Should John Smith VI MD read like John Smith III MD? #784
Copy link
Copy link
Closed
Labels
designquestionvocabularyPer-word vocabulary model and language packsPer-word vocabulary model and language packs
Description
Today a roman numeral before a credential reads two ways depending on its spelling:
| Input | Reading |
|---|---|
John Smith III MD |
suffix "III MD" |
John Smith VI MD |
family "VI", suffix "MD" (reported suffix-or-name) |
The difference is that numerals are recognized two ways. ii, iii and iv (and i, v) are suffix vocabulary; vi through x are read by shape alone, by S10's trailing-numeral fork (_vocab._ROMAN, I–X). S8's credential run asks the vocabulary whether a word starts the run, so a listed numeral starts it and a shape-only one does not; S10 then reads a numeral with a credential behind it as an ordinary name word (its Accepted clause).
Questions:
- Should a numeral read as a generation before a credential, whatever its spelling?
- Do we still want two recognizers, vocabulary for some numerals and shape for others, or one predicate that every rule asks (mechanisms.md#ONE-PREDICATE-PER-QUESTION)? AGENTS.md's spelling sweep records this split as a recurring source of bugs (Two stages hand-copy each other's trailing-word logic after #601/#602; share one predicate each #610, Should a known suffix right before a comma make it a suffix comma?
Fabio Martini II, ADreads givenAD, familyFabio Martini#656).
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
designquestionvocabularyPer-word vocabulary model and language packsPer-word vocabulary model and language packs
Activity
Ask one predicate for every roman-numeral rule (#752, phase 2f) (#886)