release: v1.5.0 — a comment narrates as well as instructs - #22
Merged
Conversation
Ships DECISIONS §52 (merged in #21) and adds §53. Typecheck, lint, build and 657 tests green; built CLI reports 1.5.0. The constraint gate accounted for 29 of 29 code-bucket fallbacks, and roughly twelve were a codebase narrating its own history rather than instructing anyone. Four files that produced nothing now reduce: 572 of 576 corpus rows byte-identical, 0 new fallbacks, 0 files that stopped reducing. §53 ends the version-reservation problem rather than resolving another instance of it. v1.5.0 was held for "Granular Sub-Query Re-hydration", and unlike the previous three collisions that release is perfectly buildable -- M5b shipped in Wave 2 and what remains is design work -- so §49's rule ("preconditions measured false -> no number") did not cover it. The general rule replaces it: the roadmap reserves no version numbers at all; a number is a fact about what shipped, assigned at ship time. Four reservations in four releases were wrong, each for a different reason, which is enough evidence that predicting the order work finishes is not something to encode in a number. v2.0.0 is retained because a major signals breaking change rather than queue position. Minor rather than patch, consistent with v1.3.0 and v1.4.0: nothing removed, but the same command over the same input emits different bytes on 4 of 576 corpus rows. The 6pp on the TypeScript bucket is NOT portable and every doc touched here says so: all four recovered files are this repository's own source, which M11 measured as 32.8% comment prose written in a what-used-to-be-true style, and Python gained zero. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Ships DECISIONS §52 (merged in #21) and adds §53. Typecheck, lint, build and 657 tests green; built CLI reports
1.5.0.What ships
The constraint gate accounted for 29 of 29 code-bucket fallbacks, and roughly twelve of those were a codebase narrating its own history — "has always read these two", "could never have worked" — rather than instructing anyone. Four files that produced nothing now reduce: 572 of 576 corpus rows byte-identical, 0 new fallbacks, 0 files that stopped reducing.
§53 ends the version-reservation problem instead of resolving another instance
v1.5.0 was held for "Granular Sub-Query Re-hydration". Unlike the previous three collisions, that release is perfectly buildable — M5b shipped in Wave 2 and what remains is design work on the response shape — so §49's rule (preconditions measured false → no number) did not cover it.
The general rule replaces it: the roadmap reserves no version numbers at all; a number is a fact about what shipped, assigned at ship time. Four reservations in four releases were wrong, each for a different reason.
v2.0.0is retained because a major signals breaking change rather than queue position.Minor rather than patch, consistent with v1.3.0 and v1.4.0: nothing removed, but the same command over the same input emits different bytes on 4 of 576 corpus rows.
The caveat, carried in every doc touched
The 6pp on the TypeScript bucket is not portable. All four recovered files are this repository's own source, which M11 measured as 32.8% comment prose written in a what-used-to-be-true style. Python gained zero. It is the corpus-bias trap appearing as a favourable number.
🤖 Generated with Claude Code