Becoming an L1 does not decentralize Starknet. It drops Ethereum’s security budget and asks the current STRK set to replace it.
Staking is still phase 2 of 4. Validators stake 20k STRK and attest with a full node. They do not produce blocks, finalize alone, or hold data availability. Sequencing and proving remain narrow roles. Reorg and censorship resistance still come from Ethereum.
The crypto agility (STARKs + account abstraction) is real and useful. The open question is whether the STRK set can carry the full security load before that floor is removed.
Questions for the team
- Users: once finality is no longer Ethereum, who can stop, reorder, or censor a payment? Do Cairo contracts keep their address and receipts?
- Investors: what economic security (STRK staked, cost to reorg or censor) do you target at launch versus Ethereum? Is 2027 a commitment or a contingency? Who holds bridge and liquidity risk during the switch?
- Validators: what exact new duties (block production, finality, DA, proving) fall on them? What hardware, bandwidth, uptime, and slashing do you model? Does the 20k attester role become consensus, and at what cost?
- Your PQ roadmap still lists bridges and data availability as dependent on Ethereum. Which dependencies disappear on day one of an L1, and which remain?
- ECDSA.fail improved a circuit primitive. It is not a fault-tolerant attack. What observable trigger (not a date) would make the L1 move necessary rather than optional?
- For current validators and delegators: is there a published migration spec, or is this still under study?
Until those answers are public, this is a transfer of dependency, not a gain in decentralization.