Template talk:Indic glyph

Why do Devanagari, Bengali, and Oriya use images, but the others text? Gorobay (talk) 16:24, 4 May 2013 (UTC)

Because there are a complete set of images of Devanagari, Bengali, and Oriya letters, and not others. VanIsaac_{WS Vex}^contribs 20:06, 4 May 2013 (UTC)

I believe that there is also a full set of Javanese, Balinese, Batak, Lontara, and Sundanese characters uploaded by user User:Alteaven. DerekWinters (talk) 01:36, 26 March 2014 (UTC)

Rebuilding template:Indic glyph

I have started a discussion at WikiProject Writing Systems on organizing this template's contents. Please add your thoughts to the discussion there if you have anything to add. Thanks. VanIsaac_WS^cont 04:59, 30 May 2020 (UTC)

@Vanisaac: The discussion 'there' has been archived, so I'm putting my question here. Is there any easy way to show three descendants? Indic glyph/cell, unless reformatted, is not easy to read, and I'm not sure if I would need to upgrade all cells to hold up to three descendants. A definite example is Pa, with three New Tai Lue descendants:

• ᦢ U+19A2 NEW TAI LUE LETTER HIGH BA
• ᦥ U+19A5 NEW TAI LUE LETTER LOW BA
• ᦔ U+1994 NEW TAI LUE LETTER HIGH PA

Unless you argue that the difference between the first is just the addition of a TAI THAM tone mark...

A more doubtful set of four descendants is Ga, with arguably four New Tai Lue descendants:

• ᦅ U+1985 NEW TAI LUE LETTER LOW KA
• ᦨ U+19A8 NEW TAI LUE LETTER LOW KVA
• ᦆ U+1986 NEW TAI LUE LETTER LOW XA
• ᦩ U+19A9 NEW TAI LUE LETTER LOW XVA

There may well be some Tai Viet sets of four descendants, but I need to do some literature searches.

The best I can do at present for what I want as

 |talucp = 19A2 
 |talu2cp = 19A5
 |talu3cp = 1994 

is the terrible hack

 |talucp = 19A2;/ᦥ/&#x1994

Help! --RichardW57 (talk) 22:33, 27 October 2021 (UTC)

• I think you are trying to show way too much with this template. It's about showing the general relation of characters, not a comprehensive list of every single variation that might fit. Low Ba is obviously a simple variant of High Ba, while Kva and Xva are just simple vertical conjuncts - one of thousands among all the Indic scripts for each consonant - that just so happens to be historically treated as a separate character. It's the same as Bengali Kṣa - ক্ষ, which is used as an independent letter in Assamese. That sort of information belongs in the article text, not this visualization. VanIsaac, MPLL ^cont_WpWS 01:00, 28 October 2021 (UTC)

@Vanisaac: It was you who first listed New Tai Lue Letters High and Low BA as separate glyphs! (It was in Ba (Indic).) New Tai Lue doesn't have much that can be analysed as conjuncts. There are about half a dozen historical conjuncts with HIGH HA as the top letter, and about half a dozen encoded historical conjuncts with LOW VA as the bottom letter, and that's it as far as the script in isolation is concerned. I'm happy to treat the old tone mark and subscript VA as though they were 'productive' - the subscript might have been encoded as such if Peter Constable hadn't misinterpreted Li's Handbook of Comparative Tai. I'll fix them when I add the final consonants to the encoding sections.

I do find the chart useful for showing how consonants have split, but it does become awkward to decide what should be treated as a nukta. Thai stretchings and dentings could be seen as nuktas - but Thai tho montho is not a modification of tho thahan, even though it looks as though it were made from it by adding a dent. Most of the corresponding 'tails' in Tai Tham can also be viewed as diacritics.

I have mentioned the issue with trios in talk:Ā (Indic). --RichardW57m (talk) 13:57, 28 October 2021 (UTC)

Tamil Brahmi

@Vanisaac:: I fear we need to include Tamil Brahmi. We need it to show the splits in Brahmi, for:

1. 𑀵 /ḻa/ was derived from 𑀟 /ḍa/
2. 𑁵 /ḷa/ from 𑀮 /la/
3. 𑀶 /ṟa/ from 𑀝 /ṭa/
4. 𑀷 /ṉa/ from 𑀦 /na/.

Possibly we need to generate images from a Tamil Brahmi font - Vinodh Rajan's Adinatha seems suitable, including its licensing. It's a well documented font. I've already generated and uploaded a similar set of Tai Tham character images.

I'd be inclined to make the inclusion of Tamil Brahmi dependent on a parameter. A lot of characters had been dropped, and the Brahmi section of the chart could get fussy. It seems that southern India has seen a fair bit of hybridisation of alphabets. --RichardW57m (talk) 13:53, 1 November 2021 (UTC)

I specifically included Pallava to demonstrate the early innovations in that southern line, as Vatteluttu and Koleluttu (two incredibly minor scripts) are the only Tamil-Brahmi descendants that might have features that don't show up in Pallava. VanIsaac, MPLL ^cont_WpWS 21:46, 1 November 2021 (UTC)

Using Unicode naming for legibility, it seems that the evidence for LLA and LLLA in Pallava is extremely thin. Too much Sanskrit but no Vedic, I guess. There may be problems with where the boundaries are drawn. I have however seen references to gaps in various scripts being filled in from Vatteluttu. Possibly LLA came to SE Asia with Pali, with the Vedic LLA being unknown to the monks of SE Asia. It looks as though the Dravidian letters may have ancestries different to the rest of their scripts. If the Pallava LLA I've seen depicted is nothing but LA where one might expect LLA to have been written, then the derivation of most LLA from LA may collapse. --RichardW57 (talk) 08:32, 2 November 2021 (UTC)