Arkency + Evil Martians into the voice reference, artifacts into both remaining posts

[pftg]

Arkency + Evil Martians into the voice reference, artifacts into both remaining posts

Paul asked for Arkency alongside thoughtbot and Evil Martians. Fetched and read
two of their current posts, and they exposed gaps the other two did not.

What Arkency does that we were not doing at all:

1. A named human writes it. Szymon Fiedler, Piotr Jurewicz, each with an
   author page. Ours say author: 'JetThoughts Team' - the faceless register we
   avoid everywhere else and then use on the byline.
2. Cards on the table before the argument. Fiedler opens a polemic about
   invariants by disclosing he maintains RailsEventStore and that the post lands
   on the Arkency blog - then deliberately argues in pure ActiveRecord so the
   reasoning survives without his product. We have no equivalent move.
3. Argues with a named peer, respectfully. Subtitled "A polemic with
   Callbacks Are Not Invariants by Brandon Weaver", linked, with a section
   called "Where Weaver is right - and what I'm not saying". We cite aggregate
   HN comment counts. Aggregates cost nothing to cite, which is why they
   persuade nobody.
4. H2s that are moves in an argument, not observations wearing headings.

Density: their knowledge-graph post runs 8 code blocks AND 4 images including a
cost chart; the invariant post runs 8-10 code examples across ~3,500-4,000 words.
First person throughout - "I enjoy", "I hate", "My claim".

Evil Martians adds two more: anchor-link navigation in long posts, and
playfulness that does honesty's work rather than decoration's - their mock-FDA
disclaimer lets them state a strong reliability number and show its limits in
the same breath.

Both written into blog-writer-reference-samples.md as Samples 4 and 5, with
the measured comparison table and two rules promoted OUT of cadence-choice into
apply-to-every-post: an artifact the reader can use, and receipts.

Posts rewritten to that standard:

how-to-audit-content-you-didnt-write - was four checks described in prose;
now four checks with the commands that run them. A Search Console CSV sort, the
case-study-heading grep with its anonymous-subject follow-up, a shell loop that
counts posts over 400 words carrying zero external citations, and a grep for
version-numbered titles. Its H2 said "none of which require you to read code"
above what are now four bash blocks - corrected rather than left contradicting
itself. 4 code blocks, was 0.

what-senior-developers-catch-that-ai-misses - opened by describing a wrong
sentence; now opens with the actual diff, both sides. Adds the Propshaft README
quote that settles it and the reviewer's verdict verbatim rather than my
paraphrase of it. 1 code block plus two block quotes, was 0.

Totals across the three posts: 0 code blocks this morning, 11 now.

NOT done, and it is the biggest one left. All three still say
author: 'JetThoughts Team'. Arkency and thoughtbot put a person's name and
face on the argument; we put a company noun. That is a positioning decision
rather than a copy edit - whose name goes on these - so it needs Paul, and I
have not guessed at it.

Gates: cadence quotas pass in every section of all three posts. Banned words
zero, em dashes zero. bin/hugo-build green. bin/rake test:links clean at 31,948
unique links, zero errors.

Co-Authored-By: Claude Opus 5 noreply@anthropic.com
Claude-Session: https://claude.ai/code/session_016PUkwFTsiv7EB2DYKogbpg

[coderabbitai]

Important

• 🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: 1d3c23e6-83e7-498f-8c62-8c33e2cee159

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}