Module talk:ko-conj

Bugs

Seems to not handle 아니 + 어 → 아니에 contraction correctly. Notaz (talk) 15:50, 1 January 2019 (UTC)

I tried correcting it, please check it out. Kanawl (talk) 02:46, 24 March 2020 (UTC)

Speech levels?

For the sake of reference, should conjugations be generated for alternative speech levels as well? For example, the 하게체 and 하오체 conjugations are missing. --187.188.14.46 23:41, 18 October 2020 (UTC)

North Korean inflections are missing

(Notifying TAKASUGI Shinji, HappyMidnight, LoutK, Karaeng Matoaya, B2V22BHARAT, Quadmix77): Hi. I got hold of an old North Korean language textbook (for Russians, image format) and now I am aware of some grammar difference

Specfically: medium-formality verbs:

1. 가오 (gao) (NK) = 가요 (gayo) (SK)
2. 먹소 (meokso) (NK) = 먹어요 (meogeoyo) (SK)

Do you think it's a good idea for templates to be enhanced for North Korean inflections, not sure if it's hard. Some note I found in this blog

@Suzukaze-c, Eirikr: Please let me know if you want to be added to the Korean notification group. --Anatoli T. ^{(обсудить}/^вклад) 04:25, 13 November 2020 (UTC)

It’s not a North Korean language but haoche (하오체), which has almost disappeared in daily speech in South Korea. — TAKASUGI Shinji (talk) 08:05, 13 November 2020 (UTC)

@Atitarev: @TAKASUGI Shinji is correct, and these are still quite common in certain registers of writing in South Korea. For example, school exam questions will say 다음 방정식의 해를 구하시오, 다음 중 틀린 것을 고르시오.

The NK standard language's verbal paradigm is effectively identical to the SK standard verbal paradigm because they're both based on the Seoul prestige dialect, so there is no need to add anything for it. (The actual Pyongyang dialect, which the NK language is not based on, has a very different paradigm.) The biggest priority for Module:ko-conj—which is still not that much of a priority to be honest—is adding dialectal conjugation, beginning with Busan.--Karaeng Matoaya (talk) 08:20, 13 November 2020 (UTC)

@Karaeng Matoaya, TAKASUGI Shinji: Thanks for responses. It seems you just don't want to make this a priority. I think we want Busan inflections but we also want NK inflections, even if they are based on the forms, which almost disappeared in SK. The textbook I have, uses -o/-so endings throughout and it doesn't even mention the medium-formality ending, so common in SK. So, it seems fair to say, they are currently equivalent. If there is an overlap, they can be labelled accordingly, e.g. haoche or NK. --Anatoli T. ^{(обсудить}/^вклад) 08:39, 13 November 2020 (UTC)

@Atitarev, they have not almost disappeared in SK; they are quite common in certain registers and everybody knows them. They are not "NK inflections" specifically—no such thing really exists—and they are currently not in the automated chart only because Module:ko-conj produces only four out of the six speech levels, presumably for legibility reasons.

In the North Korean standard language, 하오체 (haoche) is mainly male speech and 해요체 (haeyoche) is mainly female speech, whereas in the modern Seoul standard 하오체 (haoche) is not really used in speech and 해요체 (haeyoche) is gender-neutral. I'm not sure why your textbook doesn't mention the 해요체 (haeyoche) but that's a failing of the textbook. Some examples from twenty-first-century NK fiction:

과장 동무, 이번에 가면 그걸 가져 올 수 있을까요? (<5시간 40분>)

아, 유경 선생이 여기 있었군요! (<행복의 무게>)

지배인 아저씨, 목말라 죽겠어요. 물이 없어요? (<비결>)

Furthermore, the NK manual of style <조선말례절법> says more and more men are using the 해요체 (haeyoche) nowadays, so it may just be that the phasing out of the 하오체 (haoche) was just delayed by a few decades compared to the South.

If you want to add two more columns for 하게체 (hageche) and 하오체 (haoche) feel free to, but they are not something especially North Korean and must not be marked as such.--Karaeng Matoaya (talk) 09:02, 13 November 2020 (UTC)

@Karaeng Matoaya: OK, thanks for the clarification and examples. Wiktionary happens to be, as it seems, the only online dictionary, which covers many inflected forms, which is great for learners. Our tables say "Selected forms of the verb/adjective ...", which is fine but the modules haven't been enhanced in a while, like making options to suppress honorifics where inappropriate, adding different styles are good to have. --Anatoli T. ^{(обсудить}/^вклад) 09:17, 13 November 2020 (UTC)

@Atitarev, thanks for the reply. I agree that 하게체 (hageche) and 하오체 (haoche) should indeed be added into the module (just as we have past historic for French), even if the chart gets crowded. Ideally the chart should be toggleable, with a way to have the chart display only forms in the desired speech level. This would solve the space issue, and also resolve the potential confusion that comes from putting forms which are compatible with all speech levels in the columns for specific speech levels. I'm not sure if this is technically feasible, though.

In the meantime I have created an entry for 하오체 (haoche).--Karaeng Matoaya (talk) 09:53, 13 November 2020 (UTC)

@Karaeng Matoaya, TAKASUGI Shinji, HappyMidnight, LoutK: Thank you, @Karaeng Matoaya. I support gradual introduction of a more comprehensive template for Korean verbs and adjectives. Being "crowded" is not an issue, if the design is good and splits the display into various speech levels, for example.

Re: your earlier comment "... they have not almost disappeared in SK; they are quite common in certain registers and everybody knows them". By "everybody" you probably mean native Koreans or advanced speakers, I guess. It seems certain speech aspects of inflections are unavailable online or in grammar references as a resource for foreign learners. All the more it is important to increase our coverage. --Anatoli T. ^{(обсудить}/^вклад) 02:46, 16 November 2020 (UTC)

Hortative forms of 있다 are missing

@Karaeng Matoaya: Hi. 있다 (itda) is missing hortative forms, such as 있자 (itja). I've come across 가만 있자 (gaman itja, “just a minute, hold on (roughly)”). Perhaps inclusion of imperative forms, such as 있어 (isseo) is merited as well. --Anatoli T. ^{(обсудить}/^вклад) 02:36, 31 December 2020 (UTC)

@Atitarev, yes, it's strange that they're missing. Both hortative and imperative forms are as common for 있다 (itda) as for any other verb (여기 계속 있자, 어디 가지 말고 거기 있어, etc.) and they should definitely be there. Unfortunately I don't really know how to edit this module.--Karaeng Matoaya (talk) 02:44, 31 December 2020 (UTC)

@Karaeng Matoaya: Thanks. As long as we agree that this is required, we can think about how to enhance it. I can't edit either. I would really like to add these and what I also requested at Module_talk:ko-conj#North_Korean_inflections_are_missing above. Calling other Korean editors to confirm that this is what we want: @TAKASUGI Shinji, HappyMidnight, LoutK, B2V22BHARAT, Quadmix77.

If the request is formulated correctly then this could be added to WT:GP and we can ask our Lua experts - e.g. User:Benwing2 or User:Erutuon. --Anatoli T. ^{(обсудить}/^вклад) 02:53, 31 December 2020 (UTC)

@Atitarev: Yes, they should definitely be added. LoutK (talk) 04:08, 31 December 2020 (UTC)

@Karaeng Matoaya, LoutK: Thanks, so we need these forms to be added (1 formal non-polite, 2 informal non-polite, 3 informal polite, 4 formal polite):

1. Hortative: 1 있자 (itja), 2 있어 (isseo), 3 있어요 (isseoyo), 4 있읍시다 (isseupsida).
2. Imperative: 1 있어라 (isseora), 2 있어 (isseo), 3 있어요 (isseoyo), 4 있읍시오 (isseupsio).

They are probably simply suppressed for this verb. We need the honorific forms as well right, e.g. 있으십시오 (isseusipsio)? Do all honorific forms exist? --Anatoli T. ^{(обсудить}/^вклад) 06:04, 31 December 2020 (UTC)

I see that it uses |adj=, not |verb=, since it behaves differently. I can't simply replace |adj= with |verb=, it's a special irregular case. --Anatoli T. ^{(обсудить}/^вклад) 06:11, 31 December 2020 (UTC)

있다 (itda) is suppletive and you use 계시다 (gyesida) for honorific forms. So it’s not *있으십시오 (isseusipsio) but 계십시오 (gyesipsio). We should have an option to disable forms with -시-. (A module on the French Wiktionary has such an option.) — TAKASUGI Shinji (talk) 09:56, 31 December 2020 (UTC)

@TAKASUGI Shinji: Thanks, I am aware and 있다#Conjugation links to 계시다 (gyesida). It looks 있으십시오, etc. is attestable though. We do need a way to suppress -시- forms as well. --Anatoli T. ^{(обсудить}/^вклад) 00:56, 1 January 2021 (UTC)

Problem with -ㅡ다 verbs/adjectives

Generally, if the stem has more than two syllables and the vowel right before the ㅡ is ㅏ, ㅑ, or ㅗ, then ㅡ is dropped and ㅏ is added.

• 따르다 → 따라, 가냘프다 → 가냘파, 고프다 → 고파

This is well-reflected in the current module.

However, this rule does not apply to compound verbs/adjectives.

• 손쓰다 → 손써 (not 손싸 despite the ㅗ in 손), 싹트다 → 싹터 (not 싹타 despite the ㅏ in 싹)

This is because they consist of 손+쓰다 and 싹+트다 respectively and follow the conjugations of 쓰다 and 트다 (써 and 터 respectively).

There should be a way to handle cases like these. --216.197.202.130 21:37, 7 June 2021 (UTC)

Issues with the module

@Tibidibi: Hi. Are you able to describe in a good summary any issues with the verbs? I know there are a few issues, for very irregular verbs, like 말다 (malda), 이다 (-ida). User:Benwing2 has kindly agreed and might be able to apply a few fixes or rewrite the module later on (it's on his to-do list), to add suppression of honorifics, maybe add missing styles, etc. whatever is required and reasonable. There will be some work involved in creating some test cases. --Anatoli T. ^{(обсудить}/^вклад) 13:28, 27 December 2021 (UTC)

To add on, 이다 (-ida) has a wrong cause/reason form :-///, we also need to add the pure 냐 (-nya) forms for the 해체 (haeche) interrogative forms. AG202 (talk) 05:53, 7 November 2023 (UTC)

Adding explanations + exporting to Jeju

@Benwing2, (Notifying TAKASUGI Shinji, Atitarev, HappyMidnight, Tibidibi, Quadmix77, Kaepoong): Is there any way to add explanations as to what each section of code does? I want to export this to Jeju, but I can't even do it without knowing what each section does. It's so hard to follow. While we're at it, I'm wondering how we'd export this for letters that aren't a part of the Unicode Korean precomposed blocks, ex: at ᄒᆞ다 (hawda)? How would we be able to parse it? AG202 (talk) 21:04, 15 September 2023 (UTC)

@AG202 Unfortunately I haven't worked on this module, so I'd have to figure it out myself to add the comments. I agree it should be commented. The use of raw numbers and subtracting numbers from code points is bad practice, e.g. what does 28 mean on line 737? Benwing2 (talk) 21:27, 15 September 2023 (UTC)