> the “best” modular architecture includes central sequencers, horizontally scaled DA, fast censorship resistance guarantee, instantaneous economically secured validity (and eventual zk verification) that give instant confirmation.
Take the above and add interoperability (shared settlement / sequencing), application-specific sequencing, streamlined devtools (for the prior two items), zk, validating or consensus bridges to major L1s, (and probably privacy features like onchain confidential compute), we will get the pretty much the endgame ecosystem.
Either Ethereum and the modular ecosystem become this or there will be disruptors.
(And no, other L1 ecosystems like Solana cannot compete here, as none of them can horizontally scale without compromising on security yet.)
I think
@jayendra_jog point is: take whatever modular system you maybe running, vertically integrate it and then optimize and call it a new monolithic system - this latter monolithic system canNOT be worse than the modular system.
This argument is theoretically accurate.
My reactions
First this is easier said than done - it is extremely difficult to build each one of these systems and humans / teams thrive in specialization, especially when each of those components have incentives.
Secondly the “best” modular architecture includes central sequencers, horizontally scaled DA, fast censorship resistance guarantee, instantaneous economically secured validity (and eventual zk verification) that give instant confirmation. NO existing monolithic architecture follows this spec.
Based sequencing is
@drakefjustin idea to bring this architecture as close to L1 as possible providing trust guarantees from L1 validators without changing the slick UX of L2s while bringing value accrual and MEV back to Ethereum L1.
So Ethereum is the closest monolithic architecture to this design.