Should Devanagari श्री read as a name word when the name's romanization keeps it? (श्री कृष्ण श्रेष्ठ reads title श्री, Shree Krishna Shrestha given Shree) #870

श्री is a salutation in the default vocabulary (hi.toml, mr.toml: Shri, Mr.; #269). In Nepali names it is also the first element of a given name (श्रीकृष्ण, श्री प्रसाद), and the two scripts now read one name two ways:

>>> parse("श्री कृष्ण श्रेष्ठ")       # title 'श्री', given 'कृष्ण', family 'श्रेष्ठ'
>>> parse("Shree Krishna Shrestha")   # given 'Shree', middle 'Krishna'
>>> parse("श्रीकृष्ण श्रेष्ठ")        # given 'श्रीकृष्ण'

hi.toml's entry already notes there is no Latin twin, since Latin sri and shri are given names (their exclusions in hi.toml).

In the nepal-candidates local corpus (#866, fetched 2026-10-24)

The Election Commission writes each candidate's name in Devanagari beside its own romanization.

• 215 Devanagari names open with a spaced श्री. All 215 read it as a title.
• The Commission's romanization keeps it as a name word in 201 of them (Shree 193, Shri 6, Sri 2) and writes it Mr in 2.
• 386 more names write श्री glued into the given name, which reads as given.
• श्रीमती (92, all read as a title) is romanized Shreemati/Shrimati/Srimati in 62 and Mrs/Ms in 11. It looks like an honorific more often than a name, so it is not part of this question.

The question

Is a spaced leading श्री in a Nepali name a salutation or a given-name element? The Hindi reading (Mr.) is the common one in India. The Nepali register here keeps it in the name. Options:

1. Leave श्री a title, and record the cost (rules/H1's Accepted clause).
2. Read श्री as a title only when the name has no other given-name word after it, which would change little.
3. Move the reading into the hi/mr packs and out of the default (Ship hi and bn locale packs for Latin-transliterated Indic honorifics #345 is about Latin-script Indic packs; a ne view of the same word would be the counterpart).

Related: #269, #344, #345.