Bitcoin and Beyond EP2: Build on Bitcoin's Hybrid Rollup Architecture

Transcript

[Music] [Music] [Music] welcome back to bitcoin and Beyond today we have Dom the co-founder of build on bitcoin for a presentation about what they are doing with merge mining their hybrid L2 uh and some other interesting

technical topics Dom welcome thanks for thanks for being here hello welcome thanks for having me um great to meet everyone uh I'm Dom um I'm one the founders of uh Bob and we started Bob last year essentially as we saw you know orinal Ser of 20s ruins well runs were announced last year but anyway um we saw like this new found interest in building on bitcoin and that's why we deated the project Bob uh to make it a lot easier for people to build on on bitcoin and I'll I'll walk through a little bit about what we're about and then do a deep dive on on what makes Bob special with this merched

mining um uh technology that we have right so uh you know Bitcoin has like 300 million users right now and it's growing faster than the internet uh which basically means we can sit back relax and just take history takes this course or technically we might could could do that but we we actually see a big problem um where you have a a huge difference between the Bitcoin space and say ethereum and other uh l1s l2s where bitco mostly focuses on on mass adoption you know like real world impact which is

great uh but then new things around I don't know zero knowledge proes account obstruction defi stable coins that all happens kind of in other ecosystems uh because it's really hard right now to build on bitcoin actually a lot of teams are trying quite hard and and there's just you know still cool things being done on bitcoin uh but if we are quite realistic then very often uh if you meet people that are getting into the crypto space they might hear first about Bitcoin but if they're actually ending up building something they typically go uh R it to ethereum or polygon arbitrium uh one of the evm chains uh because if you look at how

easy it is to write a smart contract and solidity versus how hard it is to write a Bitcoin script um you'll quickly realize that's that's kind one of the key reasons so really why we exist is like we wna we want to bring back this Innovation um to the Bitcoin space and do that via a a L2 um as many others you know uh are doing it right now and I think there's this new um we we called it the Bitcoin Renaissance in in Denver there was the Bitcoin Renaissance event that we coasted with with bubon together and uh it's just you know grown since then over the last few weeks and months

and there's so many new projects now um that same sharing that same vision um and we think it it's really worth it to bring all that Innovation to bitcoin we think that to to do a Bitcoin L2 there's there's kind of three major problems that needs solving uh the first is the Bitcoin problem and by Bitcoin problem we mean um the L2 needs to be aligned with with Bitcoin as an ecosystem and it needs to get uh Bitcoin Security in one way or the other and the second problem and third problem is something that we we don't see as much tackled by other uh l2s out

here uh one is ecosystem um so you want to have a very thriving um set set of apps already live uh you want to be able to build on top of other things so if you if you look over at evm chains you can you know use the open Zeppelin contracts you can build things on top of other protocols uh people call it money Legos and Defi and really that that's very powerful that that Network effect that you have if you use the evm and third thing is liquidity so if you have a lot of users and liquidity already on a chain uh it's really easy for these users and that liquidity to you know move over to other projects so

if you're new project and you want to try out something uh then it's great to to launch on a chain where there's a lot of Trac already um and the way we are approaching it is kind of unique uh I think we're the only ones that are trying to do what we term a hybrid L2 so we have a rollup that's going to be secured by Bitcoin uh integrated into the Bitcoin stack uh VI like light Clines and uh we do want to have Integrations with lightning and Noster and ordinals vc2s uh runes uh but we are also a rollup on ethereum um just to make it super easy for users to come over from the e space um a lot of people are used to you know bridging over from

ethereum to arbitrum or to base or to optimism uh and and bridging between like these these l2s and that that has proven to work quite well and we think that that's kind of the way forward uh just so you know if you're a project and you want to build anything with Bitcoin usually and it's defi you want to have access to stable coins and instead of convincing tether and circle to have dedicated deployments on on our L2 uh it's really easy for us to bring over V and usdc directly from the ethereum network and that's what makes us kind of special in in this realm of of l2s that are popping

out right now and how that kind of works under the hood is uh if you look at the Bob stack like it starts at the rollup layer uh all the way down there um that's based on the stack um uh so it's really it's an optimistic rollup uh it's rolling up to ethereum it uses ethereum as a DA uh has the evm core on the layer up so you can use any of the evm wallets you can use hardhead Foundry you save multi6 so you have all this tooling out of the box and you don't need to do anything uh special so if you're new project uh you can get started right away as if you were in any other evm chain uh if you're a user uh you can use it like any other evm chain

and where then we come in and make it kind of special is with the Bitcoin Bridges so we have dedicated support for that uh we have an on-ramp that's going to launch for like an on-ramp from Bitcoin that's going to launch uh very soon um we're working together with risk zero and others for for ZK stuff um and then we have this Bob um that helps you if you're building an app and want to interact with Bitcoin to just make that a lot easier and it includes a few things like being able to prove that things happened on bitcoin by like client uh but also includes like some helper libraries for Bitcoin transaction signing utxo selection and so on so forth nice nice yeah the I I haven't

tested out main net too much yet um except for just kind of browsing over some apps but from what what what I'm understanding is that this should basically feel like an ethereum rollup from start to finish for users who are accustomed to using evm rollups yeah right exactly so it's we didn't make any changes uh to how The evm Works uh so for uses exactly the same experience as if they were using Basse or optimism and for projects it's it's exactly the same experience as well um with a key differentiator that you can access Bitcoin tooling if you also want to interact basically with Bitcoin yeah yeah exactly cool nice all right and then uh let's

um let's look a little bit where we are right now and and where we're heading and then look a bit deeper into this kind of under the hood stuff um so we we believe in this pragmatic way to decentralization so right now uh we completed phase one of Bob we launched Mina on 1 of May and that's the what we call the bootstrapping phase uh we are right now an optimistic e roll up um using the op stack and right now we don't have Bitcoin security right now we're working towards phase two uh which is being an eal up with Bitcoin Pro work security through Merchant Mining and I

we'll go into a lot more detail around that in a bit uh but it it really doesn't stop there um and phase three is moving from a being an e rollup to being a Bitcoin rollup and what we need for that is we need to be able to verify actually that what the ret does um is correct on bitcoin itself and that's not possible ible right now and there's kind of two major ways to get that to get there um one is bit VM uh so for those of you who don't know bit VM uh it's essentially a way to do frop proofs on bitcoin

um and it's a it's right now it's mostly a research proposal uh where few people work on it uh we don't have a a finished implementation yet but us and a lot of other people hope that in a few months we can get to a a good understanding of how a protocol would look like to do bit VM Bridges and how to do bit VM rollups um and from the other you know the alternative if you don't use bit VM at some point phase four for us would be being a ZK roll up uh if if Bitcoin would add like the necessary up code or up codes uh to verify K proofs on

bitcoin um for phase three we think that maybe in a year from now is we could have some sort of production implementation for BVM um but phase four is realistically I think like I don't know at least three to five years out um there's not you know one universally accepted ZK standard there um and it will see you know there's there's a good chance of so that Bitcoin will never actually add a ZK op code um because we can see right now for example with opcat uh how lengthy the discussions are and and how difficult it is to change

Bitcoin core um also because it's Bitcoin is made to be quite resilient right and and uh cautionary about changes um so I I think like it will take a long time until we have some sort of ZK op codes on bitcoin if that that ever happens could you give us a quick sidebar on op cat and what what is that op code right uh opcat uh is quite the quite the discussion right now so everybody talks about opcat um opat is a is a actually an an OP code that there used to be in Bitcoin but it's then was removed and now uh people talk about adding it again

um the the key thing that upcat will allow is basically to concatenate two hashes together that's very useful if you want to uh verify for example a Merkel tree um inside of inside of uh Bitcoin script um and right now you can kind of do that or well you can do it in bit VM it just it it ends up in very long scripts and there is a maximum script size basically in Bitcoin with upcat uh we could reduce the size of the quite a bit because there would be now where you need like a lot of up codes um to emulate like this um concatenating uh two hashes uh we could replace that with

a upcat with like a single up code um but there's there's a lot of discussion right now what kind of that would introduce to bitcoin and um it's a bit there's there's very loud voices on Twitter going forth and back against this um so yeah that's that's a currently ongoing discussion it would help bitm uh but looks like Robin is now kind of against it um so yeah I don't know um but yeah opat is like is is is essentially a new OP code that would make BBM more efficient and what's the tradeoff what's the what's the downside um so the downside is You by introducing a new up code it's kind of

unclear what implication that would have directly on bitcoin uh so since you could do new Bitcoin scripts uh with opcat um very similar to how you know uh we had tab rot and then people discovered oh we can do aals now and inscribe arure data to bitcoin um and then okay we can actually do bit VM arbitrary computation on bitcoin upat could enable like new use cases and there's a there's a crowd of Bon core users that are opposed to you know enabling it because they say it's basically unclear what kind of new things that that could introduce and if it could add some sort of new uh me vectors uh versus there's the other kind

of crowd around mostly around CH WIS right that that want to add that so we can create more efficient scripts um yeah okay got it thanks uh thanks for taking us down that little rabbit hole no problem um I think I mean there's a lot of good ongoing discussion online and you can follow some on the uh on the GitHub issues as well uh but it's quite a involved discussion at this point um cool all right let me uh let me dive a little bit deeper now into um you know what we have done so far and what we're going to do next so right now as I mentioned we're on mainnet uh with the

phase one like the boot shopping phase optimistic e roll up so it's uh it's all quite simple you know it's op stack so you would just use it like optimism or base uh and it used for Bitcoin bridging and uses rap BTC and tbtc from ethereum mainnet um so those are rap BTC is a centralized version and tptc uses like a set of of signers basic distributed set of signers to secure the Bitcoin um so there's you know no decentralized Bitcoin Bridge as such but uh the these are really battle tested life for years um and that's where we are right now uh and oh yeah just to to you know kind of recap quickly on on how that

works if people are not familiar fly how optimistic rollups exactly work uh you basically have um a sequencer and that sequencer produces blocks every two seconds uh for the Bob L2 um you have a ethereum bob Bridge um and you would finalize the data directly on ethereum and the interaction with Bitcoin is basically via a light client and what a light client does is you can read state from Bitcoin directly so uh if you want to for example do a peer-to-peer swap between say us on Bob and and Bitcoin on bitcoin uh you could essentially lock some usct uh and say

I'm going to sell I don't know 60,000 us for one Bitcoin and then somebody could go in and say yep I'll take this um and then submit a transaction inclusion proof from Bitcoin to prove that the Bitcoin actually moved over uh to the person that was selling the 60,000 uh us so it it allows you to do uh trustless reads from Bitcoin uh and the BTC R BTC relay is live on on B main app so other person can use that we are submitting block headers there uh so yeah you can you can do transaction Pro on that right now and and

um right now the the the only Bitcoin available on Bob are these are these effectively these custodial versions um which have been trusted for some time but once we once you have the the Bitcoin bridging aspects implemented is there going to be something more like a native Bitcoin implemented into Bob or like a a less custodial more Trust trust minimize version yeah so we are working uh for fast BVM is the key enabler to do uh kind of trust minimize bridging where um pretty much everything right now is some

variation of having multiple signers so you can rep btcs like uh you basically trust bit go and their signers um so it's like a small multi-c uh tbtc uh is a lot or centralized sorry decentralized in that way where they uh have a 100 different signup um but it's still like a restricted signer Set uh but with bit VM we can actually go to a one of n trust assumption uh so you just need like one honest entity in that in that set of signers and you know you could join yourself if you're technical enough and and would have the necessary funds to fraud proof and we think that's you know the

future but uh as mentioned like we are not BVM isn't in a state yet where you can't can actually Implement a bridge cool and then um for now for for Bob sequencers it's the same sequencer type as um like optimism like you said or base it's like a single a a single sequencer um with effectively with uh no fraud proofs at the current point yeah that's correct uh so right now single sequencer uh no FR proes right now um the same as like optimism and base um the op stack has its FR proof right now on the suola test net um so we're

basically waiting for that to be finalized and being able to merge it on mainnet um so that we can get uh permission as FR proves on B uh the moment uh that's ready on on optimism as well and that's a great question um because we don't like centralized sequencers either um and that's where actually like this phase two now comes in um so the you know Bas one like uh it gives you a basic chain right so uh if if we started the project is called Builds on bitcoin right so um we actually need to do the

connection to bitcoin but at least the building hard like people can build now something um and we decided to go with merch mining uh for phase two and what does merch mining do is it's a very old kind of Technology uh so dates back to 2012 um and uh that's practically ancient in in blockchain terms uh I would say um but before like diving deep into how it works uh give you like a why uh my merch mining is interesting and especially we have this uh new protocol we call optimine which is optimistic merch mining it gives us some

interesting property so compared to standard ethereum rollups uh we can get actually faster finality um by you have the centralized sequencer even if you have decentralized sequencers um you basically verify that the rollup is operating correctly and Bitcoin miners do that verification and there's a proof of work associated with that so they they have a proof of work that they actually validated this rollup and then you can settle the rollup quicker uh and you can you can get finality over the over the transactions and uh we remove a lot of trust in that centralized sequencer um

or even if there's multiple sequencers uh by having independent parties validate what happening and compared to you know Bitcoin set chains out there uh we can with this optimistic merch mining we can get a lot faster block times than uh current merch mining chains so the biggest one uh being rootstock at the moment uh they have 30 second block times uh which I guess is okay but in terms of like user experience it's kind of nice to have fast uh transaction confirmations and with this optimistic merch mining we can do the op standard two second block times and maintain those um if you do merch mining uh the traditional way like your lower bound is

around 15 seconds uh below that uh there's there's limitations um in the Bitcoin mining the way the Bitcoin mining Works uh where where like your uncle rate becomes way too high um merch mining is also our first step because it doesn't require any Bitcoin Forks or updates so it works right now uh which you know with VM as I mentioned a couple of times like is not ready for for uh for production yet and uh ZK uh up code or ZK verification of Bitcoin also isn't possible right now and will require Fork um you also get like a nice little property of like consistent block fees so uh for miners because they will mine actually multiple

Bo blocks at the same time um you get like evening and out payouts which addresses some of the concerns around me um and as a general comment uh we are reusing the proof of work from Bitcoin so miners Bitcoin miners use exactly the same proof of work to mine Bob uh as they mine Bitcoin so you don't have any extra electricity um that needs to be spent uh when when merch mining Bob so you don't you know miners don't the energy impact basically uh doesn't change from just mining Bitcoin to mining Bitcoin and Bob together um cool so uh let's let's dive a little bit into this optimine and how

it works under the head um with optimine which is the our optimistic M mining protocol we we're separating block production and block finalization so in step one uh you have the sequencer uh process all the transactions and and and generating blocks every two seconds uh that is optimistic so that there is no change to that compared to like traditional traditional op stack uh so that functions exactly the same way now where we do a little tricker and changes is on the Block finalization so in uh without optimine um you basically uh

block finalizes once uh they are the the proposer uh in the op stack submits the L2 block hatches to ethereum where it settles and then you have this challenge period So within seven days you can challenge FR proves Liv you can challenge the the L2 blocks essentially and say like oh I think this block should have been different uh and then you submit this fraud proof and you start this onchain game uh of of proving fraud and uh what we basically now do with optimine is uh Bitcoin miners run a full note of Bob very similar to how Challengers would would run full nodes

of Bob uh in in the plain op stack case and and do FR proofs but in our case with optimine uh the miners actually include a hash um that refers to a whole bunch of Bob blocks into into a Bitcoin block into the into very specific part in the Bitcoin block and they put this together they merge mine and merch mine basically means uh they are trying to find a Bitcoin block that matches the difficulty of the Bob chain and they submit that to ethereum and the smart contract on ethereum verifies that

the proof of work is actually correct and then we could we would consider uh B to be finalized if we have enough proof accumulated um that gives us two things so first of all um you get Bitcoin security by uh merch Mining and it's um the amount of Bitcoin security you're getting is based on how many miners would opt in to merch mining Bob uh so the more miners you get like the more Bitcoin security you get um if you get 100% of Bitcoin miners then you get exactly the same proof of work security as Bitcoin itself uh if you get 50% of the miners you get 50% of the per work security from

Bitcoin uh as for ethereum security um you since a smart contract actually for the Bop chain verification livs on ethereum itself and not on Bop uh you get like the full ethereum consensus security because um it's like all the ethereum validators do validate uh the correct execution of that smart contract gotcha okay so a couple things here um first why uh is is there always going to be miners from Bitcoin to verify Bob blocks without any uh you know like without problem and if they're or or like what would be a situation

where miners may not be as incentivized to mine blocks from Bob than they typically would a different type of transaction or a different uh you know or or if there's a lot of transactions happening like yeah like I I guess is there a situation where they would not mind those blocks or verify those blocks uh yeah I would imagine that if there's basically no fees that Bitcoin miners could earn they wouldn't be as inclined to merch Min it um but in that case like uh I'll come to that actually uh in a bit but uh if if Bitcoin miners would like if there was no minor mining on Bob like the rollup

will basically halt like because at that point we wouldn't have any Bitcoin security anymore so with optimine like it can be configured right you can uh you can just do it informational like okay right now we don't have Theon security but still uh continue to Bo blocks optimistically uh but the way that we think about the protocols there should be like a timeout so if we don't have a uh Bitcoin miner very like validating the blocks for like uh say I don't know four hours um and I say four hours because that's similar to how ZK roll ups on ethereum uh validate on ethereum um so if we don't have any much mining security for four hours probably

something is wrong uh and then we should hold the roll up and see what's actually happening if that okay yeah it does um and then when the when the uh Bob smart contract on ethereum verifies the the proof of work from the Bitcoin set how long does that process actually take oh that's that's one block yeah yeah it's just one block it's just one transaction basically so uh you what you do is um I I'll come to that actually in a in a bit but um you the Bitcoin miner uh submits basically a Bitcoin block

that has a reference to Bob uh in the coinbase transaction and that submits that and it gets checked in a smart contract so that's like a single transaction basically on ethereum and that does two things like it first verifies that the proof of work is correct and valid and then um it submits it um for the minor and then the minor would actually then have a claim on the fees that are generated in the B net workk Co and I I think the last one is how much how much more decentralized ation do you think using uh Bitcoin security and using uh miners versus uh like proof of stake in ethereum's um validator Network provides like I

understand that they you kind of get the best of both worlds but I'm curious like do you think that Bitcoin is more sufficiently decentralized in terms of like verifying transactions than ethereum interesting question um I think it's just a very different approach to validating that things are correct uh right so um I think ethereum is pretty set on you know Pro of stake Works quite well there uh and it's it's well it's a proof of stake validation on the ethereum side uh which and is a game of kind of capital uh there's a lot of stuff going on on the ethereum side with mbv and specialized block Builders uh

versus Bitcoin is quite different because it it uses is a different uh mechanism with with proof of work essentially um so it's a lot more you know Miners and Bitcoin usually run large Farms of of Hardware actually in data centers and searching for cheap electricity um there kind of like a different crowd uh of of organizations and people uh and I think by bringing them basically together you you get a a more decentralized set uh of validators for your rollup because you don't only have the proof of stake or only have the proof of work crowd but you actually bring them both together uh

to both validate um the Bob roll up yep yep makes sense cool all right um yeah I think we can we can go forward here one more question Dom on on merge mining um for my understanding there's a couple rollups or protocols that that already rely on merge mining I'm I'm curious if like a couple things like I If you if you are aware of like what the current adoption rate is and how many miners are are merge mining those those other rollups and I'm curious is there like an upper limit as to the mining power that uh we can merge mine like I

if I choose to merge mine to one roll can I can I do so with a 100 Bitcoin rollups is that possible or do I have to sort of split up my resources if I am going to merge mine multiple rollups yeah that's a great question let me actually um answer this the second part of the question uh and then get back to the F like let me answer the can I merge my multiple roll up side chains at the same time and then how does adoption look like and let me quickly jump uh forward to this slide um sorry that's a a t bit technical so uh but to answer that question like it will make a

lot more sense so on the left hand side you have like a merched m Bitcoin block and merched Mining basically means uh you as a Bitcoin minor what you do is you have a Bitcoin block and the Bitcoin block is like uh consists of like a block header which the version previous blog reference the Merkle rout the current time bits nons and then you have the transactions in there which is a very special one the first one is the coinbase transaction which gives you the Bitcoin block rewards and then you have more transactions in there uh now to um to go t bit deeper into details for for actual mining so mining follows a p process and that means that if you

change uh so what bitco miners basically do right they uh if if all the transactions would stay the same all the time they would just change the nons in the in the block header um until they find a a Bitcoin block that has enough leading zeros to meet the Bitcoin difficulty that that's what mining is right you you try to find a a block hash that has enough leadings zos for Bitcoin difficulty uh now because it's Ault process so your probability to find a block uh doesn't change if you if change out things inside of the block so that's that's why basically Bitcoin miners all the time they they swap out like what I wrote here with transaction one two and

so on so on like they change out these transactions which are actually included in a block um by which transaction basic pay pay them the most fees um in return that also means if we change things in the coinbase transaction then you know the the your probability to find the next Bitcoin block doesn't doesn't really change um so let's have a quick look inside of that coinbase transaction so coinbase transaction is the very first transaction inside of a Bitcoin block and it has exactly one input which is the block subsidy so uh that's the the minor reward so when we in April talked a lot about the having the having was exactly um the the block subsidy the minor rewards were slashed by 50% um and

now we have the outputs of this com transaction so utxo based model one transaction is inputs and outputs and coinbase transaction is exactly one input and then we can have multiple outputs on the coinbase transaction and the first output look if it's a standard uh non-merged mine uh Bitcoin transaction typically the first there's often only a single output which basically transfers the block subsidy to the Bitcoin miners account wherever they want to have that um or well account is not necessarily the right word in a Bitcoin context but like the the address uh of the Bitcoin miner but now we can actually add uh a whole bunch of extra

outputs to that coinbase transaction and that's what happens in merch mining so you you would typically uh add a op return output op return in Bitcoin is a way to include uh small arbitrary data um um it's usually most miners only accepted to 32 bytes uh you can in theory add a lot more bites there uh so for example Marathon runs this kind of service where you can uh mine uh transactions with larger op returns but anyway um don't want to get too sidetracked um and you can so to answer your in a very roundabout way to answer your first question uh can we merge mine multiple chains at the same time yes you

basically only need to add uh more outputs uh in the coinbase transaction so you would have one return to merch mine on rootstock One op return to merch mine CIS coin uh One op return to merch mine uh say namecoin and then one more output to merch M Bob and those would all be valid now the last step we need to do uh for Merch mining is you you know miners do this Mining and they're trying to find a Bitcoin block but um uh all the Bitcoin uh like the rollups and the side chains they have lower difficulty than Bitcoin itself so let's say um just for sake of argument you need like 10

leading zeros to find a Bitcoin block uh maybe you need only seven leading zeros to find a valid block for rootstock and for Bob and they have the same difficulty uh then uh now you have a you found a block that has seven leading zeros but it doesn't have 10 now what you do as a minor is you take that Bitcoin block with seven leading Z and you submit it to rootstock and you submit it to Bob and then you can claim um basically the the transaction fees uh on rootstock and on Bob and it it can be like the verification basically then on the other side is um is this uh you

verify um that the merch M bitom Block meets a b rollup difficulty or rootar difficulty um and rejected if it doesn't have uh enough difficulty leading zeros and you verify that the oper turn output uh so the one that we put in the in the comass transaction here that this actually refers to Bob block bundles uh in our case or or for Roar it's that they have slightly different checks um so you can merge mine multiple rollups at the same time and there's no extra cost for mining basically because you're reusing the same proof of work uh for all of those rollups at the same time or

all of the side chains at the same time got it okay and um so let's say let's say I am running a minor and I include the merg Mind outputs for some various Bitcoin rollups uh and let's say Andy's running a minor but he doesn't include the outputs for those pull ups and let's say you know we're competing to try to find like the the Bitcoin block right for the next hash uh if Andy finds the block and he doesn't include the outputs does that mean that uh that for that Bitcoin block

there's no verification of the Bob rollup uh yeah so um yes um the this is one of the super common kind of slight that that makes merch money a little bit hard to understand um so actually having the opper return on bitcoin itself uh isn't necessary for merched mining to be valid um what so in in the case where where Robbie you're a minor and Andy You're A minor and Robbie only you do the merch Mining and and he doesn't do it uh Robbie what you're

basically doing if you find a block with these seven leading zeros uh then you submit it directly to rootstock and Bob and it these Bitcoin blocks will only ever be stored on rootstock and Bob uh and they will never be part of the Bitcoin blockchain because they only have like seven leading zeros and the only case where uh like a op return output of like merch mind side chains or rollups actually end up Bitcoin is if by chance you're finding a block um a Bitcoin valid Bitcoin block uh with say the 10 leaving zeros um and then you would put that onto Bitcoin but uh you would you would always check uh the

merch mining security from the perspective of of the side chain or the rollup uh to which these Bitcoin blocks are submitted and and so yeah I think that makes sense but but so does that mean that the difficulty of mining a block on Bob does not equal the the difficulty of mining a block on bitcoin yeah that's correct that's correct and it always it always should be lower than the Bitcoin difficulty um because uh uh well unless unless you would have a lot more fees on Bob than you would have on bitcoin um like miners would never spend like let's say they find a Bitcoin block right uh

but it's not yet a valid block in Bob uh so they need to continue mining like that that doesn't make sense unless they would get a lot more money for for mining on bitcoin so your difficulty on on any side chain and rollup uh would be upper Bound by the Bitcoin difficulty do you think do you think it's possible that uh like after the block subsidy is depleted and there's only transaction fees do you think it's possible that a Bitcoin rollup like Bob generates more transaction fees than the Bitcoin l one uh it's possible yeah I mean it's possible uh there's there's a lot of unstable regions when the block subsidy basically runs out um where uh yeah so

Bitcoin miners might actually uh you know get more fees on on rollups or side chains um and Bitcoin miners might actually stop or hold their miners altogether if they don't give get enough fees on bitcoin so it's yeah it's it's possible okay but okay this is this uh by generating fees on Bob it could persuade miners to keep their miners running because not only would they get the Bitcoin L1 transaction fees they would also get fees that originate from B yeah exactly so uh it's it's a way to kind of help miners essentially keep

keep Bitcoin mining running and uh using merch mining is a is kind of a decent way to uh make sure that Bitcoin miners are incentivized to keep on you know running your mining Hardware even if it wasn't directly profitable uh to mine on on bitcoin directly but they would get the additional fees from Bob or other s rollups that that would use mer mining um cool and to kind of go back to your first initial question around um minor adoption uh what we seen so far uh I think root has been the one that had the most minor adoption with about 80% of the mining power I think uh now I think it's back down to uh I I'm

probably going to not tell exactly the right number right now but I think it was around 50% or so um so it's like merch mining is essentially op in and uh the more miners you have the more Bitcoin security you get um to go quickly maybe back on one of the first points that I made uh with like this consistent uh block fees so uh there's one like major difference to our optimistic merch mining protocol to kind of traditional merched Mining and that's um that's on the slide here so uh on most or traditionally when merch mining

was conceived like um the task of a uh of a Bitcoin miner was actually to uh also assemble the the blocks of the side chain or the rollup but in our case we are using the the sequencer for this so um we have this you know the thing that I mentioned here we have the split between block production and block finalization and we only get on the Block finalization we do much mining but the block production actually doesn't use any proof of work um in um in so in our case like in the lower part of this of this little graph here uh we have the in Blue uh from one two 3 and minus one and those are the

Bob blocks and then up there uh there's in Orange there's the Bitcoin blocks and then what we include is a block bundle and a block bundle is a hash over multiple uh over multiple block uh B blocks so uh let's say block 1 2 3 n minus one and N uh those all have their own hashes and then we put these hashes together and and derive a block bundle hash um and what that gives us uh let's say a Bitcoin miner wouldn't verify like a single block V Merchant mining but they would uh verify say a 100 or a thousand blocks uh all at once um and and the key interesting difference there is if if miners are

allowed to produce blocks as well then and um it's a big worry if you don't have enough miners in the system like if you don't have enough uh mining power then um there've been cases where kind of uh other miners attack chains so uh if uh in in our case basically all all the minor can do in terms of attacking is they can just stop merch mining that chain uh which would then halt finalization but as long as we have some minor uh validating it uh we're okay um but if you also do block production via merch mining uh miners can uh create Forks of the side chain or the rollup

and that means if you have I don't know maybe just like 1% of the Bitcoin mining power then a miner with like 2% uh can come in and create like competing forks for your chain uh which then need to be resolved and and that's has created provid some issues for for Merch mining project uh in the past um and yeah that's kind of a issue that's that's solved through optimistic merch mining by exactly doing the split of optimistic block production versus miners only on the verification part and um this is where we basically come in like in terms of the consensus role so we in terms of liveness

failures uh we require merch M BLS every X minutes we still need to figure out exactly the parameter uh but otherwise we would halt the chain and then uh we would resume once merch M blocks are submitted again uh but really all miners can do is like Hal this uh thing and for safety um there's no additional safety concerns introduced uh via miners in that sense uh so you could still fraud proof things uh but in our case you would if you have sufficient proof of work being submitted uh or well you have minus basically valid in the rollup so if they disagree with what the sequencer uh was putting on chain on ethereum they would trigger

this the FR proofs okay so um in a in a uh in a pref fraud proof environment would would disagreeing with um with uh the miners um would this would this cause a Halk like is there a way where where somebody can understand this consens model and just purposefully disagree and just cause halts yeah so um again this is depends a little bit how we initially I think we would uh it would be just flaged like hey miners don't agree with this we probably need to resolve the situation because um in uh uh it might just be like a software

bug uh but in the in the like proper final production version of this uh it will actually halt uh even pre FR proof like um but we kind of the way we want to actually implemented is that we do this together with fraud proofs because we want to actually integrate that into the kind of fraud proofing logic uh into the op sta but essentially you would got it cool all right um yeah so um optim like just to summarize it it's it's a kind of new uh optimistic merching construction and it it allows

you to do some cool and interesting things to get in Bitcoin security basically right now without doing any hard forks and we're currently working on the implementation so there's a few um there's a few implementation details that I glossed over a bit and also some things that we're still figuring out in terms of like the the mning mining timeouts uh and then the actual implementation issue of getting this properly implemented into the pack uh so you know if there's uh there's anyone who's super interested in that topic uh we'll we'll share kind of we have the protocol on our website but we will share some more of the implementation details so if people want to look at it um that's really cool for us and we actually encourage that um cool um do you guys have any

more questions on optimistic merch mining otherwise I would nope I think we're good there awesome cool uh right then uh quick recap so basically um we're phase one now phase two is a merch mining that I talk quite a lot about right now that's what's currently been implemented the research that we're doing um at the moment is towards SP PM um but given that I don't want to talk for for another hour or so I'm I'm skip skipping the bit the M part right now but it will the step one for us is like get a bridge working with BBM so we can

optimistically um you know bridge and do FR brof uh so that you can have I think one very important property uh of l2s is that you can withdraw always to the L1 and right now that's possible to ethereum but it's not really possible on bitcoin and then and bit VM based Bitcoin bricks would allow us that and uh that's the you know towards the end of the year uh next year here uh is one we think that can launch into production and then eventually maybe uh ZK if Bitcoin actually adds the up codes for that and uh yeah in terms of I mean just to give the quick Glimpse on the bit VM um is basically we would still have

merch mining uh uh we will still have like uh kind of sequencers and they would run this bit VM stuff uh the merch mining is actually super useful um in bit VM comp text um I think people are sometimes surprised by that um but uh if if you do bit BM you want to have a like client on bitcoin um that that can verify the other chain or rollup and merched Mining gives us a very good notion about finality so if we have uh enough proof of work accumulated uh like a certain percentage uh like a certain difficulty we could say that the

that Bob is finalized under the perception of Bitcoin mining and it helps quite a bit with light clients um because for example uh Building light climbs for eum right now is quite difficult uh with the um with the ethereum kind of signature aggregation and to reason about finaly and that concept is pretty hard uh and uh merch mining actually helps us with that uh because we can use that as a as a form finality in BVM and uh yeah so if all ever goes right then nothing's really needed on the bit VM side you just you know have the BVM part uh and if something goes wrong uh you have

disagreements then uh you would basically do FR proofs inside of bit VM on bitcoin and then you would relay the outcome of that FR proof back to ethereum so that you keep both um Bob is it like a hybrid roll up that you would s synchronize the prop moving State between Bitcoin and ethereum um and uh this is this is a very open research problem this is there's a lot of details that need to be figured out uh once we start working on the implementation um in terms of like when are we actually allowed to settle um on ethereum uh depending on bitcoin's State and um this will be this will be very

interesting project um and yeah uh that's the kind of the bitm Outlook and yeah so we're basically doing ethereum L1 uh it's already in h sorry we're doing ethereum L2 it's already in L2 obviously with a lot of limitations right now um as op stack continues to evolve uh we want to adopt the fraud proof but right now they're not alive and Bitcoin L2 uh that's the vision um and once we can actually verify the roll up on bitcoin uh we'll make that a reality but the way that we see ourselves is uh we want to essentially be the platform in between Bitcoin and ethereum and if you have

assets from ethereum they should have ethereum security if you bring in assets from Bitcoin they should have Bitcoin security and you should be able to use both on Bob and yeah we're on minute uh we launch May 1st so uh you can go check it out there's a bunch of apps already deployed um yeah and let me know any more questions yeah D I mean I think that's about it from our end it was an awesome presentation I appreciate you Fielding our questions and um showing us of the technicals of how Bob Works under the hood cool um you're welcome and you know if you guys want to follow up or anybody wants to reach out like uh I'm available on on telegram Twitter just kind of text

me always happy to talk about research and Tech sweet and yeah as far as our community getting involved they can scan this QR code and find you guys on Twitter and Discord Etc awwesome all right sir thanks a lot