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`

[derek73]

A credential run written after a suffix comma is read as a suffix only while every word in it is suffix-shaped. Once its last word is a bare member of suffix_acronyms_ambiguous, the run is no longer wholly suffix-shaped, so C1 reads the comma as a FAMILY comma: the whole pre-comma name becomes the family and the run's words fall into the post-comma name slots.

>>> parse("John Smith, PhD MEng")
ParsedName(given='PhD', middle='MEng', family='John Smith')     # every release: suffix 'PhD MEng'
>>> parse("john smith, phd meng")
ParsedName(given='phd', family='john smith', suffix='meng')
>>> parse("JOHN SMITH, PHD MENG")
ParsedName(given='PHD', family='JOHN SMITH', suffix='MENG')     # capitals do not save it
>>> parse("Jane Doe, MS LAc")
ParsedName(title='MS', given='LAc', family='Jane Doe')           # a title-listed front word reads as a title, and NOTHING is reported
>>> parse("John Smith, PhD, MEng").suffix
'PhD, MEng'                                                      # a second comma keeps the run

Since when. The path is older than #540: john smith, phd ma reads given phd, family john smith on 2.0.0 through 2.3.0, and Jane Doe, MS Ma reads title MS there (1.4.0 read suffix MS Ma). #540 (PR #543) routed meng and lac into it and recorded it as an accepted cost in decisions.md#suffix-acronym-collisions (2026-09-25), with case rows comma_credential_run_ending_in_meng_re_reads_the_comma and comma_lower_credential_run_ending_in_meng_re_reads_the_comma pinning today's reading and fix(#540) ledger rules classifying it at every baseline. The silent case conflicts with two rules as written: C1 (the count decides first) and S2 ("either reading carries the ambiguity flag"). A related silence: an ambiguous member inside a multi-word post-comma credential run never reports (Jane Doe, MS MA, Jane Doe, PhD LAC).

Proposed rule

Rationale: a comma followed by a run whose every word is credential vocabulary is the writer's suffix comma, and an ambiguous member at its END has an unambiguous credential in front of it to lean on, which is more evidence than a bare member standing alone ever has.

Statement: after a comma behind a full name, a run of two or more words in which every word is suffix vocabulary and the last is an ambiguous member reads as the suffix run, reported SUFFIX_OR_NAME at the ambiguous member. A run whose FIRST word is ambiguous (John Smith, MEng PhD) is out of scope here and stays as it is.

Examples (target readings; today's readings are the ones above):

"John Smith, PhD MEng"   →  suffix="PhD MEng"     ambiguity: SUFFIX_OR_NAME
"john smith, phd meng"   →  suffix="phd meng"     ambiguity: SUFFIX_OR_NAME
"Jane Doe, MS LAc"       →  suffix="MS LAc"       ambiguity: SUFFIX_OR_NAME   · MS is a title word too; the run decides
"John Smith, MEng"       →  suffix="MEng"                                     · unchanged, C1's name-word count
"Smith, John MEng"       →  middle="MEng"                                     · unchanged, a name word in the run
"John Smith, MEng PhD"   →  as today                                          · boundary: the ambiguous word leads the run

Accepted consequences: John Smith, Ed Ma and every ma/ba spelling of the same shape move with it; the differential ledgers gain a rule and the #540 rows are re-recorded.

Open questions:

• Whether the reading should also hold for a single-word post-comma run written with periods in a chunked spelling: Wang M.Eng. reads family M.Eng. today because S2's period gate counts a member as unambiguous only written one period per letter (M.A.), and MEng and LAc are the first members whose conventional dotted spelling is chunked. Accepted and recorded in Should meng and lac be ambiguous suffix acronyms? wang meng has had no family name since 2.0 and renders Wang MENG on master #540 (Derek, 2026-09-25) rather than widening _vocab._dotted; decide here whether a listed member's chunked dotted spelling should pass the gate (Wang M.Eng. → suffix), with Wang M.A. as the control that already does.
• Whether the silent title-dual reading (Jane Doe, MS LAc with no report) is worth an Ambiguity of its own if the rule above is declined.

Exclusions: nothing leaves any vocabulary set.

Found by the #543 review (2026-09-25); measurements on the tree at that PR and on the 1.4.0 through 2.3.0 wheels.