Why Layer 3s are Inevitable | Steven Goldfeder, Co-founder at Offchain Labs

Transcript

it's becomes a lot easier for a lot of teams to say you can build on a layer two or layer three but ultimately be secured by ethereum building the ethereum ecosystem and you don't have to sacrifice you don't have to make those sacrifices today because I get you I get that if like it was like 10x more expensive you're probably better off on an alter Al layer one because it's better for you like like we get that and and I never judge a team for saying that but our job was okay how do we narrow that Gap how do we meet people and say okay you can actually have your cake and eat it too you can be an ethereum but you can also run a business and not have to worry about you know being competed by someone else hey every Ron if you have been

listening to Empire you know that Santi and I are fed up with unaffordable fees and frustrating transaction speeds that make the onchain experience basically unusable so the arbitrum team reached out and they showed us the platform they showed us what you can do on arbitrum whatever you're doing you can experience frictionless transactions at lightning speed on arbitrum so head over to portal. arbitrum doio and check it out this episode is brought to you by moned which has not only the highest performance evm L1 architecture ever built but also the wildest and craziest community in crypto monads internal devnet is live and public testet comes out soon so make sure you join the monad community today at

discord.gg slad m o n a d monad are you tired of high gas fees I'm excited to let you know about scale a zero gas fee modular blockchain that's become a perfect fit for gaming and AI apps because of their instant finality and lack of me explore the scale ecosystem today at scale dospace ecosystem and stay up to date with the gasless blockchain on X scale Network that's scale Network big thanks to scale for sponsoring Empire all right folks we we promised we'd do this episode we got the episode lined up we have uh Steven goldfeder back on the Pod uh co-founder CEO of uh

offchain Labs Stephen welcome back hey thanks for having me great to be back yeah um and if folks don't recognize the name I'm I'm sure they do offchain Labs is Affiliated associated with arbitrum what is the relationship there so we um initially built the arbitrum technology uh obviously today the public arbitrum chain so that's arbitrum one arbitrum Nova are decentralized and controlled by the arbitrum down Foundation um but we are still a core contributor to the technology and um you big Believers in the arbitrum tech and are building out U arbitrum Orit and other programs that utilize the technology as well cool so I'd love to start with just um almost a state of the union actually the Santi and I were talking on the Roundup last

week all about l3s and we were debating them and why do these things exist and didn't we just get l2s and l2s aren't even there yet maybe you can walk us through like how we uh how we how we got into this world where now everyone on Twitter is talking about l3s and what are l3s and should we be focusing on them when l2s aren't even fully developed yet like take us into this world of l3s how did we get here absolutely so I think l2s um they definitely still as as you mentioned l2s generally as an ecosystem have a way to go um until their sort of their final completion not that they'll ever be final people say to me is ethereum and it's final State and I say no of course not so but like L2 still have some you know development I think arbitrum is uh

that's l2's generally I think arbitrium is by far the most developed L2 from maturity security e system perspective um and just core technology perspective that exists today but that's not to say that that that it doesn't have still things that are are in the works because it absolutely does um but the problem is or the good thing I should say is two two ways to look at it is l2s have gotten really really popular and you have um this sort of Desire or need for more block space um secured by ethereum and more block space Maybe where the applications have more control over the fees and the pricing so I I'll give you an example you know there arbitrum Nova is the gaming chain controlled by the arbitrum Dow and um the just you know past couple of weeks it's been doing

like insane number of transactions it's actually you know gone down a little bit now but it was doing like by far the most transactions than any other uh ethereum um rollup or Layer Two solution and um what was happening was the fees were going up because of it and this was sort of problematic for some games that were on there they're like hey like you know uh I'm glad that you're succeeding but like the congestion of the network is is sort of everyone's paying for this resource and some have said hey we want our own priority Lane we don't want to give up an ethereum we want to be in ethereum we love the or from technology but can we be in a sort of in a chain that um has our own sort of uh we have more control over the environment more control over the fees and more

consistency there that's one uh area where we've seen some people that you know some push towards chains you have chains for example um like uh wonderful teams like proof of play building pirate Nation they actually are team that went from arbitr Nova uh were doing fantastically well actually they initially came from polygon uh went to arbitr Nova and their fees were slashed by like I don't know 30X or something like that and then they said you know we're doing really well here and now they actually have their own arbitrum orbit uh chain that they're that they're using there are also other places so that's one so it's like fees and predictability but there are some other core core features that have also pushed others that love ethereum love the arbit technology but want to be in their own environment one is a fee to fee token so

there project I said hey I want to build a chain in which my token whatever that token is um is the fee token of this chain so you pay in in in in this token that's something that aring orbit stack supports a good example of one of those recently um there's a chain called dgen chain this is a chain actually not on arbitrum one but on the base blockchain so it's an L3 on base and they wanted to launch a chain with their own custom token and they chose the arbitrum orbit technology the arbit technology stack to do that because it's the only technology stack today that's customizable on this way that allows you to do that you have chainss like rable RAR RAR chain that they want to enforce at the chain level enforce chain royalties Creator royalties at the chain level and they

can take the orbit technology and do that you can't do that on a public chain it's just not feasible it's not viable you can't sort of build your app into the rules of the chain but you have your own chain you can get all the benefits of the technology but also customize it and I'll round out with just two more examples of like and just to give you sort of a flavor of the different types of customizability people are doing in orbit chains we have a team called Kinto and they're focusing on institutional investors they are building uh sort of kyc institution F institutional focused Arbitron chain an environment where institutional investors uh can use the arbit technology but in in a sort of um environment that they're C comfortable with and can get all the regulatory Comforts that they need and then you have a chain called Phoenix uh here

Phoenix spelled with an fhe fhe is also an acronym for fully homomorphic encryption which is uh cryptographic primitive use for privacy in this case um where everything on the chain is encrypted and they are taking the arbitr technology stack but not just using it vanilla and you know slapping a brand on it but they're actually adding cryptography into the core uh chain logic and adding privacy there so you have like this like wide wide range of of uses for L3 some are like hey you know we just want our own customizability or we just want our own branding on the Chain all the way to no we want to like kyc at the chain of En force create a royalty or add like super deep cryptography into the chain level and the unifying factor of all these is they want the security of ethereum so

they want to be on ethereum and they also want to use the arbitrum stack which is the most advanced uh Layer Two or layer three stack and they're able to do that by building a layer three blockchain on on on arbit one or on base or you know wherever they are right so you have layer threes elsewhere some of these chains are actually building layer twos so you can use the arbit orbit technology to build layer twos layer threes I'm sure one thing we'll can get to is okay so I'm a chain for one of these like many reasons like I want to build a custom chain why L3 vers all2 I'm happy to get into that but let me just pause here because I've been rambling for a while I warned you guys before you know I warned you guys beforehand that I can ramble and I think

I've proven myself to do that so I think our listeners would agree that we one of us two out of three here ramble I want to get to actually you read my mind that that last piece can you can you go into that the distinction between building an L2 using the um AP stack versus an NO3 yeah so first I'll say a little history here which is the whole arbitrum orbit program um is just about a year old so a year and a month around it launched in March of 23 which is when the Dow launched and the idea was together with that it was Community own I like to call Community sourced right so in there's like a classic there's like open sourc and closed sourced but that's sort of a binary it doesn't really exist in crypto because in the in the regular software

world you have to make a choice which is like okay do I allow people to use our software or do I not and that's basically it and you have in open source World there sort of a tragedy of the commons that emerges sometimes you'll have like libraries that are used by like every Big tech company and these companies are generating like literally billions of dollars in Revenue off this and it's like supported by like three developers that can't get funding right so this is like unfortunately something that you see and you'll find these bugs and then you'll say why was no one giving them any money and supporting this and you know this happens over time and we start supporting more projects but I think in crypto you have a third option which is like it doesn't have to be like fully open which anyone can take

it and not contribute back it shouldn't be closed which people can't use it it could be what you call Community source which is hey you can use this but you just have to become part of the community and contribute back and support its development and that's where um arbitrum is so let me explain what that means so when ARB so the history is anyway when arbitrum orbit launched it was just for layer threes on top of the Arbitron blockchain but uh it quickly became clear that uh and and that's not true it was self- served just for layer 3es but if you wanted to build a layer two you had to go and ask the for permission it quickly became clear that there was a problem because there was a lot of teams that wanted to build layer twos um but also wanted to use the

arbitr technology but they didn't want to sort sort of go and air their dirty laundry public it's like hey we're doing this super secret thing Dow can you vote on this it's like no they that wasn't working for teams so now today there's a self- serve option for orbit which you can build layer a layer two chain or layer three chain and they're all aligned they all have some sort of align with the arbitrum Dow either in the case of layer three is and arbitrum one you are um building on the Chain contributing to the chain that way or you're you know in the case of layer 2 is you're explicitly asked to contribute back to the chain to the Dow to sort of uh support the funding of the chain and the ecosystem and technology that you're using but but anyway so that's sort of

the history so you can use today you can do l2s and l3s both in an aligned way with the arbitrum Dow um but the question is why would a developer want to choose a layer two or layer three so the answer is uh on the layer two side there's um some I would say um people like The Branding of it like there's a little bit of stigma still associated with layer three sometimes people like I don't want to be a layer three on top of arbitrum or on top of some other chain I want to be my own thing I want to be my own Layer Two now there's some irony there because even in a layer two you're still like on top of ethereum right and I it's funny because if you go back six years ago I had literally the same conversations people want to layer down

people said what you're building arbitrum as a layer two are you crazy don't you want to be your own thing like you're just going to be always secondary to ethereum so today like everyone's accepted a layer two is are quote unquote their own thing and now that conversation has emerged a layer three it's like uh we need to be like a an independent Layer Two we can't be a layer three so there's some stigma and I'm not trying to like you know I clearly like don't think I think that over time that stigma will go away but I'm also not trying to minim it if a team feels that their Community won't like that they Layer Two like you know that's their prerogative and they or I should say if a team feels that their Community doesn't like they layer three then maybe a layer two is better for

them but there are some technical benefits and really some pragmatic benefits I should say of being a layer three and there are several but I'll just focus on two because I think that they're really uh illustrative of of you know the core Taste of why being a layer three is just makes so much more sense for a lot of teams number one is the cost of onboarding users so say that you are a layer three um uh block say you want to have your own blockchain and your users need to get their funds there so you're a user let's say in coinbase and you want to get your fund to uh this new new app chain okay so if you're a layer two and let's assume also on day one you're not going to have support where like coinbase and binance and like other exchanges directly on board to your

chain you're going to have to go via the chain beneath you right so ethereum has support from all the all the top centralized exchanges arbitrum one has support from all the top centralized exchanges maybe your chain will eventually get there and actually we've seen orbit Chains Get There Z xai is an example of an orbit chain that does have a direct integration from binance and it was the first one that got in it's like a super big uh Milestone but like most chains won't have that on day one so you have to think about how am my users getting to my chain and your users can be in ethereum they can be in arbitrum they can be in coinbase let's just use the coinbase example so if you're building a layer two what you'll do is you'll go to coinbase said hey give me my eth on ethereum and then I'm going to

go over the bridge on ethereum to new layer 2 Chain that's two ethereum transactions right coinbase has to do one ethereum transaction that you're going to pay for then you have to do a second one uh on the bridge or the other thing you can do is build the layer three and then you say to coinbase hey coinbase give me my funds on arbitrum one that's a layer two transaction and then from arbitrum you do another layer two transaction and go to your chain so it's two transactions either way one from coinbase to the chain beneath you and then one over the bridge in the first case you're doing Layer Two transactions in the second case you're doing layer one transactions and it turns out that today Layer Two transactions are something like 99% plus cheaper than layer one transactions which means the cost of onboarding your

users has gone down by 99% right so that's like the bottom line if youve if I've lost you in my in my like previous like monologue the idea is 99% cost reduction in onboarding users um because rather than have to do expensive ethereum transactions you can do cheaper arbitr one transactions and that actually makes a lot you know will actually matter for a lot of people so that's reason number one yeah um let me pause there well I believe you have more reasons on this point on on user acquisition how do you envision that playing out over the next you know next year or two where bridging and connectivity between these chains just becomes far easier presumably it's a really interesting question so I think

bridging connectivity between chains will become easier and in fact that's something which we at offchain labs are act actively working on which is better drop technology we called it arbitrum orbit you know to signify like the chains are like moving together and they're coordinated it's not just like random chains um in the atmosphere and the idea is that over time as you mentioned interrup will get better and better that's from the work that we do that's from the work that Partners like espresso do that's in the work like bridging projects like layer zero hop connects across do you know there's a lot of different people that are sort of eating away at the same problem and the results going to be fast forward in a couple years uh we we'll we'll will uh have much better interop that being said um interrup between new chains will

always be a little bit harder right so what I mean by by that is like some of these features will just be okay you build an orbit chain and you get in drop and we're working on those things but some of it will be like you know layer zero has to support you or like espresso has to support you or you know or some piece of infrastructure has to support your chain and the idea is there like you're always still going to think about like the same way you'll think about eventually I'll just have centralized exchange Integrations but you're still going to think about like okay but like how long will it pay for infrastructure to support my chain and um will I be able to survive that period right and that's sort of and so the cost of onboarding particularly in the early days I think is not going to fully go

away even as bridging gets gets better I think eventually it will eventually be to a point where any chain you could just launch a chain and it's greatly interrupt but I think we're farther away from there than to a point where for the selected chains that these Partners select you know choose you'll have better interrup between those and then the question is okay just like eventually I'll get centralized exchange support eventually I'll get better support as part of this and again we're working away at this problem to build general inter General interrup but I just don't think that it's going to get to like you know zero costs soon I think you were gonna go into more number two number three number four yes I'll just do number two for now um which is and the second reason is um chains have to post data back to etherum and

depends how they're configured they might post all their transaction data to ethereum if they're a rollup maybe they'll send some of their data to a data availability Community like arbitrum any trust or maybe they'll send it to you know IG layer Celestia espresso any other of the You Know da Avail You Know da da solutions that are coming online but in any event you're going to have to post data back to ethereum you're gonna have to post either the actual transaction data or some representation of the transactions data is back to ethereum and it turns out that in any case there's a fixed cost of going back to ethereum right so it costs money to go back to ethereum and there's like an overhead essentially of of of post to ethereum and if you are a you know if you don't have many transactions it's actually going to be

pretty expensive like if you're doing like transactions like ARB from one which is tens of transactions per second okay it doesn't really matter right the the the overhead is going to advertise away but if you're like a new chain it's GNA have spiky usage or spotty usage so you might have like a 100 transactions during this minute and you'll go like 10 minutes without a transactions or maybe overnight in whatever region you know you're not popular in whatever you'll have lower transactions you're going to have times where you have very few transactions and but you know if there's one transaction that comes in but they have to but the cost of posting just to blob to ethereum it's a fixed cost of like 50 bucks it's going to start you know getting really expensive either for you or for your users and the analogy I

like to use here for layer 3es versus layer twos layer twos is taking the car over the toll so you have to go you know might be the only person in your car but you have to go over the to pay go over the bridge pay the toll and that's going to be expensive for layer threes you're actually going to take the bus and what that means is you're not going to ethereum yourself you're just talking arbitrum one and all of your data is getting packed together with all the rest of the arbitrum one chain and all the other layer 3es that are doing the same thing and you're going to etherum it doesn't matter like if you have spotty transaction because you're just need to buy your bus ticket and that's sort of you know would be my most two compelling reasons so to wrap them up

it's you know 99 plus percent uh reduced cost of user onboarding and take the bus to ethereum rather than the car carpool lane which is going to be really helpful especially when you have one person you don't want to drive a bus in with one person you know you sort of get on the bus everyone else from um value acral to arbitrum and maybe by extension ethereum L1 does it matter like how do you think about the like teams using an L3 versus NO2 what's more beneficial for Value cruel to you guys and than to ethereum so this is basically what I was sort of hinting at before and I mentioned like alignment with arbitr and Community Source I think it's critical that whether you're building Layer Two

or layer three I think the AR ecosystem should be agnostic right it should want to support you in whatever is best for you and the way to get it agnostic is to sort of give it you know um make sure that you're contributing back uh um in either way so in the layer three case if you're building a layer three at arbitrum one you're actually not asked to uh give anything explicitly the idea is you know arbit one is sort of your Hub you're transacting via Arbitron one you're settling on to arbitr one so like or arbit Nova and therefore you know free to use free to customize do whatever you want and and and and you don't have to you know contribute anything explicitly in the case of hey you're you're taking the technology and building a a layer two an ethereum which isn't necessarily contributing back to

arbit one by by its own nature anything right it could just be you know it could be a competitor CH says hey we like the Arin technology we're going to compete with arbin one using its own technology or you're building a layer three on a different blockchain that's not controlled by the Dow and DOW doesn't get the fees for that's I think where um you know you have to build these sort of explicit alignment so the way that the what's the arbitrum license program is set up if you want to build a layer two or you want to build a layer three elsewhere um you're asked to contribute back 10% of the Chain's you know net revenue so like you take all the everything you pay out on layer one and subtract that you don't you don't pay that you know your cost whatever your

sort of L2 you know profit if you will chains profit it gives 10% of that you know 8% to the arbitrum Dow and 2% to the developer Guild that's sort of you know this group of developers that will hopefully emerge and you know it's open actually for for teams to join that are building the core technology just to make sure the technology is uh is is is funded and the idea here is that it's fully customizable you know the the meme is arbitrum is freedom which is you can take the technology you can do what you want you don't have to ask anyone for permission but you do have to either explicitly or implicitly contribute back to the arbit arbitrum Dow contribute back to the community that's building it no one's stopping you you don't need anyone's permission but you do need to contribute back but the interesting

thing is if you're like hey I want to build a layer 2 but I don't want to contribute back you can go to the arbitrum Dow and ask for a special case you're like hey can you like uh let me build a layer two and I I only want to give 9% not 10% like you can go ahead and make your case of the Dow they have the ability to um offchain technically does too but you know we we don't intend to you know we we' much rather um um you know have people go through the Dow route um that's you know seems much more Community aligned and um the idea being that you go to the Dow you can ask them for whatever special deal that you want and they can consider it and um you know it's really up to the Dow how uh enforcable is

that um in terms explicitly or or implicitly is have there been cases where someone uses your stack and just kind of doesn't pay like how are you enforcing the 10% at the contract level I assume not it's more like soft I think you know I it's it's early day so I I don't know of any cases of teams that haven't you know you know complied um but I imagine that any reputable team or any team that sort of you know um wants to build something interesting and build something Community you know you know build something for the community will uh will

comply that you know all early signs that I've seen and again I I'm um a lot of this goes through the down the foundation so I'm definitely not privy to any everything but what I've seen as part of the community is that teams definitely um are uh are um interested in in you know getting getting a license and this is something that's like always you know always the case like sure it's possible someone somewhere you know will take the code and not comply and then you know it's up to the Dow and Foundation to figure out what they want to do about that um but like I think that most reputable teams will will want to apply and particularly here when when when you know the it's not like it's going to the community right so it's not going to uh offchain Labs it's it's not

even like going to actually to the arbitrum foundation it's going actually directly to the 8% uh 2% is going to the developer Guild which hopefully will build up to many teams but 8% is going directly to the arbitrum Dows onchain wallet right so you're going to see this interesting thing which is today the arbitr from Dow's wallet has like a ton of arbonet also a ton of e because there's like if you look at last year just like the chain profit was like $50 million in eth so a lot of eth uh billions of dollarars of AR and you're going to start seeing all these like weird tokens if you will like going to the because like remember some of these chains are collecting fees and their in their own tokens and so the AR from Dow is just gonna have this like interesting onchain portfolio that the Dow itself

can vote on how to spend so it's like very very transparent like it's hard to argue and say Hey you know we don't want to contribute back to this core Community that's funding the technology of the chain it's not like you're enriching a particular individual you're literally contributing back to the community and saying hey guys you know we spent years building this Tech it's Community owned to today all we're asking you to do is just you know contribute you know you keep 90% of it just contribute back something to the sustainability of the community and align with the community or do it implicitly or just build a l three at arbitrum one that's another you know option which you can do or if that doesn't work go to the Dow and ask for a custom deal and maybe they'll give it to you if you have a case but I think it's

there's just so much optionality and flexibility there that I think it's you know most teams will find an option that works for them Stephen how do I make the decision let's say we want to launch blockworks some some blockworks app or whatever how do I make the decision between just building an app without a chain just building an app on you know what on like an L2 or an L3 vers building an L3 versus building an app chain like something like dydx has done how do I make that decision yeah so so it's a great question so number one is the question you should ask yourself is do you need a chain and you know we sort of like when we started gave all the reasons that people might want to customize a stack you know to have their own chain but the truth is for many applications

they don't need their own chain they don't need to worry about infrastructure they might be want to be on a chain with someone else like if you're a defi app today you're going to want to be an arbitr from one you're not going to want to like launch your own chain have fragmented liquidity Etc even if you have great bridging you're still gonna have some degree of fragmentation so number one is the question is do you have a compelling reason to launch your your own chain right there are some out there that are sort of trying to push everyone out to their own chain I don't think that makes sense the way I see the ecosystem forming is there's going to be the general purpose chains that are long-term relevant long-term very important this is where arbit one is is leading is leading today and then

there's going to be the application chains where chains where you know for applications that it make and it could be ecosystem chains it could be like okay we're building a gaming chain doesn't have to be a single application more special purpose chains if you will um so that that's the question which is like why are you building your own chain and does that make sense and you might have a good reason but you should be able to articulate that reason um is number one and then the question is okay so I want to build my own chain um what do I do and I view L3 and app chains is actually very like you know basically the same uh in the sense of the L3 if you take the arbit morbid technology you can do whatever you want in the chain but you can also sort of customize it as

you as you as you want and you can restrict it as you want so for example if you want to say hey we're building the blockworks chain but and know I'm not trying to I have no knowledge here so just just uh just um you know just spitballing here but we're building the block the block Works chain and you know but we actually want to restrict it that only blockworks can launch apps on the Chain like imagine like it's an open chain permission chain anyone can go ahead and use it but like these are the only applications on the Chain so you can actually restrict it in that way without compromising the secured by etherum part without compromising the anyone any user can interact with it and you can basically have you know a restrict more restricted application there and why might one want to do that

because you might say like well I really want to control you know the congestion of the chain so I don't want someone launching you know something else on the chain and then like all the Lock Works users get congested because this other random app that doesn't even belong here like is now used using up all the block space or you can be like that would be awesome like I don't care who builds here we just want it to be blockworks branded that's fine too but my point is you have a lot of customizability with the ARB and warit stack so um I would say you know the difference between open open L3 versus like app chain is very much how you configure the stack but it's the same stack is it fair to say this might be an overly broad categorization but the L2 is for kind of

general purpose scaling and L3 is more for I guess you could either say customize functionality or customize scaling so that might emerge in how people are using them but actually like from a core technology perspective the L2 stack and the L3 stack are are on the surface like on terms of the developer and user surface literally identical right so there's nothing you can do on an L2 that you can't do in an L3 and vice versa so that that that from an ecosystem development perspective we may very well see that those who have a more broad Vision to have you know it be a general purpose chain to be like okay we're going to be a layer two because we think that that's like how you position yourself for a general purpose and those who have a vision to sort of be vertical specific or application specific might

uh tend towards L3 although we've seen a lot of l3s that are very Broad in their their Ambitions you know you have for example in gaming arbitrum uh has a ton of gaming l3s um that are not just that one game but are actually sort of you know looking to build gaming you think about like game Publishers where they're attracting you have like treasure chain Z AP chain um muster hopia you know there's like like 10 or so different um gaming chains in the ARB ecosystem as an example so that might emerge as a pattern of usage but from a technical level there's like any any chain that's built as an L2 could be built as an L3 any chance buil an L3 buil as an L2 technically if I'm playing this out far

enough like not next one to two years but like five to 10 years if I'm in your if I'm in arbitrum shoes at some point you got to make the you you got you got to consider actually this is going maybe seem crazy but getting off of ethereum um and I'd be curious if you've thought like how you think about that um as an L2 and like how you think do do you think that l2s and arbitrum specifically or anyone else will eventually try to launch their own L1 and basically say look we use ethereum for security today but we don't need them anymore so that to me would be like a big violation of like everything you know that that I care about and the reason we set things I can tell you so earlier I alluded to you know this whole

conversation we're having now between L2 versus L3 I had L1 versus L2 so you know to to give you a little more context there that was in the context of like 2018 fundraising when we were trying to fundraise say hey we're doing this like radically new thing at the time building a layer two blockchain like who else is doing that a general purpose Layer Two blockchain no one else was doing that at that point and we were raising funding for that and we saw a lot of push back uh which you know particularly from investors uh some investors obviously we found some really great investors that believe in the vision and here we are today um but we had um some you know investors and some very smart people push back and say like it doesn't make sense like you know uh you should build

your own chain why would you always you'll always be subservient to ethereum that doesn't make any sense what are you doing and our response was like and it's very aligned with what later became vitalic rollup Centric road map which is no we believe that ethereum has you know is the most secure um and and has the strongest story to be that security layer and that secure blockchain we also think it needs to scale and that's the that's what we want to do here so um you know my point is like it would have been a lot easier to just start there and build build a layer one um today that seems funny because layer TW are all the rage and all the layer on mov to the layer twos but like it was always a principle based decision of we should add to scal the

scalability of ethereum because we believe in ethereum security because we believe in ethereum decentralization so I personally um would fight very hard against U be very I should say I'd be very against um you know chains like arbitrum becoming um their own blockchains I think ethereum is contributing a a lot to them and I actually think ethereum you know is also doing a lot to reduce its its fees and its costs to make this environment more competitive for rollups and make it that rollups what I mean with that is you know an environment where rollups can use ethereum for security security and compete more but the other thing is I do think that we have to be careful as an ecosystem because yes there definitely probably will be chains that fiz that hey they can bootstrap as a layer two

Gain a community and then sort of uh I'll say rug you might agree or disagree with that term but like um you know sort of you know drop ethereum I think that that's a very dangerous thing which is why like I've always been an advocate of like hey guys we have to and I'm I'm not suggesting that any particular you know other other chain is doing this today to be clear I'm not like pointing fingers at anyone and I have no knowledge of anyone that is I just I just think as as we peripher to 20 30 40 50 hundreds of chains you know will one of them have this idea absolutely and that's why I think building out the SEC the core security it makes a lot of sense because if you're arbitrum today you're like hey we've been you know this chain has been

Pro posting fraud proofs on ethereum has been like actually secured by ethereum you know since its launch where if you take another new chain that launches today and it doesn't have anything security you know anything related to security ethereum it's a lot easier to go to your community and say Hey you know it turns out we're paying all these extra fees to ethereum but it's actually not providing us anything in you know in this case because you know some might not have the core technology that actually provides it so it's a lot easier to go and say hey let's turn that off but if you're actually getting the benefit of ethereum I think it's harder to make that argument because I think there's a ton of value from there so again I'm not suggesting that any chain that I know of today will do that or is

doing that or I suspect him of doing that but I do think that um as you have chains that are not actually benefiting from ethereum but are paying ethereum for a service that they're actually not utilizing um that is something which we as a community should should um you know be careful about and make sure that those that are actually building on ethereum are um you know actually getting the benefits of it and have a community that cares about the benefits uh of it and if they don't then they can go build a layer one but like you know anyway that that's my my thought which is I'd be very against arbitrum R other big layer to doing that I think ethereum provides a ton and I think it's it's worth the the premium that you pay for ethereum security how would you articulate the

main benefits of building on ethereum and could you point to certain examples where a chain not arbitrum may be using e theem but may not be getting the benefits that you are getting I think so the core benefits of using ethereum are um there actually two uh number one is the security security and decentralization so actually building a layer two and actually being secured by ethereum and that means it could be a a validity roll up like a ZK roll up where you're uh posting proofs to ethereum and having ethereum verify them ZK proofs it could be an optimistic role garbage from one where there's a fraud proof mechanism that if engaged will get resolved on ethereum and that's where ethereum is is you know and also chains like arbitrum oneand that are

actually using ethereum as a data availability layer where ethereum is actually giving guarantees it's a lot cheaper today because of 4844 but still ethereum is providing that core service I think that's where um these chains um are using ethereum and they're actually benefiting from it and it's a type of benefit like you don't think about it on a day-to-day basis but you know as there's a proliferation of new chains trying new things eventually you'll you'll understand you know we'll see okay like I see what happens you know I hope nothing goes wrong ever but like the point is having security is a thing that like you don't appreciate until you actually appreciate it until you need it hopefully that time never comes for any chain but I think it's something that's uh worth paying for and um something that like I'm very very uh supportive of

and you get that by actually having um a proof system on ethereum data availability on ethereum that's not the say to be clear that I'm not a favor of I don't think there's a trade-off space like I I'm not anti at all the data availability space I think you know we even in the arbit ecosystem built arbiture many trust which is probably the first live alternative da there are a lot of great teams like espresso igen layer Celestia um n ail you know really a lot of them zero gravity a lot of these teams that are building alternative Das that um I think they're they have a point they have a you know there's a point in the tradeoff space where it makes sense to use them and um but I just think there's a a gamut of services that ethereum um that e that ethereum

provides and the case of arbiton one it's data availability and also settlement and just core security that if there's some dispute ethereum will um will um you know uh refer that and and and resolve it fair to say that if you don't have fraud proofs you're not getting any of the benefit as a user from the security of ethereum L1 yeah you you may still be getting the You Know da so so those are you might be putting your data ethereum so you know your dat is there but if you don't have fraud proofs built out ethereum um there's going to be a centralized generally centralized in in the ones that I've seen a centralized let's just call it uh entity or sequencer in there in this case that basically decides the

state of the chain and if they decide to be dishonest or if they get hacked in some way where someone else decides to be dishonest um that now is now empowered um you can't go ahead and say hey ethereum like this is clearly wrong the rules don't allow this can we can we resolve this because um that mechanism just literally isn't there can we go into this because you and I have talked about this at length um there are certain l2s that haven't um don't have fraud proofs and they've been deployed for many years um or just been around for a while or are securing a lot of value you guys just what is it about fraud proofs that makes it so difficult um to build and to

deploy and then contrast that to uh centralized sequencer because I think people can flate the two they're vastly different in my opinion um but let's go down this Rabbit Hole yeah absolutely so fraud proofs are actually kind of funny um because on the surface they're like very very simple like you look at like ZK proving and you look at fraud proving and fraud proof just looks very simple but there's a lot of um different edge cases and protocol cases that make them you know very delicate and you know we've been studying these systems for um years now you know close to a decade actually from the earliest days and there's still things that we're learning and figuring out you know AR from Bold protocol is uh a really nice protocol but that's like you know the third or

fourth generation of AR from fraud proof protocols and there's a lot that we fig figured out along the way and there's a lot of you know tricky cases and gotcha so there's definitely a lot of complexity in figuring these things out figuring these things out um but um it's a lot more the Nuance of the protocol you know and uh different classes of attacks that could Happ happen and and what happens you know uh if you know if some bad actor comes in you how do we know that the good actors can can win are there any edge cases that's where a lot of the complexity uh comes in there but you know to your second question there are two roles in the arbitrum ecosystem or really in any rup ecosystem that people often confuse and I'll use the arum terminology but again um in any

optimistic rollup there's going to be these are two roles that exist even even if they're called different things uh one is the sequencer and the sequencer job is to as you said sequence transactions um right so it orders transactions transactions come in the sequencer sort of gives an ordering and uh sequencer post those transactions on chain uh it posts the these batches of transactions on chain there's a second role called the validators and these are the ones who actually provide the security of the chain and these are the ones who will engage in the fraud proof system so what the validators do is they don't actually talk to each other they don't even talk to the sequencer they just look at the chain and they see this feed of transactions the sequencer posting and they sort of execute it and get the result and and one validator

goes ahead and posts their result on chain and the other one say do I agree or disagree if they agree they do nothing if they disagree then they'll go ahead and like engage in the fraud proof mechanism to come out to the right decision and figure out what the correct execution is but the idea is that the validators keep everyone in check including the sequencer all the sequencer is doing is it's posting the transaction roots on chain it's not telling you what the result is it can't take money out of your wallet it can't you know post a b transaction if it includes a bad transaction it'll just get ignored by the validators right the sequencer is kept in check by the validators a sequencer is just sequencing and ordering transactions now people always say decentralize a sequencer and there is something to be decentralized there and I'll get into

that but there I think a lot of time people misunderstand and think when you say decentralized sequencer they think the sequencer is the dictator and can do everything in actuality there's the validators that keep the sequencer in check in the arbitrum case today there's about a dozen validators including like larger Institution like Google and consensus and offchain Labs right so web 2 and web three big players um that are that are validating the chain and the goal with bold is to open this up to everyone um but even today there's already a a really nice distributed set of validators you just have to trust one of them and you know as long as you trust that all these 12 don't go ahead and collude you have you know ethereum foundation in there you have a lot of different like you know um uh representations of really strong

companies with strong beliefs again opening up is better but arbitrum today validation is already distributed the sequencing is still centralized and what that means is this ordering is centralized so so what does that mean it means the arbit sequencer if it wanted to reorder transactions it could do that it doesn't in practice uh it does a first come first serve ordering but it could go ahead and reorder transactions um Can it censor it can it can it can delay you but it can actually censor longterm because you can always go on chain and say I'm going to bypass the sequencer and post my transaction on chain now that's if you have the but that's only if you have pro pro so um well the force inclusion that you can actually so if the sequencer doesn't and you can force it to include right

you you could technically have um a force inclusion probably without fraud proofs but I don't know how you exactly you'd enforce it but there may be some design there but certainly the validators are are only um certainly the validation is only if you have fraud proofs the fact so so in some of these other systems that you that you you sort of alluded to before that don't have fraud proofs it works very differently there are no validators because there's no mechanis for them to engage or or become or you know there's no recourse so there's no system of validators the sequencer is the dictator in that case right so the sequencer doesn't only say hey here are the transactions sequencer also says and by the way here is the result so it turns out the sequencer saying yeah Steven's wallet you know

minus one eth Santiago's wallet plus one eth that's all done by the sequencer so the sequencer decides to just like make stuff up and like take money out of my wallet it can do that if the sequencer decides to push bad transactions that aren't valid it can do that if you don't have a set of validators so that's why I think and and that in these other systems that role is played by the sucer and that's why I think when a lot of people say decentralized sequencer they're not thinking about the important but more narrow issue of hey can we have a distributed protocol that you know makes the sequence or not reorder and by the way we're working on that we're collaborating with espresso systems on that it's a high priority to get that but like I think that's like a relatively smaller issue than you know

not having validators and a sequencer basically serves as a dictator and I think people uh sometimes confuse those two again and and why would you attribute to like I appreciate the some of the criticisms I've heard about arbitrum is this point around you have to trust this group of of validators which is a fair point look I mean you're progressively um but it's way better than systems that don't even have these this group of validators again the question is like why hasn't like op optimism some of these other players like why haven't they built this group of validators and I mean I appreciate maybe they have recently but it took them a long long time like I'm just kind of curious like what makes it they have I think I think on Main they

still have oh no no it's on B it's on testnet I believe yeah so right so um on Main net they don't have um any fraud approves that are deployed to my knowledge and I think they're working on it and I and I hope that they're working that they're they know that they'll be successful I I really do and I know that they have um a lot of you know uh good good people trying trying to work on this so I think there are there are two things to say number one is it's actually really complex um you there was an interesting threat on Twitter which uh I know because I I lived it but actually just uh um I saw a few days ago which actually went through some of the history of fraud proofs and the answer is actually if you go back two or three

years they they did build it um there was this historic debate between two different types of fraud proofs there was what we call Interactive fraud proofs that's what was in the initial arbitrum paper and we've always been building and then there was what's called One Shot fraud proofs and um we were always in the interactive camp and they were in the one shot Camp they actually built one shot fraud proofs and deployed them on M that I think in like 2019 2020 2020 or so maybe 2020 2021 um but they had like critical bugs in it so they had to turn them off um so so that's and then and then they sort of you know the community uh and and everyone sort of aligned to okay arbitrum method of interactive fraud proofs is the correct way to do fraud proofs and everyone's aying than that

and then I think they've been building ever since and I know that they've had you know uh some progress recently again I'm hopeful that they do this but I think it's like if you actually look at the history um you can find like passionate debates between me uh Harry I think vitalic definitely gotten involved at times um on Twitter on Reddit uh going back to like some good archives I should I should pull up one day um um where like you know people were passionately debating these two different ways of of building fraud proofs and we were always you know and everyone sort of come around to today to okay this is how you build fraud proofs which is interactive fraud proofs the fraud proofs that sort of you know we've been building uh since the initial arbitrum uh sketch in 2014 um but um it's not that they

haven't tried or even deployed something it's that sort of you know they I guess in the consensus you now went down the wrong path and retrace and are now going down a different path uh which is you know the path that we went down does anyone try to ju have you open thought about open sourcing this interactive fraud proof Tech uh have people try to look at it and copy it as opposed to try to build their own flavor of it so the the code has always been public and Source visible right so anyone can look at it um I have no knowledge of anyone trying to to copy it but I you know I haven't really like dug too deeply and to try and figure that out but not not to my knowledge um but definitely some of the core ideas um which are not part

of the Koda so I'll give you an example um you know one of the um like you know one of the big protocol breakthrough that we figured out back in 2021 was to use chess clocks as part of this protocol and I remember like you know being very involved and I you know in this in this and having this idea of hey use use chess clocks the idea is you have two Challengers and they have like variable timers before that we thought that everyone sort of had had a fixed timer and it became problematic because the protocol just got very long and everyone had this clock and then it was like this aha moment I remember it like very well I was like oh why don't we just use chess clocks and actually um this came to me when I was watching this movie I forget it was a Netflix like

chess documentary about this I forgot what it was called you remember I'm talking about there's like this girl the yeah yeah well it's not a documentary it's a it's a show not documentary sorry sorry the short Series yeah yeah yeah Queen Gambit queen gamb queen Gambit so I'm watching the queen Gambit and that's when like this idea came came and like okay chess clocks that's the answer boom we got it we figured it out and now if you look at like every competitor's protocol they all talk about chess chess clocks which is fine right that's not like you know uh you know it's part of the code it's not a patented idea it's just the idea but my point is there are definitely like um you know the ideas have been shared you know like arbitrum has has been pioneering interactive fraud proof since 2014 and like you know

it is the ideas and the years of work we put in it is no question contributing to every protocol that's building interactive fraud proofs today but that's fine right that's that's actually a good thing in my opinion um the exchange of ideas and I think you know as an ecosystem I'm I'm in favor of collaborating more and and Building Things and Building Things you know Finding sh resources so that's like no issue but like in terms of the the other question of so so Cod code is visible today but it's and you can use it as part of the arbit from orbit program which means if you're building a layer two and you want to use it all you have to do is join the AR more program you can use the nitro technology stack you can use the fraud proofs and you know no

problem you can use it so yes it's visible yes we thought about open sourcing it but we believe the best model is community source so um and and dozens of chains are using this today U these are layer 3es these are layer twos are using the ARB and orbit technology and if a competitor wanted to go ahead you know if a if a competitor on a different stack today wanted to go ahead and say Hey arbitrum you know we just want to use your fraud proofs they would absolutely be able to do that if they just um complied with the orbit licensing program I guess so we need to find validators they'll need to find validators yet yes uh for it or they can launch ARB from Bold um which will sort of be open to open open to everyone all right I mentioned them in

the pre-roll now I'm going to bring them up again it's arbitrum Santi and I are really fed up with these high fees and we're really excited to have teamed up with arbitrum for the next couple of months on Empire as the leading ethereum scaling solution arbitrum now Powers hundreds of decentralized apps across defi perss nfts gaming and a whole lot more the team has showed us everything in the ecosystem both now and what's to come and we're really really excited about it arbitrum allows both daily users and developers to interact with ethereum at scale with low fees and faster transactions the way the team got me excited was through portal. arbitrum doio so my call to action to you is to get started by visiting portal. arbitrum doio go experience onchain like it was

meant to be this episode is brought to you by monad mad's thesis is simple the evm is here to stay similar to JavaScript in web 2 but unfortunately today's evm lacks the high performance and the scalability that developers need to make certain applications possible monad addresses these concerns in these bottlenecks while preserving seamless evm composability for application developers there's a seamless transition to monad as the ethereum RPC API allows for really easy portability and for developers monad can support 10,000 real transactions per second with their unique parallel execution environment and of course there's full compatibility with evm bike code mad's internal devet

is live public test net comes out soon you can join mad's Journey in two ways one go follow them they're on Twitter @ monad m o NAD dxyz and also join the monad Discord it's discord.gg monad big thanks to monad for sponsoring Empire hey everyone Jason here I know you've been hearing a lot about the intersection of crypto and Ai and that's why I'm really excited to share our newest partner and sponsored Empire scale a high performance modular blockchain that's revolutionizing gaming and AI with zero fees and instant finality scales unique architecture allows for massive scalability and has already saved users over 6 billion in gas fees the five main takeaways that

you need to know for scale one zero gas fees two scale has instant finality and lack of muv three the zero gas Fe model of scale is really important for blockchain gaming adoption as Gamers don't have to pay transaction costs or have the skl token four scal is multi-chain this design allows for nodes to be combined to create chains and for individ idual sub nodes to actually be removed and relocated which creates greater security and collusion resistance and five scale has become a really good fit for AI daps because they're fast automated zero fee transactions big thank you to scale for sponsoring Empire we're excited to partner with you guys bridge over to scale if you're listening to this at scale dospace EOS follow the journey along with scale on X at scale Network

that's at scale Network big thanks again to scale for sponsoring Empire I want to take us back to just the Builder mindset um we've discussed a lot about the tradeoffs the decision Tre process of building your own app chain your L2 using arbitr mer just your L3 what would you say to someone that may just come in and say hey like this kind of L2 L3 bridging sounds really complex from the user standpoint want to just go build in some of these higher througho chains like salana suie say Aptos monad down the road like um what would you say to that yes so I say a few things number one is like I think the entire space is

one of experimentation so you're not going to hear me say a bad word about like these guys you know they're taking a different approach they're trying a different thing I can tell you why I think the ethereum approach um I think is a better idea why I think it will will will live out longer but like you know salana is a differentiated experiment there's no question about it um they're doing uh they're trying something else and they have um a really great you know Community that's that's building this with them and there there's no question about that that I had you know I think I think that they are definitely doing something interesting do I think it's the quote unquote right thing no I think that you know ethereum uh finds itself on a different point of this trade-off when it comes to decentralization you know

the requirements to run a node um the stability of the core uh layer one technology I like those benefits of ethereum more and I think that um there's uh you know there's it's worth the sort of tradeoffs and but the point is I'm also pragmatic about this so you know in 2022 or so we were having a lot of conversations with gaming teams and they were saying like we can't come to arbitrum one it's just too expensive right because we were at that point say 90 to 95% cheaper than ethereum and while like couple of cents per transaction fee is like great for a defi app because you know you can trade $100 and like your your Fe is 5 cents that's perfect right so even for like moderate

size trades it was like really really great um but like if you're game and you're paying a couple cents every time you do an onchain transaction a gaming transaction that gets very expensive very quickly so we understood that well and what happened what these people were all going elsewhere they were leaving ethereum um and going elsewhere and that's where sort of arbitrum Nova came in that's where we said okay we need an alternative da solution we need something that gets them the most of the benefits of ethereum but also gets them to the price point where they can be and the idea was because it wasn't like we were sitting in like our Ivory Tower saying ethereum guys you got to build an ethereum I don't care if it's more expensive like you need they're like all right well I can just shut down shop

because no one's gonna play my game if I'm gonna pay it it just won't work it's not fun it's not interesting I won't have a community you have to sort of meet you know those two things together and my point is like you know if you look at ethereum before ler twoos like arbitrum or even if you look at you know ler twoos like arbitrum before EIP 4844 I think we're about to do it again with arbitra stylist you know about to reduce fees significantly again for those after you stylist like any each point there there there there are points where teams can say like you might have a great reasons you know Stephen why I should use this technology and you can like lecture me on that but like if it doesn't meet the price point of my users if it doesn't actually get me to where I

need to be on a business stand point it doesn't matter the conversation is not worth having and that's why I think we've we've narrowed that Gap massively and today because of the work we've done you know as layer two teams but also as layer one teams you building EIP 4844 building the ethereum RO of centric road map building arbitrum you know building out the core technology it's becomes a lot easier for a lot of teams to say you can build on a layer two or layer three but ultimately be secured by ethereum building the ethereum ecosystem and you don't have to sacrifice you don't have to make those sacrifices today because I get you I get that if like it was like 10x more expensive you you're probably better off on an Al Al layer one because it's better for you like like we get

that and and I never judge a team for saying that but our job was okay how do we narrow that Gap how do we meet people and say okay you can actually have your cake and eat it too you can be an ethereum but you can also run a business and not have to worry about you know being out competed by someone else all layer one I'm not saying that you know we'll see how the fees ultimately uh land with ethereum there's still some market dynamics that you know may may change over time particularly as more usage comes up and more demand for blop space there's a lot of open questions but I think we've done a very good job narrowing that Gap so teams that care about ethereum don't have to make the ultimate sacrifice if you will to to build there they're actually able to uh

succeed and build there at the same time how just uh my last question on this point yeah know I'm sure you others but how important do you think composability is I think composability is very important for a lot of applications probably less important for others and I think this is you know too your question before if do I want to build my own chain or do I want to build an application on ethereum um so one of the things which I said before is why do you want your own chain but the other thing you question you should ask yourself is do you need composability what level of composability do you need do you need synchronous composability do you just need asynchronous composability or really good bridging um but I think it depends so I'll take defi for example I don't see defi like fragmenting because

I think there is a network effect there the fact that you have you know 19 billion uh dollars in tvl arbitrum today is very attractive for for D5 protocols the fact that you know you have every other D5 protocol in arbitrum one today is a great really attractive I'm building a def5 protocol I want to worry about one thing how do I build the best product that will attract you know this existing user base to my CH to my app what I don't want to worry about is I want to build a great product and now I have to figure out how to get users on the chain in the first place and get liquidity here and then once they're here I gotta you know it's just like you don't want to deal with that you want to uh be composable with other apps you

want to be with the user and the same thing for gaming by the way too um there's a trade-off here right so some games might be like okay you know we we want our own chain because we don't care about composability others might be well we want discoverability or of our game we want to be Co line so it really depends on what the game is you know how in ingame assets are used are they bridged you know where do they need to live there are lots of different questions but ultimately um some game some applications need lots of composability others can live on an island and um defi certainly in the composability uh first bucket um gaming can go either way um but you can definitely imagine a game that doesn't care about being on sort of in a little bit a little bit isolated right it

doesn't want to be too isolated but like if just like an extra click and then like users can get funds there okay fine you know that's pretty pretty you know pretty fine for many games um so it really depends uh on the application but also on the flavor of the vertical right so different verticals will have different tolerance for composability and I said it was my last question but I am curious just to get your thoughts around um why do you think coinbase chose uh the op stack particularly all the things we've talked about prop proofs and and everything else like how how's that give us some insight into that process because uh people are really excited about it obviously coinbase playing a big role and and they built some really good Tech their wallet and whatnot but I'm curious uh why they didn't build on arbitrum so I I can't

speak for them I don't know why they why they didn't uh build an arbitrum um I I don't believe that I I um you know before the decision was made you know I wasn't involved in a conversation where they were seriously considering it so I don't really have um th those details I don't know you know of a diligence polies we were a part of um but I'm sure that they you know took on consideration interally the other thing is you know which which they've been very public about to to their credit um there there was you know coinbase um got um I think I think it was something like 2.75% of the op token supply for that decision so was that there was definitely a financial aspect to it which again you know they've publicly disclosed and uh I could be wrong in the exact number but

that's that's you know in that ballpark of my recollection that's know a lot of money today in the you know hundreds of millions of dollars so I think there were probably a lot of factors at play but interesting thing is like moving forward um you know coinbase obviously is on the op stack and you know base is built on the op stack and that's sort of you know just is what it is but I think that there there are really interesting things happen on layer happening on layer three where the arbitrum arbitrum stack is not only powering a layer threes in arbit one it's actually powering as you know I mentioned in the beginning beginning of the Pod the biggest layer threes on base like dgen chain is an arbit from orbit chain so um I think there actually you know what happened in the the past I would call

what are under under the bridge at this point but going forward I think that there are um a lot of compelling cases where um yeah you know and it could actually make sense right there are certain you know coinbase obviously has a great brand and there are certain applications you know certain verticals they're going to go after and and you know you might see some developer it says Hey I want to be uh on base but I also want to have the best technology from my layer 3 and I and that's fine um I think Arbitron one obviously has you know its strong points and it's you know the leader in defi the leader in gaming today but there definitely could be and will be chains like dgen chain that say hey we want to build on base and you know what we say as a community as you

know me as me as part of the developer Community but also part of the broader Community is we support that and we know we'd love ways to uh scale ethereum more generally so it doesn't have to just be um you know our goal is to scale ethereum and to really make the arbitr stack useful for uh you know um scaling ethereum generally and uh super excited to see the many many ways that that it's been used s you want to know the reason I'm gonna get a lot of for saying this study Andre Horwitz yeah I mean yeah yeah I mean I think coinbase ventur made Evol invested in optimism I forget but yes certain a6c actually invested in us as well yeah yeah which is yeah but yeah I still

remember like it was they certainly rolled out the red carpet for optimism and withdrawals and it took them longer for optim arbitrum uh but not to go down this rabbit hole further I do just as a followup in this scenario of Dean CH chain in a world where like if you're at app is deploying an L3 using the arbitrum stack but using optimism as an L2 how does the fraud proofs in the current state actually work uh because optimism doesn't have them but you may be using an L3 where you're using the Ted has it so kind of like my brain Fri so so the number one is it depends on the configuration of

your chain right the orbit orbit stack says you can use fraud proofs um but like you know you can also deploy your chain where you you just have one validator who's yourself I'm not saying that you that you do by the way that they do I'm just saying so for any orbit chain you have to not just look at the text stack says what you can do it doesn't say what what what is automatic right so just because someone's using the orbit chain doesn't mean that they have a dozen validators that are independent so you have to look at that always I actually don't know um exactly the configuration the Gen chain so this is not like a comment on them it's just a comment generally that you should look at the sort of the deployment of the technology um as well as the core

technology but assuming someone does you know um you know doesn't want to launch with like live fraud proofs and the like and they have a nice set of validators or you know they'll use bold um soon and they'll open up validators to everyone if you're what that means is if you're if you're layer three on a layer two that doesn't have that so you're just basically um relying on the layer two to resolve your disputes but but and and they and they become sort of ort of um you know your ultimate referee as well so right the layer two may be centralized so you have a whole you'll have a decentralized protocol that um when it um sort of has a dispute it will ultimately be resolved by the rules of the Layer Two And if that's centralized then ultimately that means your referee

will be centralized so it's kind of could lead to lead to some uh you know funny cases but I think you know in the long run I ideally the layer twos that are you know powering layer 3es will themselves ultimately be decentralized and everything will sort of be flowing down to ethereum that's at least the goal but certainly you can imagine combining these Technologies in in in ways where the decentralization ends up at a centralized Point yeah yeah you always look further down at the at the stack where the settlement is happening and if there's the security that's like the the weakest link so to speak there are no security further down the stack you're kind of screwed exactly your path to ethereum is the weakest link from the layer three you know down to down to ethereum um

that's exactly right Stephen if we fast forward a couple like when so we have layer 3es now people are going to build on the layer 3es so then they'll be l4s when does that game end or does it not end and that's totally fine so I think a lot a lot of the core benefits so if you if if you remember those two things I rambled on earlier on those two core benefits of building layer 3es the interesting thing is those like you basically get all that from being a layer three um if you build a layer four or five or six it doesn't really buy you much and just to re recap those two things were the 99% reduction in in onboarding cost that's because um you know layer TW are already already get you there layer 3es aren't 99% cheaper than layer twos um and you know there

are other reasons to get you there like this right that what what so so the idea is you're not going to have that benefit on layer four or five or six you'll have the same benefit as layer three but it doesn't sort of compound that way and similarly taking the bus to ethereum the same thing happens you can join the bus from a layer three or a layer four or layer five you're not getting like a compounding benefit so I actually think that layer three is um where those benefits stop and I don't see us going to layer four five six seven eight I do think you'll see layer fours and I'll tell you why but I think I think that's where it basically you know will likely stop you'll see EOS systems building because the one thing we didn't discuss is the other reason why to have layer 3

is say you're building a gaming chain today and your goal is not to just build a game it's build a universe of chain or say you're doing this in whatever it could be in any space right where you're right we see we see a bunch of chains like treasure for example plans to have many chains um um proof of play talks about a multi-chain world as well So eventually um you're going to have sort of many many different chains that are part of one ecosystem you know it could be gaming could be something else it makes sense to sort of glue those together right to have sort of one base chain uh for these and then to have all these other chains that are part of that ecosystem maybe every game has its own chain right so they're all sort of like like Nintendo would be the L3 and then

all the other franchises would be The L4 exactly some something like that and the idea then becomes that like so you might see that happen where some chain deploys as an L3 because they want that 99% cheaper on boarding because they want that taking the bus to ethereum but now they want to build an ecosystem of chain so it makes sense to go up to a layer four but it's not like the core benefits that the compounding lower cost won't happen right so that's why I think you're you're not going to be like Oh I'm building a layer four because it's 99% she than layer three that I don't see happening that's true for layer three to Layer Two um I don't believe that it really extends up so I think it's not going to sort of go on forever but this ecosystem thing you might see

so layer three might have lay layer fours and maybe at some point you'll see one of these layer four is like Nintendo like you know has like a hundred of these others and then there's some other game that now became a platform so you might see things like that happen but I think that will be driven mostly from ecosystem Dynamics more than technology Dynamics I actually think analogy could like the way I've been thinking about it is like this development of from from SAS to platform so for people who use like Salesforce or HubSpot as their CRM for example like these were just SAS brands that eventually had so many users and so many customers they started to transition from a SAS company to a platform where you could build something on top of them and in this case it's kind of like maybe AWS is the L2

Salesforce is the L3 they're fine being an L3 for years but once they get big enough they then have these apps or l4s or whatever on top of them exactly and maybe who knows maybe 10 years down the road one of those apps will be big enough that it becomes Z exactly exactly but that's gonna be driven exactly by those Dynamics as opposed to like like and this is something that a lot of the like you know uh alt1 folks on Twitter like don't understand like every time you say L3 they're like haha L4 L5 L6 which is like like they don't understand that like no like L3 actually makes a lot of sense um from a core like dollars and cents perspective all fours and five like it could grow that way if a team wants it to but it's not like you just

get this like infinitely comp like if that was the case we just like jump straight to like L 100 and just like okay look we just like reduce fees to like negative one you get paid to use the chain right that's not just seems so I asked this question and then there's a few uh I think Keith Maxis just thought that it was a stupid question so I'll ask it anyways because I don't care what they think like at what point does Nintendo realize that it's just better to launch their own L1 and the reason I asked this is because we're P we're asking these questions in a state of the world where we don't have like mainstream applications right we have like I don't know like the there's always a strategy of the commons right so like in a world where you have you're

an L3 you have a ton of demand millions of users then you're like okay well what's stopping me from going down to the L2 and then why do I stop there why do I not go to and build my own L1 and like I know this is a broad debate which may go down like a huge rabbit hole which is bootstrapping what what are the requirements of like starting an L1 finding validator security all that but my I'll front run and just say if you have like a ton of demand ton of users you're probably going to be solving some of the other stuff fairly trivially I think if you have that because you'll attract validators because the fee pot's huge and people want to validate on your

chain like so I'm sure you thought about this I'm curious if you have an opinion so from like L3 to L2 actually might make sense because those arguments that I gave you before and there are more arguments but let's just focus on those of like much you know 99% cost uh reduction uh on onboarding that was because of the Assumption we made that said you're not going to have centralize exchange support in day one but if you're like a massive brand if you're Nintendo I'm pretty sure you can get you know if Nintendo actually wants to launch an arbit morbit chain I'm pretty sure I can get centralized exchange board in day one like that would be like pretty pretty solid I take that bet uh and the other one which is like you know reduce your get on the bus that's only again the the benefit of that is in a

world where you're going to have spotty fee spotty usage you don't know your usage if Nintendo comes down and build a chain they probably know that they'll have you know pretty pretty good consistent usage of of their chain they'll be able to sort of guarantee that or you know um pay the cost up until when that happens like I don't that's going to be a massive worry for them so it could be for a Big Brand like Nintendo those two things don't apply and they'd be like hey it makes sense for us to be a Layer Two or there's no reason not to be Layer Two when you get to layer one though I don't think they're you really you know want to get you know you mentioned like bootstrapping like I don't think it really makes sense to want to get into that business of like now building you

know dealing with all that you sort of it's like you can take ethereum's cons senses out of the box and use it and just focus on building the the chain or you can now try to like build like um you know uh proof of State program and like track validor you could use tendermint uh like do X and then you know in a world there's a lot of restak you could kind of maybe use even ethereum as part of your security like like like what ethos is trying to do into Cosmos right taking I layer e three stakes and then use it to boost drop the security to incentivize validators and presumably lower the cost um to get your own app chain started you absolutely could so yeah there you could and that's

not going to be quite you know we'll see how these Services develop I can tell you today um that definitely not quite as simple as just building a layer two and and using ethereum but definitely you know the ecosystem could develop in a in a way that that is but I'd argue that in all likeliness though there's still going to be um there's still going to be an uphill battle for you um to sort of get that level because even you know reaking ethereum means you get to borrow some of the security of ethereum but you're not going to be on the level of the entirety of the security of ethereum you you'll get to borrow some of ethereum security to sec to secure your chain ethereum will be more secure but but anyway maybe maybe you don't care that much about security maybe that's not your you being this

hypothetical chain maybe that's not your top thing but the thing other thing I would say is other way to look at this is like I think it's kind of think about the internet for a second right the internet like you know we're here um you know streaming on the internet and there are like multiple layers of Internet stack uh beneath us right so uh we have like you know https and we have you know tcpip we have like you know lots of protocols going on underneath us and layered to us and at some point you know I think these things should be abstracted away that you don't think about them that much right so you say I want to launch my chain Nintendo is not like thinking about like I want to launch my chain on the internet which in this case means on ethereum and that's

just sort of the core infrastructure but like from a user whether it's an L2 or an L1 or an L3 the users don't really have shouldn't have to engage with it that much they should just be okay it's it's a secured by ethereum Chain but I sort of interact with it as a first class citizen right I don't think about like today you don't have I got okay I got to go to coinbase I got to go to ethereum I got to go to arbit you're very much interacting this ler I think one of the things of interop we need to do in ux is to sort of abstract it away from users the way we like we don't we don't go ahead and say okay I want to build like you know a streaming service so so let me why don't I reinvent the

internet that's like yeah no one's building the entire stack I hear your point I was thinking about that while I was asking the question the distinction here is though is that the value acurel at different parts of the stack it's much more lucrative and attractive from an incentive standpoint that wasn't possible in the internet that that's and that might change the dynamic I do agree like it's not your core comp it's your y build and https 2.0 like no right at some point these things become standards that that's true and you will need to choose a part of the stack but my my I guess my point is like I think the value crew on Layer Two is good and I think it just sort of abstracts away a lot from you and then like we're going to get to a point where like again like this whole conversation

of like the stigma involve different layers that's because like the layers are so much in your face so if I'm a Shane that's thinking about building a layer three today their users are going to be like hyper aware of like okay I have to climb three three three rungs to get here actually we're building like something internally a tool which will let allow you to jump directly from ethereum to layer 3 it's like a bridging tool but the idea is anyway like today these things are very early so your users like are very hyper aware of where they are of what they are where they sit on the stack but I think if we're successful and this is ironic but if we're successful it's ironic because like for me to say because I know I sit here and I'm building one of these sort of things that are sort of in the

infrastructure level but the truth is if we're successful like Mass Market successful um which I think you know we as a community have been but we still have ways to go um a lot of this will get abstracted away so the user won't be thinking about like am I on layer one layer two or layer three all they're going to be thinking about is I'm on Nintendo right now and I know the Nintendo chain is is is secure I've read that I've I trust you know my source this a secure chain but I actually don't even know exactly what layer I'm on right now I just know that I open up my wallet and I can get my funds Nintendo chain in one click I can get them back to whatever other CH I want to in one click as well I want to end this

discussion with just your overall uh take on one particular thing that I've thought about a lot is um like how would you define Network effects in crypto yeah so I I think you know we're obviously learning a lot um definitely as an ecosystem as we see you know different uh application application verticals join uh application verticals up and you know sometimes get disrupted but I think there are definitely lots of network effects on the um infrastructure layer uh for sure on the sort of chain layer I'll say um where does that happen liquidity is definitely one um and and so I'll give you I used to say this like

um I always talk about security and the truth is though that um maybe only 20% of users care about security I made that up but like let's just assume that 20% of users care about security but there's a network here um because of synchronous composability right the fact you can build apps to talk to each other in a single transaction use multiple apps the fact that shared liquidity there's a network effect here so if my better security solution can attract those 20% of users and your 80% but your solution will alienate those 20% but only to be the 80% then we're better off using the more secure one because we're actually going to be able to get a much larger Network so I think particularly you know one of the magical things about um you

know defi for example right you know defi summer was just the fact that you can just take building blocks and compose them anyone can go ahead and just like launch an app that directly ties into this monetary system and they can also you know tie into not only the core liquidity but to other you know they can you build a pool on unisa right they can actually tie into other pieces as well so I think there are very very big Network effects on the Chain level um uh particularly in application that require liquidity um and and you know this is something which um we're obviously seeing play out um but that that that's that's very very clear to me and I think that's actually as we're as we see now increased um CI if you will

or like trafi coming on board as well they're also looking very very much to to to this to this right so you know arbitrum has seen a bunch of launches from you know more traditional finance companies including uh Franklin Templeton Robin Hood centrifug is building their liquidity pools there and I think that the ultimate goal for trafi is to tie into some of these really nice um um you know Network effects and things you can do synchronously and I think over time we'll probably see a merging there between trafi and uh defi not a full merging obviously but like you know we'll see them uh merg in a significant way I think that's a really good response and a good way to end it there are folks that have said that like most crypto networks have anti- Network effects and I think they're focusing on

the fee degradation the more users that come even if you have composability even though you have security it when you look at it on on the lens of fees right we're all go back to the days where N1 you're paying like thousands of dollars in ethereum to settle a transaction then obviously that has anti- Network effects as you scale but with recent improvements uh you know 4844 and other scaling Solutions you know like l2s l3s then I think that takes the backseat and the more important thing is the user experience right where fees at some point like the marginal Improvement in fees is is not so much as important whereas the network effect uh becomes like more so the the composability that the ecosystem of apps that you're seeing

and so I think uh I think it's a good way to put it and and seeing kind of how what has been built in in arbitrum right but like a year or two ago you would have said yeah most Network most crypto certainly ethereum had anti- Network effects at the L1 like it was just really bad yeah I think this is where like actually separating out separating out verticals is really important because the problem was if you go back a few years ago was everyone was competing for this ethereum L one block space and like you know as soon as as crypto kitties or one app got successful it sort of you know ruined the experience for for everyone and know the analogy that I like to give is like TV channels um which is obviously a dated analogy at this point but I think it's useful um

which is if you had like one channel and every every TV program was supposed to go there every show was going to give this channel like it would be terrible right it wouldn't work you you know just you'd be alienating you know You' never get find someone that that speaks to your audience and it would just be terrible if if you had like and in The Other Extreme if you had every TV show had its own channel it would also be terrible because there would be like no discoverability I can't just go and say hey I I I don't know what I want to watch now but I like game shows like well there is no Game Show Network you can go to the particular Network for the this game show and then you know good luck finding anything else or even like think about like advertising deals and

distribution like every show would have to sort of negotiate those for themselves we'd have is like most of the day you have empty programming what we actually have is you have grouping by by vertical so if you like cartoons you'll you have the cartoon Network you know if you like nature you'll have you know uh National Geographic um and and the idea is there might be like overlap here you can have like a cartoon about nature which could go either way it will never be a perfect um system where it's like obvious where everything goes 100% but users will know where to go when they want to find something and developers will know where to go I think that's what we're doing here so the idea is like when you and you know this in response to like sort of the anti Network effect is we need to sort of

self Shard ourselves I think that's a lot of what we're seeing in this um in this world where l3s and more l2s are taking off where you're going to have chains that are sort of spe specialized and specific for certain purposes you have to find the right balance not that every app has its own chain but every vertical sort of has its own chain or two and I think that's going to help with a lot of those things because yeah in a world where everyone's competing for that same single Channel like yeah the fees are going to kill it it's going to be a terrible you won't be able to scale um but in a world where you have multiple channels and paths and you can sort of self and self organized around the ones where the same user set wants to be and is willing to pay for the fees

associated with supporting that set of applications um you're going to probably find the right balance and and obviously we're still in middle of this experiment so we don't know exactly what will be but that to me looks like um you know the most promising path and indeed I think it's the path we're on with the proliferation of specialized um blockchains that we're seeing today we got four more minutes so I'm going to squeeze one last question sorry just cuz monad uh and told I'm to inv both you and monad but I'm curious there what would you say to the Chism like the evm is flawed it's single threaded uh let's paralyze it let's have local fee markets uh kind of like a monad approach um what what are your thoughts on that yeah so I think

um you know there was this interesting study um this this uh post I saw from polygon where they basically you know it showed like for their workload on their chain like what percent is paralyzable and it turned out like is like 50% which means like even if you have perfect parallelization like um you're going to get like a a 2X reduction and that's fine but I think that's number one where like a lot of the hype that I see and maybe I'm sure this team understands that I'm sure that a lot of you know sophisticated people understand that but I think probably most people sort of they think paralyzation think okay we're gonna take one core and now we're gonna have 64 cores and like things are going to like go crazy but it turns out a lot of the work just isn't paralyzable and

the reason for that is by the way um the reason for that is is that um a lot of times when something gets hot it's the same thing that gets hot meaning if there's like a particular you know if the chain is like you know kind of dead and not no one cares and then it heats up it's because there's that one nft drop that everyone wants to hit that same contract right now and paralyzation is not going to help you there um you know a localized Market is not going to help you for that because everyone's sort of going right there and then but I do think that there is you know a case uh for there there's definitely pros and cons for something like localized few Market um again it won't help you in this case but but I also don't know that

paralyzation and localized markets have to be connected that's another interesting things I think you could build um you know um technology you can you can you can build you know the stack to sort of um you know you can imagine a simple design where like you just have different allocation for different applications you can't use too much I'm not saying it's it's it's a better or worse design I'm just saying these two things aren't tied at the hip you can have parallelization you can have local fee markets with or without parallelization um um is another thing and I do think that you know the the real answer is it's a broad question around changing the evm and the communities uh conversation around that which is I think one that we need to have as a community because there's no one today that I don't think anyone in the world not

falic no one in the world that would say I'm giving you a clean slate we're going to roll back you know to 2015 uh let's launch evm again do you want to change anything and I think everyone in the world today would say yes to at least something right there's a laundry list of things that we can't change because of backward compatibility so there's a broader conversation around you know right now you know a lot of l2s are including arbitrum one are what we call ethereum equivalent uh that they literally don't deviate by Design from the evm and there's a conversation there but arbitrum is actually taking a different approach where it's building arbitrum stylist to what we call evm plus so the evm is staying exactly as it is not breaking not changing at all but in the same environment you'll also be

able to write rust and c and C++ contracts and interact synchronously with evm contracts and you won't even and get a 10x reduction 10x reduction in uh cost for computational cost for doing that so we are very much thinking of how do we expand and how do we uh afford more to developers and users without changing the evm by adding to the evm and it might well happen that fast forward 10 years no one's using or most people aren't using the evm on arbitr anymore they're using something else but the evm is not going anywhere that's our our commitment because like we're still not sure as developers and the community community conversation not mine but like it's still not clear to me like that the community is ready to sort of you know change the evm fundamentally um so that's the approach we've taken which is

yes we want to innovate we also want to leave the evm so Stylus is in the realm of evm plus it's keeping the EV where it is yeah you can't change it because of backwards compatibility there's certain things you just can't change right exactly I mean you could in theory but yes it's it's yes social consensus you would have to exactly like it wouldn't be good yeah I think that's a good place to end it I would love to have you on I mean just have a as I think about like the next time because you're a repeat guest and I think there's a few people out there that can actually understand the history involve you but around this broader topic of resource allocation I'd love to explore just how you think about like local fee markets um like load balancing these are things that like in

the internet like Netflix Pioneer like microservices like a lot of things that have to happen in order for you to watch the same show Queen's Gambit everyone to watch it when it releases how do you balce that load like this is networking thing like these are problems that you know Engineers have to face when the internet was being built and as you scaled and so I think it's no different in crypto like resource allocation and um pricing right salana is having issues now because they need to figure out a better pricing mechanism uh right and and kind of research allocation mechanism so uh these are growing pains for every Network as you get scale and the exciting thing is we're getting scale like we're getting more and more users in using this deck so it's not a bad problem to have but I think to your

point I don't think anyone has solved it uh but you know the teams that are more agile like perhaps you guys are higher probability of solving it over time yeah absolutely and you know happy to obviously you know come back and talk all about that and um yeah there are a lot of open problems in the space and a lot of a lot of shared ground we're all trying to solve some of the same problems we you have like a list of problems and like scaling was like number one and I feel like scaling ethereum we've done a pretty good job there and it's not like the go goes away so now number two becomes number one and uh you know some of these problems have just uh elevated and are now you know some of the most important things to think about in space awesome Stephen as always pleasure

having you on thanks for taking the time and uh yeah for bearing with our um simplistic questions here thank you for having me always love coming on everyone Jason here thank you so much for watching today's episode wanted to take a quick second to thank today's title sponsor arbitrum we know you are tired of on chain experiences that have unaffordable fees and frustrating transaction speeds and that's why we partnered with arbitrum you can experience frictionless trades lightening speed and LaGree transactions all for pennies per transaction explore arbitrum expanding ecosystem at portal.

arbitrum doio that's portal. arbitrum doio see you for the next episode [Music]