Template talk:sa-IPA

Documentation, please!

@JohnC5: could you write a documentation page, please? In particular, what does the parameter a= do, and how do we indicate Vedic pitch accent? —Aɴɢʀ (talk) 13:09, 9 January 2017 (UTC)

@Angr: Of course! I completely forgot to do this. Is this documentation sufficient for the present need? —JohnC5 21:15, 9 January 2017 (UTC)

@JohnC5: Yes; thanks! —Aɴɢʀ (talk) 08:38, 10 January 2017 (UTC)

Wow

This is pretty cool. A lot better than my MOD:hi-IPA that I don't use ever. —Aryamanarora ^{(मुझसे बात करो)} 03:51, 14 April 2017 (UTC)

@Aryamanarora: Thanks! If you have any changes you'd like to see, please tell me. Also, I'd love if you started adding these to entries. I made it but have not added it that many places. We also had some questions about what form should be used, be it the lemma (which is often not a real form) or the nominative singular (which often is not the page name). —JohnC5 04:25, 14 April 2017 (UTC)

This template should give the pronunciation of the page name, whatever that may be. It should be able to be used on noun lemma forms, nominative case forms, genitive case forms, all verb forms, and so on. —Aɴɢʀ (talk) 06:45, 14 April 2017 (UTC)

@Angr: To be clear, this template can render any input (theoretically). The question whether we should use it for the lemma forms, despite the fact that the lemma forms are often not real forms occurring in the language but a grammatical abstraction. —JohnC5 14:21, 14 April 2017 (UTC)

@JohnC5: So what are we using? I've seen you using the lemma form, but I've been using the nom. sg. for the past few days. —Aryamanarora ^{(मुझसे बात करो)} 13:04, 10 May 2017 (UTC)

@Aryamanarora: See there are two primary factors here. On one hand, the stem-form of nouns is a real form. On the other hand, I am lazy and don't want to keep copying the nom.sg. The answer is that I should probably stop being such a whiner and just copy the nominative singular. @Angr, CodeCat, what do y'all think? —JohnC5 14:39, 10 May 2017 (UTC)

@JohnC5: I'm not sure what you're talking about. Do you mean that in the ===Pronunciation=== section of वाच् (vāc), for example, you want to list the pronunciation /vɑːk/ (representing the nominative वाक् (vāk))? Or do you mean that वाच् (vāc) should have no pronunciation section at all, but only the entry वाक् (vāk) (when it gets created) should? I think it would be very confusing to users for us to say that वाच् (vāc) is pronounced /vɑːk/; I'd be opposed to that. The second option is doable, though we should still give the pronunciation when it does happen to be a real form (e.g. वृक (vṛka), which is the vocative singular in addition to being the lemma). —Aɴɢʀ (talk) 14:51, 10 May 2017 (UTC)

@Angr: Yeeah, this distinction is starting to get too subtle for the average user. I might be in favor of just having the pronunciation always be the entry name regardless of whether it is real or not. Another idea would be that, if the override parameter to {{sa-IPA}} is called and it is different than the entry title, we display the form whose pronunciation we are showing. So, if on the page for वाच् (vāc), we showed the pronunciation for वाक् (vāk), the module would display वाक् (vāk) somewhere so the reader would know what form was being shown. —JohnC5 15:19, 10 May 2017 (UTC)

(edit conflict) @Angr, JohnC5: Why not just have a qualifier and show both nom. sg. and stem form? E.g.

• IPA^(key): /vɑːt͡ɕ/
  • वाक् (nom. sg.) : IPA^(key): /vɑːk/

preferably in a dropdown. —Aryamanarora ^{(मुझसे बात करो)} 15:22, 10 May 2017 (UTC)

I think John's objection is that /vɑːt͡ɕ/ doesn't represent any actual word of the language. —Aɴɢʀ (talk) 15:27, 10 May 2017 (UTC)

Pronunciation of Diphthongs (Other improvements)

Although it is great that there finally is a pronunciation template finally available for Sanskrit in Wiktionary, this template needs some improvements in its coverage of Vedic pronunciation. Although most cases of "o" and "e" were indeed original diphthongs, they do not seem to stand for original diphthongs in cases where *s is lost before a consonant, e.g. पुरोऌआश्/पुरोडाश् (puroḷā́ś/puroḍā́ś), मेधा (medhā́).

The template also seems to fail to mark non-independent svarita vowels that fall after udatta vowels. Further, it fails to note whether the pitch on accented long vowels is rising (Greek accute), level high tone, or falling (Greek circumflex.) Please answer back if you need me to help explain better.Nayrb Rellimer (talk) 00:29, 25 March 2018 (UTC)

@Nayrb Rellimer: Thanks. I'm fully aware of all of these problems. I need to massively refactor this code, in part to take IAST inputs and also to better show recitation practices for Vedic and Classical stress accent. The only one of these that is a little problematic is the Proto-Indo-Ayran *z/*ž becoming lengthening the preceding vowel, in part because it's difficult to know how to input that correctly (mezdhā́? mĕdhā́?) but also because it's hard to know what the synchronic prunciation of such a section might be. Looking at Sandell, one might expect something like [əː] for both cases. —*i̯óh₁n̥C^[5] 01:28, 25 March 2018 (UTC)

How does one know they still were diphthongs in Vedic anyway? Guldrelokk (talk) 10:53, 7 June 2018 (UTC)