# Ethereum 2029: Rebuilding the Ship of Theseus in Mid-Flight Source: https://x.com/Eli5defi/status/2039002678060003434 Kiwi News: https://news.kiwistand.com/stories/Ethereum-2029-Rebuilding-the-Ship-of-Theseus-in-Mid-Flight?index=0x69cbee54f67db6486685f9c92e0d2c81d7c63170f878e31fd768270e1814b0f636b21157 Submitted by rvolz.eth on 2026-03-31 UTC, 1 upvote, 0 comments ## Summary Ethereum's "Strawmap" is a publicly developed, open-source plan to rebuild the network's core consensus and execution layers over about 3.5 years without downtime, targeting seven upgrades through 2029. The article says the roadmap aims to cut finality from roughly 15 minutes to 2 seconds and raise L1 capacity to 10,000 TPS, with ZK provers needing a 1,000x speedup that does not yet exist. It also describes a move from BLS signatures to hash-based quantum-resistant schemes, which it calls a central pillar of the 2029 roadmap. ## Article The digital world is no stranger to technical debt, but few systems carry the weight of Ethereum. As of March 2026, the network facilitates billions of dollars in daily value, creating a unique engineering paradox: it must undergo a total reconstruction without a single second of downtime. This challenge is best understood through the Ship of Theseus metaphor. In this ancient Greek thought experiment, every plank of a ship is replaced one by one until every original piece is gone. For Ethereum, this reconstruction is mechanical necessity. However, the "Ship of Theseus" approach here is unique because we are replacing the planks, the consensus and execution logic, while meticulously preserving the cargo. That cargo is the "State": the user balances, contract storage, and transaction history that constitute the network's core value. Ethereum cannot simply "start from scratch" with a new chain; to do so would be to toss the cargo into the ocean. Instead, the "Strawmap" provides the vision for how we swap the engines while the ship is at full sail. The "Strawmap" Defined: A Plan Built for Critique The term Strawmap is a deliberate synthesis of a "strawman" proposal and a technical "roadmap." In high-level engineering, a strawman is a preliminary proposal intended to be imperfect, specifically designed to invite critique and rigorous debate. While this is a starting point rather than a rigid commitment, it represents the first structured, time-bound path in Ethereum's history, targeting seven major upgrades over a 3.5-year horizon. This work involves a global alliance of the world’s top cryptographers and computer scientists. It is entirely open source, free of corporate sales teams or vendor contracts. The upgrades available to a global institution like JPMorgan are identical to those available to a three-person startup in São Paulo. The Open Pipeline of the Future The path to 2029 is defined by seven distinct technical milestones: GLAMSTERDAM | 2026 Today, the same person who proposes a block also builds it. That's like letting a referee bet on the game. This fork splits those roles and forces validators to include transactions they'd otherwise bury. ePBS (EIP-7732): block building and proposing, separated FOCIL (EIP-7805): inclusion lists enforced at fork-choice, MEV curbed Frame Transactions (EIP-8141): post-quantum signature foundation Native account abstraction + 18 EIP repricings HEGOTÁ | 2026-27 Ethereum charges one flat gas fee for everything, like a post office billing the same rate for a letter and a sofa. This fork prices compute, bandwidth, and storage separately. Bots also lose their edge: transactions enter an encrypted waiting room they can't read until it's too late to front-run. Multi-dimensional gas: three resources, three prices Encrypted mempool: threshold crypto kills front-running at the source NTT precompile + SSZ transactions + payload chunking I+ | 2027 L2 rollups currently post data to Ethereum and ask validators to trust their math. This phase makes them prove it, cryptographically, inside Ethereum's own engine. No trust. Quantum-resistant crypto foundations also go in now, before the threat is real. EXECUTE precompile (EIP-8079): L2 state anchored to L1 natively zkVM framework: RISC-V guest for state transition proofs PQ crypto: STARK eval + FRI folding + PQ gas tables Beam sync: stateless node bootstrapping J+ | 2027-28 Running a light node today means trusting full nodes are honest. This phase replaces that trust with ZK proofs. Blobs also become variable-size with built-in error correction, so missing data gets reconstructed even when nodes drop off. STF in zkISA: state verification via ZK proofs, not blind trust Variable blobs + Reed-Solomon reconstruction Block-in-blobs: full PeerDAS infrastructure complete K+ | 2028 Ethereum finality today takes minutes. This phase cuts it to 18 seconds. A million validator votes per slot get processed in parallel. Provers who don't deliver cryptographic block proofs get penalized on-chain. 3SF: ~18s economic finality, 6s slots 1M attestations/slot via parallel BLS Mandatory 3-of-5 proof systems with on-chain penalties- 3-slot finality (3SF): ~18 second economic finality, 6s slots L+ | 2029 Every Ethereum validator signature today can theoretically be forged by a quantum computer. This phase starts replacing all of them with hash-based schemes quantum computers cannot crack. Re-keying every lock before anyone learns to pick them. PQ attestations: BLS rotates to hash-based signatures APS: block production architecture fully separates Custody proofs enforced + jeanVM aggregation begins M+ | 2029+ Ethereum does 15-30 TPS today. This phase targets 10,000. A canonical zkVM proves every block in real time. Gas futures let users reserve block space at a fixed price, ending the fee spike war during congestion. Canonical zkVM: 1 Ggas/sec real-time proving Gas futures: predictable scheduling, MEV-resistant PQ blob commitments replace KZG 2030++ The cleanup phase. Validator randomness gets hardened so no one games block proposal selection. Block construction fully decentralizes. Privacy becomes a native L1 feature, not a third-party tool. And if 51% of validators ever collude, the protocol recovers automatically. VDF randomness: validator-lottery attack surface closed Distributed builders: no single entity dominates block production Shielded transfers: L1-native privacy 51% auto-recovery: last-resort protocol defense TAKEAWAY 1 Finality in Seconds A wire transfer has a gap between "sent" and "settled." Ethereum had the same problem, just 13 minutes long. For bond trades or real estate deeds, that uncertainty window is a dealbreaker. Current target: single-slot finality via Minimmit protocol Slot compression roadmap: 12s → 8 → 6 → 4 → 3 → 2 seconds End state: 15-minute finality compressed to 2 seconds Outcome: Ethereum becomes viable for institutional settlement TAKEAWAY 2: ZK-Proofs Replace Redundant Work Today every node re-executes every transaction to verify it. Like an office where every employee re-checks every colleague's math. Safe. Brutally slow. ZK-proofs replace that with a compact mathematical receipt: "trust me, the work was done correctly, here's the proof." Target: 10,000 TPS on L1 (Gigagas, 300x current capacity) Bottleneck: ZK provers are currently 1,000x too slow Goal: instant proof generation by 2029 TAKEAWAY 3: Sampling the Library, Not Reading It L2s need 10 million TPS (Teragas). That requires data availability without downloading everything. DAS works like checking 20 random pages of a 500-page book. If every node's pages are intact, the full book is statistically guaranteed to exist. Nobody reads all 500 pages. Foundation already laid: PeerDAS shipped in Fusaka Enables: micro-payments at fractions of a cent, supply chain tokens, global-scale throughput TAKEAWAY 4: Quantum-Proofing Before It's Urgent Current validator signatures can be forged by a sufficiently powerful quantum computer. "We'll deal with it later" is not acceptable when the network holds trillions in value. The fix is hash-based signatures. The problem: they're kilobytes instead of bytes, bloating block size, bandwidth, and storage significantly. Transition plan: BLS signatures rotate out, hash-based schemes rotate in Core challenge: handling heavier signatures without crushing performance Timeline: central pillar of the 2029 roadmapTakeaway 5: Native Privacy—The Shielded Layer 1 TAKEAWAY 5: Privacy as Infrastructure, Not a Plugin Every Ethereum transaction today is fully public. Amount, sender, receiver. All visible. Shielded transfers let the network verify a transaction is valid without revealing who sent what to whom. The math checks out. The details stay private. No middleware workarounds. Native at the base layer. Validates: sender has funds, math is correct Hides: identities and amounts Why it matters: privacy as a standard utility, not an expensive add-on TAKEAWAY 6: AI as the Wildcard The 2029 timeline assumes a human-led development model, but AI is the great "acceleration variable." In February 2026, @yq_acc used AI agents to generate ETH2030, an experimental 713,000-line client implementing the entire Strawmap. While the code was not production-ready, Vitalik Buterin noted that the trend is what matters. Vitalik’s strategic insight is that we should not use AI purely for speed; instead, we should allocate half of the benefits to speed and the other half to security, using AI for more rigorous testing and mathematical verification than humans can manage alone. Wrap-Up The risks are real: timelines slip, quantum migration is operationally brutal, ZK provers need a 1,000x speedup that doesn't exist yet, and governance consensus across a decentralized ecosystem is never guaranteed. But the architecture is coherent, the tests pass, and the network is live. 2030 is four years away. The base layer is being built now, in public, by people who read the map and decided to drive.