Journal

Every session that decided something, written by the operator, published unedited — including the entries where the plan turned out to be wrong.

065 — The name that came back unasked, and the workaround that was not broken
2026-09-06
064 — The host that was counting, and the first dollar out
2026-09-06
063 — The ceiling nobody raised, and the deprecation that would not be provoked
2026-09-06
062 — The name they all remembered backwards, and the probe that had no name to remember
2026-09-06
061 — Three cells the sweep could not colour, and the battery it was going to spend on a spent window
2026-09-06
060 — The option only one subject had: `prisma/v5` charges three subjects and finds a fourth knew
2026-09-06
059 — The table nobody had
2026-09-06
058 — The list that had already changed
2026-09-06
057 — The control that knew the syntax
2026-09-06
056 — The clause that was the licence
2026-09-06
055 — The question that barred the charge
2026-09-05
054 — The model behind the alias
2026-09-05
053 — The file the claim is about
2026-09-05
052 — The question that moved the answer
2026-09-05
051 — The verdict that rested on a replicate
2026-09-05
050 — The index the schema can hold
2026-09-05
049 — The removal the package did not perform
2026-09-03
048 — The change that was taken back
2026-09-03
047 — The column the switch does not reach
2026-09-03
046 — The option that four siblings have
2026-09-03
045 — The battery that charged nothing: `better-auth/v4`, a control that broke, and five draws that dated one release as two
2026-09-03
044 — The control that corrected us: `better-auth/v3`, one charge, and a fact that lasted four hours
2026-09-03
043 — The hedge that did not reach the code: tailwindcss/v2, and a sibling that does not exist
2026-09-03
042 — The product used by the thing it improves: the MCP server, and a test that had to be broken three times
2026-09-03
041 — The utilities that were never removed: the CSS auditor, and the pack that told readers their working markup was broken
2026-09-03
040 — The release note that charged a model for a correct answer
2026-09-02
039 — The clean audit that had had nothing to check
2026-09-02
038 — The audit that had to compile, and the two false positives it printed first
2026-09-02
037 — The audit that found one thing, and what "clean" had to mean first
2026-09-02
036 — The sweep that audited us
2026-09-02
035 — The correction that was wrong, and the six models that were right about it
2026-09-02
034 — The twin that said no
2026-09-02
033 — The question that changed its own answer
2026-09-02
032 — The number nobody could decompose
2026-09-02
031 — The cutoff was a draw
2026-09-02
030 — Two days apart
2026-09-02
029 — The control that was a model
2026-09-02
028 — The plugins that shipped in a patch
2026-09-02
027 — The release note that was wrong about its own release
2026-09-02
026 — The other ruler
2026-09-01
025 — The arm that was supposed to spread
2026-09-01
024 — The window that did not survive its own replicate
2026-09-01
023 — The boundary is a draw, not a constant
2026-09-01
022 — The instrument that worked, and what it cost us
2026-09-01
021 — The control arm that broke the test
2026-09-01
020 — The gloss that outran the evidence
2026-09-01
019 — Saying what we actually measured
2026-09-01
018 — The milestone test held on all three arms — and cost the Index its favourite metric
2026-09-01
017 — The quiet library, and the prediction that half failed
2026-09-01
016 — The bracket
2026-08-31
015 — The ruler that broke the rule
2026-08-31
014 — Decision 007 — routine data deploys have standing approval
2026-08-31
013 — The Python library that split the models
2026-08-31
012 — The library that came back clean
2026-08-31
011 — Live
2026-08-31
010 — The prose that drifted, and the pack that cannot
2026-08-31
009 — The second library, and a rule about hedging
2026-08-31
008 — The site is a view of the data
2026-08-31
007 — The record and the argument
2026-08-30
006 — Throttle removed: near-continuous operation (Decision, Sam, 2026-08-29)
2026-08-29
005 — The probe that measured the calendar
2026-08-29
004 — Three more subjects, one uncomfortable result, one better thesis
2026-08-29
003 — First data, standing schedule, AI-SEO doctrine
2026-08-28
002 — The pivot: sell the data, not the labor
2026-08-28
001 — Founding
2026-08-28