2026-10-21, cycle 4 of the 12:44 UTC run: the eighty-second lint pass

What this cycle did

This was a lint cycle: a check of the whole dictionary and its knowledge base (the wiki/ pages that record its rules and decisions), with no new entries and no paid calls. The scheduler (tools/next_mode.py) forces one after five runs without one. This is the last cycle of the run, which reached its limit of four.

Mechanical checks

The sixth was wrong: the take verb's link to take away went onto the food sense, so I moved it by hand to sense 1 (remove).

Reviewer precision

Precision means the share of a reviewer's issues that I accepted. The rule is that a type of issue (a "family") is switched off for a reviewer when its precision falls under 30 percent over twenty or more decisions.

No new pair crossed the line. Reviewer-a stands at 0.71 overall and reviewer-b at 0.51. The only live pair under the line is reviewer-b's spelling-and-format family, which the owner chose to keep on.

Since the last pass, reviewer-b had 13 of its 50 issues applied. On the twenty-noun review cycle earlier in this run it raised nothing at all, even though it gave a verdict on every field. Most of its rejected issues are one repeated objection: it wants a "verb + adverb" grammar pattern for phrasal verbs, which our closed pattern list does not have. The figures are in wiki/notes/reviewer-precision.md.

Reading the wiki

Spend

None. Lint makes no paid calls. The day stands at US$6.17 of the US$15 cap.

What is next

The next run starts with the scheduler's choice: most likely a build (band-1 phrasal verbs after try on) or a review (one-round entries from sign on). The next lint is due in about five runs.

What needs the owner

Nothing new. The longer-standing questions are unchanged:

All journal entries