Will zk eat the modular stack?

Transcript

welcome to the rollup today's episode is based on an article titled will ZK eat the modular stack in this episode we cover modularity we talked about ZK proofs and why the ZK ification of everything is radically improving the scalability of blockchains everywhere we talk about the evolution of the modular stack how it unbundles the conventional parts of the blockchain and we talk about ZK proofs and why they are transforming the execution and verification of blockchains themselves I hope you enjoy this video sit back relax and enjoy the show before we get to today's video please enjoy a word from

our sponsors that help make the rollup possible fuel achieves Higher throughput by utilizing a novel utxo based approach that allows for parallelized execution this design significantly accelerates transaction processing which enables fuel to handle vastly more transactions than nonp parallelized systems thanks to advancements introduced by Fuel ethereum rollups are set to be more powerful than ever before and fuel is not just enhancing ethereum rollups it's redefining them making them faster more scalable and more adaptable to meet the needs of modern applications and [Music]

developers on guys welcome back to the rollup today I'm rocking my modular haircut hands is rocking his ZK will eat modularity glasses so that makes two of us and today we're going to answer the question of will ZK eat the modular stack this was a post that Hans put out as part of equilibrium about a month ago or so um and I read it I thought it was really interesting and so I'd like to unpack it a bit more uh with the man himself so welcome in thanks for having me yes sir yes sir so maybe just briefly what is equilibrium and um kind of what are your core interests uh when it comes to crypto that kind of LED you to to this ZK eating the maer stack thesis

equilibrium is fundamentally we do low-level protocol engineering um so we help protocols build things which are safer um faster and more useful uh that's kind of the fundamentals of equu over time uh We've expanded to other areas as well so today we have a a fund which invests um we have also built a Prov Network called galot um and then a Euro stable coin but the bread and butter of equilibrium is still protocal engineering cool yeah we know the julot guys pretty well um have had quite a few conversations about them about proving um they're always they're always on top

of it so ZK eating the modular stack um you kind of came to this idea and came to this thesis so I think I'm just going to start this off and just pose the question of will ZK eat the modular stack and then let's just dive into each part um and kind of talk about this uh more specifically after we kind of understand the thesis more broadly yeah good question there's really no short answer to this the reason why we wanted to explore this crossover between cam modularity is that we saw two big trends really accelerating or continuing to accelerate and and one of them is obviously expansion of modularity so splitting things into smaller components and being able to experiment um with

like combining different components and in that way reducing the uh friction for new entrance and the second part is really the C ification of everything um so we've seen both currently existing CK protocols um expanding and developing but on the other hand you also see kind of non CK protocols integrating some part of Zer knowledge proes into their um protocols so that that's kind of the the first observation where it started and then what we really wanted to examine with this pce was which parts of the modular stack has CK already penetrated and which potential problems could potentially be solved by

introducing some Z knowledge proofs to the mix um so that that's kind of the premise of it cool and so you guys laid out two big Ideas around modularity and ZK being bullish on both of them accelerating which I think that we share that thesis um but basically modularity uh enables more experimentation more like parallel experimentation um in Innovation kind of like this Al allart menu of creating blockchains um and then you kind of have this ZK idea of expanding this experimentation e even further which comes down to what you guys called expressivity and hardness so

expressivity meaning like what can we create uh like confirmations guarantees about um in in the stack and then hardness being like like how sufficient and how hard are these guarantees um and so maybe we can talk about how modularity is kind of pushing this experimentation forward and then how you see ZK kind of like like making it more sustainable and expansive as as well yeah maybe we take a step back uh before this like computers in general are kind of designed to do three things uh they store data they compute over that data and they communicate with other

computers and humans and the way we see blockchain networks is that they expand on this current Computing Paradigm by adding additional guarantees that these three things happen in agreed way so that's kind of the premise of it and then the two things that you just mentioned expressivity and hard Ness is related to these additional guarantees so expressivity is really about what can you create guarantees about and then the hardest part is how hard are these guarantees like um obviously in today's world a lot of it is uh is a trust based model where you trust a third party or an intermediary to honor the rules um whereas with blockchain networks we can

have crypto cryptoeconomics or crypto graphy which we'd argue at least are harder harder guarantees in that sense and the way that we see modularity and CK fitting into this picture is is is really around if you have if you have the two axis and then kind of a trade-off curve along that modularity enables you to move along that trade-off curve in in in a faster way and experiment and and kind of speedrun uh experiments around designs around blockchain networks and then what CK does is expanding the the curve outwards so providing us with with

harder guarantees and to some degree adding expressivity as well um through scalability or limited privacy gotcha okay so it's like all right so we have modularity as a concept for scaling blockchains it's like we can create better systems more scalable systems we can facilitate more experimentation um but then we can also do this in a way which is verifiably uh accurate like we we can create this verifiable computation nature of the stack um a and so you kind of see this as like like the next logical step for modularity to kind of take into this ZK realm so is this to say that like optimistic guarantees are

not as important uh in the future or perhaps are a bit more important now or kind of like where do you see this this current state of modularity and then where do you see like uh when we have more ZK kind of um enabled where do you see kind of this idea of more experimentation or more expansiveness kind of coming [Music] from cartey is your home for app specific rollups at the Forefront of developing new tools and platforms that enhance the capabilities of blockchain technology carty enables applications that are flexible and built to scale

build on a full Linux environment and use traditional software libraries but in blockchain development go to cy. that's cy. to learn more about building your own app chain today good questions I think regarding your first question around better optimistic versus CK um it's really around tradeoffs today with the cost of CK being still relatively High um the proving that is it's perhaps more attractive to some applications who are very cost um cost sensitive to go with the optimistic route but that also then has has trade-offs in in terms of like you have

a longer verification periods and all that with the cost of CK coming down through both Hardware acceleration and uh optimization on the software side we believe that that tradeoff space is is going to become more difficult to argue uh so that's that's broadly our thesis that uh zero knowledge systems are going to become cheaper and faster um and when that happens it's the the the role of optimistic systems in in that case um the argument for them is weaker in our opinion I think maybe one important caveat is also that um we're not modular maximist um we think that uh it

serves important um it has important use cases uh and enables things which the integrated stack doesn't do one of them being faster experimentation on the uh infrastructure layer and or the infrastructure stack but I think fundamentally there are some use Cas cases which fits them integrated stack better and and others which fits the modeler so um ultimately I don't think it's a binary choice of ID or yeah I agree we've been hearing a lot around the okay once you know ZK is the endgame right for for in for interoperability from all these different chains kind of settling into one proof into one kind of

like Global State um for for for for messaging for pro networks for um just a variety of of use cases and we've had quite a few people the pods speaking about this um kind of going further into this modular stack like one notion that I've been trying to like push that is a bit of a misconception is that just the modular stack is just execution settlement data availability and consensus it's kind of like a misnomer here and when I saw this article and read through it I was like okay this guy gets it like you guys understand like this is going to just be expanded dramatically so maybe I'll I'll uh I'll I'll pull this up here um where you guys have this this updated uh list of um

what you think as the modular stack is in 20124 and maybe you can kind of talk us through how you guys came to this uh this kind of expansion of this of this four parts to you know this this 10 plus parts yeah definitely um and funnily enough I mean some of the push back we got for this article was about our um more complex modular stack good good to talk about how we came to it um the whole premise was really that I think the the simple modular stack the for part that's often used is good as a maybe as a simple mental model but it

doesn't really convey the complexity of modular system or the modular ecosystem today because the fact remains that whether we like it or not um it has expanded a lot and there are now protocols in or protocols building in much more Niche cases than before or focusing on much more Niche Niche use case so how we got to this uh revised stack is is obviously to some degree is subjective uh like it's our approach it's not necessarily the ultimate truth but how we vision is is really like it starts from the top uh from user operation abstraction that's like the

layer that that users touch and then goes all the way down to kind of consens settlement security and and those things which are um furest away from the user in that sense yeah so it's like um it's it's it's almost like the the unbundling of the so like the modular stack is like like you could have a monolithic stack right here just like one little box then you've got like this unbundling with the simple stack and it's almost like this is like the unbundling further of the simple stack into what has now been created over the last six to 12 months more and more infrastructure um and and you know I kind of see like one day these all of these things will be or at least a lot of these will be a part of an option

that Builders have when when when choosing like a rollup framework and choosing a Ras like you can literally pick all of these different things to build your chain exactly yeah that that's pretty much how I think about it as well so it's like if if you envision that in a year's time you log on to a Ras platform and you want to build your own chain this is kind of the buffet that you see and then you can pick different components um so for example for proving you might want to pick gelot or you might want to pick some other solution uh data publication maybe you want to go with Celestia Avail IG da so like there are different protocols Building Solutions for these small modules of the

entire stack and then what it really gives is the freedom for the developer or the the whoever is designing the blockchain to pick and match these different modules yeah yeah and I can literally Name Pro like like projects for each of these like user operation abstraction you've got like socket particle Network shared sequencer Astria espresso then you go down to the execution layers you've got all the all the VMS and you've got um uh daa data query you've got kind of like block explorers Gold Sky Etc um uh even like Dora proving of course you mentioned a couple data publication and then long long-term State Storage you know we got

K eth storage Jackal some others you know of course consensus kind of in data publication are you seeing those as still being the same like that's the same layer like is it if it's a Celestia on data publication then it's a Celestia on consensus as well well it it depends a bit how you look at it um so you could have a rollup that uses Celestia for just data publication but then settles to ethereum or you could have a sovereign rollup that uses Celestia for both data publication and consensus so I don't think it's necessarily that one module needs to serve or one protocol needs to serve

just one model um and we are seeing this to some degree that you know Protocols are starting to rebundle yeah and Shrine more it's like exactly the monolithic thesis and the mazra thesis convert converge and rushian movement are calling this integrated we we're just calling it like like a middle ground almost it's like ma protocols will enshrine more and monolithic protocols will modularize more it's kind of like this middle ground this conversions exactly yeah cool so we've got this um extended modular stack now uh you guys went and did an amazing job of of building us out uh what a ZK ified uh

modular stack looks like which I think is super cool um so let's let's look at this more closely maybe you can talk us through some of these examples and and why in some of these cases uh the ZK Integrations uh could provide a better um kind of solution than uh what what we currently have or um at least why why there are advantages to these ZK Integrations yeah definitely um are there any that that you were particularly interested just out of curiosity or did any catch your eye yeah I thought um I I thought the uh verifying correct encryption of mempool

for sequencing was interesting um so maybe we can start there and then also on the security side um and uh yeah and we could probably talk about the the ZK light clients as well for data publication navigating multiple blockchains can be complex and cumbersome let's say you want to deposit into a lending Market on polygon because they have the best rates but you only have funds on optimism so you bridge your funds to polygon only to realize that you need madic to pay for gas maybe you use a centralized exchange or you ask a friend to pay for the gas but only then can you finish finish your transaction in a chain abstracted future powered by connects this is not the situation you can connect your wallet to

any dap from any chain with all the complexities of swapping bridging and depositing all completely abstracted behind the scenes head over to connect. network that's connect. Network to learn more about connect chain abstraction tool kit [Music] yeah definitely so the correct encryption of mempool um as far as I know at least only radius does this out of the shared sequencers um so it's kind of them trying to stand out um with with the with the feature and by encrypting

the mol um what he's really trying to do is to avoid um toxic me um and how CK is used in this case is say it's just to the user generates a zero knowledge proof which proves that the kind of the solving of the time CLE um would then lead to a correct decryption of the valid transaction um so it's not like CK is a feature of this uh of this like bigger piece but it's not it's not fully so it's just being used to prove that the encryption was done correctly yeah makes sense that's that's the verifiable sequencing rules cool um on the on the uh security side it's a

little broad mix between ZK and economic security I'm curious what you were thinking there yeah that's um it's a less scientific way to to put put things what I was really going after with this is that we're starting to see some protocols giving use users or users in this case being applications the option to choose between economic security and uh cryptographic security so CK um I think the one we highlighted in the piece is uh can't remember my head for the uh security I think it was

here shared security merge mining Ian layer Cosmos Babylon oh Bros code chain with the with the co-processors yeah that's right bra is the one um so yeah it's I think the reason why they're doing this is obviously that the optimistic one or the economic security is cheaper today and more feasible um and this is this goes back to your previous question about optimistic uh security versus CK security we think this Gap is going to close uh or close or at least reduce over time um which then maybe it's not

necessarily a question of mix between CK and econic Security in the future maybe it's just that CK is cheap enough we'll just use CK yep but yeah that that's kind of what it's getting at cool and then on data publication I know Celestia has been trying to bring on like ZK accounts and validity proofs which people have been arguing that that might might put them in like competition with ethereum for being like also a consensus in settlement um you also mentioned ZK proving uh data availability sampling maybe we can just um discuss this a little bit and then kind of come back to the original thesis yeah the SEC proving the data ability sampling I think that's um there

was some discussion about that earlier um I'm not sure if there is active research going on anymore um but it's as far as I understand uh rather than each light node doing the sampling themselves they just verify proof that um that the data was stored correctly so it's like an even lighter version of the sampling sampling cool well I think from this idea of of ZK eating the modular stack there's a ton of projects that that are working on this now and also I think it's it's kind of closer than people think um and so and so yeah I think I think what's going to happen first is this breakup of the

stack where Ras providers and rollup Frameworks will start to offer five seven nine different options and then those ZK options will start to come in um over time and yeah we might just get this ZK ified full modular stack everything is built with ZK and in those in those options um yeah I guess how far away do you think we are from from a z KFI modular stack I think it's hard to say hopefully closer than we think may maybe it's useful to try to Define as well like what we see as the end game yeah like trying to answer the question when has CK eaten the modular stack the two things that we highlight in the piece is

is kind of that when when all unnecessary um re-execution is is avoided um so that's kind of goes to how blockchains work today is that you know each node very re-executes confutation to verify that it was done correctly whereas with CK you have a model of one executes and the rest just verify um so that's that's number one and then the second one is around um moving from moving on from uh Economic Security to cryptographic security so going back to our point about stronger guarantees um if we can you know increase the strength of the guarantees then then we can move

further out on the trade of Frontier more experimentation more more fun for the onchain Deens exactly yeah and I mean a third one which we don't really touch that much much upon in the pie is around privacy um zero knowledge proofs are are good for privacy in in certain extent um but we're you know we believe that the full privacy or a privacy or or I think anoma has this term information flow control yeah which we actually prefer because it it captures the essence of what we're trying to do more like the

the idea is not that everything is just private to everybody because that's not a very useful system uh but it's more like the user has control to disclose certain information to others anyway yeah um privacy is like a less talked about aspect of of ZK it's like kind of gets thrown under the rug a little bit because of of regulatory stuff I think but yeah man I appreciate you coming on and breaking down this this article with us and kind of giving our our community a short and sweet uh kind of primer on this stuff so if if they want to dive deeper they can read the full piece and just kind of generally I think people will be excited about uh being able to see what's coming and be prepared for it and also kind of look at who some of the

players in these categories are so yeah thanks thanks for coming on man and um for putting together a great piece thanks a lot for having me and and thanks for the kind words cheers man