[derek73]

Part of #762, final round, the small-sections batch. Two other batches are running in parallel: one covers R1, R2, R3, R5, render-default and comparison-surface; the other covers differential-ledger, parse-cost, comma-suffix-arc, deviates-registry, review-agent-single-source and 3-0-reevaluations. This PR touches none of their files.

What changes

These fourteen files in docs/design/decisions/ each gain a ## Standing block and a ## Log heading, and leave tests/v2/decisions_without_standing.json:

• A1 (the ambiguity report)
• A2 (the empty name)
• D1 (the segmenterless-activation warning)
• D2 (construction raises, parse never does)
• N1 (the default delimiter pairs)
• N2 (same-character quotes and shared-character conventions)
• N3 (the lone-word nickname rule)
• T1 (separators, not joiners)
• T3 (the interpunct's flank guard)
• T4 (the bare letter as an initial)
• phd-merge
• suffix-field-composition
• two-input-invariants
• v1-xfail-triage

I parsed every example a Standing line cites on the tree. For each line I also grepped the whole decisions/ and rules/ folders for later entries that changed it.

Edit-kind-1 notes on entry heads, where a later decision reversed part of the entry:

• A1's 2026-10-08 Kim Min Do under FAMILY_FIRST is told 'Do' was "read as a middle name" when it lands in the family #626: its literal wording for the comma's credential reading, superseded in part by Reports say "read as a credential" for a period-marked title after a comma #793.
• N2's 2026-07-18 Support smart quotes (“Jack”), guillemets («Petit»), and CJK brackets as nickname delimiters #273: its per-pair scan, superseded the same day by the 05fe693 leftmost-match entry.
• phd-merge's 2026-08-31 Ph. D. Van Johnson has no surname: a leading suffix displaces the particle like a title used to #371: its revise() rendering, superseded by Should Parser.revise(name, suffix=name.suffix) be the identity? (John Smith MD PhD gives MD PhD, and revising with that string gives MD, PhD back) #511 (decisions/R1).
• phd-merge's 2026-07 PR nameparser 2.0 implementation #288: its spellings reading alike after a name, superseded in part by Should a credential run after a comma keep its comma when the last word is an ambiguous acronym? John Smith, PhD MEng reads first PhD, last John Smith #544 (decisions/S7). The merged pair is outside the no-comma company walk, so John Smith Ph. D. MEng keeps family 'MEng' while John Smith PhD MEng reads suffix.

Edit-kind-2 correction: phd-merge's #288 entry said the spaced and unspaced Ph. D. "read alike". That never held mid-name or at the head, on the 1.4.0 and 2.0.0 wheels as on the tree:

• mid-name: John Ph. D. Smith gives suffix 'Ph. D.', while John PhD Smith gives middle 'PhD';
• at the head: Ph. D. Van Johnson gave title 'Ph.' at 1.4.0 and suffix 'Ph. D.' at 2.0.0, where Ph.D. Van Johnson gave given 'Ph.D.'.

The entry now says "at the end of a name", with a dated note. Its "Scoped by #371" parenthetical became a pointer to #371's head change.

A dated Log line added for an undated bullet: v1-xfail-triage's #485 bullet had no entry of its own. It gains 2026-09-01 #485, dated by the triage commit that wrote it (d02dc41, 2026-09-01), not by the file split that last moved it.

rules/ edit, a claim the tree contradicted: rules/S2's Accepted bullet said a title before the credential costs the name its surname. That holds only behind a title run that addresses by the given name: Sir Ph. D. Van Johnson has an empty family, while Dr. Ph. D. Van Johnson and Sir Dr. Ph. D. Van Johnson keep family 'Van Johnson'. The bullet now says "behind a title run that addresses by the given name (H1)". The S2 entry in tests/v2/rules_budget.json was re-recorded: accepted_words went from 203 to 214. The new words scope that one existing clause and read no new signal.

Review findings (design-docs-reviewer, five rounds)

Round 1, seven findings, all fixed:

1. phd-merge overclaimed that the spellings read alike mid-name (decisions/S7 Should a credential run after a comma keep its comma when the last word is an ambiguous acronym? John Smith, PhD MEng reads first PhD, last John Smith #544 and decisions/S8 Should a name word after a credential join the suffix? John Smith PhD Jones reads middle Smith PhD #602 part them).
2. rules/S2's surname-loss claim was too broad (see above).
3. N3 stated M4's decision as a Standing bullet instead of an Elsewhere line.
4. T4 put Should john smith nee x x read like JOHN SMITH NEE X X? A shape-only numeral in a maiden clause #731, already marked SUPERSEDED, under Not standing.
5. T3 said "suppresses division" where the surname split is what is suppressed.
6. D2 put 남궁민수's segmentation report under locales.ZH, though it fires by default.
7. v1-xfail-triage credited H4's title-or-name report to the triage.

Round 2, seven findings, all fixed:

1. The nameparser 2.0 implementation #288 head note blamed Should a credential run after a comma keep its comma when the last word is an ambiguous acronym? John Smith, PhD MEng reads first PhD, last John Smith #544 and Should a name word after a credential join the suffix? John Smith PhD Jones reads middle Smith PhD #602 for a difference that predates both, so it became a kind-2 correction.
2. T3 said "written order" where the code keeps the declared order (FAMILY_FIRST reverses the name).
3. The Dr King Jr report dates from H3's Should a title written after the name still be a title? ("John Smith Prof." → family Prof.) #316 floor.
4. phd-merge's Dr. contrast was not Ph. D. Van Johnson has no surname: a leading suffix displaces the particle like a title used to #371's decision.
5. A Markdown continuation needed a blank line.
6. S2's wording missed title runs.
7. S8's pointer named a pointer, not an entry.

Round 3, three findings, all fixed:

1. The addressing clause pointed at P2's Keep citation comments short: quote, clause, pointer #760 instead of H1's 2026-09-08 A given-name title at the end of a title run should address by given name — "Her Majesty Queen Elizabeth" parses Elizabeth as the family name #489.
2. The kind-2 correction left out the head.
3. S2 restated H1's predicate without its whole-key arm.

Round 4: one label ("Scoped by #371") contradicted the correction beside it. Fixed.

Round 5: no findings.

Left as found

• D2 and T3 keep their Declined: blocks under ## Log. The Standing block notes that those lines stand as written.
• suffix-field-composition's entry quotes 2026-08-16 counts (11/5/24, honorifics the largest). These are a dated snapshot: suffix_words holds 170 words today, and the generational words now outnumber the honorifics. The Standing line drops the count, and the question stays open under Should suffix split into generation and credentials? (Smith, Jr., PhD) #326.
• M2-maiden-marker-rule.md's Standing block uses Not standing: for an entry reversed by a later decision (2026-10-03 #549). The reviewer flagged this as the same pattern removed from T4 here. I left it alone: it is outside this batch, and the convention may want a ruling.

Tests

• uv run pytest -q tests/v2/test_design_docs.py tests/v2/test_doc_citations.py tests/v2/test_rules_doc.py tests/v2/test_rules_skeleton.py tests/v2/test_rules_budget.py tests/v2/test_doc_spellings.py: 1421 passed.
• Full suite on the first commit: 22772 passed, 125 skipped, 3 xfailed on 3.11, and the same on 3.15. The later commits touch only docs/design/ and tests/v2/rules_budget.json, and the doc tests above pass on the head.

---

Generated by Claude Code

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.05%. Comparing base (9be7261) to head (e7e7fc2).
⚠️ Report is 1 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #831   +/-   ##
=======================================
  Coverage   99.05%   99.05%           
=======================================
  Files          47       47           
  Lines        4659     4659           
=======================================
  Hits         4615     4615           
  Misses         44       44           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.