search

notofonts / dogra

Noto Dogra
SIL Open Font License 1.1
1 stars 0 forks source link

U+093D DEVANAGARI SIGN AVAGRAHA is copied from Noto Serif Devanagari #12

Closed dscorbett closed 1 year ago

dscorbett commented 2 years ago

Font

NotoSerifDogra-Regular.otf

Where the font came from, and when

Site: https://github.com/notofonts/dogra/releases/tag/NotoSerifDogra-v1.006 Date: 2022-10-09

Font version

Version 1.006

Issue

The full font copies U+093D DEVANAGARI SIGN AVAGRAHA from Noto Serif Devanagari, ignoring the glyph Avagraha.dogra, which presumably was designed to matched Dogra better.

Where did Google’s sample text for this font come from? It is not appropriate to use U+093D in Dogra, because all avagrahas in Unicode so far are script-specific. If Dogra-script texts really use avagraha, it should probably be encoded as a new DOGRA SIGN AVAGRAHA character. Unicode should at least be given the chance to give some advice about this character before a font starts promoting the use of one from another script.

If Google’s sample text is just some Sanskrit automatically transliterated into Dogra, I would assume it is a faulty transliteration, and I would change the text instead of the font.

Character data

ऽ U+093D DEVANAGARI SIGN AVAGRAHA

Screenshots

ऽ Compare Avagraha.dogra: ऽ

bgo-eiu commented 1 year ago

@dscorbett You would be correct to point out the sample text used by Google is Sanskrit, not Dogri. Avagraha is not used in the writing of Dogri in any script

simoncozens commented 1 year ago

What we need to fix this bug is a reasonable Dogra sample text...

bgo-eiu commented 1 year ago

@simoncozens https://archive.org/details/QfgG_shri-ranbir-singh-sainik-dand-vidhi-military-penal-code-jammu-1887-takri-and-urd/

Transcribing a section from this would be a good idea. (Ranbir Singh was the king who ordered the creation of the Dogra script a.k.a. Takri in the sense used here.)

simoncozens commented 1 year ago

We now have a Dogri sample text in the language repository (a translation of UDHR), so we don't need the (inappropriate) avagraha any more; I've removed it from the subset.