docs(test-plan): record the execution baseline in the strategy and phasing documents (BACKLOG #1100) - #654
Merged
Merged
Conversation
… sites that assert it (BACKLOG #1100) TWO OF THE NINE SITES THIS ITEM NAMES. The other seven are accounted for below; five were fixed by 2119001, one is a ledger edit that is not a builder's, and one already carries its correction. WHAT THE TWO SITES CLAIMED. 00-strategy-and-governance.md and 19-execution-phasing-and-sign-off.md both said docs/BACKLOG.md is a published baseline that stops at heading 231 and "says so itself", quoting a sentence and citing a line anchor into that file. MEASURED AGAINST THE COMMITTED FILE, AND BOTH HALVES ARE FALSE: 373 item headings, numbered 3 to 1372, of which 320 are ABOVE 231 the quoted sentence returns ZERO occurrences in docs/BACKLOG.md control: an unrelated phrase in the same file returns 1, so the search is not blind it lives under item 185 in docs/archive/backlog/BACKLOG-CLOSED.md, which returns 1 the line anchor now resolves to unrelated text THE 19-* SITE IS THE ONE THAT MATTERS, BECAUSE IT IS AN INSTRUCTION RATHER THAN NARRATIVE. It told the reader a drift-pinning test "must therefore bound its check at 231, or it will fail on sound citations". A test built to that would have bounded at a number with nothing behind it and reported confidently on the wrong population. That is this item's own severity argument -- a test plan is the wrong document to carry dead assertions, because its rows are written to be executed. I DELIBERATELY ASSERT NO REPLACEMENT NUMBER. PUBLIC_BACKLOG_FLOOR = 1000 in scripts/hooks/ledger_check.py constrains which numbers may be ALLOCATED, not which are readable, so substituting it would swap one false bound for another. Both sites now say so, and the 19-* note tells whoever writes that test to settle the bound first and record why. WHAT SURVIVES, because it is sound and independent of the false scaffolding: absence of an item from a file you can read is a publishing boundary rather than evidence of completion. Only its evidence was wrong, not the caution itself. Line anchors dropped rather than repointed. The project rule is to cite by item or section heading and never by line number, and this is the worked example of why -- the anchor did not merely rot, it began resolving to unrelated text, which reads as a working reference forever. *** THE ITEM'S ENUMERATION OF NINE IS INCOMPLETE, AND I AM REPORTING THAT RATHER THAN ABSORBING IT. *** The identical false premise appears in THREE further chapters this item does not name -- 10-auth-rbac-and-active-directory.md, 16-security-phi-and-supply-chain.md and 18-interop-migration-and-uat.md -- and 18-* declares itself the OWNER of the plan-wide convention the other four restate. Fixing the source is a different and larger change than correcting two sites, and splitting an item is not a builder's call. Left untouched on purpose; handed to the dispatcher with the measurement. This is the enumeration hazard the standards name: prefer "at least" to a list. Measured, interpreter .venv\Scripts\python.exe in this worktree, -p no:randomly, base a490993: test_link_resolution, test_backlog_citation_check, test_backlog_status_check, test_ledger_check: 99 passed docs/BACKLOG.md is NOT in this diff -- the ledger is not a builder's to edit No code changed; this is a documentation-record correction with its Closing-act satisfied by content. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
github-merge-queue
Bot
removed this pull request from the merge queue due to failed status checks
Aug 28, 2026
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.
Authored by Builder 1; pushed and opened by the LANDER. Two files, docs-only.
Verification
Reported by the author: 1 ahead / 0 behind at hand-over, clean tree,
merge-treeexit 0, and off-file from the worktree gate — so it is independent ofb1-1067andb1-1072and can land in any order relative to them.LANDER checks
--not --remotes=originreturned a non-zero count while the naive--not --remotesreturned 0. Now on origin, count 0.backlog-hygiene.yml:124-133demands adocs/BACKLOG.mdedit only when the diff touchesmessagefoundry/,ide/ormessagefoundry_webconsole/. This branch touches neither — two files underdocs/testing/.I had told the author a banner was owed on three of their branches. I was quoting BUILDER.md's summary of the rule, which drops the path condition the gate actually applies. Verified by reading the workflow after PR 652 passed that same gate without touching
BACKLOG.md.