Module talk:kxv-translit/testcases

Kuvi orthography

@Kutchkutch, doesnt the dictionary use <a> as the base vowel for consonants like <aggaḍa> as ଆଗ୍ଗଡ​ — This unsigned comment was added by AleksiB 1945 (talk • contribs) at 14:25, 14 June 2023.

@AleksiB 1945: Yes, it is true that {{R:kxv:Reddy}} uses <a> as the inherent vowel. However, the adaptation of the Oriya script that is used is specific to that dictionary. In the Forward section it says:

The entries of the tribal language are given in the script adapted for the language following the spelling system developed for it.

and on page i it says:

The new letters, used only in Kuvi, are introduced at the appropriate places determined by their sound value.

Therefore, the dictionary compiler created a new spelling system without any mention of how existing texts in the Kuvi language use the Oriya script.

In the following Unicode document

https://www.unicode.org/L2/L2019/19005r2-odia-overline.pdf

there are written texts in the Kuvi language using the Oriya script and an analysis of how the Oriya script is used to write the Kuvi language. The analysis of the script shows that the inherent vowel is <o>. Using this spelling system may definitely cause some issues when compared the spelling system created the dictionary.

Also, the objective of the Unicode document is to show that vowel length in Kuvi has been traditionally indicated with the macron ¯ symbol to indicate lengthened vowels instead of two vowel symbols, and this macron ¯ symbol has been included in Unicode version 13.0 released in 2020 at the code point U+0B55 as ୕. The Unicode document also says there that there is additional way to indicate vowel length using the Oriya nuqta character ଼. Since there are three different ways to show vowel length, perhaps the second vowel symbol in the dictionary or macron ¯ U+0B55 ୕ could be omitted in the page title using Module:languages and instead be shown in the |head= parameter on the headword line. Kutchkutch (talk) 00:31, 15 June 2023 (UTC)

Using the macron is really good as vowel nukta can be confused with nukta consonants (<ṛ> is common) also does the article use the visarga to represent the glottal stop? the visarga is used commonly and before a vowel but there are no ' or ?/ʔ like in the word "ଏମାଃଆଁ" AleksiB 1945 (talk) 14:11, 15 June 2023 (UTC)

@AleksiB 1945: Yes, the Unicode document appears to use the visarga for the glottal stop. Since ଓ is the same as ଅ perhaps the following page moves can be made.

ଦୋସ୍ସୋ → ଦସ୍ସ

ନୋଓ → ନ

ସୋଓ → ସ

ପଆସୋ → ପାସ

ଓ could be retained in borrowings from Oriya that have ଓ in the Oriya spelling. Kutchkutch (talk) 23:18, 15 June 2023 (UTC)

@Kutchkutch, should <sō nō> be ସ୕ ନ୕ or ସଅ ନଅ like Odia? and <āṭai> as ଆ୕ଟୈ or ଆ୕ଟାଇ

also is there any info on the orthography of Pengo and Manda? can we use the same ortho for it too or are there any changes? AleksiB 1945 (talk) 03:03, 17 June 2023 (UTC)

@AleksiB 1945: Since <sō> and <nō> are borrowed from Odia, spelling them like Odia seems fine. <āṭai> must be an error for <āṭo> on page 5 of {{R:kxv:Reddy}}. It seems reasonable to extend MOD:kxv-translit to Manda and Pengo since they are both primarily spoken in Odisha, have similar sounds and in the same grouping as Kuvi & Kui inside South Dravidian II. Should the macron ¯ U+0B55 ୕ really be in the page title instead of the |head= parameter? Its usage may not be universal and may be frequently omitted such as in the Odia-Kuvi Sabdakosh. Kutchkutch (talk) 00:11, 19 June 2023 (UTC)

If we do find real <ai, au> we could use ଆଇ ଆଉ and ଐ ଔ for Odia loans with <oi ou>. As for the macron, it has been used in many documents since the 80s so it would be better to distinguish the length as there seems to be many minimal pairs with vowel length. AleksiB 1945 (talk) 12:05, 19 June 2023 (UTC)