Projects
![]()
No projects found
Investors
New
DAO Treasuries
DATs
Lend / Borrow APRs
Chains
Categories
Spaces
STAGE 2: How did Facet Become the First Stage 2 Rollup?
Recorded: Aug. 11, 2025 Duration: 2:08:12Space Recording
Short Summary
FACET has launched as the first general-purpose Stage 2 roll-up on Ethereum, introducing the FACET Compute Token (FCT) and offering innovative yield opportunities. This project marks a significant growth trend in decentralized blockchain solutions, emphasizing user control and security.
Full Transcription
Thank you. Can you hear me, Michael?
Yes, let's go.
Are we able to get some music in here.
That's a good question.
You do it.
Whoa.
Feedback loop.
Yeah, that was my bad.
I'm trying to do the thing where, you know,
oh, I got that input volume thing.
Okay.
I'm trying to do the thing where facets,
you know, in the thing, and mad yak. What's good, dude? It's been a minute.
Okay. Now let me, um, let me figure out.
Let me figure out.
Okay, what's good?
Welcome, everyone.
PIV, we got PIV here.
One of the original facet believers.
Let me see what...
I didn't quite understand.
Anyway, great post today. Collected on Beamer.
Hello, everyone. Hello, Piv.
Should I retweet this? We've got to get some music
on here. Michael, I'm juggling
the facet host
thing. Dude, you're the DJ, though.
I've never done so much pressure.
Ooh, what do I do?
Yeah, just point something to the
thing.
Point.
Okay, maybe I can do this.
Okay.
All right.
All right.
Okay.
Oh, no, I can't because I have it open up my computer.
And that will trigger the – yeah, there's feedback on this that's going to happen for me.
So you've got to do it.
You've got to put your thing up to a speaker.
Or one of the audience can.
Okay, so how do I read this?
This is always great.
If you're here listening to this, don't worry.
We're going to cut all this out of the recording.
You're sick for being here anyway.
You're a true lover of facet or hater,
you know, if you have passion. Okay. Let me share this. Come join our space.
Got to capitalize space though, because of the, you know, I'm sure Twitter appreciates that.
Okay. Posted. space though because of the uh you know i'm sure twitter appreciates that okay posted
okay
so there's going to be uh we're going to do this other thing tomorrow the unstoppable roll-ups
news probably and um or the unstoppable roll-ups version of this,
but this will be good as a spaces.
We'll get started soon.
A great tweet from Bartek.eth,
who's a classic Ethereum and L2B person who I've always followed.
And he tweeted, I'm not going to deny that assessing 0xFacet was challenging for the whole team.
And we love being challenged.
And we're proud of what we did, basically.
Not to paraphrase that too much,
but here I can put that in the thing here,
because this is a nice tweet by Bartek.
Okay.
Okay, boom.
Going to throw this in the nest.
I mean, there's also the post, obviously, the original post.
But let me get this in the nest.
Yeah, you know what?
I'll just leave this other one in the nest.
This is good.
Are you pinning these? I don't see one pinned leave this other one in the nest. This is good. Are you pinning these?
I don't see one pinned.
I put it in the nest.
Huh.
Doesn't show up for you?
Oh, it's in the nest, dude.
You gotta check that.
Gotta check the nest.
You gotta check the nest.
Oh, alright right i see it
okay all right what's up y'all we're gonna let some people filter in we'll get started
michael's apparently gonna be playing some music so there's no dead air which is the worst thing you can have but I have not seen a lot of can we get an update on
that Michael just put on some you know some chill music I got you I got you you know some
sweet with back I got there I got you I just don't know how this is gonna sound it's gonna sound good dude let's see
How's that
Great
Nope cut out
Okay, I actually should retweet this from Facet probably, huh?
How do I do that?
OK. Okay. Okay.
All right, all right.
Now keep playing it, keep playing it.
We'll get started in four minutes. All right. Okay, okay. 540 other listeners.
Who are these people?
Okay.
Very nice. Very nice. Nice.
Okay, cool.
Let's get started.
Let's do it.
Let's do it.
Let's do it.
Thank you, Michael.
Thank you, everyone.
Hello, everyone.
Welcome to This Spaces.
This Spaces is called How Did FASET Become the First Stage 2 Rollup?
That's a little misleading, I'll just say, you know, we're trying to obviously promote everything that we've done,
but we are, you know, we want to stay humble and also stay thankful to those who have blazed the trail before us.
So we're not the first Stage 2 Rollup, we're the first general purpose Stage 2 Rollup.
General purpose means you can do anything on FACET that you can do on Ethereum layer one. Other stage two roll-ups so far have been what are called app chain roll-ups that are
more specific, have more specific use cases. So FACET is the first general purpose
roll-up on Ethereum in stage two. And this is something that we are really proud of and we want to talk about today,
kind of how we approach this problem,
the innovations that we think we made.
And then we're also going to discuss
limitations of the facet approach
and what we feel like we need to do next.
Because stage two, you know,
the way we think about stage two
is it is the foundation, basically.
Like it is the minimum entry fee or whatever.
Like, you really should be a stage two roll-up before you were in the game at all.
Obviously, very hard to launch a stage two, but to us it's the foundation.
So we want to do that, but then we've got, you know, we're not done here.
We're just starting.
So what does stage two mean, first of all?
Okay, so stage two, and they had to, as you can see in the nest up here,
Bartek.eth had a nice tweet talking about this.
Like, FAST is a little bit of a weirdo in terms of Ethereum roll-ups,
and they had to kind of work hard to fit us into this, into this, this designation, because we approach
things very differently. So we're going to talk a little bit about that. And we're going to start
with like, what is the point of this? Like, why are we here? What did we think about when we first
created facet? What is the purpose of any of this stuff? Okay, because phase two, stage two is a
bunch of like boxes you have to check. There's a famous pizza. You can go to L2B, you can see the pizza.
It's all green.
That's great.
There are, I guess, what, five slices on there.
So there's a lot of details in terms of the mechanics of how this works.
But the most important thing to understand is the big picture.
Okay, why is this important at all?
Who cares?
So, you know, for us, it really comes down, I think, or at least for me, to the starting point
of this world computer notion. Okay, so, you know, if you use the internet, you will know that there
are a lot of platforms out there in the world, and these platforms exist on top of open internet
standards. The internet started as a very open thing, but these platforms are now controlled by giant corporations, basically. If you use Google,
if you use Amazon, AWS, if you use any of these giant platforms to get your voice out there to
reach people, you are going to be mediated by a giant organization. If you are in particular
trying to perform computations, you are trying to use computer resources, you know,
run an LLM or even have a website,
you are going to be probably using someone's cloud computing platform
in some way, right, whether it's AWS or Google Cloud.
Probably you are not having a web page run from a computer
in your basement.
It's just too hard to maintain that in an available way.
So people use cloud
computing platforms and cloud computing platforms can shut you off, right? So if you are trying to
fight Amazon, you probably shouldn't use AWS as your home base to do that because Amazon doesn't
have to let you, right? They can shut you down. And so this is like the state of cloud computing, this tremendously powerful ability to create,
to perform any computation you want in the world, launch any website, but this really
strict control over that platform admins have over you.
And so Ethereum comes along and says, we're the world computer.
We're like AWS, except no one can shut us off, except no one controls us.
No one can stop what's going on on Ethereum in the same way that Amazon can stop what's on AWS.
Now, the problem is we're a zillion times more expensive.
Ethereum is a zillion times more expensive than AWS, which is what roll-ups are trying to help us solve.
But the fundamental idea is this really, really powerful idea.
And so Ethereum is also not trying
to be necessarily better than AWS
in the broadest sense
because AWS has some nice things about it.
Like if you make a mistake in AWS,
Amazon can come help you if they want to, right?
There are admins in control
and those admins can maybe help users,
but they can also help users, but
they can also harm users. And with Ethereum, we have this philosophy that like, hey,
maybe not exactly code is law, but just like, no one's going to come save you and no one is
going to come harm you, basically. It is a neutral platform. And then in the world of roll-ups,
you see more centralizing factors, more centralizing influences,
more admin power.
And to us, what stage two is really about is trying to achieve a version
of this world computer notion
in the world of roll-ups.
In other words,
can there be a roll-up
that doesn't have the power
to shut your app down, right?
This is like kind of the most basic question.
Like if you are creating an app on the Ethereum L1, you have a lot of things to worry about. You don't have to worry
about the Ethereum L1 choosing to shut you down or choosing to shut itself down. But not always
the case on roll-ups. So it would be nice if you could build an app on a roll-up and you wouldn't
have to worry about the role of shutting it down. And that, to us, is really what stage two is about. It's about being a neutral platform where apps and users can coordinate and do their thing without the protocol being able to intervene.
So, okay, so that's like big picture.
So what does stage two actually mean mechanically?
Okay, so how do you actually achieve this on L2B?
So there are five categories here, okay? So there's these categories. Okay, so there do you actually achieve this on L2B? So there are five categories here, okay?
So there's these categories.
Okay, so there are five categories.
So there's sequencer failure, okay?
So on roll-ups, there is often this entity called a centralized sequencer,
or maybe that's kind of more the pejorative term.
People don't like centralized sequencers.
I'm not, you know, we're not totally
anti-anything and certainly not totally anti-centralized sequencing, but there needs to be a way to get
your transaction included if you are on a roll-up, right? This is necessary for stage two.
And on facet, this works because every single transaction is guaranteed included because every
single transaction is not...
Whenever you create a FACET transaction,
you are actually posting that transaction one by one to the Ethereum L1.
So for example, today, if you make a FACET transaction,
you're going to be sending a transaction to L1,
and that is going to be all call data, right?
There's no smart contract interaction on the L1 that happens,
but that is going to cost between $0.10 and $0.50 to do that, which is more expensive than on roll-ups that have centralized sequencers.
But the benefit is that you can just guarantee that if your transaction gets in the L1 block, you are going to make it into FACET.
So that's how FACET checks the box on sequencer failure.
Everyone is forced to self-sequence all of their transactions,
and we're going to improve on this in the future,
but that's how it works now.
Then there is this question of data availability.
This is the most basic thing that every roll-up needs to have,
which is the idea that all of the data required for proofs,
all the data required for reconstructing the roll-up state
is available on the L1,
and this happens whenever you create a facet transaction,
you're posting that to the L1.
It really is very, very similar to the way forced inclusion works
on Optimism and other role platforms.
The big difference between us and forced inclusion
is that when it comes to creating facet transactions, there is no
smart contract involved in any way. So if you are going to, and again, we model all of this
basically on optimism. So like shout out optimism for pioneering all this stuff. But if you go and
try to make a forced inclusion transaction optimism, you go through a smart contract and
that smart contract is upgradable. And if that smart contract gets upgraded in a way that's
adverse to you, they could censor you and shut off your ability to do forced-included transactions.
And this has actually happened in the past, blasted it. So that's the big difference is
you use an EOA address. And there's some downsides to that too, which we can get into. But that's how
transaction inclusion, sequencing, and data availability work on FACET.
Then there's this question of proofs.
Okay, so how do proofs work on FACET?
And we have kind of a weird setup with bridging that I'll get into in a moment.
But the proof system on FACET is like a pretty straightforward construct.
So if you know anything about fault proofs, right, they are a, you know, battle to, you know, fault proofs have gotten a lot of use and they work great in a lot of ways, but they are really tricky to use because interactive fault proofs, at least, depend on being able to
basically play these multi-step games where the proposer and the challenger can, through multiple
steps, isolate the exact instruction in the, say, transition function about which they disagree,
and then they can get the smart contract to adjudicate and execute it. So anyway, the bottom line is traditional fault proofs require a lot of capital
because you have to be willing to go 70 rounds deep in a game,
and each round has more and more bonds.
So we implemented, and this was based on what Succinct did with OP Succinct Lite,
is it's a ZK fault-proof arrangement where the root proposer can propose optimistically,
and anyone can challenge posting a bond, but then each individual challenge game is resolved with a single ZK proof.
So it doesn't require multiple rounds, doesn't require super large bonds.
And then we also have the feature in here where anyone can proactively propose a route along with a validity proof,
and that will not require any bond and will advance the anchor route and will invalidate any bad proposals that are targeting the same route.
and will invalidate any bad proposals that are targeting the same route.
So anyway, the point is, this is something that's made really only possible
by the work OP succinct has done,
or by the work succinct has done,
but it's actually like a really simple way of looking at this,
and it's a great relief if you are thinking about building fault proofs.
We encourage you to take a look at our repo,
which is called ZK Fault Proofs under the 0xFacet organization.
Another thing that we were happy to do that was a lot of fun was to put all of this into one contract.
Because Optimism, because it has to support a zillion complicated things across everything, Interop, everything.
They had a lot of contracts in their proof setup.
And so we had some fun getting it down to one contract. And then finally,
there's this question of the exit window. Okay, so this is, and you know, the L2Beat folks have
talked about this a lot. Shout out Tim Clancy in the audience here working on this problem. You
know, the exit window is like the hardest one,
basically, for people to get. Like if you look at all the stage one roll-ups, this is the one that's still red. And unfortunately, it's also the most important one, basically, because
if you have the roll-up depend on smart contracts on the L1, the rules depend on the L1. So for
example, in Optimism, there's the deposit contract where you can force include transactions. And so, you know, whether you can
do that depends on the state of that contract. These contracts can be changed without any
notice. And then that means protocol rules can be changed a lot. And so you can be,
you know, you can really be shut off here. And so this is like, I think, a really good time
to bring up Vitalik's recent
tweet, because I thought it was like a very interesting tweet in the context of all this.
So I should put it up there in the nest. But basically Vitalik tweeted in response to L2B
saying, hey, these L2B tweeted, good news, all these folks are still stage one. And Vitalik said,
great. But I think the next step should be fast withdrawals
or faster withdrawals, one hour withdrawals,
presumably from the canonical bridge,
not going to stage two, right?
So this is something Vitalik tweeted.
And this is something that many people
probably engaged with and thought about.
And when it comes to the exit window,
I think you can really kind of search a question, like, really, Vitalik, what are you actually
saying here? Because, you know, if you're saying, hey, we want people to be able to have,
be able to withdraw from a roll-up in an hour or less, well, then you, by definition, need an exit
window that's at least an hour long, right? Because if the roll-up can be shut down in 12 seconds,
then, of course, no one can withdraw within an hour. You can't withdraw if the roll-up can be shut down in 12 seconds, then of course no one
can withdraw within an hour. You can't withdraw when the roll-up is not working, when force
inclusion is disabled, when the gas token bridge is compromised. So without these, this is why we
refer to stage two as the foundation, without these protections, it doesn't matter how good
the withdrawal user experience is. So exit window
is super important and also gets to a little bit of the fun weirdness involved in fitting facet
into these categories because we are listed on L2Beat as having an exit window of infinity
because we don't have any upgradable contracts on the L1 involved in
bridging or proving or the protocol in general. And so, you know, if you read this, you might
wonder, okay, well, has FACET invented a new way of bridging that is a better way and has Fasett just completely changed?
Have we basically changed the way bridging works, basically? Have we been able to accomplish
something in the world of bridging that goes far beyond what other roll-ups have been capable of doing. And the answer to that, which is kind of a subtle answer as we'll get into,
is no. Facet is not trying to be the roll-up with the built-in bridge that is the best bridge,
right? Because all of these roll-ups in the world right now have built-in bridges. They have bridges
that are part of the protocol. Facet is trying to be a different kind of roll-up,
a protocol that has no built-in bridge,
that has no bridge that is what's called canonical,
that has no canonical bridge.
And this is a really weird,
but we think actually pretty important design decision we made
that affects kind of how the exit window works and other stuff.
So taking a step back. Okay, roll-ups and bridging. We all know bridging is important in roll-ups,
but what might be interesting to know is that there's no real one-size-fits-all way to do
bridging in roll-ups. If you look at the actual way bridging happens
in roll-ups, what you will see is there's a ton of bridged value that comes in the form of stable
coin issuance. And sometimes that can be referred to as natively issued tokens on an L2. You could
also think of that depending on the protocol, like USDC has a cross-chain protocol for this. You could think
of that as bridged value, but whatever happens with stablecoins, it is not going through
the official, typically not going through the official channels of the rollup. There's also
layer zero. There are many different ways in which bridging happens on rollups, and they all have
different security properties. Now, when L2B evaluates a rollup, they typically or they always look at something
called the canonical bridge, which as I mentioned is like the official bridge of the rollup.
Now, what makes a bridge official, and this is where FASTAets Weird. What makes a bridge official is that it gets special...
Basically, in typical roll-ups, the gas token that you use,
the token that you use to pay for gas, to pay for transactions,
is a bridge token called Bridge ETH.
Okay, so it's just like DL1 in that regard.
You pay gas with ETH on both, but on the roll-up,
the thing you pay gas with is a bridge token.
And because users need to be able to buy gas, there needs to be an official but on the roll-up, the thing you pay gas with is a bridge token. And because users need
to be able to buy gas, there needs to be an official bridge in the roll-up that can issue
them the gas token when they, hopefully, make deposits on the L1. So this is the canonical
bridge. It's basically synonymous with every roll-up facet with the gas token bridge.
And so this is the major departure because when it comes to roll-up security,
we think there's a really big problem with this kind of arrangement, with a bridged gas token.
And the reason is this.
If you have a bridge, okay, and there's someone who is the admin of that bridge on the L1,
then they can issue as much of the
L2 asset as they want. So if you
go to an OP stack roll-up today, or any roll-up,
whatever, anyone with an upgradeable
L1 bridge
or portal,
with enough signatures, maybe 15,
probably less than 20 signatures,
the controllers of this bridge,
the admins of the bridge, can issue unlimited ETH on the L2.
And if they can do that,
then they can make your ETH worth zero
because they can just issue themselves enough ETH to dilute you to zero.
And then they can basically lock up the entire gas market on the rollout.
And so this is a big problem.
Now, this can be solved in two ways.
So one way is you can make these gas token bridges,
you can make them un-upgradable,
you can make it so they have no admins.
But this is not really what users expect from a bridge.
Because if you have a bridge that's not upgradeable,
then that means you are locked into a specific withdrawal path.
Withdrawals against a specific version of the rollup.
And so when there's a fork of the rollup, users typically expect, my assets are going to go with that fork.
I'm going to be able to play by these new rules and still withdraw.
And so that's why you see bridges that are upgradable, because people need to do that.
upgradable because people need to do that. But when you have that on a gas token, then you have
this vulnerability for looking towards the future for all future upgrades. Now,
exit windows can mitigate this, but even with an exit window, it's still really, really bad.
You know, really, really, really bad to have your chain attacked in this way. And so for us, we think the solution is to do something like what
the L1 did. So you should be able to get a gas token experience that's like the L1, but it's
not going to be literally like the L1 because the L1 doesn't use a bridge token, right? You don't
bridge in to use the L1. I mean, I guess you do if you buy Ether from like Coinbase or whatever,
but theoretically, at least in the proof-of-work
days, you don't have to do anything. You can just create, you can mine this native token.
And that is how we consider Ether to be safe, is that no one can control the issuance. And so
likewise, on the L2, there should be a way of getting the gas token natively without anyone
being able to control the issuance of that token to their, you know, shape it to their needs.
And so that's why FACET uses a native token called FACET Compute Token.
That is a token that is mined by burning gas on the L1 when you create FACET transactions
or post them to Ethereum L1 for data availability. So this is, if you're
familiar with OP stack, this is basically heavily inspired by how the OP stack handles
force inclusion transactions, which to their great credit don't require bridging to do. But basically
there is a mechanism, a formula that is designed to issue this token in a smooth fashion over the course of several years so that no one can exploit it or get all of it.
And then once you have this token, you can buy gas and create transactions. And so because of this, we don't really have a canonical bridge. Because this is the only reason you need a canonical bridge, is to get the one token that
can't be bridged on the application level. Like you can't, you know, have a random contract that
increases people's native balance on a rollup. That can only be done in one,
you know, by one bridge.
you know, by one bridge. So we don't have a canonical bridge.
So we don't have a canonical bridge.
You know, that's what we would say.
You could also say pick any bridge that you want,
and that could be the canonical bridge.
The point is that there is no bridge
that has any power over the protocol.
So L2B chose a bridge that you can all check out and use.
It's an unupgradable bridge.
You can find it at bluebird-bridge.facet.org.
This is a bridge that's kind of at once a demo,
proof-of-concept demo of the proof system.
It's also something you can actually use.
It's not upgradable, so be careful,
because if there's a new fork of facet that you want to use,
you have to be careful about withdrawing from this first one,
or if you don't withdraw, you have to be careful, basically, because it's not an upgradable bridge. But it
demonstrates to you how you could build a bridge on this if you wanted, and the bridge proof of
concept we put out there has some additional affordances for training wheels. So if you want,
and we didn't turn this on on ours, but if you want to be able to like blacklist roots and have like a time delay to make that decision and so forth, this isn't kind of a core part of the proof system, but the example we had has that.
And so, you know, the final thing that's worth saying about canonical bridges is that, you know, we are not just not trying to change the way people think about rollups.
We're trying to reflect that.
So if you look at right now what's going on on L2Beat.com, shout out L2Beat, which is the amazing resource for this.
I don't know what we would do if they didn't exist.
You can see that about 70% of rollup assets today are not canonically bridged, right? They're either
native assets or they're externally bridged, meaning third-party bridged. So, you know,
this is because it's just like on the L1, users fundamentally want choice, right? They need to be
educated about what they want to do, but they fundamentally want choice. They want choice to say, okay, well, I want to do this faster option,
even if it might be riskier for me.
Or I want to do this option that lets me go directly from roll-up A to roll-up B
without going to the L1.
And maybe I am willing to take some sacrifices to do that.
Just like on the L1, where no one in the ecosystem is policing every single app, just like on
the L1, people want to choose the relationship that they want with their bridging provider.
And so we feel like we're kind of reflecting this decision. Now, that doesn't mean people
should have bridges that aren't secure.
It just means that there can be different use cases
and that people creating bridges
should have as much ability as possible to win users.
They shouldn't be competing with the roll-up itself.
And so ultimately what you see with the canonical bridge,
if this trend continues,
if we see on other roll-ups, okay, 70%, 8%, 90%, what you see is you have the users prefer to do their asset transfers with these external bridges.
So they have that preference.
They're going to do that.
But they also have this reliance on the canonical bridge.
on the canonical bridge.
So I think the canonical bridge
can also be this big drawback,
which is that users are exposed to the risks
without getting any of the benefits.
So a lot has been said about like,
oh, like, you know, Sablecoin issuer,
like USDC can, you know, Circle can blacklist you
and whatever, like, you know,
people should be warned about centralized assets and should choose them if they still want them, knowing all the risks, Circle still doesn't control that relationship with the user
because the protocol can use
the canonical bridge upgradability
to interfere, basically.
So that's kind of like this tension
that we're trying to work out
where users don't, in large part,
prefer to use canonical bridges,
but canonical bridges still create risk
for all of them.
And so that's why we're trying to try moving away for it.
So, okay, but yeah, so, you know, this, just to be very clear, though, this does not mean that facet is like the best, the safest, the, you know, we're not like the best safest role up in the world necessarily, right? Like, what it means is that if something bad happens on Facet, it will be because of an app doing
something not good, right? And ideally, the user should be, you know, complicit in that to some
degree. So the user researches USCC, they use USCC, they're blacklisted,
bad outcome for the user. But it's because of their interaction with Circle. It's not because
the protocol came around and said, hey, we're also going to stick our nose here and make you
worse off. As I wrote in this post, other roll-ups include bridging protocols and asset escrow protocols. They have that functionality.
We don't.
We are a computation platform, and we view ourselves as trying to enable freedom, basically, not personally securing people's assets.
We think that that should be done by apps.
We want to enable that and give everyone the tools necessary, which is why we built this whole proof system.
And so, you know, last thing I was saying, I really want to say questions.
Obviously, I'm passive or comments like challenge what I'm saying here.
But, you know, back to this platform question, like, you know, the temptation, I believe,
to have what I called in my little thing here, a thick platform.
Like if you were a thick platform, if you provide a lot of functionality to the user,
you know, famously Apple like does this kind of stuff, right?
That is tempting to be really good, right?
Because if you are going to be a roll-up
and you're going to provide the bridge,
you can make sure the bridge is as good as possible
when you provide this functionality.
We believe a thin platform approach for roll-ups
is better though, is that, or we're trying it at least, is that if you refrain from providing this functionality and focus on being neutral and stable, you can avoid the risks of, you know, overstepping and interfering with what users want to do.
And we think this is how, you know, Ethereum basically works. So, yeah, that's it.
You know, I think looking forward, you know, FACET's got a ton of problems, right? Biggest
problem on FACET is right now there's no transaction batching, so it's expensive to
use FACET, okay? Like, you know, if you think it's expensive, we know, so we want to fix that.
We got a bunch of other things in mind, but thanks
to everyone for listening. Love to take any questions
or hear any commentary
from anyone, challenge what we're saying.
We love this stuff.
Thank you all. Did I cover everything, Michael?
Did I need to motor math more?
No, I think
you did a good job.
Yeah, I think
like you were saying,
one of the big pain points right now
is just the creating of transactions.
I think that's like the UX that we need to work on next
to make Facet just better for developers and users
because even though it's a great first step
and it provides a completely trustless experience for creating transactions.
It's not really the easiest for developers to integrate with Facet,
because most chains, you just have to essentially add an RPC endpoint and a new chain ID, and then you're basically good.
But with Facet, you have to, and we made an SDK that makes it easier,
but you have to handle creating transactions on the L1 to get the transactions on the L2.
to get the transactions on the L2.
And I know that probably sounds confusing,
but basically you're posting your L2 transaction
inside of an L1 call data.
And most L2s will handle putting your transaction
on the L1 for you inside of Blobs.
But we were wanting to take out the middleman,
and we have some ways of improving on that
and providing some sort of open protocol slash service
for handling this in the future.
But that's definitely something that Facet could be better at,
and we plan on improving with time.
But it's still actually, even though it is not as cheap as using base, it's still pretty dang cheap.
And you will save like 90 plus percent, depending on what you're trying to do, on facet versus when you compare it to the L1. So if you're looking at creating somewhere
that lasts forever, can't shut down,
you're basically stuck between using Ethereum L1 or Facet,
then obviously Facet wins there as being cheaper.
It's also all on call data, which is better than blobs
because six round forever.
No kidding, I'm not like that crazy.
We want to use blobs.
Blobs are good.
Okay.
We're not, we're not insane.
Blob truthers.
Tim.
Yeah.
Hello.
I can't stay too long, but I do know that we will be having a chat later this week.
And so hopefully I'm looking forward to getting more into things with you guys then.
I just wanted to jump up and say, first of all, congratulations for Bluebird out there and getting that ranked by L2B.
You know, we see Bluebird's got that green pizza.
And obviously that's, you know, great work to you guys.
Great work to succinct.
We're both, you know, we're both building off a lot of that same stuff.
I had more of a question regarding how best we may accurately communicate the state of
facet to people. And this might get a little spicy, but do you generally agree that it is fair to say
facet bluebird is immutable and unstoppable, but it is not correct to generally characterize
facet itself as perpetually tethered to the state of Bluebird.
Yeah, that's a good question.
So I now see, I thought you were kidding, but now I see that L2Beat actually did change the name to FACET Bluebird,
which is humorous and also, I think, accurate in many ways, although I would have chosen a different name.
So taking a step back for those listening, so when you are doing a blockchain, you have forks.
So the Ethereum L1 has forks and OPStack has had many forks. Those
are the two chains I'm most familiar with. And typically what you do is you pick some kind of
name that has a theme and then you pick the alphabet to that. So Ethereum is always like
cities and stuff. And then Optimism has the earth theme, bedrock and so forth. So anyway,
we picked birds and our first one was A
and then here we are B, it's a facet bluebird.
And so this is the page now is listed as facet bluebird,
which is funny and I'm down.
But right, so the question is,
how do forks work on blockchains?
And I think it's a really tricky one.
So first I'll kick it back to you, Tim,
and just say, okay, how many, you know, there's Ethereum, right? What is Ethereum? Ethereum refers to what, right? Like there's the, obviously there's Ethereum Classic, which is a different blockchain. There's also Ethereum L1, you know, pre-Cancun, right? That was the Blobs one. There's no Blobs Ethereum. Is that a separate blockchain?
We don't really talk about that.
It's not really called ETH Classic.
We've had a discussion about this one before.
And my point, then, yes, this is where social consensus comes into selecting the fork.
I believe this is only a valid thing to do for an L1 because the entire L2 space, we
are able to outsource that degree of trustlessness directly to the Ethereum L1.
Therefore, I don't necessarily find it suitable for any L2 to rely on social consensus as we are able to outsource that degree of trustlessness directly to the Ethereum L1.
Therefore, I don't necessarily find it suitable for any L2 to rely on social consensus as a mechanism for doing its fork selection, right? Like I'm able to immutably point a smart contract
to know what the state of any of the op chains might be, whether they're upgradable or not,
and to track that upgrade going forward. I'm more curious what your response should be
for L1 attempting to do that with Facet.
Earlier in your chat here,
you gave a little disclaimer about,
essentially, you can build things immutably using Bluebird,
but Facet's long-term upgrade plan,
if you did want to add more functionality,
I don't know if you're necessarily committing
to keeping Facet as a whole immutable forever uh maybe speaking to some of that yeah
no definitely so definitely we are not committing to that because it's uh you know got all kinds of
problems now like at least we got to get some cheaper version of transactions and you know
integrate with blobs or whatever so yes definitely some stuff plan not intending to do the thick
platform thing of like building tons of functionality, but there's basic stuff we have
to do. And so yeah, what happens? And so yeah, so if you look at the way the proof system works
on Facet, it's not the proof system. It's just a way of doing proofs. The setup that we have for
it, and anyone could deploy their own, but the setup we have for it is immutable. And so if you
are posting state routes and defending them and, you know, doing the whole
validity or optimistic fault-proof game thing, then, you know, there's no way to change
how that works. And the way that works specifically is that there are some ZK parameters,
right? We've talked about this, where there's various commitments to the range and aggregate,
you know, various commitments to the way the ZK side works,
and you have to store these, and they're stored immutably in the contract,
and so that's it.
If you have different ZK parameters, you can prove, you know,
you will accept different proofs as being valid,
and so, you know, if there's a contract that has these ZK parameters
and it can't be changed, then that contract is stuck on that fork.
And because that fork is a valid blockchain and a valid roll-up, then yeah, I think it's totally natural to speak of that fork as being its own thing.
is a social consensus element to this,
where I think more commonly people will probably,
like with the L1,
think of facet as being a name
that refers to their preferred fork.
It might not be the fork that we,
as the facet random guys,
basically like we're just people, right?
We have no power,
especially like we can say,
hey, follow our fork, but people might not want to do that, they might prefer a different
fork. And so I think facet like the L1 will just refer to the fork that I like right now. And then
the goal, I think, in many cases will be, if people are using bridged assets, they will want
their bridges to follow the fork, meaning they will want
upgradable bridges that have suitable exit window arrangements. And I think that will be a very
common thing that people want. Some people might not want that. But if you look at the L1,
and you look at other roll-ups, especially the L1, like the L1, if a new fork is coming out,
and Circle doesn't upgrade their software to support this new fork,
that's going to be a major, major, major problem for users in terms of what they expect Circle to do.
Circle and a bunch of other people had to come out and say,
hey, proof of work Ethereum, we are not going to honor any redemptions of anything.
So there's that expectation.
That makes sense, right?
And then obviously Bluebird will always maintain its teeth um i guess i have two small follow-up questions with that
right so i guess the first is do you guys have any idea how long you intend um you know like
for lack of a word like facet prime like the name of just plain facet with no bird after it to uh
be equal to the state of facet bluebird or when you might be cooking out
like a facet cardinal or whatever your seabird is? Right. It's a good question. So just to
maybe take it. So facet, so taking a step back, like facet has been around for a while
in terms of our working on it. So we launched this thing that we're talking about now.
Albatross is what we call
the first fork.
Launched that in December.
Before that,
there was another version of FACET
that was non-EVM
whose state got put
into the genesis state of Albatross.
So we've had some stuff
and basically,
this past Bluebird fork
was to fix some really broken things,
specifically pertaining to exploits in the,
not exploits, but like not good functionality
in the way the FCT mid system worked.
And that was based on, you know,
a bunch of experience with how that worked
and wanting to fix it.
So that was a very specific thing
to fix a really specific problem.
We have this other question of the blobs,
but like, yeah, I don't, I think, you know,
my hope would be to, you know, certainly before Q4 of this year to have blob support out. And then that will be, I hope, probably the, the quote unquote, the end. I mean, you know, if you keep changing things, you know, that's not so good be the thin protocol has to also come with an idea that we're not going to be trying to change it a ton.
But then the underlying L1 changes.
So, you know, all bets are off, you know, there.
Because you can't use the old version of the L1.
I guess when you guys do upgrade, are you carrying, like, you want to talk about what old
state, what's going to happen to that, right? Is facet, you know, seabird going to be like completely
empty when you start it, or are you going to roll over all the existing bluebird state or?
Yes, good question. So, so the idea is keep it, it's got to be what people are used to. It can't be so crazy that it's impossible to understand. So no, we're not going to can ever achieve a positive outcome that starts with,
hey, everyone, evacuate everything. Like, that just seems like, you know, so, you know, violent
in its own way, even if people weren't forced to do it. So, so yeah, that's the idea. And then,
yeah, the question would be, right, like, what do all these assets mean if various L1 representations
are tied to other forks.
You know, and I think this is, again, something that works magically on the L1 because USDC
just always means the same thing no matter what the fork is because, you know, behind
the scenes, you know, good stuff happens.
But yeah, that's going to be a challenge we face.
Now, in terms of our withdrawal setup, we basically copy-paste,
copy-build on the optimism setup where you have the L2 to L1 message passer, and to withdraw,
you're basically just appending something to an array in the storage of that L2 contract and
improving that. And so if a user just has an asset and they want to get it out, they probably can do so on the next fork,
unless we change the rules about how that pre-deploy works and everything,
which would be very extreme.
The challenge would be if they had it locked in some protocol
and that protocol was doing something with a new rule.
But yes, this is a social consensus question where...
Yeah, that's more my problem, right?
Like what happens when a facet bluebird goes bananas in the next week
and suddenly you have 100 million Ether directly out of Tom Lee's pocket
has been wrapped and bridged on the bluebird.
Now suddenly I don't know if you would even be able to create
your your sea fort because now none of those assets would be you know what i mean like you'd
lose you'd have zombie state where it would look as if you have all these balances but in reality
like whatever bridge sent those on might be immutably tethered to the immutable bluebird
and then you would never be able to like retroactively pass
proofs back down a layer well you could use you know you could use the the i think if unless we
are like if we add blobs for example as long as that is like backwards compatible in some way
um even if someone didn't know about it like i think you could either use an old version of the
software to prove it uh on the
bluebird fork or it would probably still be valid uh like it would probably getting getting the
the message into the storage into the storage of the l2 l1 message pass or probably is still
possible right even so i think the balances will be there the question of are they redeemable yes
that will be a.
Right. That's what I mean. If you end up a situation where something's only redeemable on Bluebird and then you have a fork and now it's you can't necessarily.
You know what I mean? Like would I guess I guess that's just sort of like that could be the answer right there. Right.
willing to eventually have one of your immutable forks get so popular that you are then effectively
restricted from doing upgrades or is there going to be like the stomach among like the facet team
to go ahead and like do an upgrade anyways and abandon value on an old fork i mean i think a lot
of value if if not just looking at us or whatever anyone anyone, like, in any of these roles, a lot of value is going to be in upgradable container, you know, like, whether it's third, if you just look at all of the value that's on L2s right now, even beyond the question of, like, exit window or whatever, like, a lot of this value is just coming from, like, upgradable apps, and that is bad, And some would say that that's bad.
And it said bad qualities to it.
But it also has advantages where you can have that greater degree of control.
You can follow the latest fork and so forth.
You don't have to depend on...
But yeah, look, if you run another protocol in the roll-up
and you can just 100% trust this specific contract address to always have the
root of the correct fork and you can 100% trust that, then yes, you are going to have an easier
job building upgradable stores of value because you don't have to upgrade. You just point to that.
But the downside, of course, is, well, what if you don't want to trust them as being the
arbiter? And you can still do this in another up, right? Like you can still run an old base node, right? No one would do that. I guess that could happen to you guys if Bluebird got completely full of ether that is not otherwise able to be effortlessly bridged on a later fork.
But I mean, it's interesting.
It's a novel design space for sure.
And I'm glad that you got your bridge construct rated by L2B.
Now it'll be funny if eventually there's like 12 facet entries on L2Beat.
I would imagine that
before that happens, there would be some sort
of UI change
is my guess.
I'm ready though.
I mean, you know, hey, for me,
it was an honor to just get back on L2Beat
after everything that's happened with the crew system.
So now the idea that I could be
a part of unlimited roll-ups on LTV,
it's like, whoa, whoa, whoa, whoa.
Are you kidding? Is this real life?
Half the roll-ups will be faster roll-ups.
Yeah, shout out to...
I think that's kind of where your design goes.
If you're still intending to upgrade
but leave immutable stubs behind with every version,
I guess why not, right?
Anyways, thank you guys for letting me up here and uh congrats again oh yeah appreciate it talk soon
talk soon shy yeah hi uh so i was just going through the white paper and i just wanted to
understand as you are not giving preference to sequencer, how are you going to manage it in the terms of on the time of crisis when there is a crisis?
How are you going to manage it?
So right. So yeah, good question. So how does the sequencing work?
So can talk about like a little bit today and then the future. So today, you send an Ethereum transaction with a special payload in its call data, which
is an RLP-encoded FACET transaction.
You send it to the FACET inbox, which is an EOA address that's 0x and then all zeros
in phase 7.
On the end, the FACET node will observe all of this that has happened, and it will take a given
Ethereum block, it'll pick out the facet transactions, and it'll assemble a facet block
using those transactions in order, and some other stuff, and then that goes into Geth,
and Geth executes it. So the way it works today is, yeah, just the transactions that go into Ethereum
uniquely determine the facet block
because the ordering is set by Ethereum.
So the goal in the case of blobs,
or putting aside blobs,
just sequence transactions in general,
would be to, and forget blobs,
simpler to think about it with call data, right?
So if you want to be the sequencer
when we upgrade facet to enable this, now I can send you my transaction, so can 20 other people, and you
can take those and put them in call data and send them to the facet inbox address. And those bundles,
batches, whatever, will be extracted and put in order to create a facet block. and if you were to do it within a blob, it would be the same way. So the tricky thing, the gotcha, is that facet blocks are a fixed size, right?
They're 200 million gas limit now.
Maybe that'll go up.
They're a fixed size, and yet there is no way to reject a transaction on the L1 synchronously.
So with Optimism, a benefit of their deposit force inclusion mechanism
is that if you make a transaction to get something force included,
and that's not possible, for example, if you hit some kind of limit or whatever,
that will fail, that will revert, and so synchronously you will know
my thing didn't work and I can try again or whatever.
And so with facet, that's not the case, right?
So this hasn't really been a problem so far in terms of our usage hitting the cap,
but if the block is full, your transaction might succeed on the L1 in getting to the inbox,
but might not make it into the block or won't make it into the block if it's full.
not make it into the block or won't make it into the block if it's full. And this is particularly
gnarly in the case of contract-initiated facet transactions like bridges, for example, because
if you are a user and your thing doesn't get in, you can probably try again, but contracts are
doing it in a transaction where they receive value. So a bridge contract that doesn't have any training wheels
or admin ability to withdraw
needs to implement a retry logic.
And we actually have that in the bridge that I mentioned,
that proof of concept bridge.
So you can look at that and see how you might implement that.
So that's kind of like a gotcha there.
But besides that, I think the system will work pretty well.
We also have to make some changes to the way Geth works because typically in Geth, you can't have a block whose gas limit, where the sum of the transaction gas limits is higher than the block gas limit.
So that's a problem.
But for us, we change it so that you can do that.
And every transaction will just try to buy gas. And if it actually fails, then it'll fail. So you will get a notification or not a notification. You will be able to see that your transaction made it in but failed. But you'll still have to try it again. Is that helpful?
helpful yeah yeah that's good uh i am pretty actually i am trying to write a detailed one on
peset so i just needed this info and can you ask one more question sure
okay so i just i'm just going to the paper and i just got to know that all the uh admin keys
multi-six and pause functions are removed so So in the event of critical consensus or vulnerability,
what is the recovery model which you are going to prefer?
So in the end, okay, so there are sort of two ways of viewing this
and this is why we're not like totally aligned with L2B.
So one way to view it is, hey, if I'm using a proof
system here, if I'm using the facet proof system, which is embodied in
these literal smart contracts that have literal parameters for the ZK stuff,
if I'm using that, that's the truth, and whatever that says is the truth, and that's the thing I can prove against,
and I have bonds and so forth going into this thing, and I have bridges. So that's the truth.
That's one way to look at it.
And so even if there's a disaster
on the Facet off-chain software side,
if Facet node has some major bug or something,
like as long as the proofs are possible to work,
then that's the answer.
Another way to look at it is say,
hey, how many people are actually using these proofs, right?
Like if you look at, again, I always pick on USDC because it's the biggest one.
If USDC starts doing stuff with Facet, they are not going to be worried about a proof system,
just like they're not worried about a proof system on base. They're going to run their own
node and do their thing. I mean, they may be worried a little bit about it.
So if you look at that, then the answer in this case of a catastrophic consensus thing would be
case of a catastrophic consensus thing would be, you know, the person, the entity that controls
the largest pot of redeemable value gets to say what the answer is, right? And that's, you know,
going to be in the end state more, I mean, that's kind of a reductive way to look at it. But if it
becomes a social consensus question outside of the immutable smart contract proof system, then the people who have the sway are going to be able to say what happens.
And this is really true of the L1 itself.
If there's a complete disaster on the L1, what's going to happen?
I don't know, but I guarantee it's not going to be something that doesn't include Circle or whatever.
That's how you can absolutely guarantee we will never be on an L1 fork that Circle doesn't want, even if they don't get to dictate it.
So I think that's another way of looking at the system.
But from the standpoint of a user, you don't actually have any, you have to choose or you have to do what you're going to do.
But there's no, you can always get your transactions included.
And so, you know, there's no, no one's going to force you to choose those.
But you might feel the pressure from, you know, wanting to no, no one's going to force you to choose those, but you might feel the pressure from, you know,
wanting to have your USDC worth something.
Great.
Gotcha.
Thank you so much.
I'm going to write up, like, I'm going to do a detailed write-up about it.
I might share it in DMs as well.
Great. Yeah, DM us.
Would love it.
Sure.
Thank you so much. I'd love it. Sure, sure.
Thank you so much.
I really appreciate it.
Cool, anyone else out there?
Got Kelvin Fitcher,
not trying to shout anyone out,
but just, you know,
big fan of that guy. So, you know,
you guys can throw that guy a follow,
major influential force in FASCET.
Who else is Jesus.E? Thank you for also major influential force in FASCET. Who else is Jesus.E?
Thank you for also major influential force.
Who else has got questions?
Who else has got, you know, challenges like, you know.
Hey.
Yes.
Hey.
Oh, cool.
Okay.
I thought my mic was off.
Yeah, yeah.
Go for it.
Hey, okay.
So I'm actually pretty new to NFTs and blockchains, and the one thing that gets me is that, okay, so how exactly does the blockchain differ from the regular internet?
Or, like, that's what's confusing me. Is it like separate servers? Like, what's the word? Like, no, decentralized internet or how how does blockchain work
so yeah so i i think a blockchain um it's you know these they're based on protocols meaning
rules that are made up right so uh you know i always think of a blockchain as like driving on
the right side of the street like why is driving on the right side of the street good it's like
well because everyone wants to play the same game with blockchain it's a game that you can play uh that uses normal technologies but it's a
a game that you can play and the way the game is set up is you'll always be able to um know uh the
history of everything that has happened basically and um that's kind of the the magic because you know you can well anyway
oh okay
okay so
deposits the data in each
what was it node
or fork
yeah I think
at the highest level
I wouldn't look at necessarily as like the
the yeah the node
a fork is like a set of rules right and a set of rules I wouldn't look at necessarily as like the, yeah, the node.
A fork is like a set of rules, right?
A set of rules.
And a node is a piece of software that allows you to actually implement these rules to do, you know, to actually create the blockchain, derive the blockchain, because you can't do that with a pen and paper. But on the abstract level, it's not really something that is tied to any specific computer program.
It's more of like a set of rules so that when people say things, you can put them in a format that allows it to be easier to kind of like audit and understand what's going on.
And that is this notion of blocks where basically you can create a block, which is a bunch of things people are saying,
and you can package them up in a way where you can know which was the parent to that,
and then the children of that you'll know.
So it will form this chain so that you will basically be able to have the ordering of
what is happening.
So yeah, blockchain is basically a way of having order, and if you have order, you can do a
lot of other cool uh things uh like
create uh okay yeah because i've been i've been trying to picture it like in this abstract level
like okay block chain but okay it is um okay i so i do know about NFTs and crypto. Would this be a good place to ask about that, at least crypto?
I mean, this is kind of more of like a specific, today's about like a more specific, well, sure, ask one more question, then I want to get to smart contracts who actually went down.
went down no okay yeah go ahead um okay uh is crypto good to invest in right now like just
Oh, no. Okay, yeah, go ahead.
generally speaking no comments uh on that um but yeah i i appreciate it let me let me let we're
gonna get some other uh speakers going i appreciate it i can't comment i am myself scared of everything, personally. So that's, you can take that.
I fully understand.
Thank you for having me on.
Hell yeah.
Thanks, dude.
Peace.
Kelvin, is the bridge the roll-up?
I mean, I'm trying to figure it out, but I think, I don't know.
You know, man, I respect this this design I respect facet a lot this I
it feels a little bit like a vindication in some sense of the you know whatever
the whole roll-ups aren't real thing from a couple years ago but um
yeah i'm i'm uh i'm curious honestly about what you so you obviously changed some things from the
op stack and i know what you changed at a high level um some things I think we can probably upstream
into the LP stack.
And so I'm curious what you,
you know,
why you diverged
or where you diverged
and what you wish you could see upstreamed more
so that more roll-ups can share
some of these design principles.
That's a great question. Yeah, thank you. And I think, you know, it's funny because,
you know, we came at this from, you know, I mentioned like the first version, this wasn't
even EVM. And so we were, so we came at this from the standpoint of like, we are like,
you know, really like, ah, we're the separate people and F optimism. And then we started to
actually like look into how optimism actually works
instead of just assuming how it worked.
And that's how we were like, whoa, this is great.
Let's just change a couple of things.
But yeah, I think for us, one major thing is obviously this non-bridged gas token,
which optimism has and is dope.
So I don't think that anyone is going to want to get rid of their
canonical bridge and stop doing bridge deeds. But I think optimism could do a lot. OP could do a lot
with the deposit notion. So, you know, I think the biggest things, you know, one big question is,
could it be in a non-upgradable contract? And then even failing that, I think the big question would be,
there's all this resource allocation, resource limitation stuff,
where you get some low amount of L2 gas that you can buy with the deposits.
Every L1 block, it's like 20 million or something.
And then there's a really high elasticity on that, so the target's really low and so like if you look at like the way this would actually play out in a true
base sequencer fail thing as it did like a couple of you know days ago or whatever just to pick on
base like it would be a really grim you know uh scenario and since op is so close to having a
really good thing there that would be my uh wish would be to um either make it immutable the fact
that you can force these transactions,
or failing that, at least give the forced transactions some reasonable amount of block space.
Not like 1% or whatever is the case now, but another conspiracy.
I definitely think we can make the deposit path not upgradable.
So I'll send you a message uh there's uh there's
some good designs there hell yeah um so then i guess we can upstream more of that so i'm i'm
you know this is more of a general question i think that the obviously interest like this i
think the thing that tim was was trying to get at and and that that i actually don't
put it this way i i think tim and I have a very different perspective on these things.
But you know, I'm curious how you see, you know, what you see, you imagine the bridges
into facet will look like, you know, do you, do you imagine a world where you're going to invest in,
uh, in some kind of upgradable bridge for yourself or is facet kind of going down this route where most it's mostly relying on basically other people to build the bridges.
There's no wrong answers.
I'm just very curious as to how you're thinking about bridging into Facet
because I think it's very unique in the landscape today.
So just yet, we're up here talking about, hey, L2B protocol, stage two.
So we also do other things.
And one of those things is we are also app developers on Facet, which we're not trying to like push necessarily this space.
But we built a bunch of apps on Facet, obviously.
And one thing we built is a bridge that you can use.
And it is not a proof system based bridge.
It's a bridge that is like probably similar to, well, anyway, it's not a proof system based bridge. It's a bridge that is like probably similar to, well, anyway, it's not a proof
system-based bridge. We originally had it as a proof system-based bridge, but then the delay,
you know, users don't like delays. So that was, that was an issue. So we have that and that bridge
is upgradable and will follow the latest fork. And if someone wants to use that asset, they should
do it. I think that's what people like. They like an asset that will follow it and follow the latest fork. So I think you'll see a lot of that. I think you'll see a lot of
just what goes on, like look at Arbitrum and base and like, where do those values come from?
Like, I think like really the, it's hard to do this. So, you know, maybe we're like, like sick,
this so maybe we're like sick people but the real pitch of Facet is to a person. Again, I feel like
stablecoins get kind of this weird thing because we're supposed to hate them or love them but I
think the pitch of Facet is to the stablecoin person and say, hey, guess what? Here's a chain
that like all the other chains is going to be dependent on you and use you but here's the one
that can't F with you. You know what I mean?
That can't hurt your setup here.
Like, you know, imagine what would happen to Circle if Arbitrum or whatever
decided to shut down tomorrow, game over.
Like, it's not like Circle could just reissue those things
because the USCC is owned by smart contracts on Arbitrum.
So Circle would be in a really bad situation.
So that's kind of like the thing we're more trying to say
is like we're a better way of doing the thing
that is happening now.
But I think, you know, it would be great to have a,
you know, a proof system-based, very fast bridge,
just like Vitalik says.
We're just personally not there yet.
But I don't know, does that make sense?
Yeah, that makes sense.
Well, anyway, I'll let you to it,
but I will definitely send you a message.
I think we're going to have a lot of fun talking about this stuff.
I feel like we're very much on the same wavelength
when it comes to how to think about these things.
Oh, yeah, I love it.
So have fun and enjoy Mainnet mainnet thank you thank you guys so
kelvin mentions this uh video or basically there's this great blog post you all should look at uh
which is the john charbon outpost about what is a role anyway there's this whole huge thing that
happened like two years ago that kelvin referenced but like little does he know that like i've like
watched it like a million times like two weeks ago or Kelvin referenced, but like little does he know that like I've like watched it like a million times, like two weeks ago or something. So anyway,
what else, what else is, uh, who else is, is out there? Sasha, Sasha, uh, Jeremy, shout out to, um,
to y'all. Um, anyone else want to come up? Edward Lee, dude.
Come up, come up.
So yeah, just to reset the room,
what are we talking about here today?
We are talking about FACET
and how FACET became the first ever
general purpose Ethereum roll-up.
What is that?
Excuse me, first general purpose, stage two. The most important thing. Stage two Ethereum Ethereum rollup. What is that? Excuse me.
First general purpose stage two.
The most important thing.
Stage two Ethereum rollup.
And why this is important.
General purpose means you can do anything on FACET.
You can do any Ethereum L1.
How do we become the first general purpose stage two?
What does this mean?
Does this mean FACET is perfect in every way?
It means FACet did not chose to adopt a different approach to making this happen. And anyway, I'll stop restating the room because we got Edward up here who was amazing, played an
enormous role in this because he made the succinct part of Pappen and succinct brand the proofs.
So what's up, Edward?
Hey, Tom, how are you?
Doing great.
Yeah, thanks for the generous invite.
Let me speak.
I was not expecting this, so I'm going to go off the cuff here.
So hey, everybody.
My name is Edward.
I work at Succinct on the BD team.
And we're super excited to see what Tom and the FASA team have done.
In essence, we have a product called OP SYNCT that allows any OP stack rollup
to become a ZK rollup.
And the sister product is called OP SYNCT Lite, which allows, yeah, any OP stack rollup
to adopt ZK fraud proofs.
And this is what Tom and the FASTA team have done.
And yeah, it's really cool because your rollup can,
you know, be put on the path to stage one and stage two
with like very minimal modifications.
In Tom's case and in FASTA's case,
they had to do sort of more custom work, but yeah,
in essence it's like super cool because you can just like modify uh the sort
of base repositories that we have to make it fit your particular roll up and we're super keen to
see uh this is so cool hell yeah i really appreciate it and just to really emphasize
this shout out like we were like the i don't know if anyone out there has done proof system stuff
roll ups optimistic roll-ups,
fault-proofs, it is really
bad. It's really, really, really
hard to do.
With Sysync,
you get it all, really. You get
the battle-tested
verification and the software.
You get that. You get all
this other software written around this
notion of ZK fault- proofs and making that work.
And then you get the prover network where you can actually generate these proofs.
And it's pricey, but hey, it just will effing work.
And that's when you're dealing with high-stakes situations like this.
You want it to turnkey work.
Maybe you want to do your own GPU thing also, but you want something that will work.
And so that's kind of like a real end-to-end
treatment we got from Succinct. And I really highly recommend anyone working on... And by
the way, I got to say, OP Succinct Lite should rename itself to OP Succinct Heavy, because
I really think it's the best, because the problem with these EK proofs is they're so pricey.
Why are you going to spend all this time proving things that no one disputes?
Okay, like, what are we talking about?
Like, a good integration there.
And that's kind of, I think, what we were able to do also is really valuable.
So if you're stressing a proof system, talk to Edward.
And shout out succinct.
Who did I just invite up?
Someone.
Who else is going on? Nam? Did you whoever invited should speak yeah i i wasn't sure this is the time to ask questions or not so i just press that button
yes i'm also happy to like uh i can ask questions different uh on the like text or something but
yeah honestly super cool i came in like just five ago, so maybe a bit late to the party. But I was kind of curious, how did you guys decide,
like, I guess, have you guys considered, like,
as part of your system, right, like you already
have the proofs on L1, have you considered just,
like, making the canonical bridge for some assets
just, like, with that proof system?
And I think you could still qualify as stage two, right?
Or, like, basically like how would you imagine what would be ideal world for like how bridges develop one passive
sure so what we did in fact so l2b um who really has been really great throughout all of this
because this is annoying for them right to do a different weirdo thing and so you know they could
be just annoyed about it right because it's annoying it requires work for them, right, to do a different weirdo thing. And so, you know, they could be just annoyed about it, right,
because it's annoying. It requires work for them.
But they did a lot of work to, like, help figure out how
and this should fit into their system.
And basically their system is really super heavily,
like, you need a canonical bridge.
And so we say, well, you know, we don't have a canonical bridge,
but we want to, you know, our view is that
every bridge should be amazing on this.
So we're going to create a bridge setup.
We're going to build it.
It's going to work.
It's going to be some bad thing.
It'll be good.
We're going to build a bridge setup.
And we're going to say this, you know, we'll designate for the purposes of this review, this bridge setup as our canonical bridge.
And that's what they did.
They reviewed on that basis.
And so there's a bridge right now now and it's linked on their thing.
You can go to facet,
go bluebird-bridge.facet.org
and you can see how this bridge works.
You can read the code
and it's basically a,
it hooks into the immutable proof system
and it's an immutable bridge
that allows you to bridge Ether from the L1
and get bluebird-weave on the L2,
which is named because it is proven
against this current fork. And, you know, this is not a full thing in terms of like, you know,
arbitrary ERC-20, arbitrary NFT. It's just ETH. But it shows you how you could do it. And I think,
you know, it also has some additional affordances where you can, if you want to copy paste this,
you could add like some training wheels to it because it has some ownable, if you don't renounce the owner function, you can blacklist roots and do
this kind of stuff. So people can do that. So it's kind of a bridge building framework plus the one
bridge that we built and it works and you can use it. So that's kind of like our canonical bridge for
the purpose of review. We don't consider it to be the actual canonical bridge because there's nothing official about it. It happens
to be made by the people who made the protocol, but there's nothing
about the protocol that designates it as special. We were talking a little while
ago, how should bridging work? Bridging should work how the users
want it to work. That's really the bottom line.
If a user wants a
certain type of bridge, then hopefully the, you know, world and the market and just everything
can lead to the efficiency of that being created if it's efficient. If it's not, they should use
a different type of bridge. This is what we've seen with other roll-ups. People don't like to
use the canonical bridge. They don't like the cost. They don't like the time delay.
They don't like that it doesn't go roll-up to roll-up.
So a lot of people don't use that, and that's good because that's what they want.
And if people want to use the canonical bridge, they should.
And so likewise with Facet, if people want to use bridges that are super immutable and proof-system-driven,
like the one we made, they should do that.
If they want to use other assets that are more familiar in terms of how like their assets on like base or whatever work, they should do that. The idea is to basically say, we as the
protocol are not going to interfere. You know, it's like a non-interference type policy. Like,
we are not going to do everything for the user. But in exchange for that lower level of functionality
than you might get from other roll up, we are also never, ever, ever going to interfere ever.
So we're not going to provide the user with value
while at the same time injecting our little hands
into every single transaction the user does.
That's kind of the philosophy.
Gotcha. Thanks. Yeah, thanks for the answer.
I'm curious, like, I think when you guys did the Blackbird,
I think, Richie, if you said, did you guys consider just, like I'm curious, like, I think when you guys did the Blackbird, I think, BridgeE, if you said,
did you guys consider just, like, doing the, like,
out-of-the-box, like, ERC-20 bridging functionality?
Because that already exists, right?
And I don't think, like, if you guys did that,
that would have, like, interfered with other people building bridges.
Like, same way I think other people are building bridges
on, like, other roll-ups as well.
So I'm kind of curious how you guys thought about that trade-off.
Yeah, we could have done, and maybe we should do that.
A more sort of, like, optimism-style L1 standard bridge,
and then you have the factory contract on the L2 that mints the L2 ERC-20.
So, you know, we have some different stuff, but pretty similar, straightforward to copy that.
We could do that. Maybe we should do that.
And then, you know, NFTs are their own
thing also. I think, you know, what we
have found, it's funny, for a
version of FAST that really no longer exists
from, like, 2023, we
made something like this and
it didn't get a bunch of uptake
and I don't think it gets that much uptake on
these other things. Like, I think basically the
way canonical bridges work is Ether
is a meme, like, we have to use
bridge Ether,
but everything else,
it's like,
are people really bridging
any ERC-20s
or are people creating
native assets
and, you know,
doing stuff there?
Which, by the way,
is another really
super important thing
to think about here
because, you know,
there's a lot of talk
about bridging,
but bridging is really a small part of what happens on rollups because what people want to do on rollups once they bridge is create native assets, create native state.
So think of all the coining that happens on base.
And base and Jesse, whatever, they catch a lot of flack for this content coin, creator coin, what is this?
But hey, at least they're doing something, right?
They're doing something, and it's a new thing, and it's something you can do on the rollup. And I don't know if I like it or it's going to this? But hey, at least they're doing something, right? They're doing something and it's a new thing
and it's something you can do on the rollup.
And I don't know if I like it or it's going to succeed,
but like, hey, do some cool stuff.
And whenever you do something cool,
you're going to create state on that rollup.
And when you create state on that rollup,
that state can't be withdrawn ever by anyone, ever, ever, ever.
And so, you know, it's really important
to be able to protect that state.
And so that's another big thing that we're trying to say
is like, hey, you know, you might not have to worry about doing all this bridging because your native assets are going to be more secure because they can never be interfered with or shut off.
And I hope base gets better on this, but the sad fact of the base content coins is not what everyone's saying about them, which whatever,
who cares, right? The sad fact is, if base, meaning if the 15 signatures of the control base,
decided to turn base off, which they could do without any notice, then all of these content
coins and any other piece of native asset would be lost. So what you're left with is this really tough user experience question on the Native Asset side of how do you really use this?
Anyway, whatever, that's a longer answer there, but that's kind of how I think about it.
If that makes sense.
I have one more.
Yeah, it totally makes sense yeah yeah it totally makes sense and i i i like the previous answer on this
as well where like oh like you know circle probably cares that like all the issued usdc
on like base cannot be like with which is something that like yeah probably facet is one of the
i guess you probably only actually roll up uh that like you can do this on so that's i think super
cool maybe one last question if you don't mind.
The most valuable state, I guess, maybe,
that people, I think, talk about these days is maybe stable balances.
So I'm kind of curious if you have a particular prediction
or maybe a request for how you would like stable coins
to exist on facets.
Sure, yeah.
I mean, I think you're exactly right.
You know, facets is the chain for native assets, and facets is also the chain for stablecoins. And
that sounds kind of silly, because like, we're just a little thing right now compared to the
big stablecoin people. But yeah, I think facet is the best chain for stablecoins because we can say, hey, if you issue a stablecoin here, you have complete power over the relationship you have with the customer, with your user.
You never have to worry about us intervening.
So it really is an empowering thing, especially to people who aren't using the canonical bridge anyway.
So yeah, we think Facet is the best chain for stablecoins.
And also these stablecoins, and I mentioned this a moment ago,
but it's just really important because stablecoins, in one sense, are funny money.
Circle can do whatever it wants, and that's a feature and a bug.
That's the deal.
But stablecoins are not exactly funny money
because it's not always the...
Like, if you are...
If you're on base and you hold USDC
and then you get censored by base,
then Circle could reissue that USDC
to you somewhere else.
Like, the balances are kind of more fungible,
maybe, when it comes to stablecoins.
Like, they don't have to actually bridge things in and out.
They can just blacklist you and whatever.
So, but there are many, many smart contracts on base in any other chain, not to speak on base, that have USDC balances.
And those smart contracts don't have an independent existence on the L1 or any other chain.
on the L1 or any other chain.
And so if the functionality of those smart contracts,
like Uniswap pools and so forth,
get interfered with or halted,
Circle does not have the information required
to determine, the information doesn't exist,
to determine who actually, as a human being,
owns that USDC.
So it's a really big risk.
So yeah, I think that's
the answer. We want to be the best chain for native
assets and the best chain for stablecoins.
Boom. What's up, Sasha?
What's up, dude?
Hey, can you hear me okay?
Yes.
Yeah, just want to say congrats on the launch.
I didn't expect it to come this fast.
I guess I have like a spicy take.
I mean, I really appreciated our space on Sovereign rollups.
And for me, the spicy take here is you're kind of um like leapfrogging
in a sense you're forcing l2b to sort of accept that sovereign rollups are a thing in a way by
redefining this and i'm curious to say how you see that do you see it that way do you see it
differently um how do you see the future of respect to sort of the canonical bridge being the centerpiece of how Ethereum or Ethereans define a rollup versus something like FACET?
Yeah, that's a great question.
Like, I always want to be careful here because, like, you know, people hate sovereign rollups.
And I really think you can really get people mad at you.
So I've been trying not to.
But yeah, I think, you know, what is a sovereign roll-up?
So for those out there, so what is a sovereign roll-up like?
You could get a lot of people to say what it is.
For me, a sovereign roll-up is a roll-up that has no in-protocol layer one smart contracts.
In other words, no smart contracts that are viewed as being special in the eyes of the protocol, for example, a canonical bridge.
And so FACET is, in that view, is a sovereign role today,
and then we've gotten in.
So yes, we've changed some of the dialogue there.
But there's a big difference between what FACET was before
and what it is now in terms of who we are.
FACET was also a sovereign role six months ago or three months ago,
but we didn't have a proof system then.
And what we always used to say
is, hey,
FACET doesn't have
a proof system,
but it, as a protocol,
doesn't rely on a proof system.
So anyone could create
their own proof system
we wouldn't even have
to know about,
and they could build
whatever bridge,
whatever proof system
they wanted,
and that's all good,
and so that's not
our responsibility.
And then it was like, all right, guys, you need to build a proof system beyond L2B.
And then we sat down to sort of build it and we were like, oh my god, this is pretty hard.
So at that point I was thinking, A, it's not so good for protocols to say, hey users, just
do it yourself.
And also B, I want to pull up the ladder for anyone not doing this so that I can basically
lord it over them.
So, yeah, I basically changed my view half facetiously for those reasons, which is that I think just as a rollup should provide software that implements the state transition function, should have a node.
You can't have a rollup unless there's a GitHub link, basically, in my view.
There's got to be something I can run. I also think that you can't have a rollup
unless there is a proof system that you can run,
that you can actually use,
a proof of concept, so to speak,
that this rollup can be proven.
Because it's not obvious that a rollup can be proven.
It's not like there are details of the facet setup
that we had to change around this.
So yeah, I think you should have that.
But I don't think you should have to, you know.
And then the other thing L2B said is, like, listen, we had to build this bridge setup, which we weren't planning on necessarily doing.
And so I think that's also a good thing for a roll-up to do is show here's a proof of concept for how the proof system can be used to create an actual trust-minimized bridge.
So, you know, I like that stuff.
I think it's, if you look at it on a raw protocol standpoint, it's not essential because it's just software.
But I'm a believer in that.
But, yeah, I would say that because we still have no in-protocol smart contracts, you could still call us sovereign.
And so then we are the first sovereign roll up stage two.
I'm trying not to use that word too much because I feel like people really, really don't like
So if I can avoid it, why bring the hate?
You're going to get us kicked off L2V.
Sorry, Michael.
No, no, I mean, regardless of the terminology, I think one thing that's been really great
is to see you guys shift the Overton window i mean getting the the optimism guys to kind of say okay actually this might you know
the way we're thinking about it has maybe been limiting and if we really want to get to stage
two then we should think about it differently which i think is better for ethereum and better
for users um for me that's a huge win so congrats thank you yeah like, we are, you know, there is also the situation that says, what is so great about stage two, right?
And that is something that, you know, we're very proud to be stage two and whatever, but, like, the ordinary stage two designation already, in my view, in our view, ensemble roll-ups, right, we talked about this extensively, also has really big problems because of this whole fantasy that an exit window can meaningfully help you in Armageddon.
You know, the arbitrum's getting shut down.
$20 billion of value is getting erased in 30 days.
Everyone get out.
Like, it's a fantasy, basically, to think that that will be,
you know, okay. And so our view is like, yes, you know, there are a lot of things that are,
you know, not great about FASTA that could be better, and even our approach might want to
reject it. But even if another approach got to stage two, if it had this 30-day exit window
thing, if it had this ability to be shut down, then there's still a great limitation there. So,
this ability to be shut down, then there's still a great limitation there.
So, you know, definitely want to push this idea that, like, hey,
you might not like things about the way FASTA is doing it,
but, like, we've got to come up with some kind of answer here.
You know, we're trying.
So, yeah, hope to push the discussion forward.
I don't want to make optimism too much like, you know, FASTA.
We need to be the renegades, but I, I, I would, uh,
I would love to, uh, to get the facet spirit out there more.
So thank you. You got to come tomorrow. Uh,
I think it's going to be tomorrow.
What's happening tomorrow?
Trying to do another unstoppable rollups call,
which would be similar to this except in a Google meet with a, uh,
slide deck that I, you know, will prepare. But yeah,
I'd like to dig more into like the mechanics uh slide deck that i you know will prepare but yeah i'd like to dig
more into like the mechanics of the uh you know the mechanics of the proof system like
i don't know yeah i don't know what it'll be about we got tim luke is gonna come uh i think
it's going to be uh not announced yet but it's going to be uh 2 p.m ut, so 10 a.m. Eastern. Everyone in here watch and I will tweet about this,
but it'll be fun. Some of the same. Some different. It'll be nice to be able to present
something in here. I like being able to present, but then Twitter spaces has its own benefits.
but then Twitter space has its own benefits.
What else?
Who else is out there?
Who else?
Comments?
Questions?
You know, I can try another room reset.
Facet, first general purpose, stage two,
Ethereum roll-up.
What does that mean?
That means Facet is the only roll-up.
And, you know, I really want to try not to, like,
because we are humble in the sense that, like, we got a lot of flaws, and we're building on the shoulders of, you know, I really want to try not to like, because we are humble in the sense
that like, we got a lot of flaws, and we're building on the shoulders of, you know, amazing
people and whatever, and Kelvin was here. But like, still, it's true that FASED is the only roll-up
today that cannot, for lack of a better term, F with you. So if you are on a random roll-up,
a random stage one roll-up, that roll-up, meaning the are on a random roll-up, a random stage one roll-up,
that roll-up, meaning the admins of that roll-up, which means some group of people in an organization
that came in islands who are great people and whatever, but some group of people. It's people,
right? People are all kinds of problematic things about people, myself included.
People can F with you. People can devalue your native balance of Ether.
People can remove the ability to include for you to force transactions.
People can do things like this, and that can get in the way of the thing you want to do, which is interact with some app.
So, like, if you care about the Optimism Foundation, then you should interact with them, right?
Great. If you don't care about them, you shouldn't have to interact with them, right? Great. If you
don't care about them, you shouldn't have to interact with them, right? But that's not how
it works today. If you don't care about the Optimism Foundation, you still have to interact
with them on Optimism Chains because they control a lot of what goes on. And so, you know, we think
that's not ideal. And so we wanted to build a chain that wasn't necessarily the chain where
everyone gets rich, right? You might not get rich on facet, right? But if you get rich or you don't, or you don't care about money, what's so great about money? It will be because of what you decided to do and what an app developer decided to do. It won't be because any foundation did anything, it won't be because Michael or I did anything. It's going to be on you. And this is just like the way it works on the L1. The L1 is a, it's often been called,
you know, a dark forest, whatever. It's a book about this. Like, it's not necessarily the most
inviting place to be. You don't go to the L1 because you think everything's going to be like
super chill all the time. You go there because you don't want people messing with you. You don't
want people to be able to affect you without your consent. And that's what FASTA
is about. That's what we think stage two is about. FASTA has got tons of downsides. The biggest
downside is it's more expensive. So, I mean, that's a really big downside, right? If you want
the cheapest thing, then you shouldn't be using FASTA. We're going to fix that. But FASTA,
you know, the more expensive, there's something that you get when you use
facets. You don't have to worry about what's going to happen if the chain gets shut down,
you know, tomorrow, which could happen with these other roll-ups, but would not be here
at all, even close, without, you know, the OP stack in particular so um you know got to shout
them out and hopefully uh we can help um affect change on the op stack as well but that's what
we're uh talking about today facet first general purpose ethereum stage two roll up
facet is the first cypherpunk l2 i guess there's something there and yeah you're curious how do you
get it cheaper what's the the plan there right this is a good question so so cheaper uh means
you know sequencer basically someone who's the intermediary and so obviously we don't want like
a centralized sequencer although maybe we should have done that we could probably make more money
doing that and whatever so you know whatever our bad for not doing that, but we've got this
base thing. And so, you know, basically it's going to be a total anarchy base style. So,
you know, when base sequencing, when sequencing was first invented, Vitalik made this term
total anarchy, which came to describe base sequencing. Basically, the idea is contemporary base sequencing approaches
are reliant on smart contracts to do important things,
like tell you whether two people are proposing the same block
in an invalid way.
And we reject this because these are inevitably upgradable,
and that's its own problem.
They can shut things off.
So it's going to be total anarchy.
Similar to how the OP stack works, because OP stack, you submit blobs to an EOA,
you're going to be able to send blobs or call data batches or whatever,
and anyone can send any partial block.
This is the big difference.
So anyone can send any partial block,
and the partial blocks will be included in a facet block
in the order in which they are seen by the node,
in the order in which they appear in the Ethereum block.
And so you can end up with some cool stuff here.
You can have app-specific sequencing, so you could go to some DEX,
and maybe the DEX runs a sequencer,
and the DEX can then submit a batch of just those DEX transactions,
and then it could prevent MEV with those or something still working this out.
But yeah, the overall challenge here is how do you pay the sequencer?
And so one obvious way is in FCT, the famous native gas token thing.
Maybe there's some other innovations that we could pursue there.
I don't know how else you could do it,
but yeah, basically high level users submit
individual transactions today to an inbox.
In the future, they should be able to submit
more than one transaction at once to that inbox
and then have those transactions still come together
to form a block.
Interesting.
How expensive is it today compared to,
for example,
optimism and like how,
how,
how low do you think you can,
you can make the fees?
Do you have any,
I mean,
don't put me on the spot,
but just curious if you have any ballpark estimates.
Sure.
I mean,
you know,
so today right now you're sending an L1 transaction with a small amount of,
of call data.
And so, you you know whatever that
is like I'll log back into my computer here so let me just go to etherscan and just look as
etherscan has this nice thing so it's basically okay so right now it's... Whoa, it's really $0.08? Okay, so yeah, it's blow1gwe, so it's super cheap. But yeah, so 10 or 100 times more expensive than a traditional roll-up, which is not good.
How good can we get it? The challenge here is, I mean, you know this, right?
you know this, right? It's that if you are based, you have to post everything the next block. And so
you're competing with chains that can hold on to data and run ahead for days or whatever,
or at least a day. And so it's really, really, really hard to compete with chains that are
batching things, batching hours and hours versus transactions together. I think to get there,
there's going to have to be some like real,
like zero to one things on like blob sharing or something.
We're talking with Spire,
which is a company who's,
they're working on something in this world of blob sharing
that we're excited to understand and use maybe.
But like, hopefully we don't have to do all this ourselves,
but like to make, there's kind of two,
so one is Facet and we're weird,
no smart contracts, no kind of whatever,
but then there's also just base rollups
as their own massive problem.
And so that kind of hits us both,
but yeah, to get Facet as a base rollup
to be as cheap as these non-based folks,
there's gonna have to be a real breakthrough,
basically in terms of,
you know, using blobs. But hopefully Spire or someone does this and we can do it. But even
without blobs, we can still make it a lot cheaper. Like, the issue today is that, and sorry for
everyone who's excited to build a facet app, It's great, but it's pretty confusing because even just at a given price,
the user is expecting to connect their wallet
to a different chain.
You connect your wallet to the L1,
you send a transaction to the L1.
It's a little confusing in addition to being expensive.
So probably we can do,
hopefully we can do both of these things.
But yeah, that's the landscape.
It's hard.
I was surprised we were able to get proofs out before sequencing, but thanks to the magic
of ZK, sequencing is actually the super hard problem.
Yeah, I mean, I think if anyone can solve the base sequencing and DA price issue,
it's probably Spire.
Talking to Matthew over the last six months,
not recently, but he's incredibly bright and incredibly driven.
Encourage anyone to follow him at mteamisloding or at spire underscore labs
because they are doing incredible work in this space.
So yeah, bullish.
Yeah, I love it. We it we're talking this week i really um i think you know i think it's a shame because the base thing
is just still so captivating for me and i feel like there just has not been
you know this year as much progress as i had hoped there would be. So hopefully that'll pick up.
So if you're working on base stuff, like work harder, basically.
Like I'm, I'm a, I love base rollups.
Like let's, let's get some progress here.
Come on.
Let's make it happen.
Don't leave it all to M team.
What's up, Menev, Menev?
Did you jump up here?
Someone just jumped up here.
Yeah, it's me.
Oh, yeah.
Hello.
Hey, everyone.
What's up?
Yeah, what's up?
Yeah, I've been thinking about this, how,, like, for choice, your rule, yeah.
So I kind of, I built, like, a stack previously for, like, we used to brand it as stock-sop
and rollups.
But then we kind of now do, or, like, it's more like a branding is more as an L1, but then I feel like what I kind of see, uh,
facet as is like, it's incorporating this like creative whole choice rule from Ethereum.
So it's just like, uh, it just incorporating, like you can take an Ethereum state route.
And if you burned one ETH there, then you can min gas on the chain.
Right.
Yep.
Uh, I feel like you, you can min gas on the chain.
I feel like you can do more cool stuff.
We could add some sort of fork choice rule sort of interface for a stack so you can kind
of incorporate other things.
So you could, for example, send money on Venmo and then that mince gas on your chain.
I feel like you could do things like that as well.
That would be pretty cool to see as well.
Been thinking about that recently.
Interesting.
When it's like being your post.
Interesting.
I mean, I think you need a way to like sequence that.
But I think one thing you're getting,
one thing I think of thinking of this is like,
you know, we are trying to ride like trends, right?
Like obviously we want to create trends, but like, you know, are trying to ride like trends right like obviously we want to create trends but
like you know ride the trends and so one trend that i've mentioned a lot is the stable coin trend
is the non-canonically bridge value trend but one thing you're making me think about again
is this uh gas token uh abstraction trend right which is that oh i feel like you could do much
much more than a gas token you could do for
example do things like you could just bootstrap a chain for example like an artist could do it
and then they could basically let people who are like their top listeners let's say on spotify or
something they have some sort of uh gift to them know, or like a token or something.
And then they can kind of come together
and do things together.
So I feel like you could kind of use it
as a powerful bootstrapping mechanism as well,
instead of like launching something empty
and then being like, hey guys,
come and bridge to my chain.
Yeah.
So then you're saying,
but if someone did a,
but if you had the Venmo, like anything that was an input to the chain would have to be like something, you know, repeatable, you know, something you could play back.
I mean, this is something we've also talked about, like, it'd be cool to like have like, you know, Twitter be the driver of a Facet instance, you could have facet and you could just say, okay, it's Twitter. And obviously, you have to do something there.
But if you have any external,
any non-blockchain thing that is being observed by the node,
then can it be replayed in a consistent way or whatever?
Is that what you're-
Yeah, basically.
It could be stuff like the ZKTLS,
or it could even be just like a random T compute
sort of thing, right?
Right.
But whatever, it doesn't need to be on chain.
It could just be anything you do in any Web 2.0 place.
Right.
As long as it's like...
Yeah, that's interesting.
I mean, I think... Yeah. like, yeah, that's interesting. I mean, I think, yeah, I think that is.
Yeah, I feel like it's pretty cool.
That's something maybe I'd like to see if something like that would work.
Yeah.
And I feel like then what you can do is like basically enshrine something like this in the L1.
is like basically enshrine something like this in the L1. So you could have a verifier for like
ZKTLS specifically or like TEs and stuff, right, as well. Or like basically it could be a pre-compile
or something, you know, like especially with Ethereum like trying to do lean being more lean yeah yeah yeah I think it's
interesting I mean I think I think I mean what I was gonna say about the gas hook abstraction
yeah it's not really I guess actually not understand it related to your very only basically related, but just like I think the normal user experience of this stuff, right,
is not going to be anything relating to like, oh, I'm bridging in and here's my like native balance or whatever.
You know, it's not that is like a true advanced user thing to know about like the gas token that you're you know
people should be paying gas and whatever they want on whatever in whatever web 2 in context
they want you know with with usd and just it's sponsored for them it's paid for them so you know
separate from what you're saying i think there's a uh uh that you know that trend also you could
even like sort of get rid of gas tokens uh fully if you do something like this.
Because gas tokens at the end of the day are the way to pay the infrastructure costs or
it's more spam resistance. If you, for example, had a rule where, oh, like all the humans on world chain kind of get like a free, you know, they don't have to really pay gas.
But then if someone else is non-human, then they have to pay.
So it can be more like the bots compensating like real users kind of thing.
be more like the bots compensating like real users kind of thing yeah well i mean that is the uh
and maybe this is an advantage of based rollups finally looking for advantage for these these
things which we love but like uh you know for us right the base fee is uh is burned it's it's it's
it's and there's no priority fee now there will be but it's it's um it's like the l1 there's no
there's no implicit economic assumption underlying the operation of the chain
that someone has to get paid this money for the chain to keep working like it does.
For us, it's purely a spam prevention thing.
It's all burned.
And so, yeah, I think that gives you some flexibility.
I mean, everyone on the chain is sharing the same, you know, resources. So like, you know, for that, people would all have to use the same, you know, exemptions.
But yeah, I mean, definitely we are not seeing the gas limits being exhausted today.
So we could probably, I mean, gas is basically free now.
So although we did set a
minimum base fee, okay? We're tired of this eight-way thing. That's what it was before.
It's now a million ways, so you can't go too crazy. But yeah, no, I like this stuff.
I like this stuff.
What's the facet endgame? Like, do you see it as something that could become as competitive as base,
or do you see it as more something that pushes the Overton window?
How are you thinking about the long term?
Well, we're trying to make money.
We are trying to push the window, and we're trying to be good or whatever,
and we're not trying to make money, you know, in the way most people are.
But just to be clear, like we are trying to make Facet big and make Facet great and then be app developers on a big, great chain that, you know, produce apps people, you know, people want to use basically.
And, you know, the protocol won't play favorites towards us, but we believe that, you know, we'll be able to offer a great
user experience. So we are, you know, we can't do it for free, you know, forever. So we want it to
be a great thing that people use and we want to build apps. But, you know, yeah, the big picture
is like, you know, you got to be careful how grandiose you sound because you don't want to,
you know, turn people off or put yourself in a box of some
kind or, you know, whatever. But like, you know, we think, we think it's a better way of doing
roll-ups, you know? And this is something we've talked about extensively, obviously, in the
sovereign case. So we think if you call us sovereign, this is that version of sovereign.
We think it's a better way. And we think it's a way that the actual consumers, the stable coins
and whatever, as we discussed, are going to want more. And so, you know, maybe the ship has sailed on new
ideas, in which case, you know, too bad, you know, so sad for us or whatever. Like, it'd be great to
get Coinbase on Facet if we had launched it, you know, a while ago and the stars have been different.
But, you know, but yeah, the goal is to basically say, hey, this is like a new and maybe better way of doing this stuff.
And yeah, if you're thinking about creating native assets or being a stablecoin, you should come join.
So that's the goal.
Yeah, I mean, I love to push the Overton window.
And I think it's also very fun to do this project.
But ultimately, the goal is to have people appreciate the value and build on it
and for us to be a part of that, I think. I think there's also another question
that we've talked about, and you could probably give some insight there, is how attached are we
to facet itself, facet with this specific chain ID and everything. And obviously the answer is we are very attached to it because it's the thing that we did.
But is that level of attachment the same as the level of attachment to the approach?
If someone else were to copy-paste our code and launch a new Facet thing tomorrow,
I would be pretty... I'd say that's really cool.
I guess it competes with our thing pretty i'd say that's really cool like i guess it competes
with our thing and whatever but like that's dope like you know so another question is like
you know how much and probably not until we have at least the batching thing built out but how much
we want to encourage other people uh to build this stuff and so that's like another end game
where you say okay now it's like the we call it the unstoppable stack because we don't want to
you know even pump ourselves too much or something some other brand or the facet stack call it the whatever
stack and say okay anyone can launch one of these things and now we're going to actually leverage
some of this base roll-up stuff which has been such a pain uh and we're going to leverage it by
now people can do this synchronous composability thing somehow between us the different facets and
the l1 and so uh you're going to get like, you know, whatever people want to launch
change for app chain stuff or whatever.
So that's, that's, that would be kind of the other thing, which unfortunately
like is just farther out of like the zone of which I really am super up close to
like what would make that successful.
But I think there's, there's something really that could be interesting there as
well.
Are you talking to stablecoin issuers?
Sorry?
Are you talking to stablecoin issuers yet, or is that sort of in the pipeline?
I can't confirm or deny any giant launch.
No, no, we're not really talking to anyone super seriously.
It would be good to do that.
I mean, I think prior to today,
we were not even on L2Beat,
so I think it would be tougher.
But yeah, I think that's something
we want to start doing.
Now you have the momentum
and the mindshare.
I think now's the time.
If you want an introduction to Noble,
I can facilitate.
But yeah, I think this is now the time to accelerate.
Yeah, super bullish.
Hell yeah, I would love that.
I'll send you a message.
Send you a message.
Stable.
Is fast...
I'm sorry, go ahead, Manas.
I was just asking a question.
How fast can, like, right now,
is it basically the Ethereum block time?
Yes.
Well, strictly speaking, it's 12 seconds,
because, like, when Ethereum misses a slot,
we have, like, a blank block or whatever.
But, yeah.
So in that case, it's actually, like, pretty fast.
Like, 12 seconds is slow, but in terms of getting your thing actually on the L1, it's actually like pretty fast like 12 seconds is slow
but it's in terms of getting your thing actually on the L1 it's like super fast
so you send it it's on there so it's cool yeah 12 seconds though so can is there
like a way if you would you ever consider doing like something like sort
of like base I didn't like this sort of pre-confirmations to it right so that's a
good question what's the deal with pre-confirmations to it. Right. So that's a good question.
What's the deal with pre-confirmation?
So, you know, for one thing, I am excited to see
what the deal is with those things.
Like when is someone going to, you know, I'd love if I could help,
but like where's the actual like thing that you can like, you know, use?
I haven't really seen it. Maybe I haven't looked close enough.
I haven't looked close enough.
But overall, I think it's kind of like I, you know, for better or worse,
But overall, I think it's kind of like,
I, you know, for better or worse, for based,
I am hesitant to enshrine, you know,
I'm hesitant to like give up,
if you're going to be based,
like I don't want to give up like the good stuff there.
And so, you know, when it comes to pre-confirmations
and the designs I've seen, it's like,
okay, there are some group of people who can do it.
And then there's some smart contract that manages that. And maybe there's like a elections or, you know, you can, there's an auction.
It's just like, if you create this structure,
do you lose some of the permission lists, like kind of like magic?
It's more like you don't have to, you could just add it in a preferred way,
but then you could still keep it.
So you would still have the base part of, you know, the thing.
So you can always, like, send something on L1, like you do right now.
But then you would basically just give up the rights for, like, a certain amount of slots.
a certain amount of slots. So you can say, okay, if you send it through here, then they
get to be the first one to include transactions in, let's say, 20 blocks or 10 blocks or whatever.
Right.
So then if you send it on the one, then they would also be executed. But that's the rule of the chain.
Yeah, and look, yeah, this kind of gets to some other stuff I've said about
I'm not some based maxi insane person saying that.
What's important to me is transaction inclusion.
And so you can give a first look to someone.
I think, what is the amount reserved for the pre-confer is the question.
But I don't know.
The real thing is I just have not seen,
I just have my basic prejudices or whatever,
knowing I don't want to add trusted intermediaries unnecessarily.
That's just a thing.
Where's the good version of this that's like,
I just have not seen the actual thing where here's the way it works,
here's the code.
Yeah, you should use, you could look at like sovereign SDK. They have like a preferred sequencer kind of model
and they have good docs. That might be something you could look at. That's like a pretty cool
thing. And then anyone can come in and register as a sequencer as well. And then that's like,
that's all permissionless.
So basically there's no trust in the intermediaries.
Okay.
Sovereign SDK?
Yeah.
Like they can't mess up with whoever's like the sequencer can't like really mess with
you except like basically giving you like censoring your front actions for like a couple seconds
right
they can't
cool
okay yeah
I'll
I will check this out
alright I got it
I had a quick question
about the
the ZKA fraud proofs
so like
I've just been
studying I don't have the dog
in the sequencer fight, but I do have a dog in the
approver fight
and I'm wondering, given
the base roll-ups like transaction
inclusion is the most important thing
how you think about
malicious transactions
that are designed to mess with
approvers. This is like
you know, a thing that
I've obviously been doing research on and the TLDR is I can construct real cheap gas, but really
proving expensive transactions. And it looks to me like actually the ZK proof isn't required.
This is just an optimistic system with ZK proofs tacked on top. I just wondered if you have thoughts
about that. Yeah. I mean, that's a great question. And like, you know, I consider myself to be, you know, a well-informed person on Ethereum.
But when it comes to the thing you're saying now on the ZK side, like, you know, I, we really, you know, rely on, you know, succinct a lot here.
And so you can do tests and you can do cost
estimations and you can do all this kind of stuff. But yeah, the intuition, like mod expert,
I forget even the stuff, like the various, you know, the intuition, like I don't have like the
super, you know, it's like, yeah, you can mess things up without actually using a lot of gas.
That's a big problem. I mean, I, so the way our system works is there's actually like two tracks
to it. So one track is anyone can do a validity proof at any time in 1800 block increments. And so
go nuts, but also a whitelisted proposer can do optimistic proposals unless they are
silent for two weeks and then anyone can do optimistic
proposals. So in the case that you're saying, it could make a validity proof really hard
or impossible, or it could make an optimistic proposal indefensible, undefendable. So like I
proposed the route and you say challenge and now I have to find the proof and the proof, you know, cost me $10 trillion.
And so, you know, I can't prove it.
And so, yeah, we can't get the new route in.
So I guess, you know, not as bad as a bad route getting in, but yeah, still a problem.
And I, you know, this is something where, yeah, this is a tricky one uh basically but you know that's
kind of why we're set up yeah i think i think the optimistic system like is the the judo move around
this particular problem is like as long as you have like your honest proposal then everything's
kind of hunky-dory um and i don't think it's like i think it just delays finality with the validity
proof so i don't know i'm not really doom and gloom about it but it's like, I think it just delays finality with the validity proof. So I don't know.
I'm not really doom and gloom about it, but it's interesting to see how you guys have designed the system.
Yeah, I mean, I know this is recorded and I can say that we, you know, we did a lot of testing with this whole thing.
But, you know, in terms of the actual most pathological everything, ModX, you know, it's like, it's hard to.
They got to fix the gas cost of these things. I mean right like i shouldn't it's not my problem make the gas work
that's literally the crusade that i'm on right now so i'm uh i'm working on it is there like a
thing that i do you have like rec i mean we could i prefer not to do this but i mean
are people doing this we could be a proof of. I guess we don't want to do any big fork now,
but like, is this literally just,
is there a kind of a playbook here?
Like make, increase the gas cost of this one to this,
or is it still just active research?
You know?
Well, the issue is the entire ecosystem,
the entire Ethereum ecosystem relies on a bunch of dev tooling
that sort of hard codes all of these gas costs
for all the different opcodes.
And it's like a social coordination issue
unless it's a technical issue.
You know, it more or less requires Vitalik
to just like say, okay, we're doing it.
Just like how we had, he tweeted,
okay, we're raising the gas limit and blocks
and then everybody all of a sudden
changed their validator parameters
to increase the gas limit to 40 million.
So that's what I imagine might need to happen here
and so but we have raised the gas limit uh before though but i guess he did have to step in
yeah it's it's that's less of a social coordination issue than all the you know you got to change
hard hat and freaking foundry and pretty much everything that sort of interacts with the chain.
Right.
And I guess we want all those schools to...
Yeah.
But, I don't know.
We've got a paper coming out on the 20th that kind of describes all the different Rupert Killer stuff.
So, I'll send that your way.
Yeah.
Hell yeah.
Hell yeah. kind of describes all the different prover killer stuff so i'll send that your way yeah hell yeah hell yeah i guess this is so maybe this is like the secret double triple thing where
it's like hey like the fault fruits of all these problems so zk it's just gonna but then can you
um so yeah looking forward to this to this paper there's probably a bunch of things that
yeah yeah throw on a te and uh that that side steps the problem entirely. But then you have to
make sure that that thing is
okay or whatever.
I guess
someone can challenge that with
like you still
or I guess what is Vitalik's gadget, right?
You can have TE plus fault-proof
thing. Yeah, exactly.
anyway. If we're gonna but the zk stuff
is so simple that's the allure of this like i guess getting the te thing set up i'm sure once
i started would probably you know because zk seems super stressful but like i just like there
being one thing but you know it's it's it's a whole thing. It makes sense.
So cool.
I appreciate you thinking through it with me.
Let me...
It's a fun thought experiment.
And, yeah, okay, so I hit you with the...
Oh, you have this on your pinned tweet.
Prover killers.
Yeah, it's kind of the only thing I talk about.
And so, oh, you're a prover killer. Okay, yeah, yeah, yeah, yeah. i talk about and so oh you're prover killer okay yeah yeah
okay so i'll follow you notifications on and uh you will tell me uh what the deal is
max pain mod x i was right i you know i've heard about mod x uh but yeah cool
cool cool i'm gonna go back to work but Godspeed and welcome to the community.
Bye, everyone.
Peace.
All right, Manav, I don't know how you got kicked off, or maybe you left.
Good, good, good.
Cool. Okay, well, this has been a meaty and very fun space.
There's going to be Unstoppable Roll-Ups community call, I think, tomorrow.
So I'm going to tweet about that. You can watch for that.
It's going to be some of the same stuff, maybe different people.
Luca from L2Beat will maybe be there.
Anyone have last questions or ideas?
Really appreciate everyone who participated. Again know this is just the beginning everyone always
says that but it's true true for us for real and so it's great to have y'all's
interest and support and I really appreciate it so unless there is anyone
else last time okay thank you all good night bye bye Thank you.
