Why
The site's own claim in Tech #74 is that a faster Ethereum is a delivery chain: spec, client, RPC, wallet, user, and a change is only real at the last link. This update is a rare public view of the chain's first three links moving at once. The Quick Slots section shows how an EIP actually gets into a fork: not by argument alone but by an ecosystem article, six client teams grading it S to "do not include", a specification sprint, a prototype, and a two-week deadline to raise conviction. The AA section shows the step after two standards agree to coexist: someone has to trace a real account through both and find the gaps that only appear then. The FCR section shows research turning into a product question (who tells the RPC provider to expose it). Reading it, jay gets the vocabulary and the calendar for the next month of Ethereum core development, which is the material the Eng interview items (#4, #16) and the Rabbit and Wallet services depend on.
How it works
Quick Slots: survey, tier lists, and a two-week clock
- The ecosystem article on EIP-8198 (by @binji_x and @barnabemonnot) carries quotes from 20+ teams, from researchers to apps to protocols migrating onto the L1, and argues for laying the groundwork in the Hegotá fork so slot time can change "as soon as possible", starting with 12 → 10 seconds.
- Client verdicts the same day: Teku S-tier, Prysm and Lighthouse A-tier, Lodestar C-tier, Nimbus and Grandine prefer not to include. Six consensus clients, four in favour to some degree, two against.
- Status: Proposed for Inclusion now; the target is Consider for Inclusion at the next ACDC in two weeks. Barnabé is completing the specification with review from Jihoon Song (EF Protocol CL specs); Terence Tsao (Prysm) continues a prototype.
- The site's earlier reading (#54) holds: the change is making slot time a parameter, and the 12 → 10 step is the proof that the parameter moves.
Blobs: from capacity to expectations
Ansgar Dietrichs's framing: L2s may not be limited by current blob capacity, but by unclear expectations about future capacity. The response is a survey of L2s on actual and expected blob demand (Caspar Schwarz-Schilling, Derek, Francesco D'Amato; Derek doing outreach), feeding a report on where blob targets should go and when, alongside Francesco's blob-capacity report and a proposal to raise the PeerDAS custody threshold. The lesson is the same one Verex meets on settlement: the constraint that shapes behaviour is often the announced roadmap, not the current limit.
Native AA: two EIPs, traced account by account
After weeks aligning EIP-8130 and EIP-8141 (the coexistence the site recorded in #79), the work moved to what an account must do to get from here to there. Ansgar found gaps between the two proposals that appear only when a real account is traced through them; Mislav mapped the flows: existing accounts migrating into Frames, delegation via SETCODEFROM, and invalidating a compromised or retired ECDSA key. The output is a personas-and-user-stories draft that wallets can read against their own architecture. A second thread asks whether cosigners belong inside a new standard for accounts portable across both EIPs or beside it. This is exactly the capability matrix Tech #72 asked for, now being written by the people who will ship it.
FCR and finality: a zk proof and three speed boosts
- FCR adoption is gated by RPC providers, so Julian Ma is asking teams that want faster confirmations to say so to their providers, and is writing guidance on how FCR interacts with the "safe" tag and how to roll it out without surprising users. This is #74's RPC link, named.
- zk-provable FCR: while extending FCR to weaker observer-network assumptions, Francesco found a path to proving a block fast-confirmed with succinct zk proofs; modified Lean proofs exist; Derek is exploring the product side.
- Decoupled consensus is being formally verified in Lean; all properties currently verified, repository under cleanup and audit before publication.
- Three speed boosts from a fast-finality sprint (Francesco and Barnabé):
| Boost | Mechanism | Claimed gain |
|---|---|---|
| Check finality every slot | Today finality is checked every 6.4 minutes after all validators vote, though 2/3 may have voted much earlier; with decoupled consensus each slot can test the 2/3 threshold | up to 1/6 of time-to-finality |
| Heavy validators vote first | Consolidated stake crossing 2/3 sooner | another 1/6 or more, depending on consolidation |
| Early justification | Process justification as early as finality, then start the finality round immediately | compounds the two above to more than 50 percent overall |
These are research claims from a sprint, stated as "potential" in the source; none is in a specification yet.
Reading it with the galaxy-brain test
Ethlabs is funded to grow Ethereum and the author co-founded an AA company, so the bags are on the table, and the update is honest about them ("part of our role at Ethlabs is to bring ecosystem needs into these discussions"). What survives the opposite-conclusion test: the tier-list verdicts (they include the two "no" votes), the status of each EIP, the named people and their deliverables, and the finality arithmetic, which anyone can check against the 6.4-minute epoch. What is advocacy: "as soon as possible", the 20+ quotes (selected), and the "more than 50 percent" headline, which compounds three unimplemented ideas.
Where it lands in Jayverse
- Rabbit (#1) and Wallet (#5): the AA migration flows are the spec to read next. Migration into Frames, SETCODEFROM delegation and ECDSA key invalidation are the three flows the Rabbit AA account and the Wallet's simulate-before-sign must handle. When the personas-and-user-stories draft is published, it replaces the site's own #72 matrix as the reference. Cosigner placement decides whether the Wallet's guardian sits in the standard or beside it.
- Devnet (#6): FCR guidance and the "safe" tag. The devnet's RPC is the one place jay controls the RPC link of #74's chain. Expose FCR-style fast confirmation there first, following Julian's guidance when it circulates, and record the confirmation rule in the Auditor's evidence (which rule said "confirmed").
- Auditor: finality arithmetic is a rule with teeth. "Finalized when 2/3 of stake has voted" is checkable; "fast" is not. The Auditor's settlement invariant should name the threshold and the clock, in the spirit of #33's three clocks and #125.
- Knowledge Notes: a fork calendar. Quick Slots to ACDC in two weeks, the L2 blob survey, the AA wallet-support plan, Julian's RPC guidance, the Lean repository publication. Each is a dated follow-up; #54, #72, #74 and #79 get a line pointing here.
- Eng: #4 and #16 have new facts. The tier list, the two-week ACDC clock and the 6.4-minute finality check are the concrete numbers an interviewer wants when Jay says "Quick Slots does not make finality faster, decoupled consensus might".
Verified and unverified
Verified on 2026-09-21 from the article itself (full text via the public tweet API; the X page requires login): the author (Mislav, Ethlabs), date 2026-09-20, every status, name and figure quoted above, including the client tier list, the Proposed/Consider for Inclusion status, the 12 → 10 second target, the three AA flows, and the three finality boosts with their fractions. Not verified independently: that the tier-list rankings are as the author summarised them (the client teams' own posts were not checked), the claim that all decoupled-consensus properties are verified in Lean (repository unpublished), and the "more than 50 percent" finality figure (research estimate, compounding three unimplemented changes). The ecosystem article on EIP-8198 and the Milk Road episode were not read. The reading of the update through the delivery-chain and galaxy-brain lenses is this note's, not the author's.
Sources: Mislav | Ethlabs, X Article, 2026-09-20 · related: Tech #54 (Quick Slots: slot time as a parameter), #72 (two native AA standards, capability matrix), #74 (fast Ethereum is a delivery chain), #79 (EIP-8141 and EIP-8130 coexist), #33 (L2 finality is three clocks), #31 (ERC-8141), #125 (galaxy brain resistance); Eng #4, #16.
Key expressions
| Expression | 뜻 · 쓰이는 자리 |
|---|---|
| ecosystem survey | 생태계 조사(여러 팀의 의견을 모은 글) · EIP 지지 근거를 만들 때. "An ecosystem survey for Quick Slots" |
| tier list | 티어 리스트(S·A·B·C 등급표) · 클라이언트 팀이 EIP 우선순위를 매기는 방식. "client teams gave their own 'tier lists'" |
| Proposed for Inclusion / Consider for Inclusion | 포함 제안 / 포함 검토(포크 포함 절차의 두 단계, PFI → CFI) · ACDC 용어. "moving to 'Consider for Inclusion'" |
| ACDC | All Core Devs Consensus(합의 레이어 코어 개발자 회의) · 격주 회의 이름. "by the next ACDC in two weeks" |
| Hegotá | 헤고타(예정된 이더리움 포크 이름) · 슬롯 시간 변경의 목표 포크. "groundwork in Hegotá" |
| slot time | 슬롯 시간(블록 제안 간격, 현재 12초) · Quick Slots의 대상. "enable slot time changes" |
| blob | 블롭(L2가 L1에 데이터 가용성을 위해 올리는 데이터 덩어리, EIP-4844) · L2 비용의 핵심. "Should we keep scaling blobs?" |
| PeerDAS custody threshold | PeerDAS 커스터디 임계값(각 노드가 보관해야 하는 데이터 샘플의 최소량; Peer Data Availability Sampling) · 블롭 확장의 조건. "increasing the PeerDAS custody threshold" |
| native account abstraction (AA) | 네이티브 계정 추상화(프로토콜 차원에서 스마트 계정을 지원하는 것) · EIP-8130/8141의 주제. "turning native account abstraction debates" |
| migration flow | 마이그레이션 플로우(기존 계정이 새 표준으로 옮겨 가는 단계) · 지갑이 구현하는 단위. "AA gets concrete migration paths" |
| Frames | 프레임(제안된 네이티브 AA 계정 실행 단위) · EIP-8141 계열 용어. "existing accounts migrating into Frames" |
| SETCODEFROM | SETCODEFROM(계정 코드를 다른 주소에서 위임받는 제안된 연산) · 위임 메커니즘. "delegation via SETCODEFROM" |
| invalidate a key | 키를 무효화하다(탈취·은퇴한 서명 키를 쓸 수 없게) · 계정 보안 플로우. "invalidating a compromised or retired ECDSA key" |
| ECDSA | Elliptic Curve Digital Signature Algorithm(이더리움 기본 서명 알고리즘) · 기존 EOA 키. "a compromised or retired ECDSA key" |
| cosigner | 코사이너(거래에 함께 서명하는 두 번째 서명자) · 사전 보안 장치. "whether cosigners belong in a new standard" |
| personas and user stories | 페르소나와 사용자 스토리(대표 사용자와 그 시나리오) · 제품 스펙 형식. "a personas-and-user-stories draft" |
| fast confirmation rule (FCR) | 빠른 확인 규칙(최종성 전에 블록을 안전하다고 판단하는 규칙) · 확인 지연을 줄이는 연구. "Fast confirmation rule (FCR)" |
| the "safe" tag | "safe" 태그(RPC가 정당화 기준으로 표시하는 블록 상태) · FCR과의 상호작용 대상. "the 'safe' tag" |
| succinct zk proof | 간결한 영지식 증명(짧게 검증 가능한 증명) · 확인 사실을 증명하는 도구. "using succinct zk proofs" |
| Lean | Lean(정리 증명기; 형식 검증 언어) · 합의 프로토콜 검증에 사용. "modified Lean proofs in hand" |
| decoupled consensus | 분리된 합의(가용성과 최종성을 분리한 새 합의 설계) · 빠른 최종성의 기반. "the new decoupled consensus protocol" |
| justify / finalize | 정당화 / 최종화(두 라운드 투표의 각 단계) · Casper FFG 용어. "the first to justify a block, the second to finalize it" |
| time-to-finality | 최종성까지의 시간 · 가속의 측정 지표. "reduce time-to-finality by 1/6th" |
| consolidated stake | 통합된 스테이크(한 검증자에 합쳐진 큰 예치금, EIP-7251) · 무거운 검증자 우선 투표의 근거. "heavier validators (with more consolidated stake)" |
| bring needs into the discussion | 필요를 논의에 들고 가다 · 생태계 대변 역할. "bring ecosystem needs into these discussions" |
| raise conviction | 확신을 높이다 · 반대 팀을 설득하는 일. "increase conviction by the next ACDC" |