in things that matter for you so um it's not about like some super Mondo thing it's like it's a different way thing to pull together these these parts and so um you should think about it like a uh so the cosmos stack and like SDK and like the Comm bft plus Cosmos SDK um have a very particular way to build in mind which is that you like build on the ABCI or directly in Cosmos SDK it is it handles a lot of the complexity for you

you just work on like the execution part you can think of us as capable of doing the same things but with a very different approach to building where you pull together like every piece that you want to be part of your application and turn it into some sort of blockchain or some sort of application so you should put us in the same bucket as things like that um now in the short term we think that a lot of the people coming to us will be coming from like the world of avss that like want to build stuff that like is so different than like what tooling is available to them um but in the medium and long term uh we think that people will uh start to create cre applications that don't really feel like

the blockchains they know and so the idea that like we're just for building this particular type of blockchain that is like on a tone is not what we do instead it's about like here's a really good consensus module consens like a P2P module and storage thing and then you want to pull that together in some cool way for like your application great like we'll call it a blockchain today but like I think anyone that's used forecaster for example would say it doesn't really feel like a blockchain um and behind the scenes It's like got this really cool P top stuff and they like use um I think it's the op chain and like optimism as like a reference for like addresses but like a lot of stuff happens like P top via these hubs we think the universe of like the constructions that people will actually

use are just going to be so different from each other that like we're not just trying to replicate the cosmos SDK or something like that this is um built from the ground up for that wider Universe of things that really really want people to build so think of it as you want to build Crazy blockchain applications here's a bunch of pieces to make your life easier now in terms of like what it means for us and like I said at the beginning this could have been like some sort of like nonprofit thing and instead it's a company well like where's the the money or like the business related to this and like how does it help and accelerate the the work of developers and like where does it all fit in so most um Open Stacks uh are predicated on one of two approaches one is

you are part of our ecosystem and thus we give you this software and but the expectation is it's it's open but it's so tightly coupled with our ecosystem that while you could not be part of it you might as well or we'll charge you some sort of lensing fee to use this open software and like we'll collect either some percentage of Revenue um from every transaction that that flows through the software or like we charge some flat licensing fee so um those ideas I think uh were not appealing to me nor were they appealing to like some of the developers we talk to um I think I mean when you talk to an engineer like

there's few things that like frustrate them more than like they can see the software on GitHub and then someone says they can't use the software that they can see like if it's open on GitHub I should be able to do whatever I want with it and so all of commor software is that so like if you can see it it's Apache 2 MIT you can do whatever you want with it you can rip the name off of it you can change it you can whatever the whole point is that we have these Primitives that like you should not be afraid to use because like they come back to us the whole idea again with the original approach is these are neutral things they're ecosystem agnostic you want to go build

something we think this is the most compatible for many people with the types of things that they want to build and you don't have to worry about you know being part of some sort of like commware ecosystem that that becomes your life um now uh the way we think about how we support developers is a lot like how nextjs and versel uh work together so um verel really cool product makes it really easy to like launch different web applications and stuff like that they give next JS you know a library to build like websites and all sorts of stuff away for free you can run it anywhere it's MIT like put it on cloud flare put it on AWS put it wherever um but forell

offers a number of different products uh and like infrastructure related products that supercharge what you can do with nextjs and if they gave you the open like the software that they were running you wouldn't know what to do with it anyways so like for example um most websites when you launch them uh you need some sort of CDN for like images or like what it's called a Content delivery Network to like put the images near the user so that it loads really quickly now cdns are super easy to use but are extremely complicated so like if cloudflare was like you know what like we're going to give away our CDN software and if you have like 150 uh you know collocated boxes all

over the world like you can run this software and like all the stuff that comes with with it it just it's tough to I wouldn't know what to do with it I don't have that um and I think that the idea that um everyone has to run every part of their infrastructure stack um is running into a bit of a wall with the Boost in like application TPS so in a world of like 10 to 100 TPS when you want to like run your own stuff you know you just run a node you put a nice disc on it on AWS and like you put a load balancer in front that's how like people serve a lot of eth traffic um salana I think really started to change this because you can't do that anymore like you need to have

more infrastructure components to offer like really fast access and like helus I know for sure definitely doesn't do that they instead have like more of a purpose built infrastructure so when you get up to this world of like massive throughput um we believe that there will be a transition from everyone running everything to offloading certain parts that don't differentiate your application um to different infrastructure providers and we think we have a way to augment different sort of people building on the common Weare stack with what we call the EXO stack with all sorts of um really cool uh augmentations that you can't really do with um like locally run

software um and so we think that that will really accelerate the Journey of a lot of people from going from hey this thing runs to like oh I can actually serve you know 100 thousand people interacting with this application um and I think the theory behind all this or like the thinking is standardization has driven a lot of our development or like the development of the space for a long time because you can benefit from all this stuff out of the box right like you know you you you you deploy an evm right like all the tooling works all the Explorers work all the apis work what we're proposing is a deviation from that and so this isn't like some Rosy picture that's like oh yeah all like it's going to be the exact same thing it's a

totally different way to think about the problem um and we think that uh even in this world or like broader universe of the way people do things that there are like um particularly infrastructure services that can still apply more than just like rpcs like deeper PTP stuff deeper consensus constructions things that are storage related that really like uh push an application forward but are not specific to the app uh itself so uh we have some really cool ideas for what that'll be and so we think that um most people using the commware would benefit from from different things like this and so we're working on um some cool augmentations to the commare stack that we hope to offer

as services so Patrick so basically what I you know if I could try to repeat back there that um there there's no like licensing business model here or anything like that all the code that you're building just lives in GitHub it's open source anyone can use it that said you know the expectation here is that because this per this code lends itself to more high performance applications there's going to be some some infrastruct infrastructure stuff later down the line which probably the teams are going to say I don't actually necessarily want to run this myself or I don't know how and that those sort of like a managed Services model is sort of what it feels like commware is shooting towards yeah I think with the one very important call out there is that you

know we're in crypto because it lets you do things you can't do anywhere else so like the idea that like the community itself can run a service so our goal is never to like make a validator as a service thing where like we run your entire chain or something like that not our goal uh instead we are focused more on like how do we actually like scale participation and scale access all secured by cryptography using infrastructure things now that's a mouthful so yes in in nutshell TPS goes up your life comes a lot harder yeah we're here to make it easier when that happens so that more people can use your product um we think that we can do it cheaper and we think that we can do it

faster and the idea is that um most people today run their own infrastructure because it's a differentiated service they offer it because it's better than whatever they can acquire or like buy we think that we can uh offer something that's really good that would be difficult to replicate at that level of throughput um and as a result most people hopefully will just think of it as like oh yeah nice thing that I can use I don't have to build it myself it's like dramatically cheaper and dramatically faster to like than whatever I could do on my own um and I don't have the resources to like rebuild this entire thing for my application and so we think that there's a number of different services that work like that so it's different than just like running

the open stuff think of it more like uh for developers listen mean like think of it like Dynamo DB um like it's a purpose-built cloud native infrastructure storage thing it's not like running postgres um and I think that there's a lot of like blockchain focused infrastructure products that could be built like that that really today are just running open code and so we think there's a lot um a lot of really interesting stuff to do there that um we think we can do in a way that doesn't mean that you just trust us entirely to like serve all the correct stuff and whatever and that we can secure it with cryptography but um make your life a lot easier now again not required you're like hey Patrick like cool cool

pitch um I'm gonna do it myself I would say like cool like I would love to see how it goes and like if anything doesn't work as well as you hope like we'd love to help at that point but it's not required to interact with anything that we do it is more for folks that uh don't see how they could do it and provide additional value than what we're doing so Patrick let let me uh I'm going to be very reductionist here and just try to but so you would not call common wear an L1 necessarily or there this is probably something that you can't say but I'm gonna ask anyway plans for token at some point in the future like if how how does that all look from your standpoint commware you could use it to build an L1 like all The Primitives you could pull

together and turn it into something you want to have like subsec finality and slap W into it and turn it into a chain like you can do that um our view is that uh the bigger business is around and like the the better way to make the whole project sustainable is to help and accelerate the work of developers next next to them launching things like l1s or l2s or whatever you want to call it uh as the space develops now our view is that um it might be beneficial at some point for us to like serve as the tip of the arrow and like create some application that like you can only do with an exost stack to show how it all comes together um and the idea for that would be you know some sort of network

of some kind doing something um can't say exactly how that'll come together but um or like what it'll look like but uh we have some interesting ideas for uh cool Crypton native applications uh that we think are uniquely uh made possible with some of the ways that we're building and um you know I think there's a a really interesting Nuance there so we really really really really want to avoid a world where people think of us like apple coming in to replicate your application as soon as it's successful so if you build something really cool and then

we're like oh yeah we're going to build this like common wear branded app that's like a version of it just like how Apple music you know replicated Spotify or like Apple news replicated you know pattern that they continue to use um we don't want to replicate your success we want to help you we don't want to like take the best ideas that people use and build with common Weare and make them common Weare branded um and so there's a a delicate line that we would walk if we ever did anything like that where we would be there to either like accelerate people by offering some sort of service they could tap into or like some application that hasn't been built yet um but we are not interested in being like you know the 10th person

who like takes the best idea that we've seen so far and makes it even better um so Nuance there being you can build l1s with it you could build the stuff like that uh we may uh to help get things started uh create one of one of our own of some kind um but our preference is that we uh find a a number of different teams that get what we're talking about get excited by it and want to collaborate with us to build something of their own and that we would love to just collaborate and support you know you building something magical along the way so how do you think about the I think this is a little bit of an interesting Trend that you're starting to see which you could call it first-party apps you could call it dog

fooding your own infrastructure whatever it is like pick your there's sort of igen layer building Igan da right is the first uh y AV an example of an AVS there's you could look at jeto doing a little bit of this with uh you know their reaking product in me infrastructure you could look at um at L the new svml 2 from Ellipsis you know having like taking the lessons from Phoenix there's there's sort of this melding of you know I think a lot of the best infrastructure is actually being built by people who've built applications and deeply understand and know the pain points the you know the trade-off there a little bit is I guess there's an operator trade-off standpoint and maybe an optic standpoint there's uh hey we want to be known as someone that

you can build great infrastructure on not as a competitor and then there's also that there's a very dual Focus there which would just as you know putting my operator hat on that's maybe what I would be a little bit concerned about so how do you think about those tradeoffs I mean I think this is a good point I like so I um I haven't talked much about like the actual semantics here but like uh common Weare uh fundro like raised $9 million co-led by Han and dragonfly so I've had a lot of great conversations with both folks from Han as well as from dragonfly um through a lot of those conversations what became super clear to me is like um these sort of decisions I believe are born more from practicality than like trying to like capture all so if you're

building something that is made for other people you have two options you can basically sit there and say like if you build it they'll come and like hope someone comes and like At first you're probably going to have some pretty rough deals um where like you're just trying to do anything to get someone to use this thing and a lot of people that are building infrastructure products or like low-level things like something like an exost stack would be um you put a lot of faith in someone coming to you to like take it to its uh you know it's it's something that makes it look great or like a Showcase of what's possible and it's really hard to find that person or that team of people

that get it so deeply that they can make it shine as good and better than like the people that actually uh design that lowlevel part did um and so you see this pattern play out time and time again because I just think it's a terrifying thing for like a founder or an entrepreneur to put so much faith in uh at the early days in someone else telling your story better than than you could and so a lot of people are like you know we're trying to create some way for help people to build um no one cares about it as much as we do like we should you know showcase what it's possible because by definition we're creating this thing because it lets you do something you

can't otherwise do so we should be able to also create something that is like crazy like that has value that is like interesting um and so if we do that we can you know the theory is that like as soon as you have a success story then everyone like comes in it's like oh yeah I get it like I can use it I don't think it's as simple as that but uh I I really think that's born from a the need as a Founder to drisk like the idea that um you know the that your entire future is out of your hands like I think that that is a tough way to sleep at night if you're trying to launch a company you have employees and you have people you're trying to work with that um sometimes it's more than just if you build that they will come you need to uh

show them what it looks like uh to to like uh to to Al together so um we are open to that perspective but uh you know I think you Ideal World you get people that are like yeah I get it like I'm all in like want to do it like and that's not something we need to do um but uh if uh you know this is a you know a lowl thing pretty interesting like infrastructural approach and um you know it may take some time to for people to digest it and like how to build it and we'll run into rough edges I'm sure um and the idea that we could help smooth those rough edges by doing it ourself in the process I think could really accelerate a lot of different people so

uh it's something we've thought about but um our our real focus is on like how do we take people that have these that are bumping up against these abstractions and make their life easier and we think that making their life easier is more than just here some open software we think that there are other things that we can do um and we think that if they run like an exost stack like structure something like the commonw library that uh it becomes a lot easier for us to to help them out I I I want to end here but we you're right we haven't really talked about the semantics of what we're announcing here the Rays or uh or anything but one question before we get there um you know I want to unpack just the context of if common Weare is successful what does that mean for devs in the industry in

general you said something at the top of the episode which resonated with me which is people end up deciding an ecosystem really before they understand or if deeply thought about the technical architecture of that ecosystem and oftentimes it kind of goes hey I like ethereum or it feels like the dominant place where everyone is launching and there's a user element to it as well and a distribution element which I don't want to under sell and but part of it is just a values element as well and you know also Founders get told by VCS hey don't take platform risk and and you should launch here but oftentimes once you get pretty deep into building what you need to build you realize that the ecosystem that you're in doesn't support your application or you need to make serious compromises and trade-offs in

terms of ux and so you know with something like common that spans multiple different ecosystems what does this mean does this offer developers much greater flexibility in terms of where they want to end up building does it actually mean that the technical considerations of how uh an ecosystem like ethereum or Avalanche or Salon or whatever is built today matter even less and then you can actually just build where you want to build because they're users or because they align with your values help me unpack the context here it's exactly the idea is that like you uh so early on like you know it's a new new frame like you know new approach to building new library like we don't have a thousand different Primitives and things but the idea is that over time uh people release Primitives that make it

easier to connect to different things so like you could have like I said an EVS adapter you could like have a your own stake adapter you could have whatever so that when you're a developer you say hey I want to build on chain because of the properties that that brought us all here which is that like I want the community to run the product and like I think that that really is a a better way to like return the power of the internet to the users um and it benefits me to join or like partner or collaborate with people in a particular ecosystem for some reason that's important for your application that you can then shape whatever you're doing with different Primitives within the commer library to connect to whatever matters to you so

you focus on the application you want to deliver and then when you need to actually interact if you do with someone else that you can shape it for that and so the idea is that you build something that's great that users would want to use and then you can connect it however you see fit and instantiate it with some sort of security model however you see fit like we're not so attached to the idea that every one of these has to be its own chain with its own token like I think as much as I wish that was possible like I just don't know if there's enough value in the world to secure every possible application interaction um as its own token and so the idea that you're going to have all sorts of things like hooking up to each other in interesting ways different ways

based on like what benefits the application I think is a much more realistic View for the for the foreseeable future um and in that world you want to build what's best for for like your users and then you want to actually bring that to life in a way that makes sense with an ecosystem that you care about or that matters to you and so uh I think it's just the flipping of you come in and build something and then you can bring it in so the idea is that we just stop having these conversations about TPS we stop having these conversations about like my Chain's better than your chain all the great stuff that you would want is just there open as a default primitive from things like the conare

library and then you bring meaning to it and what we talk about instead is the meaning of like what your application is bringing to people and like why it's a great product we don't talk about like Netflix's CDN architecture we just don't we just use Netflix because it's great the same sort of thing will happen with crypto and we need to move past like the TPS era and like all this fighting about infrastructure things and instead just as soon as it's readily available by anybody I think that that will help us move past that because like the differentiator of an ecosystem being like some sort of particular primitive or like some sort of particular scaling technology no longer be becomes the cell

the cell is like what that EOS system together can do together and if it makes sense to be there because of that or not um and the hope with common Weare is that we can cut through a bit of the like the fog of war between these like ecosystems that uh are looking to be like the ecosystem and instead put the power back in the hands of developers uh that just want to create cool stuff and then connect it to the things that make their application better so hopefully it becomes a thing where it's like I want to create something cool I'll look at commware first because like they just have a bunch of different ways to do that and then if I need to go to a different ecosystem where I want to create my own I can make that decision um as I go along

so um Patrick I want to I know we're in our closing minutes here but uh you know probably should have you know talked about this at the top of the episode but you know the big announcement here as well is that you guys have raised $9 million from a great um array of uh VCS but also angels and so you know kind of give us the the details of the r here and then you know if there any you know calls to action for developers or applications that should be leveraging common wear you know what's the sort of closing closing statement as well that you want to leave listeners with yeah so um at different early on there was this temptation that I you know go to my garage and just like SP spend a bunch of time coding and just that that's it I

have some like repo that I post stuff too um but what became clear is that it could be a lot more impactful of an effort if it had uh a lot more backing behind it and uh had the like think of it pretty frankly right you're building some new application or business if I told you like yeah it's kind of a best effort thing like there's that diagram of like that little sliver of like a a block and like this tipping over Tower of like people that maintain open source if the idea is that you know you're just betting on like the faith in me to just kind of like keep whipping up random stuff and maintaining it it's a tough thing to base your the stability of your business on or like the

application or community on like you need to know that we're around uh and will be around for a long time maintaining the commware library and supporting developers and so the motivation to uh raise uh some funding was to uh ensure our future uh for the foreseeable future so that people that were building on us knew that it wasn't just you know some cool cool project that you know Patrick decided to cook up and we'll see what happens with it it was this is enough uh Capital that we think we can uh you know maintain operations for some time without any you know risk of anything and that uh as a result people should have confidence

that like what we're building will be around and that if you decide to build on it it's not just going to disappear or like stop being maintained in six months um the other thinking there was uh partnering with two you know great investors uh that could help us navigate bringing this thing to market right like these are complicated uh go to market like there's just a lot out there right so like how do you decide which verticals to focus on how do you decide um how to collaborate with the right teams right I think the risk is if we're like yeah we want to do everything all at once like it just ends up being nothing to nobody right like and so the idea that we pick

one or two verticals that we're ourselves super excited about and focus on supporting and collaborating with those people um is something that I think you can get a lot of uh support from from different folks that spend all their days thinking about all those different things so uh yeah so the Rays I'm really excited to work with folks to like uh to with that working with the investors that are backing us and super fortunate that they decided and believed in in what we're doing um and what it means for us is that we feel stable and well equipped to go collaborate with teams that are ambitious and want to see what they can do on chain and really rethink um some of the patterns and design patterns that people use I

think the f I gota be careful here but a lot of people when you interact on chain I guess like the the driving factor for everything that that's going on here is like we maintain standards and keep things um the way they are because it makes it easier for everyone to inter that's the guiding principle with a lot of the way different ecosystems and ways to build our BAS the funny thing is though is if you ask people like do you feel like it's easy to interact with the with a blockchain I don't think you're going to get yeah it's all standard it makes it so easy to interact like I think we may have prioritized standardization with how we

see it instead of what actually makes a great experience um and I think that um prioritizing really like something that gives developers more choice at the cost of some definition of standardization may actually take us a lot farther than um than we might think um and so I'm I'm super excited about uh having the resources now to uh pursue that view and collaborate with with great folks that see see it the same way Patrick man this has been uh great um congratulations on common r s super exciting and definitely looking forward to a world where uh the maybe we can ratet down the tribalism a little bit

and we can scale applications and have that be the central Focus so um if folks want to find out you know this will be live um you know when the big announcement comes out but if folks want to find out more about the work that you're doing on common Weare or follow along or there are abdabs out there they're listening and saying hey I could maybe use some of this stuff what is the best way for them to follow along and get more information yeah so uh somehow by the grace of God we got common Weare doxyz so it's exactly how it sounds go there and that points to GitHub All The Primitives we've built uh Twitter whatever like that and uh you know if you uh this sounds interesting and you know you've had something on your mind that you haven't quite been able to get

off the ground because you know it's just been really hard with what's out there uh you know we'd love to chat about it and see if there's a way to bring it to life excellent Patrick this has been great thank you so much for coming on and uh congratulations we'll have to do a check-in at some point in the future just get the get the update excited for it yeah thanks awesome all right cheers [Music]