Ethereum Sequencing and Preconfirmations Call #16

Transcript

okay uh welcome everyone to sequencing call uh number 16 this is the very first of 2025 happy New Year everyone I think 2025 will be a a very very big year for for base rollups um what I want to do uh before we have uh people present is give a very very quick update on uh some of the things that have happened in the the the last couple months so let me just share a couple slides um and then we'll get going with the with the presentations okay so

um can you see my screen yes perfect so one of the things that I wanted to to highlight is that we've we've had uh sequencing week in changai it was a a huge success in my opinion uh we had about uh 50 different people from I think 30 different teams come together we made I think a ton of progress at the social layer because everyone got to uh meet in person spend quality time we we did this this this caving uh here uh on the very first day to to to get to know each other and then we we worked uh for for four days and these are some of the topics that that that we worked on um and uh if you if

you'd like to kind of dive in into one of these topics um I've shown the the various team leads for each of those that you can uh that you can go contact now the other very quick update that I want to give is uh basically a refreshed version of the this Slide the the the base EOS system and um really there's been um a lot of of uh changes I guess uh at the at the base rollup uh first line here so we we have now finally two uh base apps that are in production uh Tao has been for several months and facet has been for several

days or maybe a couple of weeks and uh what I'm expecting to happen is that in 20125 there's going to be maybe six maybe more base rollups that will reach uh reach reach main net and then there's uh some of the updates here that I want to talk about the the ones that are new maybe to you are those that are uh in uh in in yellow here the the five of them so the the first one that I want to talk about was uh when announced at at Devcon uh name chain um they haven't made any kind of public commitment on what type of of sequencer they want to go with but in in my personal opinion the only thing that

really makes sense for for for name chain is to to launch as a base rollup I think they want to launch towards the the end of 2025 so you know they have a lot of time to to flesh out the exact design that they want to go with uh they are partnering with with Linea to reuse some of their their ZK stack and we'll see how how that evolves the other name that I've added to the slide here and actually it was previously uh hidden uh behind the the stars emoji here so this is a bit of a of a reveal is uh scroll um scroll is uh working on this uh based rollup SDK um and um I think they're trying to uh basically have the SDK be

as much of a public good as possible they haven't made an announcement but they have given me permission to to disclose this here to this group so if you're interested in working on on on sdks uh for for baseall uh do do go ahead and and ping them and uh We've also have a new rollup that I'm uh still uh keeping secret for now uh which is uh you know potentially interested in in in becoming uh based and then on the uh I've added a new category here pillars um people who are basically uh interested in in in pilling other teams to go build base rollups uh one of them is etheral lie which is

a a very new team and basically their focus is specifically institutional people so for institutions the the the the pitch for base rollups is relatively simple which is that it's you know maximum security maximum credible neutrality you don't have a a a a choke point uh which could be a regulatory uh choke point um it's uh you know as as close to the metal as possible um and you know for a lot of big institutions that are moving very very large amounts of money what they're trying to do is basically reduce the settlement times from you know one day or multiple days down to a few minutes or a few seconds and for them the fact

that Bas World apps have a a 12C native block time is is not at all an issue and then um there's a final uh project here which I think you more details will be will be made public uh in in in the near future uh is called based um and it's kind of this this play on word it's a it's a temporary name uh but just to let you know that uh there's you know yet another team that is that is forming out there and what they're trying to do I think is to focus on the the public goods aspect of of based infrastructure trying to have a a stack which is you know as much of as possible a public good and try and encourage maybe as well

some public good base roll apps to to be uh to be launched in parallel to some of the the more commercial rollups so overall a lot of you know exciting things that are happening and uh yeah again I'm hoping that 2025 will be a big year for for base rollups okay with that said I'm going to stop this the sharing and uh open up the floor uh to the folks from facet we have uh the the the the two Co co-founders and also the very first employee from from facet so that's um Tom and Michael co-founders and and Jeremy the first employee and I think we're going to have uh Tom uh present today so Tom the the

the floor is yours yes thank you hello everyone yeah I'm Tom Layman one of the two co-founders has just mentioned I'm here to uh give you a quick intro on Fast first am I okay here I'm not screen sharing but you can see like my video this is fine okay it's fine so fastet is the second based Ro roll up uh as Justin mentioned but being Bas is kind of of a piece with our overall uh strategy here the overall idea which is to be a a rollup that can withstand Hard Times we're going to talk about that and also uh the based aspect specifically so we're going to talk about first why do we exist what's the point of any of this uh now of course how it does will work and arguments against facet okay why is

facet bad why shouldn't you uh use facet we're going to have Q&A so starting with why facet exists so I think these tweets by vitalic are a pretty good framework so vitalic raises this idea of good times and hard times and how the L1 has to withstand hard times and how l2s can build for good times they don't have to withstand Hard Times necessarily later CL clarified the operaters can so l2s don't have to build for good times and we think it would be good to have some l2s that build to that are built to withstand uh hard times and um you know our question is could you have an L2 uh a rollup in this case that is uh as resilient in hard times as the L1 is as metallic wants uh the L1 to be

so let's let's explore this so what are hard times well hard times when everyone turns against you basically it's when trust assumptions break down times of great stress people are getting hacked and so you know a lot of bad things can happen but one bad thing uh that we can explore a really bad thing would be a hostile government or another powerful party trying to stop a rollup right trying to stop all user initiated transactions deposits withdrawals every activity that is happening uh on the rollup so how might this happen and how can we defend against this so what do you want to do if you're the Hostile government so the first thing you want to do is get the admin keys of the rollup okay by admin keys I mean the set

of signatures required to do everything that is possible to do uh with the rollups L1 smart contract track so you know mess with the bridge mess with all the functionality and um in practice this turns out to be about 15 signatures for the most secure uh rollups like op mainnet the stage one rollups and it's less for others uh one interesting thought experiment is the number of unique signatures required to shut down every rollup which is definitely less than number of Roll-Ups times about uh 15 but you want to get the admin keys and uh it's kind of a a loose notion that that you know changes par rol up but it's the number of signatures they're required a number of individual people that have to sign and so if you have that on an optimistic rup at least

on op and this is analogous to other rollups you can switch the sequencer to the zero address okay so you switch the sequencer and now the sequencer centralized sequencer here is uh not uh sequencing user transactions and so the rollup is paused right can't can't do it but then of course we have this notion of forc inclusion so maybe people can get around the sequencer using a special uh technique and still get their transactions in okay but actually we know now uh due to this great article at least this what introduced me to it just the extent to which force inclusion just doesn't work right because if you are trying to accomplish something that has contentious State like withdrawing from

a protocol uh that requires certain preconditions like there being enough liquidity the sequencer can sandwich you uh and prevent you from achieving your goals even if you're using Force inclusions this is a big issue basically the handoff problem recommend people read this article I'm sure many uh have so okay Force inclusion doesn't work so now what about Bas sequencing we bring in base sequencing uh this is way better than Force inclusion because there is no uh Power that has Monopoly over sequencing around the force inclusion so it can control you know the sandwiching so base sequencing is absolutely necessary but then you have this question is it sufficient okay do we need more okay well it's not sufficient in the case that the base sequencing

aspect of a rollup can be disabled by the very same admin keys that were just compromised okay so the admin wants to shut down all transactions okay the admin that's compromised basic when he says no we're going to actually prevent that because we have a way that you know we can we can submit transactions without any permissions but then the admin says okay I'll flip off uh base sequencing and so this is possible uh this is um you know the way it works in you know Tao I think Tao is an amazing pioneering rollup that was first and led the way but I think this is a part of their you know design that I would disagree with which is you know if we say Bas sequencing you know if we say a rollup as base sequencing what do we mean do we mean that it as of the very

last time we checked was using Bas sequencing or do we mean it's baked into the to the protocol itself and can't be changed and so I think that's kind of the question here so it's really not about phase sequencing in the way we use it it's about the power of admin Keys okay so if you have 15 signatures that that can shut off transactions uh this is a problem but we have this other notion called stage two which is to hey don't worry so much about op mainnet today they're making great progress and I believe they are uh they make great progress but still they're not there but what about when all these rups reach stage two are we going to improve the role of bman Key's situation because if we are why criticize this let's just wait basically and let progress roll on so stage two does not change this

fundamental uh tension this fundamental issue what it does do is it says you have to give users 30 days notice before you shut down uh the rollup and uh this is from L2 bead's medium article uh they lay it out here they describe it as an apple time frame you know I love l2b like they're the only ones out there basically doing anything like this talking about this in a rigorous way so I don't want to pick on them uh at all but I think it's worth asking the question okay is it an ample time frame what do we mean when we when we say that how good is 30 days notice how excited should we be when stage two rolls around because of what it gives us here in terms of a notice period so looking at this now you know

we can ask the question well what happens okay the uh hostile government does their hacking they get in there and they trigger a shutdown but because it's stage two we have 30 days so what literally happens in this case well first the happy news of course is the EO a held bridged assets get out so we're talking about base sequencing here right we can't be a base sequencing it can't be turned off so users can force their transactions and get their eoa held bridged assets out and that's the great thing about stage two and that is a great uh thing if we can achieve stage two it's great on the other hand native assets everything that doesn't have a corresponding state in the L1 that you can kind of withdraw into uh all these

assets go to zero which is which sucks you know uh meme coins maybe we don't care about be coins nfts you know the x copy nft on base right like things that people might actually like value uh they go to zero which is unfortunate and then we have contract held Bridge assets so here we have black and white up here EA held bridged assets and all Native assets contract held Bridge assets assets that are bridgeable but not held by a user and so what happens to them well it depends on the logic of the contract right because these contracts are uh you know turn complete systems and so uh you know we can't predict what's going to happen so I think a good

uh starting point might be to illustrate it with uh this um with this with this contract so this is a very simple time lock contract right this contract um allows the Creator to lock up value okay so you in The Constructor here you send in some value it remains in the contract and it sets a locked until date uh to now plus 31 days and then you can withdraw it once that is passed so uh in the in the roll out in the in the shutdown of this this rollup that we're talking about unfortunately uh These funds are lost right because you can't withdraw them

for 31 days and all transactions uh are being shut down uh in uh 30 days so we have this kind of bizarre situation where you know you get to stage two uh which is the one that we're all shooting for and now you can't really write this contract right like you can write it uh but it's not going to function as intended because you know functionality of this contract is doesn't say anything about the L2 shutting down doesn't say anything about it getting hacked it doesn't say anything about anything it's just uh it just expresses a very simple kind of relationship here so we would expect this to not depend on things outside of the contract just as it would on the L1 okay so again to emphasize here we're talking about uh rollups at with sand

hard times can we do it as well as the L1 and quite obviously the L1 can lock a token for 30 days as long as anyone still believes you know in ethereum anyone wants to follow the protocol uh and if we are pursuing a rollup strategy that's trying to scale the L1 and we can't do that in our strategy then maybe there's a problem if we are really worried about hard times right in Good Times maybe it doesn't matter so what are the solutions ions here so obviously facet is a solution wouldn't be here but there are other Solutions so uh one solution is stage three right we invent a new stage and we really uh double down on everything that works in stage two and amplify it so one thing we say is

it's got to be based right because Force inclusion and hard times doesn't work well enough for us and the other thing we say is really we're not we're going pass any sort of notion of signatures or admins really the the uh there's got to be some kind of structure called a dow where we can just Define the behavior of the rollup as the will of this structure so if if the rollup sets shuts down then that's actually fine and if people lose money then that's actually fine because the way this is set up is it's the will of the Dow and so we just make the Dow really expensive to attack and if the Dow gets attacked then we say well that is is what it is and so depending on the cost to attack this Dow uh this can be pretty darn secure you know this is not

a bad solution it solves some of the problems we talked about before uh but we're still not doing as well in hard times as the L1 and it's not coming you know now it's going to take a little while to get to stage two let alone stage three we also have in Shrine rollups right this is a major Topic in itself but the core element here is that you can have an enshrine roll up and this is how I think they're supposed to work where uh there are no L1 contracts to attack so uh you know there's no way to attack the system uh aside from proposing a fork getting people to go to your fork resisting a fork fixing a bug and that's generally like a lot harder than doing something with admin Keys it's always a lot harder so you get the

full L1 security and these are great but when are they coming right one year out three years out 5 years out we don't know a lot of things have to happen between now and then um I actually didn't know what number to put here because you have to have a lot of context uh in the ethereum road map but I put three and then once they're here it's obviously hard to iterate because everything has to be tied up in you know the L1 road map and so it raises these questions about okay what is the role of the L1 do we want the L1 to take over all of this or do we want the L1 to focus on the things that only the L1 uh can do so another great solution again though not here so now we have facet so facets got one really big uh thing going

for it here which is it exists today and it's impossible to shut off so if you want something that is out there today that can be used today on main net and can Implement a 31-day time lock smart contract on evm there is literally one option right so in three years one year there might be other options but for now uh there is one option and so um you know this is not free there are downsides to facet and we're going to talk about those but uh if you are singularly focused on being impossible to shut off and existing you have no other choice so what is facet okay so this is a little bit there's some words here this is like a little bit of like the breakdown of the

specific labeling and details of fastet then we're going to talk about how it works from a normal standpoint but for this audience I wanted to make sure how we get into the details so what is facet so facet is a based Sovereign rollup so it has the word based in there that's recognizable there are other there is another base rollup there many Bas rollups plan Sovereign a little weirder you don't see that really that often when it comes to ethereum rollups and then you know rollup you see but is actually a little bit um there com different definitions of rollup we're going to get into that so facet is a rollup so when we say rollup we mean something very specific we mean a an ethereum rollup is a blockchain whose

state is derived by applying a state transition function stf to ethereum's transaction history so if you have that you have a rollup and rollups are desirable because anyone can know the state roll up without any external information aside from the ethereum L1 and without using uh any when and just with using software that is Source available or open source that they can get so there's never any dispute about what the state of a rollup is because it only relies on uh uh decentralized information and um deterministic software so that's what we say when we mean a rollup l2b means something a little different which we're going to

see on the next slide another way to look at this is we use ethereum for da like a rollup ethereum rolls rollup uses ethereum for da bcet is based so of course a base rollup uh is a rollup where uh rollup blocks can be permission are permission in permissionless included in L1 blocks so uh there is no one with more power than anyone else no gatekeeper and um anyone can propose blocks uh help create blocks and so this is super important you can't be resilient against hard times with the centralized sequencer even with um Force inclusion so uh in kind of another way to look at this is we use ethereum for sequencing right

when you're looking at the big picture R thing and then finally kind of the curveball one right facet is Sovereign a sovereign rollup is a rollup without a canonical Bridge without in our case and I think this might be true overall without any enshrined L1 contracts in other words enshrined here meaning recognized by the protocol as special and playing a role in the protocol itself so uh the facet world the facet ecosystem has smart contracts Bridges and so forth uh but all of those are uh equal in the eyes of the protocol there's no special smart contracts and we think this is like a super important thing because special smart contracts uh

you have to upgrade them and then who has the keys you know to do that so uh no canonical Bridge right means that there is no special Bridge but it does not mean that there are no validating Bridges to the contrary any number of validating Bridges can exist and they can have the exact same security level as any normal rollups canonical bridge in terms of proofs uh in terms of literal code as we'll see uh so this is not a question of the role Bridges Play It's does anyone have a monopoly on Bridges so stf is enforced by social consensus meaning that uh no one has any power to make you use version one or two or five of facet or any s rollup uh it's just like the L1 in

this way uh and then finally reiterating this you know users can validate the state by running their own full node and there are other ways to do that again using proofs and so forth like but the base line way the thing that the proofs are trying to get to uh is what happens when you execute a full node which anyone can do so I mentioned this a moment ago like what is a rollup uh I told you our definition but I just want to emphasize for clarity here we are a rollup but we're not a stage and rollup according to L2 beat and so um let's look at why that is so what are the criteria for being stage zero for being a rollup at all well first criteria this one's pretty easy we we nailed it the project

has to call itself a rollup okay so we uh do that the inputs for the stf are posted say transition function posted to ethereum L1 so we do that as well a source available node exists that can recreate the state yes we have that and we agree that these are the uh criteria but um uh L2 be has one more which is State roots are posted to the ethereum L1 and so we don't do this although we actually do facet the company actually Post State routs but that is an activity that we're doing because we want to not because the protocol you know requires it which gets to this idea of how you know sovere rollups allow all these

great things but they don't enforce it because they don't want to have uh privilege surrounding it so uh anyone can go to Bas or op mainnet and start posting State routes and 100 different people can do that uh on their own uh Oracle uh and likewise for fastet anyone can can do that but it's not part of the protocol and the reason for this is that you know we don't think it's good to wrap up Bridge security which is what state roots are fundamentally about with uh rollup security um when you do that bridge security is harder right you don't want to uh it has some necessary elements of privilege in it and when you

give the bridge controller the canical bridge controller power over the whole roll up you end up amplifying the power of the bridge controller to a point where they can jeopardize the uh uh the big picture so we're trying to separate them and that's why uh we don't do this or anything else that's related to uh a pro protocol recognized official Bridge Okay how does fastet actually work okay so the first most important thing to understand about fast is an optimism Fork so we looked at optimism and we looked at other things too optimism is the best optimism is amazing software we love optimism and so we looked at optimism we said what do we have to modify in optimism to get the

results that we want and so if things look familiar then that is intentional and the reason that we try to keep as much familiar as possible is so it's familiar to developers and users so it's an optimism Fork to create a facet transaction you send an ethereum transaction to this inbox address okay so this is different from the usual conception of a base roll up and we'll get to like kind of how we're going to maybe evolve this or definitely involve this in some ways but just like the Baseline level this is a very inforced inclusion type idea in other words instead of a full facet block being proposed at once in a blob or call Dat whatever individual users propose their own transactions one at a time much as in the force inclusion uh

case okay now you have to pay for gas to do this and this rollup needs a uh this rollup needs a native currency all blockchains all evm blockchains need it and so we have a bridgeless native currency called fastet compute token uh that is necessary to um to buy compute and then finally we have the support for optimistic bridges built by anyone and so really what this means is we have the classic L2 L1 message passer predeployment built-in way to kind of do you know proofs so you don't have to prove the whole state you prove the state of this contract and so forth so uh that's the

big picture here are some details okay you start with an ethereum trans action all right and so what you do to create a facet transaction is you create your facet transaction object which includes only a few Fields right we don't need um other fields on this and we take this transaction object we put in this envelope which is the ethereum transaction so uh the data call data etherum transaction is the rlp encoding of that fcet transaction including the transaction type gets sent to the inbox burn address and the from on the fastet transaction when it's interpreted by the node will be the from implied by the signature the person who

signed the uh ethereum transaction okay and this is how EAS Set uh create facet transactions contracts can do them and we'll see an example of that in a moment so here's how it works so once the transactions are sent uh right go to this facet inbox okay now they're in an L1 block so this is up here this is nothing to do with facet this is just how the L1 works you can send transactions but now our facet node fetches these transactions that were in this L1 block using you know get block by number and get transaction you know get Block receipts normal RPC stuff and then builds blocks by extracting the facet transactions that are in the L1

block bills blocks and sends it to facet geith which is uh a geith instance using the engine API and the geith instance you know computes the state of the blockchain and the history and and so forth and this is very similar to optimism facet geith is a fork of op geith uh facet node is not a fork of op node but it probably should be maybe we should we're going to try there's some details about why we couldn't do that but uh ideally it would be a fork of of op note it's very very very similar and um trying to keep it that way and then finally users can query data from fet geith just as it were any other RPC endpoint you know any read function get whatever you want to get from from that you can't uh write using at uh rpcm

point you can only write by sending these L1 transactions so this is uh kind of the flow of how it works so if you go to our docs you can see this part of the docs which is how to run a note and so the the key here is um again everyone thinking about rollups whether it's us l2b we all agree you need a source available node and that's what this is so if you want to get the facet state any time guaranteed 100% guaranteed you go to uh you clone this repo and you run doer composts and you have to add your L1 RPC which is you know part of the theory here right you need an L1 RPC but that's it you need

the software L1 RPC you're done it will work every time so just to emphasize that anyone can you know do this so now we have this question what about gas because on normal rollups the gas token is bridged ether and we don't have uh you can have Bridge ether but it won't be the uh the native token right there's no you can't have that because if you have that then you need one bridge to be special in the eyes of the uh protocol and we can't have that so we need a different gas mechanism and so the way this works is basically uh when you create your facet transaction when you're putting something in the call

data you can put this amount in the call data a little bit or you can put a lot so this would be maybe the example of a um contract being deployed on facet right so you have a bunch of bik code there here you have less and then you take the call data size gas units you multiply it by the uh fct mint rate which is dynamic and you get out your fct Min so if you're 100 gas units and the rate is 26 million 2.6 million gay per unit you'll get 03 fct and then after that you will burn a portion of that because you need to create your you you pay for your fastet transaction which burns fct and you'll be left over with a balance this is very

similar to how optimism handles uh paying for deposit transactions which are also paid for using L1 gas uh but we have various differences here most notably that you can maintain a balance uh of fct which is not possible in the optimism case and the purpose for that is for um future uh being able to pay for for batching and so forth so we have this what about bridging so I want to make sure we have enough time for questions so I don't want to get too far into the bridging thing because this is by Design not meant to be interesting this is just us saying we do it you can build a perfect copy of an optimism bridge on fastet and work in the exact same way so

I'm not going to go through this this how deposit works this is a quick shot of how you send transactions from a smart contract um let me keep it moving this is how deposits work excuse me withdrawals work state Roots everything we don't have fault proofs right now because that's more modifications you have to do optimism system uh but we're going to get to that but yeah withdrawals complicated process the exact same way optimism works you can build the exact same Bridge it'll all work in the same way the only difference is no one has a uh Monopoly and the gas token is not bridged okay so what is on the uh what's on the road map here so number one most important thing thing from a user

standpoint is we need a way to send transactions that is not so expensive right if you're sending a transaction on fcet in an ethereum transaction it's going to cost you a dollar instead of you know $15 whatever cost on the L1 but people are used to less than a Cent and so you have to take things out a lot so you know how can we do batching uh the main idea we have here is to share it as much as possible in other words users should be able to hitchhike on any batch in addition to any formal sharing arrange that we create on a protocol level you should be able to put your transaction in a batch on base and you know maybe there's a time delay there of course but that's what you want to do that should be recognized uh as part of

fastet you should put something in a tao batch you should put something in uh any batch that makes it and then in and ah but the final thing is we want to preserve this forced inclusion thing we have today where any user without creating a blob or type3 you know anything like this any user can just get their get the transaction in ball prooof so this is engineering challenge but not conceptually uh difficult just changing the way you know uh stealing optimis amazing work basically and and using it and changing it uh notably though we can share settlement with other rollups other base rollups there's no reason why uh someone couldn't be uh involved in proof on uh

facet and other rollups and have a bridge that does both um again there's no Monopoly when it comes to the bridging aspect so anyone is on equal footing to build the best Bridge arguments against facet without an enshrined Bridge how do you know who to trust right you go to use base you know you got to trust coinbase that's it you you say Boop trusting coinbase at the door and then you don't have to worry once you're in there that's like a waiver you have to sign and there's something nice about that you know it up front uh with fastet and with ethereum L1 you don't know who to trust you want usdt well they might be they're unaudited and they have a problem but what about usdc I actually heard that

they couldn't they halted withdrawals because of the Silicon Valley Bank thing you have to figure it out and it's annoying and so if you want to get something as Unstoppable as the L1 you you need to take on this negative aspect of users every user basically having to figure this out uh but we think the trade-off is is worth it in hard times so another question is like you're going to have some too big to fail bridge and you're not really fixing that problem cuz you your too big to fail bridge is going to be the same as coinbase not to pick on coinbase here just any example it's going to be the same as any other rollups andrine Bridge once it's too big to fail it's going to have a ton of power and uh its security is going to drive the rollup

security and uh you know we agree right uh even on the L1 this is the case if you want to Fork the L1 and circle opposes you you are going to have a really tough time doing that and that's how it is on facet and that's also how it is on any normal rollup if the most powerful person says play by these rules you know you have to also the powerful person can lose the money or steal the money but what a bridge can't do is stop the chain using admin keys right you we've given that to bridges in normal rollups they don't by default you don't get that Circle doesn't get an admin key vitalic doesn't get an admin key right no one has an admin key on the L1 and we've given this to Bridges and also Bridges don't get Monopoly right Bridges

have to fight but we've given them a monopoly in the normal l l2k so this is an accurate criticism in that there will be no one not everyone will be equal just like the L1 but I think there's real differences that make it not the case that bridge security and rollup security are the same uh you know do users want ethers in Native currency probably right fct I don't want another token I don't care about this uh unfortunately we can't do it but yes this is a downside you have to do a different thing than you do on every other roll up in the L1 um but there's no way around it given our architecture and finally um this whole we call Atomic based sequencing where you send individual transactions and that is assembled later into the block

uh it's expensive and confusing you don't want to connect your metamask and send call data garbage to a garbage address that's not like what we expect from metamask we expect it to show us a preview of what our assets are going to do so this is a mess right now basically uh and again we think it's worth it because of the benefits of being Unstoppable but uh it's going to be a big problem for wider adoption so we need to get the based excuse me we need to get the batches uh to work so this is definitely a very valid uh criticism there are probably other arguments against facet but I want to open it up to Q&A so thank you all for for listening to this portion amazing thank you Tom so much for the presentations um I've even had

someone uh DM me privately saying this talk is is amazing it should be public and it will be public on the uh pre-confirmation and sequencing uh YouTube uh playlist awesome I guess one one of the more of a of a comment than than than than than a question uh you mentioned the andrine rollups um one of the things here is that we've renamed them to to Native rollups um and uh today actually we had the very first Native rollup uh call so it's a series of calls similar to to to this one uh it's being led by by Matthew from from Spire if you're interested in

joining the native rup call and telegram group uh do do send me a a message oh yeah we'll do so what do you think I mean you're the you know because I was trying to guess and I you know what's the what's the what's the correct way to give this slide and this part here how many years out uh is it so I can have a good slide yeah that's a great question so um if you had asked me a week ago I would have said something like three years uh but if you ask me today um I would say something closer to to two years maybe potentially the fusaka fork in 2026 and

the reason is that we realized that uh there's a way to deploy a version of native rollups that doesn't depend on SNS so just like uh Proto dang Shing is kind of Faking It by asking all of the validators to download all of the blobs as opposed to doing data availability sampling we can have the validators fake native rollups kind of do it the naive Way by just re-executing the the state transitions um and the main constraint here is that we would need a fairly small gas limit so we we wouldn't have the amazing scalability benefits of snars uh but at least we would have uh Native rollups for people to to get comfortable with and play with and uh

and start familiarizing the the the C with and this this small limit would basically correspond to the the six blob per per block liit that we have that we have there um and uh one of the things that I'm hoping to do is basically uh hire someone full-time at the foundation to lead the rollup effort and uh Shepherd it and accelerate it so um if I am successful in in doing that and I do have some some candidates in mind then uh then yeah hopefully uh 2026 is kind of the the earliest possible Fork where it could come in awesome that's I'll update the slide and I'll change the name I'll change the name very

cool any questions thank you in the chat uh I do see the handoff problem has been dropped in the chat I think uh really completely obvious in retrospect in a way but like super eye opening when I first read that I was just like whoa like we are in trouble right now with this whole uh centralized sequencing uh thing so um highly recommend that um ores you have your hand up do go ahead yeah um I'm just trying to wrap my mind around this is one of the many validating Bridges not becoming the def facto enshrined bridge at some point

thing or like Outlook and I'm wondering could you describe for instance for um assets such as stable coins how um fungibility would work and like how this scenario might play out sure so maybe jumping back to the to the bridge slides so uh if you look at like the players on this slide you see L1 standard bridge and L2 standard Bridge right and so if you try to bridge on an optimistic and op rollup you will see these same contracts and L1 standard bridge will be you know controlled by the admins controlled by the protocol

and uh it will be the one that everyone is directed to to use but the same EX code can be deployed by anyone so for example in the stable coin case Circle if it wanted to could deploy the same exact code and say Here's the usdc uh bridge and it's the same exact code but the L1 standard Bridge supports all erc20 tokens in one bridge and so no reason to to do that because the standard Bridge always gets all these gets all these benefits uh anyway so this would be the criticism of facet you'd say well if it's chaos then there going to be six bridges that are usdc and they're all going to be incompatible tokens on facet and that's bad you know

you could say well you know would they be incompatible maybe there would be ways to make them compatible but I think more likely the answer would be someone would Emer you know there's Network effects here obviously and so someone would emerge as the um as the best one or the best two or three and they would um they would would they would fight it out so usdt and usdc on ldl1 are not totally fungible although you can make that work using defi uh and that's how it would work on um on facet and the hope will just be you will have the right number of these things where uh you know you don't maybe have one really huge Monopoly but you don't have 500 people uh and that's kind of like the

the question really is how to see it play out but the overall idea is you pay a huge price for giving someone a monopoly a huge price and so you know that's the choice do you want to get that convenience at that huge price or do you want to pay our price which is not zero to to get you know the equality type I don't know is that does that help so I guess you're saying let's say St Bridge stable coins are not fungible with one another anyway so why not accept fragmenting the liquidity a bit more and having different versions of bridge tokens in exchange for an oligopoly rather than a monopoly is that

is that the trade of you striking um I don't wouldn't necessarily say an exchange for olop I'm more so just pointing out like the way it works on the L1 given the same preconditions basically where on the L1 there's no enshrined or chosen or Monopoly given granted provider of any uh stable coin and bridging is not the same exactly than having real world assets so it's not exactly the same of course but uh given those initial conditions we can see how it played out in the L1 and draw conclusions from that and so I'm proposing like you can kind of draw conclusions it's the best shot we have at at predicting how something like this might uh work um and I'm saying you know whether I'm saying if that happens it is

still better than there being a monopoly but beyond that it's better than uh it's better than sorry I'm reading the chat there it's better than um giving the bridge controller you know so much power right where uh you know the the the the the admin on the L2 controls the bridge and that is connected to controlling whether or not you can even send a transaction uh at all but um I mean argue one takeaway from the L1 world and the rollup Centric road map is that the fragmentation is already too large and I

see Greg just dropped a message in the chat I think reiterating exactly that like different Bridges even connecting the same two chains are inherently um non fungible as in security parameters so how would this not fragment the the the the the Spectrum even more right so I think the fragmentation yeah I think the fragmentation is a great question I think right now we see tremendous fragmentation you know uh today where there are a million different versions of usdc because there are a million rollups and so each version on each rollup is different it's not the same it's not funu because the security situation on these different

rollups is different who signs you know whether they have a doubt you know it's different different Bridge security means the tokens have different relative you know positions so uh we have that what we are saying is if you have to have fragmentation and no one's really proposed a great solution to this in my opinion it's better to have it on the app level than on the protocol level in other words uh if you have it on the protocol level you are dealing with like a fundamental incommensurability between what is going on on two rollups because they are controlled by these external factors like you know someone's Keys whereas on facet everyone knows they're playing by the same rules everyone's running a node they know that you can deal with that fragmentation you know more uh easily uh because users can

switch from app to app and these apps can can work together uh but the overall issue that you say exists on op mayet today because anyone can create a bridge for usdc and if that bridge became popular you'd have the fragmentation again and so what you're saying is no one's actually going to do that but you don't know that you're just saying based on the Monopoly based on social forces based on consensus based on all these factors you are really predicting its low percentage likelihood and I'm saying if you look at facet what's your prediction like we don't know how it's going to play out on on facet are there going to be 5 10 100 1 we have to see but there's no technical reason that when it comes to Bridges fast it would be any more or

less you know fragmented because anyone can build and equally good you know validating bridge on on a on a rollup an ethereum rollup which is not say you know it's it's an issue the the fragmentation thing is definitely an an issue we have to we have to deal with I I think um you know whether it's more giving someone a monopoly makes it less likely that they'll be fragmentation for sure so in that regard I think we are more vulnerable Ellie um so yeah I think this is a really cool idea and I'm just asking this question to be Double's Advocate I

don't mean it in a negative way so if you are submitting transactions to facet through an ethereum L1 transaction and you kind of have the same security as the ethereum L1 why not just use the ethereum L1 itself um great great question and the reason is because when you submit a facet transaction all you are submitting is the inputs to the function you're trying to run you're just submitting the the data required to do that execution in Dumb call data on ethereum the actual execution of the function uh happens offchain in the uh using the facet protocol so for example if you wanted to deploy next deploying a

contracts not such a great example but there might be like a Unis SWAP SWAP for example and say that swap C cost 500,000 gas units it's complicated the L1 reproach would be to actually spend 500,000 gas units in the Unis swap contracts but the fastet approach would be you put uh the amount of uh data requir ired to um you know the you put the transaction data I.E the contract you're calling and the parameters in call data you send it here and that's maybe going to cost 20,000 uh gas units because none of the execution happens uh on on ethereum so that's the thing that you are uh saving which is also what happens on normal l2s uh yes sorry

another really important thing to say is the facet inbox is not a smart contract it doesn't it's just a EA burn address basically so it doesn't do anything with this uh stuff so really it's the same as normal l2s normal rollups actually and um the difference is on facet you have to pay this overhead of 21,000 gas units every single time which is the thing that makes it uh bad but it's still way better still 10 or 20x better than than doing execution on the on the L1 yes thank you Jeremy uh Luca hey

so um uh first of all I I wanted to say that I spent some time uh this morning uh looking into facet and I actually added pass to Al to beat uh so like you can go I can maybe link it already in the chat if anyone is interested um because we introduced uh this new category which is called others uh which is Loosely defined on purpose to like um add all the projects that um you know either don't really fit into the current definitions that we have or like experimental stuff um another example is hyper liquid how that we add

it uh in any case I have a few questions I'll try to prioritize um so the first thing that they don't really uh get uh from from your talk is what you mean by L1 and Shrine contracts because um really not like let's pick the opack since you're an OP Fork right the L1 ther bridge is not enshrined into like optimism or base in any way in a sense that uh the L1 standard bridge is just an escrow that uses the arbitrary message bridge to like communicate to dl2 uh the something happened and then

they mean tokens right the only definition of enshrined that I can see is for gas token right uh you need e to transact on optimism and you need to bridge e through a spe through a specific contract which is the optimism portal right um and um but in optimism today you don't need to Bridge anything to use the system like you could use deposit transaction burning the gas onal one without ever like bridging anything so like I don't really see any difference

from like what you're doing here with OP stock like normal op stock like optimism base you can already transact on optimism and base without bridging anything by burning G on one with deposit transaction and the fact that anyone can build a different bridge is true um for all the systems so that's also another part that I don't fully understand like you can build like suent could deploy an escrow for optimism this validated with sp1 um so I don't again I don't really get the difference between like your system and another one with the

exception of the gas token that I understand but um again um you can already transact uh like on optimism and based r one and like apart from this like there is a bridge like that you developed uh that is the one that is described on on the Alit page which is an EA Bridge more or less like you discuss that you don't want to have like it is not okay to have systems controlled by 15 addresses but like the bridge that you build as one EA that can like steal uh all the all the tvl uh today

so okay so let me respond so uh I think you're wrong about a couple things so one the biggest thing you're wrong about is you can't transact using Force inclusion on optimism according to the protocol you can do that only when the admins have it enabled and admin can disable Force inclusion by upgrading the contract and getting rid of it so this is the huge thing that all about here right which is if an admin can disable forced inclusion what is the value the value is only in the case the admin is not hacked but we are talking about hard times here so I don't think you can you can depend on that so that's like the really really big difference is an optimism is not the case that you can use forced inclusion you can use it at the pleasure of the admins does that

make sense uh yeah I mean because you because you need to call the deposit transaction on a smart contract in that sense yeah but like like the other the other part that I don't fully understand is like when you say that if the bridge is acted um you lose native assets which I don't really believe you know that's true like even if the bridge is compromised like you can still act as a sovereign rollup like there was a big discussion in the past on like that native assets are not affected by uh the chronical bridge if act uh because of

you know two things either you can use the system as a s rollup or you can even withdraw like if the value of the token of the meme coins or the nfts is in The Ledger you could in principle withdraw such Ledger on one or continue using dots as a as a sovereign roller even if the bridge so I didn't say that the bridge being hacked would cause thetive assets to go to zero I said that the role being shut down causes the native assets to go to zero and so the shutdown case is the one I'm talking about a moment ago where the uh Force inclusion gets shut off the sequencer gets shut off and no transactions are possible at all which is a much stronger result yeah but like if you act as as a sovereign

rollup it's trialed to fork and use another contract or use another eoa for first transactions like if you have S rollup you can do anything sure I mean forking is a second order question here though like anyone can Fork anything at any time and if you are really you know but forking requires a lot of work you got to say here's the fork this is the Right Fork social consensus hey everyone come upgrade your software open C everyone upgrade your software come actively join my Fork right that's a lot of work what uh op mainnet admins get is the ability to Nuke the rollup without causing a fork on the current Fork people's existing software will break will stop cause having transaction so if you're saying well Robbs can Fork to uh

to save this then sure I I agree with that and I'm saying the downside is that's a lot of work and also the uh admins have a ton of power even if that can't happen because they nuke the old Fork right so in the case of normal forking you have this old Fork that can work so it's like chaos right they nuke the existing Fork so yes I think the problem is is not that what forking can or cannot do it's the problem is forking is a second order thing the most important thing to look at is uh what the admins can do without anyone doing anything on the current fork and just to just to respond also to the to the bridge thing so our bridge is

not controlled by a single uh eoa um you know I would say our bridge is absolutely not as in line with uh you know it's I wish it were more like op mayet it's like op mainnet has uh you know Great greate Bridge um you know are we have a history of um you know having some uh leeway for bridges that don't have proofs which is a bunch of things happening now that you know don't have proofs so we want to improve ours you know really quickly but uh it is a big challenge to uh to create a proof system that works which is why optimism spent so long on it and uh it's going to take us some time to um you

know modify their changes to get it to work so no not trying to at all say uh that our bridge is uh where we want it to be just in the same way that you know 6 months ago op wasn't trying to say that it's not controlled uh by a single eoa and um you know I appreciate being on l2b like l2b again like I uh obviously I'm pushing back on some of your criticism now um but again I want to say I love L2 beat I love the work you do you know it's an amazing site it like let's be real like I'm sure people criticize it a lot it's the only thing out there right that's doing this so I want to show True Love for for your work on on L2 be and and not uh uh push back uh uh too too hard on on any of this uh

this criticism which which I do appreciate and you know if I'm pushing back also it's I probably need to be clearer when I uh when I describe it um in the first place so thank you for for that so uh just quick real quick seeing in the chat Ellie asks fct should in theory always be pegged to eth the price Dynamic of fct it's interesting right like you're giving up eth to get it so yes but the rate of fct changes based on how many people are uh uh minting it at a given time so that rate fluctuates which affects the price and then also the you

know gas price is relevant because the fee you pay per fct is related to the gas price but uh aside from those Dynamics yes it's the the only way to get it is by either buying it or burning it and the person you buy it from burned eth to get it so it should be related to the price of eth uh for sure maybe as a quick follow-up question uh maybe you could expand more on how the minting works because you said it you know you mint it at the uh gas price rate are you looking at literally the gas price that that transaction on the L1 consumed and then minting accordingly no that's a good question it's a little confusing so uh so you take the call data size so one thing is this is not

the full size of the transaction this is the size of the facet portion of it so maybe it's 100 gas units fct mint rate this is part of the facet protocol this has nothing to do with any rate on the L1 uh the facet protocol adjust the mint rate using something analogous the way Bitcoin does difficulty adjustments where uh every 10,000 blocks it changes this based on mint activity and then you multiply these and you get the uh you get the mint so the factors that determine the eth price are the eth price of this gas and the current mint rate from the the facet protocol and then you

multiplies these things and you get the amount of uh of fct does that does that make sense am I yeah that makes sense uh could you explain like the purpose of the fct mint rate and why like why you want that to fluctuate so the idea okay so why does this exist at all because we were talking a moment ago about the optimism setup and optimism as is force inclusion thing which can be disabled but they also allow you to pay for stuff in L1 gas and that's what we're kind of copying here optimism doesn't let you keep a balance though when you do that and we think it's important that you keep a balance because we going to want to pay for things using fct like

inclusion in batches that are going to be getting into the L1 so we want else fct to have some kind of like scarcity to it so it has some kind of sort of value uh notion and that means we don't want people just be able to like mint unlimited and so we try to come up with a system where we would want it to be somewhat smooth and regular so that uh you know you wouldn't see huge spikes based on uh demand changing like the rate should track that and then also uh the rate or ingredient in the rate gets cut in half every year like a Bitcoin style having so uh that comes into to play so the idea is have it be available enough in the early stages for people to

go nuts basically and then as we move towards batching as we move towards greater uh use cases for this this token have it be a little bit more scarce so it holds its value and kind of represents what ether I guess originally was maybe supposed to represent which is like bottled up compute that's what we call fasted compute uh token but um it is very new right now so we have to kind of see how it shakes out uh it is live it is not um yeah I don't think it's listed by anyone or maybe it is but I don't not surprised

not by those two you can see it if you want to uh use our Explorer uh this is a good way to see you can just see top accounts and see everyone who has fct I don't know if you all know uh block Scout but it's actually like basically better than ether scan so shout out to block Scout shout out to l2b like these are uh great resources but anyway it's it exists you know you can get it using RPC stuff and block Scout uh puts that together but there is um I don't think it's listed anywhere it doesn't have a representation on the L1 like so you would have to I mean obviously not every coin on coin gecko is is on the L1 but you would have to have some kind of like

it's a little counter intuitive in that in that way but it does have this it's wrapped fct it's cool but yeah hopefully it'll be listed amazing Tom thank you so much for the presentation