les rollups bitcoin n'ont pas de sorties de secours
je ne fais pas de correspondance directe entre ces choses, car CR par rapport à la couche DA, ethereum gagne
janusz
janusz10 janv. 2025
Can upcoming bitcoin rollups meet @l2beat's Stage 2 requirements? The tl;dr is almost. But to meet all requirements, you'd need something like OP_CAT. Bitcoin rollups can have 1-of-N trust assumptions related to their bridge. But there’s some nuance to this. There’s two things you trust: - Operators: You trust 1-of-N operators to process withdrawals and not freeze the bridge - Verifiers: You trust 1-of-N, N being anyone, to secure the bridge In the proposed implementations of rollups, withdrawals are processed by permissioned operators. Operators of the bridge front liquidity to users to satisfy withdrawal requests and then reimburse themselves from the bridge. While not incredibly efficient, this could work. Now, stages. There are a number of requirements for a rollup to be “Stage 2” per @l2beat’s framework. Let’s see if a bitcoin rollup can meet them: Bitcoin rollups can meet all of L2Beat’s stage 0 requirements. You can use bitcoin for DA, post state roots there, and people can reconstruct that state of the rollup based on this data. Stage 1 is where it starts to get tricky. Bitcoin rollups can meet all requirements… except: “Can the users exit without the operator’s coordination?” The system should be designed so that user withdrawals cannot be blocked by the rollup operators. The rollup must implement mechanisms that allow users to exit independently, ensuring they can always access and control their assets. Users can bypass the rollup operators theoretically (we haven’t seen it yet, so I’m not entirely sure how it’d work). But, they can’t bypass bridge operators if all of the bridge operators are offline. Therefore, the condition that users “can always access and control their assets” is not met. This trend continues in Stage 2. As we mentioned earlier, these rollups can have permissionless fraud proofs. Anyone can ensure the integrity of the bridge. And, they can likely have a mechanism to bypass rollup operators in the event that an unwanted upgrade is implemented. “…This ample time frame allows users to react to significant changes in the system that they may not agree with and withdraw their assets if needed.” But, requirements related to permissionless withdrawals cannot be met because users trust at least 1 operator to fulfill withdrawal requests. So, to clear things up: - Bitcoin rollups are possible - They can have permissionless fraud proofs to secure the bridge with a 1-of-N trust assumption - Instead of relying on a 2/3s federation to process withdrawals, you can rely on just 1-of-N operators - This is not the same liveness (exit) assumption as optimistic rollups in ethereum. The security assumption is similar A bitcoin rollup, with a Strata or Clementine-style bridge, can meet all of L2Beat’s requirements except those that require permissionless exit. While users may be able to bypass rollup operators, they must trust that 1-of-N bridge operators will fulfill withdrawal requests. But, if you have something like OP_CAT, you can construct state-carrying spend conditions and onchain verifiers. This enables a bridge program to unlock funds if certain conditions are met. Like verifying a SNARK proof attesting to the updated state of the rollup, or a fraud proof challenge period passing. So if additional opcodes are added, then bitcoin rollups could meet requirements related to permissionless exit and be a Stage 2 rollup. Hope this clears up some confusion!
même si un utilisateur pouvait se proposer lui-même une sortie, il fait confiance à un opérateur de pont pour l'honorer.
Je ne dis pas "pas d'échappatoire" pour dénigrer les rollups Bitcoin, je le dis pour que nous sachions, immédiatement, que ce n'est pas une comparaison 1:1.
Les rollups Bitcoin permettent aux opérateurs de pont de fournir de la liquidité sur les marchés émergents du Bitcoin où ils ne font confiance qu'à eux-mêmes pour être honnêtes. Cela s'étend encore aux utilisateurs qui n'ont besoin de faire confiance qu'à un seul opérateur plutôt qu'à une majorité.
est-ce que cela fonctionne complètement comme ça en pratique ? non, il y a d'autres considérations (c'est-à-dire des conseils de sécurité). vaut-il la peine de construire et de voir si ce principe offre un avantage concurrentiel ? bien sûr.
1,38K