DeFi by Design EP#113: Adaptive Scaling & Modular Thesis with Injective CEO, Eric Chen

Transcript

[Music] welcome to D5 by Design where we talk all things blockchain and cryptocurrency while striving to educate Empower and enrich what's going on guys welcome back to G5 by Design episode 113 get ready to get modular pilled because there's nothing else on the agenda because we're going to istan buul and I'm feeling istan bullish I got my more layer shirt on are you kidding me we're scaling Rob more layers please how you doing Rob and joke with me because I posted a tweet yesterday that said layers don't exist anymore because we're in modular world now and this is no longer layer one

layer two layer n blockchain space we're now in this world where modules of blockchains just exist in space and we're going to the Moon of scalability we're definitely scaling um comes with tradeoffs with interoperability so I'm definitely curious to hear more about how injective handles those tradeoffs I'm just gonna I'm just going to sit back and uh and learn that's what I enjoy most about these podcast and Eric um I'm gonna pass it over to you to to just kind of introduce yourself how you're feeling um you really need no introduction we've all seen injected on the top uh top 50 of coin market cap so kind of speaks for itself nowadays but we won't we won't speak too much about the price uh and the token um we we are really here to to learn more about the

the technology and uh the contri contributions you're making to the blockchain space so welcome to the show and uh it's a pleasure to have you on yeah thanks so much for having me um so yeah my name is Eric Chen I'm the co-founder CEO of injective labs so we're like a core contributor to the injective E system and so what injective is is since basically get a l one blockchain uh that focuses on financial applications um it has a lot of you know uh modules that allows Avers to Plug and Play and synergize with each other like the or of module to support spot and derivatives exchanges Oracle modules Etc and injective itself is you fairly scalable with uh you know roughly uh 20,000 TPS if that sad the matter C says

um and also you know uh have done you know very very uh uh tremendous amount of growth for the past year or so uh with you know helixes uh you know clocking in I think you know 13 or 14 billion dollar volume to date and also more than 300 or 400 transactions uh you know to date yeah and that's that's uh very very uh Great accomplishments thus far and I'm sure that you're proud of yourself but also looking at the road map like damn we have a lot of work to do um and just to kind of rewind a little bit uh Rob and I were actually part of the University of Florida blockchain Club um you know and and when we were getting ready for this podcast we we we saw that you were pretty involved with the NYU

blockchain labs and we're kind of like you know doing some risk management assessments at various funds and things and I'm just curious like how you how you went from like okay I'm GNA work you know in different funds and kind of do like uh kind of Quant things okay getting like the university education angle and then you know now we're here in injected it's just it's quite an interesting path yeah so so I think my student Club days back in the day was M mainly focusing on like research and stuff like that I don't think the space was like mature enough where uh you know alberth make trading on like Quan trading I remember you know back in the day the real like you know automative strategy uh that everyone participated was really

like the Kimi premium so I think back then you know um um most of M share and most of participation w space are generally on a research Direction and a lot more heavily on you know the cryptography uh side these days you know like unfortunately you know given how much sure like spaces relative to what it was like 2018 or 2019 certainly cryptography has been you know kind like the you know uh on the sidelines you know behind the curtains uh and you know it's it has already you know like uh uh provided and realized uh so much of the infrastructure that we use uh in Troy today yeah super cool um appreciate that yeah Robin I had similar experience with

with the uh while all be probably less Technical and less in depth we were quite young at the time 177 16 18 something like that you know we were kind of doing this this education angle on on in the club and that led to just a whole different you know insights into this faith and just what's possible and now there's you know there's there's tons of college Dows and uh you know there's there's Berkeley ysh and E Etc participating in optimism governance and a governance and and some interesting things happening at the college level but we won't we won't dive too deep into that uh but I'd like to go to to to the library and open up your mind a little bit as to where um the role of the VM

has started to play in your uh kind of Journey with injective uh you guys are currently using SPM which I think our audience is probably pretty new to and the community is as a whole because it's quite a bit of a new development and also evf so obviously compatible with kind of both so um yeah I guess let's just start at at the foundational level and then kind of dive deeper and deeper but how does how are these pieces put together um you know with these with these different VMS and you know where does injected fit and then I think we we can dive further yeah so so all these uh you know recent launches around like Cascade which is the you knowm RAB and also like IMM which is you know like evm rollup

are all you know part of this initiative that's called like electron chains and this means that you know it's a alternative environment or R layer or you know uh a network that you know connects or a secure cured by injective in some way what are injective Ser servicing like the verifier D layer uh Etc and you know these rollups are kind of like these specialized environments or you know uh kind of like distinct and separate environments from you know what's typically offered within the D injective core chain um and acting as kind of like these surrounding surrounding like electrons that you know uh um kind of like a uh provide a lot more utility provide a lot more like uh scale and also provide a lot more you

know interesting use cases uh for you know the overall injective ecosystem and I think what's interesting about you know uh injective role in this uh and why you know like uh you know we had injective Labs decided to pursue this direction of you know adding out uh adding on more roller lers and you know pursuing a you know more like a pathway towards a modular approach is that basically uh injective uh as a chain of cell it's a fairly specialized environment and has you know kind of like a kind of peculiar like smart contract by you know environment called you know km which is uh cly you know primarily rust based um you know isn't really like a you know full full on like

transp H from evm or like VM um and you know it's a it's a bit of a you know work to you know get used to it and you know create like scalable and you know interesting applications for it and now on top of that there's you know all these like highly efficient and optim Iz modules like exchange module that can support a lot of uh you know uh traits uh for every single block being able to match them the fishing rate uh in like as you know a single like aggregated way and you're going to see you know a lot of these uh adaps within the COS LM ler also you know jamming in orders you know uh um you know creating different types of like oning liquidity uh um kind of like protocols or you know creating

different types of uh uh Financial products or protocols uh to support and synergize UNS and this basically means that there's going to be like a huge am of you know uh traffic uh coming through uh into like the exchange modle and first of all on like the developer friendliness side um the these type of R environments kind of serve as this uh I would say like more like a high level um like a onboarding environment for different types of developers are like hey like I'm not super familiar with ch and I'm certainly not familiar with rust but hey here's like a you know egm environment where you know I Jame used to now if you look at it from like the most Astra World level way uh it just

feels like a you know like a alternative envirment with different Preen tiles but um in reality uh from like a you know high level developer perspective it's like hey this is something I can get started with a you know lower risk and then the same case for like SDF which is you know another set of uh um kind like an environment that has very very interesting properties uh distinct from like the EPM layer um that is actually you know a little bit more aligned with like the injectives networks offering given that you know it's more I would say relatively uh computation uh uh uh heavy and you know storage uh obviously andje is a little bit more focused on the space aspect of things by supporting like a large order book Etc um but I

think you know like that's also you know where you're going to find like uh the most amount of uh uh you know hft or you know uh on kind of like uh Advanced uh developers are working on you know very very cutting much uh Financial applications and stuff like that and all these you know R environments with different types of like security properties different types of interoperability properties uh and more ordinate you know uh different types of scalability properties um allow for these sub developers to you deploy the application in the easiest way possible and I think more importantly in a secure way because they're familiar with solity or they're familiar with stdm and uh interrup and uh kind of compost with all the existing applications within the injective chain itself so right now

obviously injective doesn't really need uh uh uh or isn't really hitting the sale delity bottleneck but you know later on U it's going to transition from a onboarding tool into more of a how do I call it like a a scaling got it and you you're tying in so many of the components that we've seen exist in the space so you mentioned svm and is an svm maxi today that's obviously the Solana virtual machine and then you also mentioned that it's evm compatible and that you're leveraging cosm wasm which is inherently a cosmos infrastructure piece so I'm I'm curious what is what is special about injective that pulls all these Technologies

together yeah so so I think that's both the you know unique situation that uh you know intive uh you know electron layers are uh facing essentially these sort of tal and also like un offerings is that typically when you look at like a rollup or like a modular kit um you know unless it's you know very much an application specific layer um it's typically served as like a scal layer uh in the case of like optimism in the case of like arbitrum in the case and case um they're pretty much identical um you know environments with you know different parameters different like uh kind like tradeoffs in terms of security uh when it comes to interoperability but at the end of day it's meant to scale the layer um but for injection case it's

a slightly different problem like scalability isn't necessarily like a a problem for injective uh just yet it's going to take a while before that becomes a problem um and um the the real problem is you know inter uh interoperability between you know these very very distinct environments uh and figure out a way to like you know uh unify and uh you know pass messages you know secure messages because you know uh token transfer T swap are all just messages at the end of the day um and then like uh you know making sure that all these applications can compose with each other in the most natural way and so through these rollups it's almost like a you know uh um kind of like a battle battle testing of you know building these interoperability Suites

uh building all these you know like U kind of like a base layer tools that developers can leverage to make sure that you know like their solity contract can you know send very direct instructions into custom lawm contracts or or you know they uh uh you know so programs can you know create very very high frequency signals towards uh you know orders on top of the injective and basically you know like it's a slightly different Direction with like a slightly different Focus where you we worry less about you know the scaling aspect of things um basically injective a oneish second block time um kind of uh uh alleviates a lot of the concern around like challenge period uh and the size of block space Also you know allows for it

but I think how wasum also had a bunch of like added benefits from it but at the same time it's like hey like this is a completely different language this a completely a different environment if you go down to the B Cod it's like pretty much the same but like you know um uh how do you make this developer layer um uh which effectively speaking it's like almost like a uxx but like devx uh layer you know uh much more efficient much more um uh usable and this kind of uh and fly you know eventually like a very very much uh seamless frictional uh you know borderless you know multi-chain experience where users might not even realize they're going from like inject J

to salana to ethereum to whatever later um or you know from like arbitrum to salana or like a salana rollup solution um uh using the same side of keys by probably like a account abstraction or something like that uh and hopping between five applications in the process but in from user end it's just like oh like after five seconds like you my my my uh transaction is all said and done and I describe and a pool or something like that you You' shed a very clear light on the tradeoffs in the design space for for modular uh blockchains it it's something that we've heard time and time again that these modular designs allow for this this great scalability but at the at the tradeoff of of interop

um and it's something that you know I think we've we've kind of seen this like Cambrian explosion of chains that have have launched in the modular space and they're leading to an increase in the amount of in of block space and the amount of like throughput that we can push through some of these some of these networks uh but again it's like it's further fragmentation and then and then you guys are now building the interoperability sweet I I I think that's a really a really cool term for it too because it's it's almost like these these like pieces of infrastructure structure that pull together those fragmentations and kind of make sense of those those modular chains so I'm curious like kind of looking at this lens this modularity lens where you can now interrupt across

these different modular apps I'm curious how this how this may change something that we we've spoke about and I hope you're familiar with with which is uh the fat application thesis and the fat protocol thesis I'm curious if if you've heard about this if you have like clear thoughts and and how do you think the modular landscape changes the fat protocol thesis and and where value flows to the top of the stack the bottom of the stack it's like layers no longer exist so like where does Valley flow in this modular landscape so so uh I for for the longest time I've been kind of like not heavily subscribed to the fat protocol thesis um because I

remember when it came out when it was like a kind of like like hotly discussed Topic in like 2018 or 2019 where you know this is a new paradigm and like you know it's a fat protocol and stuff like that and like you know application doesn't like uh gal value the underlying protol uh do I think it's just the same phenomenon of seent Industry where like infrastructure first in an application because if you look at like uh you know uh the internet era like to you know uh um you know what it is today which is primarily like you know a bunch of social media apps or something like that um that are uh you know driving in terms of marchet value um there are like obviously like exceptions

in the sense but the the point being that like back in the day um primarily like those who achieved very very high valuation and those uh who have you know like who who guard like the most activity are server Farms you infrastructure providers um you know all those like uh you know uh TMT uh uh you know companies um and then you know now it's primarily like application sided like um you know that's infrastructure side has matured a lot and the valuation driver is you know mainly resing on application so right now we're also kind of seeing like a transition from fat protocol to Fat application um you don't see as many you

know like uh uh like a ethereum killer out there you don't see that many you know chains trying to be their own like stand alone ecosystem infastructure with their own wallet solution like that um what you do see is a lot of successful applications getting a lot of user interest you know driving like a lot more you know user Focus adoption rather than saying that we're building an infrastructure for applications that's going of goes a lot easier um and for for inje case that was it kind of like uh baked in uh pretty much since day one where it's like um for a lot of the application on top of injecting they get this very interesting benefit of uh uh connecting with gabas bya you know uh uh um basically signing a message from the

ethereum uh you know main net and you know that becomes a Val transaction on top of injective about relay and uh they're sing like uh gas fees uh where you know basically the gas fees is front to by the applications themselves uh so users will have to pay for it um and also you know like all the it uses the same signature scheme so on a surface level it means that the user using like a application on top of injective if that if the developer chooses the user will never realize that they're using application on top of injected for them it's just like oh I deposit which is a bridging process um and then it becomes like a balance on this uh platform and then you know I can just you know do whatever transaction I want primarily s

like training and Stu like that and uh uh I can just you know withdraw which is bridging it out or bridging it back to Source chain uh and then you know call a Daye but in reality you know in the back end was you know your application paying for the gas fee Florida user you they're relaying you know like the sign message uh from uh sio data from like ethereum M that to uh injective um and you know they're uh they're you know supporing the entire infrastructure along the way um so yeah I think fat application is going to be like uh kind of like economical term pretty soonish and it's G to be interesting to see um because I think that's really where the excitement is and also interestingly enough

objective as this ecosystem really started like kind of application tokus right where you know uh it's really just you know one single application that was presented to the general user um uh as you know even though the infrastructure is like pretty generalized you can support a bunch of other things and so like that but initially it interfaced uh the users with like a a sample application basically as in to exchange and this is uh you know users treat uh injective as ecosystem initially as as as that application and there are many reasons why you know uh us said injected WBS you know approach injected exchange that way but I think you know starting off as uh application focused building like an infrastructure or friendly to um

application developers rather than you know inuring value to the underlying prol itself is first of all a significant advantage in growth against you know like uh uh other know eniz environment but more importantly it's uh it's to you know understand you know the needs and hauns of a developer in the most intimate way possible and I think the modular stack these days it's another step towards you know empowering application right because uh historically one of the biggest problem uh for you know modular architecture is uh the symmetrical security uh um amongst you know different layers and more importantly you know the data passing between and coincidental uh that's basically been like the whole

thing that Cosmos was trying to address himself and you know they present a very very elegant solution that us IBC and it's you know very very scalable from like an infrastructure maintenance level uh um and it's also you know permiss in terms adding new layer and most importantly it's you know uh almost like a alart style where if you trust this uh you know like a bir directional uh chain to basically this destination you're going uh to then you know you can trust this BR that goes between them because it's practically speaking secure about the BR themselves so secure about gen themselves so that actually you know using that it actually solve a lot of interoperability blockers that enables like a modular layer um and you know like a lot of the toolkits and Suites uh

that we uh adopt uh for you know interoperability oftentimes are also powered by IBC right and you know a lot of our work uh that we're trying to do is to create kind of this interface that allows for you know ethereum contracts uh to you know create and you know transmit like IBC compatible messages and now the more like ia uh you know compatible messages uh to basically go through this you know Highway a secure Highway of uh message passing where we don't have to worry about too many things per se and not saying that it's the easiest but it's you know it's uh it's you know one of the biggest um kind like Step Up in terms of building these

interoperability Solutions yeah I mean to to put this into perspective uh we we had the fact protocol thesis with how like a fear and these and these uh of ideas where like these underlying layers would would through a bunch of value thousand almost before we had def5 and a bunch of like popular apps and then now I think to your point uh you know when you think of popular apps you think of social media applications you think of applications that your pH you think of different things that you use on the internet those AR know you don't really think of the https or the email server or the technology layer under it uh while that layer is definitely aing value in the sense what the what the consumer what

the where where all the economic activity is happening is within that app obviously it's if if it's on a c if it's in a certain Network that Network thus can can you know kind of show F picture thing and say hey know this is all the activity that we have but from a perspective of like investing I think of a perspective of uh onboarding I think it's definitely a matter of the apps and then furthermore kind of all your point about where where um you know injective lies in the in the uh overall ecosystem of of of inop me there's some really really good points there and one specific that you made was uh regards to ethereum uh as well and kind of like this this idea of being a

settlement layer um you know alongside in injective so could you please uh walk us through what a modular stack using injective either cring looks like or could look like so okay inject those this L1 it's the settlement layer if I'm not mistaken then there's different you know you can have different VMS different rollups uh data availability are all these compatible in the sense of like the big players in in the game now what is a current stack if there is one now and or what could one look like separating from monolithic change to to a modular or does injective just do all this itself yeah I I think that's you know like starting off um uh you know kind of

like a uh almost like a you know up to the you know a rollup developer um in a sense like you know injective w space isn't you know that expensive be like etherum Etc so often times you know like we see like inje types not only C elements you know verifier so but also like the TA layer um because you know like posting all the proofs some you know the S transition isn't you know too cost prohibitive and you know currently I I believe like IBM is uh you know doing that um but you know later on you know it's it's a very very like hands off SL natural Dynamics for it's like you know let's say the cost of block space within injective goes up a lot more and uh you know submitting these

proof becomes more expensive and uh you know like the the rollup stock can easily migrate to like a DA or that's you know like a off chain essentially um then you know it's uh it's up to developer themselves so you know try to like move it off um so so I think the advantage of building like a ro player where you know scalability isn't you know like the primary uh objective here per se uh but rather you know offering more features and tool set in expanding the offering um is that you know you you don't have to have any like uh uh uh difficult problems to you know to heavily optimize for and instead you focus on your experience uh because often times let's say you know there are

10 you know very very talented Engineers working on uh a specific rooll of solution for evm ETC um the time allocation could be like 2080 uh between you know 20% on building up like a well uh out of the box uh you know uh tool Suites and you know contracts where developers can utilize so you know uh inter and you know building these Bri Solutions uh or like Atomic swap Solutions uh so that you know it's not just message passing but also you know like the endend solution uh but the 80% is you know figuring out like a economically sensible uh you know like a secure own you know highly scalable layer um that doesn't have you know liveness or correctness Maes Etc um and

currently you know like uh the the majority of a lot of these scoll up layers you know trying to uh build these rollups on top of if the and has been you know mainly competitive on that side on like how do we you know make sure like the challenge period for let's say like optimism is on shorts possible or how do we build solutions to like you know uh navigate that and like uh or like you know like uh how do we lower like the cost of maintaining like a z rollup or something that and um yeah like uh um obviously celesia and other you know like modular architecture also gets to you know focus on leer experience of We Are started off with let's say like this highly modular you know symmetrical and secure layer where

it's basically you know uh uh separating you know all these you know like components within the blockchain itself and you know allowing you know a certain like environment and later to like specialize on this single thing say da Etc and at the same time you know keeping them like so closely sync to a degree of almost like uh Atomic um then you didn't like worry about like you know how to make the developer stents seeing how you be all these linkages and Stu like that um and yeah like I I think you know um that is uh that is one of of the greatest things about the Moor sag is that you know when you don't have to you know initially consider like all the scalability issues and all the economic factors of that you free up a lot of

times to build a truly you know like a userfriendly developer friendly application yeah two pun fact say did you know did you guys know that there's not been a a a fraud proof yet on any RS because there that's exact no yeah and there's no there is highly unlikely for a centralized sequencer to be able to commit any any malicious action or really be incentivized to um and then the other one is that that sday optimistic fraud proof window uh was actually an arbitrary uh time it was just ah yeah it sounds good and like you said Eric there's um there's been some proposals I think from Stephen goldfeld there or from um one of the arbit guys

offchain Labs who's starting to push on trying to to lower that which which I think is super interesting and well uh it's not like super arbitrary per like you know between like what say like five and like nine days seven days probably just pre- arbitrary but like you do have like a minimum where like the you know the challenge per needs to like the challenge process like you know for some that's interactive for some that's not interactive need to fully uh play out and that's depending on you know the block space Plus like a little bit theway because you need other transtion as well um so to make it happen so yeah like the debate around you know lowering the CH the I mean like at the end day Pro proof is permissioned yeah you can

make the challenge pretty much just like one block but like um uh you know like in the case of like lowering it and opening up um there you know like it's so it's better for it for ferum to wait for um you know trial proof to be Fully live um and you know battle tested with malicious actors rund did um before you you know lower that definitely definitely so the uh kind switch Hees back to the uh injective side I mean from my perspective it seems like you guys can handle the da layer uh because the the call data storage is are effectively not very expensive right now when compared

to something like ethereum um you guys can use VMS other VMS Salada ebm kbm Mo VM uh W your VM here um and then uh teams could come in and build a top different rollups to using uh injective as a settlement layer and also as a DA layer as you said as long as the cost don't become too prohibited now specifically on the da layer side as as how uh injective handles it we don't have to get too too technical but I just want to understand that the trade-off that you guys make if we haven't already established it but on the on the basically gas per bite situation and the onchain and offchain storage so are do

you guys have an achain data storage committee secured by the ejected nodes or do you guys store these data o chains and something like a validium or like some other mechanism and then where if you can draw like a I saw this graph of celesia had like this x AIS of gas per bite and Y AIS of decentralization and roll up were like at the very top vums were like at the bottom and then in the middle was DC a data ability community and then uh Celestia was kind of there in the middle uh I'm curious where you guys where you think that you find yourself in the basically cost and decentralization parameter on your data availability situation I guess yeah um

so basically the interesting part about injective and this is one like nice little add of benefit for you know inj as of blockchain as inem not being uh generalized is that uh without sounding like you know like a you know a car washing guy but like you know it's it doesn't really need to subscribe to that type of bound um being like the point being that like you know the cosm smart contct envirment still substracts a lot and as a matter of fact you know like it's pretty similar you know to any type of standard like smart Hing environment like ethereum um and it might be slightly cheaper uh just given that you know it's Wim based and you know uh

there are a few parts are optimized um for example like tilting transfer utilizing a module which you know uh doesn't require that like smart contract operations Etc um other modules and that's why like the out of Bo and play module is like so important is that it has its all like customized you know like a gas Tran uh where you know a specific application is kind of deterministic in terms of its computational costs um you know in in you know Cosmos is basically like a t of message uh and this message has um kind like a Bound in terms of you know or like an expected Bound in terms of like how much operation uh or like the cost of it and this means that you when a developers Implement these sub of modules and stuff like that they have a

very good guess on you know the cost of uh that specific message the lower upper bound if they don't have a good guess they probably need to like uh um you know like like adjust that because that would would be kind of a mistake basically but um yeah the point being that like um that's why like you know for example like creating like a trade posing a order canceling order this really cost much uh on injection and that's why you know applications you know can front the Gast stre for an user because it's not expensive to do it it's just a little bit it's just you know you don't want all the users to go through a hassle of acquiring the native to to pay for gas um and then on the other hand you know but C launching uh uh you know uh contracts and transactions oftentimes

other than you know like a very specific hook to interact with invaded modules uh all subscrib to the same type of like down environment and there's you know no easy way around it and if there is like you're still making like a very significant trade-off that you probably regret later on and you have to adjust um so yeah like that's why like uh you know role of is has a very very important strategic you know position within the injective ecosystem is that you know injective as an ecosystem is you know more specialized more sector focus more specific where you know you can totally do everything you do on ethereum on top of the cosm L but that's that doesn't have like a significant Advantage uh over top um you know it's not going to get you know uh amazingly

cheaper unless you want to sacrifice decentralization it's not going to you know have a lot more feature offer per se um especially you know as ethereum itself like improves right um but what you're going to get is basically the set of specialized modules that has you know like a um kind of like almost a constant scale uh uh gas cost Etc that um you know you can always have a very stable estimation and you know stable uh um kind of like uh cost parameter for application Developers for the users and more importantly because these are belum modules for uh within the chain itself um it's almost like a specializ PR file where you know the entire chain and the

state orders are aligned to you know uh uh um basically you know optimize for it and self-evolve in a way like if it uh if it can be scaled better if it can be optimized further they will you know try to make these messages more optimized whereas you know if let's say you know like some specific contract in ciz them uh because of the you know the way the gas economy is designed it's not exactly most friendly for them and they say that oh like this is super unfriendly you know like I want to make a change on like how the gas is calculated um then like you know it's it's going to fall in De of years like it's it's not not really going to be like uh uh prioritized by all the stakeholders um as a result and that's kind of General problem within you know generalizing

margins um so yeah like like uh that's kind of an interesting situation but that's also why like you know the trade-off here is that you don't uh you you you you don't have like a generalized Focus you focus on you know like the financial application part yeah yeah so I mean uh from from that explanation the only reason for me to understand as to why one or one of the key reasons would be to use a SE data de like a Celestia XYZ would be to just add a different piece of the puzzle to the equation or to this alicart menu I really I don't think you're describing module you said that but I really lik it because it's almost like a pluging play a pick and choose what what do you want it's like a four course meal but but it it would be using

an alternative da just in the sake of uh likely reducing single pointed failure and just expanding decentralization uh as well as onboarding that community of you know potential users or fans or marketing etc etc um in there which is which definitely interesting um yeah I mean I think we can get out of the weeds here and uh and kind of go more to where where you guys are are working on I just had one last thing about the VMS um it's a lot of VM powerful because it allows as you know rust to build daps within other ecos systems or is there a technological reason as to why uh you

guys are bullish on svm and why you are so keen on kind of getting it into injective yeah so so so I think Salada VM um is powerful in a sense like at the end day like everything converges and everything utilizes right um there's no there's nothing that you know like uh on uh um evm you know can like like there's nothing like you know just fundamentally you know better for evm more like svm it's just say you know like that they're just picking a tradeoff at the end of day now there could be like some optimization or some improvement that can make it marginally better but uh uh there's no like fundamental Advantage but what's fundamental about like uh the SDM layer is that it between like kind of like the space and computation um um

like a slider uh they picked uh you know very much uh through the uh you know like bias towards uh computation and this brings a lot of added B FS Etc because as long as you manage you know your storage uh uh again like you know space correctly uh these applications will have the fundamental like advantage over like uh EDM applications um and this basically means that uh for sure obviously for evm games you know they they're designed to be highly generalized they try to like pick the mean uh or like the middle ground between every single like consideration and trade-off and as a result you know you make uh a certain like uh vertical

application kind of unhappy and for Sol the reason why like initially it had a lot of uh you know uh U kind of passion or a lot of adoption amongst like you know hft uh developers uh you know building their def applications or you know a lot of uh you know like a D5 protocols is that uh they they just allow for a lot more uh uh you know computation to go through like a reasonable cost especially with the you know parallelization aspect of things across like know like a different types of tradeoff that you know they don't necessarily you know need to worry about and uh yeah like ra like a whole new sector of like uh you know def applications that totally would have been like economically like inviable on

top of the ethereum unless you built like a you know a customiz mup which loses composability and yeah so so so so by picking and choosing uh you know procing the tradeoff Spectrum svm you know kind of created this uh room for like a lot more like opportunities and like uh um a lot more like uh kind of like a sector of application to really Thrive and I think interestingly now it's you know uh also like nft which uh uh people have been like very passionate about it's also kind of ridiculous how like Ming ft costs like so much during a Heyday on ethereum hundreds of dollars in gas thousands so if I was build the fastest modular stack would I build injective as my settlement layer and my

da layer spvm as my as my BM and then you know build build a top that stack yeah if you want to build a layer that's uh you know financially focused but computational like hftt speed if you're building something Financial related certainly like injector is the best fit obviously like you know like if you're building something that doesn't necessarily have to do much with like the core daps on top of injective or the modules um you know it'll equalize eventually um even though you know we are at an advantage in terms of like speed and also like uh block space and that's because you know like we have a sector Focus right um and uh you know like if you like try to

apply something general like it's always going to be like a musical chair of chains where it's like oh etherum is too expensive let's go to BC BC is too expensive let's go to you know like like somewhere else right so so I think you know that part it it doesn't have like a long-term like uh you know like uh kind of like moped around like your advantage for like oh like there's you know like you only do it on top of in if you're building something that's you know like non-financially related but you know if you are building something that's you know uh whether it be like a E5 protocol or you know something more institutional Etc like you thereby have a vote where like you know there's literally a lot of things that you can't find or would be economically inviable on other chains I got a quick question like at

aside from so so the VM narrative uh you guys are are all over in the execution space with uh with the salon of V thinking about about inject from the from from the settlement perspective I'm curious like do you guys given that you you host Your settlement layer that that settles execution from these VMS do you have proposer Builder separation and collect me is that something that you guys do at at the chain level or the settlement level oh so so that is a bit under it works with uh collaboration with skip um on that aspect of things that becomes kind of like an interesting question because it also depends on you know like like the uh amount of loading

power that are subscribed to that type of uh PTP Network um if it's like a low amount you know like uh it almost has like a know threshold sh where it's like after a certain threshold um that that uh that basically always becomes a canonical and the standard expectations of it so I would say like the the the uh the current answer is that just like you know there's no way to tell kind of preemptively um uh because is an interesting dynamic because one on one side like the uh the majority of transaction on top of injectives are you know orders and trades and inherently there's little to no um you know like value in you know having MDB production on these trading uh and and that's just

for context here that's gu exchange module uses uh batch option to remove uh any Tye of front running or you know me issues and so that kind of uh you know makes so that you know validators have somewhat of a partial incentive to you know like subscribe into those uh protocol but at the same time you know it depends on your heal as pick it up Etc so that that is kind of like a you know we'll see type of situation um but I'm overall very very excited about like uh what what SK is working on because there are far more you know type of applications and we we already bring some of them a lot like for example like liquidation auctioning Etc that can be like you know super super useful um

through you know their their mechanism um but yeah like uh um it's it really depends on you how the ecosystem plays out yeah it's it's fun relative to the fat pricol thesis too because it it was also a value musical chairs there for a while like especially in times of the bull uh the the applications were collecting so much in trading fees and then they had pay so much in in in gas and the blockchains and the settlement layers ended up collecting all of that me and it was almost like the trading fees were just getting shipped right over the chain so yeah again I think the modularity landscape just breaks this fat protocol thesis and we're starting to see a lot more reason for the fat

application thesis yeah is literally change of landscape us uh you know kind of like an application blockchain like uh you know like a musical chair where like you just keep moving uh lower and lower within a costu Trum almost like know fairly identical like a contract environment and then flipping a game uh on its head where it's like now it's like you know full on customizable and it really would compress the cost of uh you know uh applications trying to build um you know or you know cost of users interacting with application but more importantly um it creates kind of like this uh um bottom line where it's like no matter like how empty your block space is on like you know like a certain like out Lund or evm um you know like

you're still not going to be as uh as cheap or as you know like as optimized or efficient from tier of the utility function um as you know like a modular stock yeah yeah I tried to get out of the weeds but the weeds just keep pulling me in um I'm curious um furthermore on on on this kind of idea here we didn't't touched too much on like like app cic versus generalized we got a couple of minutes um so I just want to posed the question that we had on the panel in uh nebular to you uh L1 Cosmos app chain or rollups yeah um I I obviously am quite biased on that answer but I think like

one misconception people have about uh app chains and and you know it's it also could be the case or you know this is like mainly my opinion you know um and it's certainly you know not necessarily agreed by a lot of other Foles building like uh you know uh chains within costos or like you know subscrib fully to like that apch definition but I think you know the notion of apch might be a little bit misguided uh given that like uh it's almost like a evolutionary stage um where you start off as uh you know you you start off you're searching for like environments or implementations that can satisfy your product offering and your uh applications so you know I can't build this contract on ethereum or

like this is Impractical economically inviable uh to build this on ethereum and there it makes a significant difference for this application to be its only like decentralized nwork to be its own chain uh and it makes sense for me and the value is so high that I lose all the aity the composability uh the adjacency stuer like like face the roast and more importantly you know um um deal with like a kind of like a codar problem of like a zero to one of infrastructure and like you know all the surroundings uh tool stack but like you know it it's worth it for me to like pursue that direction um then you know like uh um uh

it makes sense for you to be like an app chain and often times um The Next Step dra is that this is say uh by being like a network by being like a chain um you are no longer like an app chain because you don't have to be the only app you can be like a you know like a castle SL management where like there's you know like one big uh application and there's a lot of surrounding applications father arguably but you know uh service value to it uh you know synergized with it Etc but you really just like a you know like a sector spec specific chain or like a you know like a you know vertical Focus uh chain so I think that that's blast stage of evolution is often time like

you know uh I know a lot of people are working towards it but often times people kind of stop uh at you know hey like you are a application but have that to be a chain but you know don't really think much after like their uh next stage of evolution and um oftentimes there's also you know like challenges where it's like yeah let's just build a chain and you know uh not thinking about like you know the real advantage of it over like uh you know just being like an application with so much adjacency so uh so much dis Discovery we certainly you know uh during the early days underestimated how much work there is uh you know even after we try to make everything as you know ethereum Coke and ethereum possible to make sure that you

know we take advantage of the existing mind sham with them ethereum we also underestimated just like you know how much uh how much like uh stuff that we lose out on like you know it's almost uh it's highly costly to like keep track of you know L of staffs and public objective by like public explorers like d Etc because you know they simply can't decipher like the stats and to this day it's like you know still um and uh arguably uh incorrect um and then you know like there's still a lot of uh you know other places where you know they they built it to be ethereum because they assume the whole you know eosm is ethereum so you know you lose out on all those application Suite uh or

you you know have an increased cost of you know being included in those application Suite so it's it's tough man like uh you know being an nchain it's uh it's not like uh you know just building an app and then you you just want to be a chain um it's like you know you you've considered multitude of options and then it's a final you know like uh um plan where you know hey like there's no available option that makes sense uh based on my needs and want and uh you know I I gotta you know do the hard work and like uh build like a chain around it and then my uh my my you know like like the pros behind that is that you know I become a bigger ecosystem than just application creat this infrastructure

where a lot of other applications in lb flourish yeah I mean to to kind of wrap this up here the infrastructure dilemma is is one where you know you have to make that you have to make that choice like is is the infrastructure is it Deployable from a Dev and a financial cost perspective is my application specific is there a specializ need for this uh execution enir right so those kind of like the big two biggest points that you made in in detail there right just kind of putting those into perspective for a Dev or Builder a CEO those are the important questions um and then I think uh I think there's other builders that can sympathize with the uh understanding of uh trying to build away

from ethereum or not use the infra that currently exists you know I I notice for example even uh two examples one is just the reason why these VMS are so popular uh and these alternative VMS have grown in size and in mind shares because you know uh there's there's there's a entire code library for the last 20 years on some of these VMS that evm doesn't have there's tons of different tools and infrastructures that you can just kind on more so Plum PL and then even further I I read um this article through Neil from Eclipse who's also on this panel and you know they started Eclipse thinking about trying to build AB rollups only you know they they were trying to build to steal 10,000 rollets after several months of realizing how

much damn work it is to to to build individual execution environments time and time again uh you know they decided Well this isn't for us the 10,000 rule of future it may or may not happen but we're not going to like you know really kind of build a part of it or they're gonna focus in a different L right and kind of go towards a more general purpose environment just to kind of wrap up uh we are the rollup so I think that we can agree with your perspective but but I do appreciate what happens in the cosmos ecosystem and I think um I'm a little bit biased towards in in a sense but I think that uh next to next to ethereum is is calus with their devs their Builders their kind of technology and their overall like the whole IBC and

the level of just uh Innovation there is is is always cool to see at these different conferences and and what they come up with so yeah man thank you for coming on uh that was just my closing thoughts Rob yeah thanks so much Eric it's been a blast to learn um and uh be injected filled um as well as modularity and uh all these Concepts that we spoke about I think um it's really cool to uh highlight some of the components that are out there that are kind of catch and win and then uh pull them together so that we can get a more comprehensive understanding so thank you yeah really appreciate it thank you so much