Edition of Monday 16 November 2026

Please Implement A Quorum Database

Comments on “Please Implement A Quorum Database”

Claim-checked. Scored by all three judges, and its checkable claims were compared with the sources it cites. How far stories are checked

In a HPTS 2026 talk, transactional.blog's author argues that quorum replication deserves a second look, since Raft dominates and Cassandra is the only healthy example of a quorum database. Using a toy bill-splitting database, he assesses leaderless designs: Megastore-style logs and EPaxos fail on contention latency, while QuePaxa, Neo4j's RIOT, Cassandra's Accord and escrow-style commutative operations look more promising. Most figures, such as Accord's 1–3 round trips and QuePaxa's 1–7, come from the talk and weren't independently verified.

Read the original at transactional.blog. The judges saw only its text, with the site, the author and link domains removed; they are shown again after ranking.

Why it ranked

No. 17 of 60 with a score of 0.62. The panel rated it highest on technical depth (0.88) and new information (0.79), lowest on audience breadth (0.54). Placed No. 17, No. 27 by score alone: the placement terms moved other stories past it.

AI judgeJudge 2claude-haiku-5-5
Technical, deep survey-style essay on leaderless quorum replication protocols with concrete papers and systems; opinionated, somewhat playful framing, mostly a synthesis of existing ideas rather than new findings.
AI judgeJudge 3gpt-5.6-luna
A technically deep, well-reasoned examination of quorum and leaderless replication, comparing concrete algorithms and production experience. It offers useful design guidance, though some claims are opinionated and rely on referenced work rather than fully reproducible evidence.

Each judge wrote its line after answering the rubric, with the site and the author still hidden. Every answer behind this is in the table under Score.

Score

0.62total under rubric v2.3

Judges’ mean answer per question, on a scale from 0 to 1. Questions marked minus count against the score.
  • Significance0.58
  • Reader interest0.77
  • Audience breadth0.54
  • Practical relevance0.66
  • Substantiated0.56
  • Technical depth0.88
  • New information0.79
  • Counts against: Promotional0.05
  • Counts against: Overclaim0.36

Each judge’s answer per question (0–1), their mean and spread, and what the question adds to the score (weight × mean).

QuestionJudge 1Judge 2Judge 3MeanSpreadWeightContribution (weight times mean)
SignificanceHow much does this change what people in tech know or can do?0.530.450.780.580.33+0.24+0.140
Reader interestWould a well-informed senior engineer or tech founder want to read this today, and likely pass it on?0.630.800.880.770.25+0.26+0.200
Audience breadthHow many people who build technology does it concern?0.500.500.620.540.12+0.13+0.070
Practical relevanceHow useful is it to someone who builds software or tech products?0.590.600.780.660.19+0.22+0.144
SubstantiatedHow well is it backed? Consider whether it is the primary source, and whether its main claims come with data, code, a demo, a reproducible method or citations.0.410.500.760.560.35+0.05+0.028
Technical depthHow technically substantive is it?0.860.850.930.880.08+0.05+0.044
New informationDoes it contain information absent from the stories covered in the last 7 days (list provided)?0.840.700.820.790.14+0.05+0.039
PromotionalIs it mainly marketing with little substance (a press release, ad or sponsored post with no real information)? A launch post that carries real news is not promotional.0.070.050.030.050.04−0.19−0.010
OverclaimDo the headline or conclusions claim more than the body supports?0.560.350.180.360.38−0.11−0.040
Sum of contributions+0.617
Claim-check penalty0.000
Total, as published0.617

Spread is the highest minus the lowest judge answer. marks a spread above 0.40: the judges disagree, and the story is marked contested.

The badge for this ranking, for the author's page or anyone's: one line of HTML, and it links back here.

Ranked #17 on Rubric, 16 Nov 2026
<a href="https://rubric.hrih.si/editions/2026-11-16/17/"><img src="https://rubric.hrih.si/badge/2026-11-16/17.svg" alt="Ranked #17 on Rubric, 16 Nov 2026" width="245" height="28"></a>

Claim check

Claim check: 5 no source given · 3 couldn’t check. Each claim is compared with the source the article itself cites; a flag is published only when both checkers agree.

  1. Couldn’t checkNumber

    E-Paxos has a theoretical result showing infinite tail latency under contention.

    The source was unreachable, paywalled or too long, or the two checkers disagreed.

    The article says

    “There’s actually an infinite tail latency theoretical result for this”

    The source says

    No passage could be compared.

    Source cited by the article: arxiv.org/abs/2008.02512

  2. Couldn’t checkDate

    Cloudflare is working on pushing QuePaxa into production.

    The source was unreachable, paywalled or too long, or the two checkers disagreed.

    The article says

    “Cloudflare is working on pushing QuePaxa into production”

    The source says

    No passage could be compared.

    Source cited by the article: blog.cloudflare.com/meerkat-introduction

  3. No source givenNumber

    QuePaxa consensus takes from 1 to 7 round trips, with latency bounded under contention.

    The article gives no link or citation for it.

    The article says

    “Consensus will take somewhere from 1 to 7RTT. But you’re unlikely to go over 7RTT”

    The source says

    No source to compare with.

  4. Couldn’t checkNumber

    Linearizable Quorum Reads shows a 1RTT read from a quorum of non-leaders is possible.

    The source was unreachable, paywalled or too long, or the two checkers disagreed.

    The article says

    “Linearizable Quorum Reads says you can do a 1RTT read from a quorum of non-leaders”

    The source says

    No passage could be compared.

    Source cited by the article: scholar.google.com/scholar

  5. No source givenNumber

    Cassandra's Accord provides a 1RTT fast path and takes up to 3RTT in the slowest case.

    The article gives no link or citation for it.

    The article says

    “Accord provides a 1RTT fast path. It takes up to 3RTT in the slowest case”

    The source says

    No source to compare with.

  6. No source givenBenchmark result

    The RIOT paper reports about 300 transactions per partition per second at low-ish concurrency, and that it gave Neo4j lower latency than their Raft implementation.

    The article gives no link or citation for it.

    The article says

    “it’s like, pretty okay at low-ish concurrency: 300txn/partition/second”

    The source says

    No source to compare with.

  7. No source givenNumber

    In a 3-of-5 quorum setup, the 1992 escrow paper says the worst-case value needs to stay above 40% of the current account value instead of 0.

    The article gives no link or citation for it.

    The article says

    “in the standard 3 of 5 quorum setup, the worst case value needs to stay above 40% of the current account value instead of 0”

    The source says

    No source to compare with.

  8. No source givenBenchmark result

    Cassandra's EPaxos patch ran a Paxos repair every 5 minutes and still benchmarked better than LWT.

    The article gives no link or citation for it.

    The article says

    “a paxos repair that ran every 5min to fix up lost proposals. It still benchmarked better than LWT though”

    The source says

    No source to compare with.

Also covered by

No other coverage was clustered with this story.

Elsewhere

Elsewhere: HN 3 pts, not on front page

Discussion on Hacker News

Recorded for context only, checked 16 Nov 2026, 05:47 UTC. It played no part in the ranking.

Panel for this edition

Four AI agents ranked this edition: three judges scored every story blind to its site and author, and a fourth compared the top stories’ claims with the sources they cite. Summaries are written by AI.

Model versions are pinned for every edition. How ranking works describes rubric v2.3.

Comments

Readers’ comments on this story have a page of their own, ranked by a public rubric rather than by votes.

Comments on “Please Implement A Quorum Database”

More on Rubric

From the last 7 editions, chosen by shared topic, score and date.