Fat Apps: Aggregation Theory in Crypto Networks | Zaki Manian, Sam Hart

Transcript

all right everyone welcome back to another episode of bell curve before we jump in quick disclaimer the views expressed by my co-host today are their personal views and they do not represent the views of any organization with which the co-hosts are associated with uh nothing in the episode is construed or relied upon as Financial technical tax legal or other advice you know the deal now let's jump into the episode [Music] all right partner episode 2 of the app Gene thesis I'm excited for this one yeah this should be a great one we're talking to Sam Hart and zakimenion today and we're going to be talking about kind of this idea of fat apps and for for listeners a you know an important

analogy or framework that we draw on in this episode is uh this idea of aggregation Theory which was originally pioneered by Ben Thompson and just to give you an overview you know aggregation theory is kind of a framework for how to understand you know a very successful kind of aggregator business model that's appeared in web 2. so you can kind of think of it as those famous two-sided marketplaces that everyone sort of uh knows and loves and uses the Ubers the airbnbs but also even the Googles and the Facebooks where an enormous amount of value has been captured by aggregating sort of consumer demand um around like a specific uh supply of something that looks commoditized so I think that using that analogy is not super perfect for for web3 but it was

interesting you know to hear zakies and and Sam's thoughts on where it applied versus where it didn't and ultimately I think it was a good way to reform you know sort of Supply driven investment arguments that get made in crypto a lot over like demand driven investment sort of rationales and you know that's what we're very focused on in in this episode I think we'll we'll get into this in the app but a lot of these Protocols are marketplaces um both l1s that are you know a Marketplace for black space with applications as well as the application layer which is you know often a Marketplace for say liquidity with with Traders and just you know kind of teasing out where this is going based off of what we see in web 2 and the aggregator side

um and and maybe you know to think about give the benefits of getting closer to the user to have more control over the demand side um is a big you know Tailwind behind this whole application specific uh thesis so um yeah I'm excited to dig into it all right uh without any further Ado let's just let's hop right into the episode all right everyone welcome back to another episode of bell curve today I'm joined by my co-host miles O'Neill and we've got zachymanian and Sam Hart on the episode so fellas welcome to the show it's great to be here thanks for having us yeah guys thanks very much uh so we owe a special debt of gratitude to bozaki and Sam who gave their thoughts on this season so guys thanks so much for uh not only appearing today but also helping to uh steer us

hopefully in the right direction for the trajectory of the show um you know today we really want to get into this idea of like really explore kind of fat apps uh versus fat uh fat protocols and really try to examine this idea that value May accrue to the app chain layer which is closer to sort of consumer uh you know sort of consumers that are actually interacting with with blockchain so Sunny was getting into this at the tail end of uh at the first episode that that went live last week but guys if I had to turn it over to either of the two of you just from sort of a high level to begin with and then we can drill down um you know if I ask you to make sort of the case that value will accrue to sort of the app chain layer as opposed to the

the fat protocol layer um do you want to you want to take up the mantle for that argument so like I view app chains as a way of enabling sort of product Market fit Acro us blockchain technology as a whole and so what the version of that world that sort of exists uh in my imagination and it's slow I think slowly starting to be validated uh by the market is that as you get increased conviction in entrepreneurs about what the market wants what is needed what what they want to build um you want to be able to like sort of vertically integrate that whole experience right uh and usually you know

if you're building something on a blockchain you're building some sort of two-sided Marketplace uh where there's a user base and there's a a there's a sort of a supply chain uh or uh providers on that on that space whether it's like you know you're you're bringing together nft creators or users or defy uh investors and strategists on some um you have this two-sided Marketplace you want to be able to optimize the value flow between all of the parties um and one of the things that is complex about l1s is actually in many ways there are a bunch of established uh value

flows in the L1 that you have to contend with whether it's you know uh the like Mev Searchers or the or the or block producers or miners or all of these things who are uh not really uh aligned with your particular application and who you can't necessarily harness uh to uh to uh as part of the creation of this two-sided Marketplace and so what I am what we you know what I think is going to be the design patterns that will succeed success um is applications that um uh do this now one other thing I will say though is I think if like for

instance if you look at the evolution of the ethereum ecosystem you can start you can start your ecosystem as a fat protocol oriented ecosystem and then application and then like things can be built on top of that fat protocol that start to align it better with the application specific thesis um and like I would say both like the emergence of like eigen layer and the op stack uh have been aligning the ethereum ecosystem a lot better with the with the uh application specific blockchain thesis uh uh in a way that was like organic but somewhat unexpected so so I think that um that you know to maybe sum up what you guys said there are a couple

different angles there which is one they're sort of the the entrepreneur angle right which is that eventually you know entrepreneurs are going to want more control basically over the full stack that they're building on so app chains have a natural sort of Advantage there Zach you were starting to you know your description of app change sort of a two-sided Marketplace is one that I think is very important uh can you elaborate exactly on what that means because I think there are some a bunch of questions or analogies that we could draw there to how that played out in web 2 and I want to get your thoughts on whether or not you see the same Dynamics being prevalent for what we're talking about here yeah I mean most applications inside of blockchains whether or not they present themselves

this way have this sort of two-sided Marketplace characteristic right you know and the classic Web Two two-sided marketplaces are Uber or Airbnb Uber's you know matching a pool of drivers with a pool of of riders Airbnb a pool of hosts with the and so you know we you saw you know uh a sort of strike goal there was definitely a hype cycle of this in general um but in general you know you have you have the classic you know even you know your your D5 apps are just you know we you know the whole uh peer-to-pool sort of thing like what this is like you know

the defy summer Innovation uh which was that you could have these smart contracts that can pool resources and then act as efficient coordination mechanisms between uh you know essentially makers and takers in a in a in a in a market in uh like you know a trading application so you you know so you're uh so like you could have uh the taker side essentially filled by this like LP pool um and then makers on you know who want to buy and sell being able to draw from that liquidity or you know people being able to aggregate their funds and like a lending protocol and then you know people on Demand Being able to pull and borrow from that you know sometimes for

you know in a Flash loan for one block uh you know these are all these are all essentially uh the great successes of cryptocurrency have been the creation of these extreme the pro so the problem of uber right the problem above Airbnb is it's not a neutral marketplace right is is you know they are both rent seeking on top of that and the dream of cryptocurrencies is basically to create these like extremely neutral platforms by which you are able to you know match part you know the two sides of the market adding on to that a little bit uh the analogy like one of the reasons that that Uber works is because it makes the supply of drivers like more uniform for the

um for the ride buyer and it's kind of exactly what what's happening to LPS it it kind of gives them a uniform strategy so it it's a much kind of easier interface for a Trader to interact with so this is kind of a good um you know a good place to introduce because in in those Dynamics right for uh for web 2 right so we're talking about the classic platforms of of web 2 right like the Ubers or the airbnbs even before that you had kind of like the Facebooks and the Googles some of the best sort of aggregator businesses uh of all time this is where you know I think it would be cool to kind of layer this idea of aggregation Theory which is this is Ben Thompson of strategories thing he wrote a blog post in 2015 that initially

gained a lot of traction and has sort of since blossomed into this whole way of describing uh it's kind of a framework for describing these these very successful businesses in the internet era and most of them are the sort of marketplace businesses and we can get into the exact definition of how he defines an aggregator but essentially it's these businesses that we're talking about and really the the key distinction I think that he that he makes is that in the internet you can if you're looking at sort of a supply chain you know where that has like kind of production and Supply over here and then demand over here the best businesses in the web to internet era actually optimize for owning Demand right that's what these aggregators sort of did so do you broadly so it was interesting exactly to

hear you kind of describe uh some of the most successful protocols in crypto is actually having a similar Dynamic how appropriate do you think that lens is to apply to crypto where does like optimizing for taking demand kind of work in a crypto Paradigm versus where does the analogy sort of start to break down uh from web 2 in a crypto World from my point of view I think this has basically been the quest this has been the question right is the aggregation of demand at an L1 layer or not um and I think that like for what like there you are okay so you do see aggregation of demand at L1

layers I mean I think there's just like no doubt that like there is an aggregation of demand that they're you know they're that like networks of like telegram channels and communities um form around uh like L1 and L1 like objects like so like you know like arbitrum right now is like where the D5 cool kids are all hanging out and there is an aggregation of demand that is occurring around arbitrum and it's like um and you know that and there was an aggregation of demand that happened around ethereum and there has been an aggregation of demand around BNB chain um so you know you do um you do perceive that that aggregation of demand does occur the question that like the or like the

skepticism around the fat protocol thesis from my point of view has been that that aggregation of demand I believe is like largely temporary and transitory and not particularly sticky and you've seen the aggregation of demand kind of flit from L1 to L1 over the last like three Cycles um and and you know generally nothing sticks um what has been sticky is like people use openc and uniswap and they use it across multiple multiple chains uh and are like the core of the like belief of this is that the actual aggregation of

demand from the end user point of view is not going to be like oh like because like like let's be honest with blockchains are critical infrastructure and ubiquitous it's the user base is not going to be like oh we hang out in like this Discord or this telegram group and like we just we're like oh like that's not our user base right that's on our log trip user base our long-term user base is like barely aware that there's a blockchain involved and are just like this is a marketplace where I can buy the things I want to buy or and use the stuff that I want like I need compute I need uh uh you know I need an AI model I need something to be rendered like whatever like I they're just you know using using Computing for the services that it provides and blockchains are

part of the infrastructure so in that work you know I need to send money to my friend like they're like all you know these are the different things I need to hedge my Forex risk like all of the like all of the ordinary things that people are going to be doing none of those people are going to be hanging out on telegram channels uh and discords uh to like pick a to aggregate demand they're going to be like this is the app this is the app that does the great job does it like does this really well I use it I believe in that I like that's where my relationship is I've had this app player um and so that's that's the thesis it's like we we are we continue to be sort of a little bit modeling the future of cryptocurrency based on and like the

future of our space based on the users of today who are not really representative of what the future user looks like I think that's fair and how much of that do you think is you know I guess the difference between I think maybe another analogy is like an L1 is sort of like you know an app you know Apple's app store right or or Windows right in that it's a Marketplace for supply and demand of developers a supply of block space um and you know it gives them the ability to rent seek throughout that entire you know I guess value chain and squeeze on both sides um versus other platforms are really owning that unit user relationship right like uber Google you know they are they

are the the touch point of the user um and so I think are we do you do you agree that you're seeing something where you know as as apps actually start to gain more leverage and traction on these general purpose l1s that's that's part of the friction here or the tension right to maybe move at least to their own app chain or get closer to the user and be able to you know extract more of that value versus the L1 um is that a fair analogy you'd say where this is all going is is really all to the app's motivation to get closer to the user and control more of that value stack I think it depends on where you're starting um so if you're starting out as as uniswab

um you have a great relationship to the user and you're trying to like expand the addressable Market into other chains um if you are starting out as osmosis and you're you have an L1 like uh they have expanded into kind of adjacent categories because they want to um I mean one there's a smaller user base that they they have access to at the moment so they kind of want um to get access to users that way um so this is that question of like um you know what move you make uh and and how that kind of impacts your

your decision about you know your architectural decision about the application and you know whether you're going to be um doing kind of further vertical integration it really is highly dependent on on where your Beach head is I think that beachhead sort of point is is I think a good one Sam and maybe I mean there's a lot of work that gets done in traditional businesses of sort of mapping like a customer Journey right um and again to borrow sort of this uh sort of helpful analogy at least for for me to understand this within the context of kind of aggregation Theory and where value gets captured you know Ben has basically used this analogy to describe how people consume information right so you know in Olden kind of times when you

were trying to read the news you'd go right to the newspaper but in event and then there's sort of like a value chain of people that actually produce content and then there's sort of this um you know that there are Publishers but now with the Advent of these aggregators in the form of like social media people actually first go to like Twitter or Facebook or wherever you get your news and that's why there's a tendency to sort of bypass the the middle step for um you know which is Publishers and that's why traditional Publishers are sort of struggling so when we think about uh that sort of customer Journey for crypto I'd be curious to see are there any like opportunities and what what is an aggregator maybe look

like in the in the future end state for crypto maybe it's like a wallet is kind of the most obvious thing that comes to mind uh Sunny made the observation in the last episode that people aren't users of ethereum they're users of of metamask so I'd be curious like do you guys view wallets as being potentials for for aggregators or like are there other apps or like what is that sort of uh eventual Journey for people into crypto look like and who has the value capture to take a kind of sober look at this like 99 of applications right now are are like day trading investment applications um it's all like Financial rails so of of that single category

um wallets and and potentially kind of like an integrated dap experience yeah absolutely that makes makes sense to me um I I think if we're like really successful here I we're going to be entering into different markets and and you know we might kind of white label some of this infrastructure for for other use cases or or kind of deploy uh you know separate instances that that really help us you know do non-financial use cases the finance Finance is just the back end so um what the what owning the customer means in that context I think is um you know to be written exactly

exactly I don't know if if you have thoughts there about just like potential maybe even even looking at it today I mean even if we just look examined the you know Financial use case or what crypto looks like today I mean you could even view decentralized exchanges right as a potential aggregator like if you think about that customer Journey people will you know first engage with crypto you know via um you know trading on a DEX although I guess the wallet even comes before that but you know I guess Jack any any thoughts to build on on what what Sam said in terms of potential aggregators today or even just apps that own sort of the customer relationship or the consumer relationship I mean I think like one of the things that has um

sort of been most interesting to see right is you so you see both nft marketplaces and sort of token marketplaces having been remarkably sticky you know the uh you know the numbers that you see are like you know three I think I I think there was a block Works piece that came out recently that was like three million of the like you know uh you know very well which is a total very large percentage of the total a solid base of medevas are essentially just open to users right um and then after you go from uh the the you know just open C users there's a lot

there's you know it's you know I remember you know I have no idea this number is still is true and like what the actual number is but I remember something like uh maybe a year ago someone saying that 40 of the uh traffic to the unit swap protocol is sourced from the uniswap front end right um in spite of the fact that there are alternate protocols inside of the fact that you almost always get a better price from something like one inch you know you you still have this like sticky uh connection to users um and I think we see this in in honestly in Cosmos as well like with osmosis like you know the cosmos Community is there

um that that is where you know and there are there are numerous other dexes in the cosmos ecosystem today um but the sticky aggregation place for uh user demand in in the cosmos ecosystem continues to be awesome it's that really come from an early focus on on supply of liquidity and and I guess that is really when we think about you know both sides of the marketplace what's more important in the long run um you know I think it's very interesting to look at osmosis and how they're able to bootstrap themselves by you know using a lot of incentives frankly um and we're starting to come down on that right yeah but they built the Skinner box is what they actually like I think like what the most

uh what the thing that osmosis did that has what is that Zaki sorry what is it yeah so a Skinner box is a system of like very predictable rewards so like you push a button you get the you you get the reward right um and the um and the question becomes you know you are uh you are doing the you are doing this the the um you're doing this question of you know what is this you know what is the Skinner box in this scenario um and the

um in the uh the question then becomes you know you are um uh so like you know what osmosis built was this you know every every day at you know uh at the epoch time you know they distributed a bunch of of LPN rewards um and then people needed to end you know so there was this like long period of time of like really like of like daily engagement with the osmosis protocol um among a large population of of early Cosmos people and I think that was probably the thing um that built the uh uh that built that user base and I think the the other the

other thing that I think has been successful about these exchanges right is probably more than any other application in crypto it's like it has you know it on board you know the users built a model of like what a centralized exchange was and then we eventually built dexes that fix fit that model well enough that people were like okay I get what this is too like and so that that conceptual understanding of what it is that you're interacting with I think has been a very powerful feature of these systems it is an interesting observation because you know if you're a builder in in crypto you definitely have to think there's a little bit of a you know and I think Cosmos frankly it's like way ahead of the curve in terms of ecosystems in

terms of like how can I build the best product but I think a lot of Founders also think of where they're building in terms of L1 ecosystems as distribution as well and right now ethereum has the most the most users right it's an L1 but it's a very interesting observation to make I think at least to me that maybe there's less lock-in on kind of the L1 ecosystem if you look at it over time right like users initially like Bitcoin was the first kind of uh you know software you know ecosystem that attracted the initial developers and obviously that's not necessarily the case anymore and you know today ethereum is kind of the the Juggernaut but ultim like Cosmos and you know Solana on other l1s have attracted a lot of developers

over time and really some of the stickiest relationships are actually like like metabask or coinbase or you know potentially open C so it's just a very interesting observation as to where you know which apps like might be sticky in a crypto environment because it's hard to predict ultimately it's still pretty early days for that um I I guess you know just uh uh to to wrap up this sort of line of thinking I'd be curious you know there are some when you look at some of the stickiest web 2 products especially the the sort of marketplace products um it's kind of hard to tease out you know how much is aggregating Demand versus aggregating Supply obviously you

have to when you're bootstrapping those two-sided marketplaces you need both to happen at the same time right so like Amazon with their sort of network of of merchants right or Airbnb with its network of hosts and suppliers or Uber whether it's you know with its drivers uh you know Etc um I'd be curious uh you know do you how do you see those moats sort of playing out in crypto like when you look crypto is a little bit of like an open source like lowering switching costs uh type environment like how do you guys have you seen any outside of like the exchanges that we've talked about uh have you seen any examples of successful modes uh in crypto around aggregating demand or what do you think that might

look like in in a crypto environment as opposed to web2 I think it's worth mentioning that the the bootstrapping phase is like pretty different from from the kind of like maintenance phase um there's I mean my I'm sure you had a similar experience like the first time you used uniswap like there wasn't really a comparable experience you're kind of just playing with it it was kind of the only place to get random coins and 2017. um so you were pretty insensitive to anything but just any Supply whatsoever being there that looks pretty different today where I think that they have like a very

strong customer relationship and and and very strong brand that allows them to to be more kind of mobile across different different chains different l2s um and so they they've carried their their customer relationships so that you know wherever they go um yeah so um the moat I mean kind of building the moat is you know in a creative process that happens over a long time and um I think it starts with like you know getting building the Skinner Bots box early I guess and then uh and

then um kind of building some kind of liquidity flywheel or truss flywheel that allows you to to kind of retain users um yeah I mean I think that there are there there are relatively few Mo like applications with moats uh in uh cryptocurrency which goes back to the general reality which is that there are very few applications with like users in cryptocurrency um like the application you know you know the application where has it's continued to lag behind uh uh all of the

other like hype and bits and pieces and and and things going on in the in the crypto ecosystem right um and so you know we are um there's like there's there is to my mind uh not really a a doubt here that the um that like you know one of the like one of the biggest challenges of Imagining the future of crypto is that like you know that at some point if this whole industry is to succeed the future has to look a lot less like what what the president is yeah I mean today just the absence of bridges has been a big move right

um yeah uh I mean yes yeah go ahead Zacky well I mean I I would say uh there have been a variety of things that are have been sources of aggregation of demand at the L1 layer that are breaking down right like uh usdc also has like availability of stable coins and specifically usdc has been a big moat in a lot of ecosystems um but you know with uh cctp you know and like sort of circles vision of kind of usdc everywhere uh that aggregation of demand layer is starting to break down and I think you have you have uh you know so you have and you know the

the other component has been like that yes you know with the exception of of Cosmos most other ecosystems didn't come with any bridging built in so like once you got exchange a critical mass of exchanges onboarding onto your platform you know you pretty much then that was where everybody stayed um you know it was hard to you know that that wasn't you could aggregate demand from on that layer um but like all of those those modes are sort of temporary to the image maturity of the industry rather than a permanent features of the of the of of the system um and so you know ultimately what should own like the only thing that you know

the last you know 60 you know 40-year history of consumer Computing has told us is that applications are really where people like where where the where where the stickiness occurs right you know going back to like the ultimate sticky application like Microsoft Excel and I think you know a lot of the value of L1 is based off of like how differentiated the apps built on top of them are um but I do think that that presents a natural tension right because as those apps become more differentiated then they're going to want to own you know more of the stack and take those users with them um I'm curious how you guys think about block space and and this idea on the on the supply side right is block space

completely is it going to be completely commoditized or do you see some Merit in this idea of there being like a premium block space to say ethereum L1 that will always you know retain a large amount of I guess market share you know in terms of value to the amount of trend transactions versus other you know sort of environments um or is block space just going to be completely commoditized across the board um and and yeah how do you think about validator validating or invalidating that thesis I love this question I've been thinking about a bunch um if you look yeah so my opinion is block space is not a uniform resource

um it is uh so box based on osmosis is not the same as like block space on ethereum first of all obviously um and block to block like and even with you know interblock your intra block um block space is not uniform so it's a highly context sensitive resource and um the main thing that application specific chains are doing is is basically giving you more kind of granular control over the the shape of the resource in the first place like what you can do within a block or what you can't do within a

block and how how the market is structured so um he was able to buy what the auction or or kind of marketplace looks like um and um I I think that there absolutely is some commodity block space um I and I think particularly for for kind of time and sensitive transactions so if I'm like sending a payment to my grandmother or something I kind of like don't care if it takes like five minutes to get there like it'll get there she's not gonna check for another like 10 days um better hope your Grandma's not listening to this Sam just trash around

like that uh but there are very there there are transactions that that you send that you know can get front run or sandwiched and um that's a that's a very different kind of mode of interacting with the the base resource so my opinion about the value of block space really comes down to my perception that there is that block space is primarily valued by the accessible network of counterparties on in that box space

um that like the reason ethereum block space is valuable is because of the large number of like counterparties and Integrations and whatnot but the problem is but the real but the the flip side of that is that there is a there are there are tipping points in every block in in sort of in net in Block space or like networks of blockchains um that where you get to a critical mass of counterparties and then the uh the advantage uh erodes dramatically um and so the so I you know

I think uh in a lot of the fat protocol sort of thesis that we're here to in you know we're arguing against and we're trying to invalidate here there's a tendency you know there's a tendency to ethereum block space has the most Economic Security therefore uh it is the most valuable box base and it's this tautology right and I think the actual the answer is ethereum has enormous first you know guys first mover Advantage um has you know deep Integrations with lots across you know lots of different counterparties is essentially the de facto first integration point for many services um and therefore gives you has Network effects around that but the end you know

the likely future is uh is is you know a more of an evening out of that where there are other places where like large number of counterparties are willing to are available and engaged in a system allowing them to um and so so you know how does that translate into the like future value of of you know the future aggregation of demand and the future aggregation of value on ethereum L1 I think it's an open question uh uh it's I'm not trying to be completely down on ethereum I'm just saying that it's a bit of an open question uniswap is is the largest consumer of ethereum block space I don't know the exact

numbers but it's you know it's pretty significant um I think openc is like number two or something like that um and if ethereum is ex or sorry if unisol is expanding into layers used in other chains and likewise open C is doing the same like is like should ethereum be worried about the value of its block space being eroded because the amount that you pay for a layer 2 transaction is like you know fractions of what what you might pay for a an L1 settlement um okay and those are questions that as as an

app chain uh developer like you start to ask yourself yeah I totally agree I mean I think number of counterparties and Integrations is is one kind of you know lens to look at this and I think maybe something more lasting you know for ethereum L1 is that maybe there will always be some class of transactions that are just so important right that people will not trust any sort anything other than ethereum L1 um but on the other side of the spectrum you know that that to me I could see as being very high value but on the other side of the spectrum I think Samuel are alluding to it is being able to really optimize the block space for your application uh both in terms of you know ux benefits and value capture wherever

you think about like Mev that's where it gets into this sort of stuff um yeah I don't I don't know if you know what what the opportunity is for something that's in the middle of that but you know I think in general that's where I see the most value in terms of the block space guys I've got a um uh you know sort of a concept that I want to introduce to both you and the listeners as well and this is sort of the last analogy we'll we'll borrow from from aggregation Theory but there's a there's a concept that maybe even is actually a little bit adjacent to to aggregation Theory but it's one that uh sort of describes where value accrues in different Supply or value chains so I'm actually gonna share my screen here

quickly so people who are following along via YouTube can actually get a little bit of a visual representation but the The Smiling curve is is a framework that it was it was actually introduced from I think Acer miles the name of the company it's a it's a Taiwanese company um and and it was used to describe kind of the early PC value chain and the reason why it's a smile is he sort of uh bifurcated into three three different areas which is the the base part of the value chain which is kind of R and D right and path which is secured by patents and Technology then there's a middle which is kind of fabrication and then there's uh the part that's closest

to the consumer which is marketing and that the the lock in there is kind of brand and service and all that sort of thing and if you look at where value accrues the reason it's a smile is because at the very start of the value chain there's a lot of um you know Enterprise Value that gets created uh with patents and technology and kind of at r d that's very hard to do then there's a lot of value that gets created around like the brand and service right in kind of the consumer lock-in but it lags in the middle which is around fabrication which ultimately ends up being commoditized and this is an analogy that has been used to describe like a whole bunch of different Industries I'm wondering if there are places in crypto that this is applicable

to to because a lot of what we're describing right now like I can sort of see the smile beginning to form like when you think about like app chains or app dedicated block space right which ends up being very valuable because it ends up having like a relationship with that consumer right there's some kind of brand attachment and stickiness maybe it's at the wallet side maybe it's at the decentralizing we don't know exactly yet right there's an end state but we can we sort of see that then maybe there's kind of this premium blocks block space right uh that maybe it's ethereum maybe it's not um but there's you know kind of Premium block space here I'm wondering if there's like kind of a middle layer to be honest in in a lot of discussions I see about like

that the easiest analogy that I see here is like layer twos frankly like generalized layer twos seem like they are highly at risk of being commoditized um but but I'm curious like when you know when we look at that sort of that that smile um does anything stand out uh in terms of like analyzing sort of crypto value chains or what do you guys think about that analogy as it applies to crypto you know the ethos of the space is so uh fixated on uh is so or like one thing that's like very unique compared to the ordinary technology industry is how much of the our most R D heavy parts of the industry eventually get manifested as open source

systems um where the where the um where the uh technology is um where there's no IP protection around the technology or very limited ipt protection around the technology or honestly even if you know uh uh the like you could put years of r d into something and uh you know once you release it it's pretty common in this industry in crypto that like if you have something that's a substantial Innovation people are able to cream room re-implement it relatively fast um so like there's a lot because there's like there's no patents in this space

there's just copyrights really um and I don't expect you know patents to to start playing a major role in this space anytime in the near future so in that world you know it's like all basically all you're doing with your you know there's this weird feedback cycle where what you're earning with your like very intensive computer science r d is essentially like a marketing and a brand position and then the thing that like is the monetization layer and the distribution layer that you earned with doing a lot of intense r d is the brand marketing networking layer distribution layer you already kind of mentioned this

but the The Smiling curve basically defines like where the monopolies are in a traditional business um like patents are you know intellectual property Monopoly and Brands a kind of trust Monopoly with the user and the fabrication process in this instance is like is commoditized um although it's kind of changing with tsmc and just the kind of unique fabrication capabilities there um that's actually a cool they tsmc really flipped that on its head frankly for that that story but that's a story for another time sorry go Sam yeah we could talk about that offline um

uh so I completely agree with with Zaki um like patents kind of aren't available to you um and uh so the the Monopoly on that side if you're gonna try to kind of retain an r d Monopoly you basically have to have like a sustained organizational like uh knowledge Advantage um so I do think that some of these ZK um kind of shops that that have like you know have aggregated the best ZK researchers in the space um could could potentially have a kind of sustained Advantage there

um but it's you're not locked in in the same way um uh as well as some like you know Asic hardware and um and now some of the like ZK A6 things like that um if you're going to be doing a bunch of research there you and and roll out a bunch of those kind of um Supply constrained machines I could I could see you potentially getting Monopoly um but yeah the the vectors of competition are just like a little bit different with the the kind of Open Source mentality that's interesting because when I I looked at this graph and I was trying to think about this Maps over to our space I I honestly thought it from the seed of a

application developer that you know the left side the r d side is something like moving you know dydx moving to its own app chain because it can build a you know a differentiated product that way um versus any sort of you know order book Decks that is built on a monolithic chain um and the right side I I saw more as like uniswap moving to its own wallet right getting closer and finding a new ways to build brand and improve their marketing abilities um I guess do you see in the end game where you know these super apps will actually be both um you know or or potentially you know if once you move to your own

full stack app chain then you're going to be motivated to build the wallet or once you build the wallet then you'll be motivated to eventually move to your own app space well I guess I would say that DUI DX already is that in a way like the only reason it's able to move is because they own the customer relationship um right they have a fantastic brand they are the the D5 perpetuals protocol and you know users would not go through that kind of pain if they didn't already have that um that lock-in and like fantastic customer relationship so um I think there's probably again just kind of getting back to like where you're entering into the market like there's probably ways to enter in with like an r d Advantage before you know people have

kind of understood what's going on I think ZK is this like there are very few ZK researchers until you know a couple years ago um and if you can get a first mover Advantage there you know you can kind of move into more of like a con consumer facing um relationship that's great um I guess speaking of dydx before we came on we were talking about uh you know validation or invalidation signals um of this thesis to you longer term um we're talking about the importance of dydx within that context right um if this works I think it's a you know a massive validation signal I would love

to just kind of hear you expand uh either of you guys uh on that and and other you know if it's if it's dydx or if there are other sort of validation invalidation signals that you're really paying attention to I will say that the d y d x and like I've been working closely with them um and many many people in the cosmos ecosystem have been um working closely with them um so this is not purely an outsider perspective but it is without any doubt in my mind that what we're doing with the uidx will will either show like will be either like a tremendous validation of the app chain like thesis or a demonstration that you know the

pieces are not in place um and the you know because that migrate you know that like we have to succeed they have to succeed we have to succeed collectively in taking that and migrating that you know a very high threshold of their customer relationship which is currently on a stark X platform L2 on ethereum and migrating them to Cosmos um and you know if we're not successful in doing that you know uh it's a huge demonstration of the like this is uh this is a really like you know this is a a problem um on the other hand one of the other things that I think is most interesting

about DUI DX is they are legitimately building an option um I think a lot of the uh decentralized exchanges in Cosmos and a lot of the users of the cosmos SDK um are building what are effectively l1s that benefit from IBC Network effects um but are not truly app chains um and I think we have a lot of things coming um in the cosmos ecosystem uh I like and I think like it's actually a relatively small population of Cosmos genes that have like been going down the app chain route I think we've been trying to do that with some we are actually been doing that with sommelier I think that

uh um that the uh that the dydx team by like you know essentially building a custom mempool building all these pieces uh you know building a thing that serves is basically their to like allow people to have like the centralized exchange scale and Order experience on a decentralized order book um yeah or like an order book that like teleports around the world every you know three or four every six seconds or every five seconds uh is such a is like an it's such an amazing Vision it's I think it can be like people if it works will blow people's minds about what is possible with an app chain that you're like no you can't do this on any L1 no

matter how fast the L1 is um you can't do this right I think it's a it is genuinely an opportunity to blow people's minds about how easy it was to migrate the customer relationship and how uh uh and like what is possible in a decentralized setting um that I don't think has ever been done before and so I am I do think that like that is our right now I very much are like the ecosystems like unlike even the entire app chain thesis is like sort of Flagship Endeavor right now Zach and I have both done our fair share of like trying to chill the cosmos SDK and like the cosmos stack to people and like you usually whereas

the kind of two things that I um you know first first you talk about IPC and and then you basically talk about um kind of transaction control at two levels like one is the fee model and the other is within consensus and um really just this year I've been really encouraged to see some extremely um creative and kind of high performing teams uh doing kind of innovating on both of those levels um and just these are these are things that you

can't do with a kind of commoditized or or monolithic L1 applications having complete control over every step of the the kind of mechanism um so that would be um kind of early validation for me um but I completely agree with was Accu that um dydx just in terms of kind of magnitude of liquidity and and the the experience that they're trying to deliver to users um would be major validation although one instance I don't think is is enough to to Really um

to kind of prove out the thesis that makes a lot of sense and kind of a good segue into my my uh next question we've talked about this on on the first episode but kind of look at this as a as a spectrum from you know a full stack app chain like you have with Cosmos uh app chains to maybe a Celestia Sovereign roll-up um and then on the other side of the spectrum we have you know app specific Roll-Ups on ethereum um and I think Sam you just got into this a little bit but what what would you point to as as the real like marginal benefits in terms of customizations that are possible when you own the full stack as a app specific

L1 but you know versus becoming an app specific L2 uh or an L3 right um I think a lot of people will look at this dydx move and it's successful and they'll say all right was that because they own their own you know block space uh and they have more in it you know can I does that same benefit map over to an app specific L3 which is maybe cheaper for me um so I'd just love to kind of unpack what you would if you were to you know show the full stack app chain and why why full stack over you know app specific roll up or um you know both today and and also maybe you know in the future where some things like interoperability are taken

care of to kind of like couch this a little bit I I think all these are like valid ways to build an obligation um I that is kind of the cosmos ethos is like ethereum is an app chain it just like snap chain that's trying to do lots of things for lots of people um and but but I do think that there is a certain class of applications that you know really would benefit from Full stack control um I I think the the two kind of largest privacy applications or the ones that I've kind of worked with most closely penumbra and anoma are great examples like it's just stuff that you can't really do in

even on an L2 is very challenging um there's Solutions like Aztec but Aztec has like a 10 minute like finalization time or something and penumbra is building a fully private decks with like I mean just the the level of like functionality that the users can have available to them is is absolutely wild um so you know and and you need the ability to um have the consensus mechanism recognize like the fee token and um

and be able to you know validators need to be able to like validate proofs um there's just like a lot that you need to do in integrating the the execution with um you know with the entire App application experience um so um yeah and then and then in the middle there is is like modular applications and I'm I mean I'm definitely like bullish Celestia for sure um and I Jack and I have been talking to them for a long time um I think there's an enormous class of applications that you can build on Celestia there there is a segment that

you cannot um and basically the idea behind um you know modular modularization is that there are kind of defined boundaries between the different um uh kind of utilities so data availability consensus settlement execution um and the idea there is you're not supposed to be breaking those boundaries um or they have kind of all these different names for like you know a sovereign role of versus an enshrined roll-up and like like what they're doing there is basically they're they're saying that okay in in this instance you're allowed to to kind of like

uh to break a single boundary and and uh have this part of the application interact with with this other part and um like a fully sovereign application is basically says that okay I need like full stack control over every part of every module because there are optimizations that that require um you know boundary breaking um and yeah so I think that there's lots of different valid ways to build to an application yeah I think that makes

sense I think you know people think about app chains and often the first thing they think of is vertical integration for higher value capture but um you know what I'm what gets me more excited is more more about the applications that have that wow factor right so something that is you know a a differentiated user experience because it's built on a full stack app chain um and I think teasing out some of those you know true differentiated capabilities you know as you go across the Spectrum and what you lose as you go across that spectrum is really what you know we're trying to get a good sense of this season that's going to tease into to next episode as well where we really get into the weeds of like what are the

different different trade-offs right between these different models and again there's no like you know freezing by analogy is sort of imperfect by definition but like there are lots of good reasons you know when you look at traditional Supply chains of like where to vertically integrate and where it makes sense versus where it doesn't makes sense and you've seen examples of companies for and Supply chains vertically integrating you've seen examples of Supply chains intentionally unvertically integrating as well right um to try to give them some relief from uh you know different fluctuations in in demand I I do think that there will be Cosmos applications that like Outsource a very specific part of their you know their application stack to another entity like like philosophia for

instance um in order to cut costs and kind of focus on their their kind of core business I also wanted to put one like last little bit into the whole thing which is I and I don't think we didn't really we haven't really talked about this uh in the in in this conversation but I think it's an important thing to cover which is the basic idea of the minimum viable application specific blockchain stack um and basically uh you know the way that like I kind of uh uh uh view it as is like the minimobile thing is some way of sequencing or producing blocks some way of bridging in and out of the ecosystem

um and some sort of toolkit that gets you network effects where you can like integrate with like wallets or uh uh uh ad like other Android bulk explorers and stuff like that you don't have to like essentially create the entire universe for nothing um those that's what I kind of view as the minimum viable uh stack and the question like what you're seeing is you know the cosmos SDK was the first bit of a viable like the cosmos stack was sort of the first minimum viable application specific blockchain stack you are seeing right now the rise of other minimum and so I don't know that I would say that like there's a big difference between like I expect there to be big differences between like different

instances of minimum viable Stacks but what the problem is that we're still at a point where we can't really deployed yet to anyone else having a minimal viable stack and like compare and contrast it against the cosmosis it's a very good point very good point Zach and a great place to end and guys uh Zach and Sam thank you so much for this has been a great episode where I think we've really dug into the Nuance of some pretty some pretty fascinating topics and and I think you've given miles and I some some ideas uh for that we're going to explore for the rest of the season so um guys if folks want to find out more about you or follow the work that you do or what's the best way for for people to to find out more about you both well um you can find me on Twitter at hxrts

um I'm around uh also you know pretty available in in TMS if you want to learn more about the cosmos ecosystems um I'm at z-m-a-n-i-a-n uh on Twitter uh that's usually the best place to find me still uh for as long as Twitter stays up oh boy all right ending on a dire note there guys thank you so much this has been a fascinating conversation yeah I'll have to do it again soon all right buddy that was a great great episode what'd you think really enjoyed uh really enjoyed talking to Sam and Zacky um I think you know we we dove into the weeds quite a bit on the first episode with with sunny um and Dimitri and I think you know this this balance it out a little you know pretty well by by kind of zooming out a

little bit um and thinking about you know how to apply a lot of these Concepts to what we see of the forces of traditional companies um in marketplaces and you know I think aggregation theory is a great framework to think about this stuff um and you know Israeli kind of talked about I think in this crypto native space well I think we came into this episode hoping to sort of reorient our our listeners to from this sort of implicit bias maybe towards like from like towards like fat protocols to one to a lens of the world where really you know value might have accrue to the the kind of app chain you know close proximity to the consumer and then everything below might get commoditized but and I think we did that but

ultimately I really enjoyed this episode because I think we took a pretty nuanced look at that whole value chain and examined it through the like a more traditional lens that you might look at Supply chains and exactly those those different forces that that interact on Supply chains and you you get so little of that in crypto because there's kind of this you know for so long it's been this magical internet money and normal forces of supply and demand don't uh you know necessarily apply but I think you and I don't don't have that view and that you can apply you know somewhat traditional Frameworks or analytical Frameworks to where values can accrue to this space and I think we we had some pretty fun uh you know discussions around that in this episode yeah yeah

absolutely I mean we I think I totally agree that that most crypto protocols um both applications and NL ones are marketplaces to some degree um you know matching matching supply and demand and we got into really I think what is you know an observation that I think the focus to date has really been about uh controlling the supply side um whereas where we think this is going is is moving towards these apps wanting to control and really the apps that own the demand side will eventually have the leverage right over the general these general purpose platforms um and and how that plays out and you know whether they're expanding up into wallets or down into chains you know I

think that teasing out a lot of this stuff has been was very very interesting I think there's been an implicit uh there's been a bias in crypto towards making Supply driven arguments when it comes to investment and maybe you can trace that to bitcoin and it's hard cap 21 million dollars 21 million coin Supply and you sort of see that barred a little bit I know you think about this a little differently but you know in ultrasound money the eth meme and there's been an enormous Focus towards restricting the supply of eth and you know frankly if you look at some of the most compelling outcomes in web 2 they've been demand based stories and controlling demand as opposed to supply yeah 100 and I think I think speaking

about this in the context of block Space is really interesting um because yeah I do see you know a future where there could be a premium of block space which I consider to be you know Supply on the uh for say ethereum L1 but at the same time you know the apps are going to want to get more opinionated and specialized with their own block space uh both to improve the product the user experience and and also to improve their amount of value accrual um and so I think there's going to be you know it's a natural tension that is kind of uh the motivation for this whole season right um and it yeah and block space premium is a pretty interesting way of framing that and there's sort of generalized

block space you know which ethereum sort of has and then there's app specific block space which is what works blowing with app chains and frankly you know I think the jury's still out on which which bet which type of block space accrues a higher premium right you could see there being like Ultra Premium block space for for decentralized exchanges for instance and maybe that's ultimately because it's specialized in you know that block space you tend to have more granularity and control over fees maybe that ultimately ends up funnily being more more valuable than than any sort of generalized block space we just don't know yeah I think I view premium block space more first of all there will not be more than one I think environment for premium block space and I think about it

more on the on the axes of like security and censorship resistance um and really making no trade-offs right to the benefit of the applications because that's what you're prioritizing um and I do think that there's probably a space for that but how much of you know the Val how much value that premium block space captures versus the rest of the market um and you know how much activity moves to these specialized you know app environments where they can optimize the Box space for their use case uh also impacts you know how much of that general purpose ethyl one you know uh it captures um in terms of the the overall Market Gap with the space yeah you know the the last point that that uh you know I kind

of make on on this episode was I I really agreed with the the PO I forget if Zacky or Sam made this point that we are just so far from sort of the end state of crypto and the vast majority of of the use case uh is is people just doing Financial speculation today so it's you know I think there's this sort of consensus agreement that this is not ultimately how people are going to be using crypto but it's very difficult to get a line of sight to what that end state does look like because we're a little ways off from there right now oh no I totally agree um and you know say we were in a world where there was thousands of applications decentralized applications that that are used by billions of users um you know I think I think at that

point it does you can see like in an obvious aggregator in terms of something like a super wallet right that is basically in the same way that search helps people sift through the internet and basically go from intent to action and faster right you can see a wallet kind of doing the same thing but today there's just not enough valuable activity to aggregate um and so you know I think there's still a lot of a lot of Green Space there both you know at the application Level um and and maybe the applications become their own wallets right um so you know I think that we are still so so extremely early and and it's hard for an aggregator to exist when you're

this early 100 you know the the aggregate the wallet idea is an interesting one because you and I sort of went into this episode thinking if there was something that looked like a web 2 aggregator in web3 or crypto it's probably wallets J you know Jason's been talking a lot and he did an episode on Empire about uh wallet Wars and it's just funny because hey you can go all the way back people were talking about the wild Wars in 2017 or 2016. uh you know you can actually find threads on Twitter and on the one hand you haven't really seen that play out like a lot of people thought that wallets are going to be more important than they are on the other hand you look at something like metamask and it has early indications of it being sticky for no offense to

metamask being there are a lot of flaws with the product I would say like nothing nothing you know like uh it definitely could be improved on and yet it still tends to be a very sticky thing that like I'm the same way like metamask was my first wallet and I use it yeah I totally agree and that's why I think that you know differentiating in terms of a general purpose wallet hasn't really been able to you know unseat metamask but the real like signal will be the success of uniswap wallet and openc wallet right if some large percentage of all metamask transactions are attributed to those applications and then those applications build their own wallet right that that

will be a a big signal to see just how much how many of those users migrate right to the apps that they're using most um and the wallets for those apps right in the same way that I think dydx is also a big you know uh validation or invalidation signal um and we knew dydx was a big deal going into this this season it was interesting to hear from from Zacky and Sam just about how much of a uh a Bellwether or an indicate forward-looking indicator this is for sort of the whole app chain thesis so no pressure Antonio exactly we're all counting on you got the hopes of a whole sector and idea riding on you buddy so uh yeah good luck

miles you want to talk a little bit you know we we started to get into the very last portion of the discussion about sort of trade-offs between a full stack monolithic app Chain versus one that more modular sort of the Celestia model via versus one that's sort of built on ethereum we were kind of getting into that uh design space uh difference and that's that's we're going to be focusing on primarily in the next episode I don't know if you want to tease it a little bit yeah yeah I think you know as I've said before that's really kind of digging into um the differences in terms of what you can do with the product uh what you can get in terms of value accrual between a full stack app chain uh to a modular you know app specific roll-up uh to an app specific roll up on ethereum either L2

or L3 that's one of the biggest questions for me in this show and I think that you know Zaki I liked his framework that he used around you know a minimum viable stack right and and so far we've only seen you know apps and action that are using the cosmos uh stack but soon we'll see you know a basically a stack emerge on the modular side um and we're seeing lots more I would say roll-up Stacks emerge on the ethereum side whether it's Bedrock from optimism or slush that's you know doing uh building a an SDK for for Stark net um Roll-Ups and just you know kind of mapping what the benefits are from the

cosmos SDK full stack uh to those other minimum viable stacks and seeing just how many you know you can bring with you um because frankly like it the cosmos full stack is expensive and it is risky um and we don't have a lot to compare it to yet so you know I think we'll we'll talk to some folks on the next on the next next episode who are in a great position to kind of dive into that and I'm really excited it's it you know it's expensive and it's risky and it's it's hard you know the same Sam kind of made this point right at the end of that last episode of of you might actually see some some chains that started out trying to be full stack you know monolithic app chains they might actually say Hey you

know if you've got you know kind of execution and product and then there's like all these things that need to happen underneath settlement data availability consensus maybe we don't really want to do all that because like bootstrapping a network of validators is not what we do really well we want to just focus on that that top app chain and lots of examples of this again in traditional Supply chains of like just wanting to focus on the one area that you want to play and you have a confidence and you'll outperform so um it'll be interesting to see if we unverticalized or unbundle unbundled unbundle thank you very much yeah I think I think it'll be interesting to see how much of the you know the the stack is unbundled and and also I think which uh Zachy got into is how how much horizontal expansion will

app chains you know be able to accomplish without you know I guess dealing with scope creep and and really deteriorating you know the ux of the core product um like we're seeing a lot of sector specific you know app environments or app chains now um so yeah I think I think lots more to lots more to come on this season but next one should be a great one too agreed buddy all right I'll see you for that episode sounds good see ya [Music]