оновлення української розкладки

[schreik]

То вони як раз пишуть, що U+2019 варто використовувати як пуктуаційний апостроф, і в наведеному прикладі вони показують це "We’ve been here before", що є скороченою формою "We have been here before", тобто апостроф був використаний для об'єднання слів, а не є частиною слова.

В українській же мові, апостроф є невід'ємною частиною слова, що за своїми властивостями не відрізняється від літери.

[DalekiyObriy]

гмм, можливо таки MODIFIER LETTER APOSTROPHE краще
обіцяю подумати...

[DalekiyObriy]

поки що залишу, як є:
хоч семантично "MODIFIER LETTER APOSTROPHE" мав би бути кращим відповідником для української мови, на відміну від U+2019 з основних програм що я спробував жодна його нормально не опрацьовує, в інших технічних аспектах (упорядкування в локалі тощо) я теж переваг не знайшов

беручи це до уваги, а також те, що наявність двох різних апострофів, які виглядають абсолютно ідентично (принаймні, в більшості випадків) створить величезну купу проблем, думаю тут є сенс принести технічну правильність в жертву здоровому способу життя

[DalekiyObriy]

Оновлено статтю про розкладки на Вікіпедії
, зокрема додано опис і зображення варіанту «Unicode»

[samjavvat]

I'm working in a library and this debate is very interesting for me :

Last year, our library decided to put the original cyrillic caracters in our e-catalogue. I'm trying to do that for the ukrainian, but there is this problem with the apostrophe.

I've a preference for the "MODIFIER LETTER APOSTROPHE", because Ukrainian apostrophe seems to go with the it's Unicode definition, but as I understand, the Ukrainian practise is U+2019 or U+0027.

Have you any suggestion for our library? Do you think, that normal Ukrainian reader would type (or not) an apostrophe when he is searching an Ukrainian document?  If not necessery, the best for us would be definitely U+2BC.

Thanks for attention!

[noddeat]

As very few people use ua_unicode layout, normal Ukrainian user would type English apostrophe ('). But advanced users may also use this symbol (’). As for the library database, i suggest that your search engine will ignore the apostrophe (for sure this wouldn't spoil search results). So, if anyone will type "П’яний дебош", your search engine should search for "Пяний дебош" (and ignore apostrophes in the database, so that both "П'яний дебош" and "П’яний дебош" will satisfy the search criteria) but display the results with apostrophe:

1. В. Петренко. «Дивоцвіт каміння». К: Чорна кішка, 2012.
«І Сергій хутко влаштував нам п’яний дебош, ото була забава...»
2. П. Василенко. «Дискредитація неінтегрованих 4-вимірних ніштяків». Л: Мазох і сини, 2025.
«в такому разі, наша функція буде інтеґрованою на всьому просторі, і п'яного дебошу не влаштує.»

I'm working in a library and this debate is very interesting for me :

Last year, our library decided to put the original cyrillic caracters in our e-catalogue. I'm trying to do that for the ukrainian, but there is this problem with the apostrophe.

I've a preference for the "MODIFIER LETTER APOSTROPHE", because Ukrainian apostrophe seems to go with the it's Unicode definition, but as I understand, the Ukrainian practise is U+2019 or U+0027.

Have you any suggestion for our library? Do you think, that normal Ukrainian reader would type (or not) an apostrophe when he is searching an Ukrainian document?  If not necessery, the best for us would be definitely U+2BC.

Thanks for attention!

[samjavvat]

Unfortunately, your answer cannot be applicated in our library, because of the number of languages that use and cannot ignore apostrophes in our searching system (in english (with "I'm", for exemple), in french too).

But your suggestion helps me : with your explanation, I think, we could consider the Ukrainian apostrophe as we consider the transliteration of soft or hard sign in Russian romanization (ALA-LC) : it is not regarded as a normal letter, but as a special sign. User don't type it when searching, but system display it in results:
For ex.: when users are searching "Gor'kiĭ" (Горький), they have to type "Gorkii", and the result will bee "Gorʹkiĭ" (with the modifier letter prime 02B9 Unicode).
(But for an original Russain cyrillic search, they have to type soft sign : "Горький" (and not Горкий).)

It means that Ukrainian users would have to type the soft sign "ь", but not the apostrophe. Do you think this is possible? This idea is suggested using ALA-LC romanization : in this table, Ukrainian apostrophe is disregarded in transliteration:
ALA-LC transliteration of "п’яний дебош" is "pi︠a︡nyĭ debosh", without any apostrophe (users have to type "pianyi debosh").
In original scripts, display would be "пʼяний дебош" with 02BC (but users would have to type "пяний дебош").

What do you think about it?

[DalekiyObriy]

Unfortunately, your answer cannot be applicated in our library, because of the number of languages that use and cannot ignore apostrophes in our searching system (in english (with "I'm", for exemple), in french too).

But your suggestion helps me : with your explanation, I think, we could consider the Ukrainian apostrophe as we consider the transliteration of soft or hard sign in Russian romanization (ALA-LC) : it is not regarded as a normal letter, but as a special sign. User don't type it when searching, but system display it in results:
For ex.: when users are searching "Gor'kiĭ" (Горький), they have to type "Gorkii", and the result will bee "Gorʹkiĭ" (with the modifier letter prime 02B9 Unicode).
(But for an original Russain cyrillic search, they have to type soft sign : "Горький" (and not Горкий).)

It means that Ukrainian users would have to type the soft sign "ь", but not the apostrophe. Do you think this is possible? This idea is suggested using ALA-LC romanization : in this table, Ukrainian apostrophe is disregarded in transliteration:
ALA-LC transliteration of "п’яний дебош" is "pi︠a︡nyĭ debosh", without any apostrophe (users have to type "pianyi debosh").
In original scripts, display would be "пʼяний дебош" with 02BC (but users would have to type "пяний дебош").

What do you think about it?

I'd recommend to use one apostrophe for Ukrainian texts/titles - U+2019 seems to be becoming a standard but allow user to enter U+0027 just replacing it with Unicode one when doing a search (I'd rather not mix apostrophes and soft sign though).

[samjavvat]

The reason of my doubt with apostrophe (0027 or 2019) comes from my only Ukrainian author with it:

Ващенко, Юлія В’ячеславівна. (ISNB : 9789663642178)

I did a normal search with patronymic В’ячеславівна, В'ячеславівна and Вячеславівна in the online catalogue of the National library of Ukraine (keyword):

Вячеславівна : 246 results
В'ячеславівна: 353 results (with 0027)
В’ячеславівна: 0 results (with 2019)

You can see here that :
1) apostrophe is not always written (and for me, it's means perhaps not obligatory here)
2) 2019 isn't used

That's the reason of my doubt (2BC is like an apostrophe, but disregarded in search; 0027 or 2019 are obligatory for our search, like normal letters).

[Володимир Лісівка]

The reason of my doubt with apostrophe (0027 or 2019) comes from my only Ukrainian author with it:

Ващенко, Юлія В’ячеславівна. (ISNB : 9789663642178)

I did a normal search with patronymic В’ячеславівна, В'ячеславівна and Вячеславівна in the online catalogue of the National library of Ukraine (keyword):

Вячеславівна : 246 results
В'ячеславівна: 353 results (with 0027)
В’ячеславівна: 0 results (with 2019)

You can see here that :
1) apostrophe is not always written (and for me, it's means perhaps not obligatory here)

Apostrophe is obligatory. However, omission of apostrophe is common error for Russian speakers, which are about 40% or people.

2) 2019 isn't used

Site uses 8 bit encoding (Windows-1251), so it is not a big surprise. :-) It is technically impossible to use Unicode here.

That's the reason of my doubt (2BC is like an apostrophe, but disregarded in search; 0027 or 2019 are obligatory for our search, like normal letters).

You can use Google's mechanism - ' " U+2019 are mapped to common "apostrophe character", U+02BC is removed from search string.

[samjavvat]

If it's obligatory to write apostrophe, like a letter, then it's more logical to use U+2019 (because users will of course type it).

Thanks for answering!

[noddeat]

1) apostrophe is not always written (and for me, it's means perhaps not obligatory here)

no-no, aphostrophe is always written. Вячеслав and В’ячеслав are different names and are pronounced differently (the first one actually has come from Russian.)

[Михайло Даниленко]

Так, трошки по темі гілки.

Здружив ua(phonetic) із ua(unicode). Для цього довелося зробити додаткову розкладку ua(phonetic_unicode), яка є просто копією ua(phonetic), тільки як базову грузить не ua(basic), а ua(unicode), ну й доповнив її визначеннями для клавіш із цифрами із basic - вже звик до знаків пунктуації там.

Але є одна примітка. Краще за замовчуванням не вантажити level3(ralt_switch), це не дає змоги встановити перемикач мов на ralt - або порадьте як це зробити (ralt - us/ua, rctrl - 3lv). Коли я закоментував оте include - все завелося як треба (через grp:toggle,lv3:switch).

[samjavvat]

Вячеслав and В’ячеслав are different names and are pronounced differently (the first one actually has come from Russian.)

I understand it so : "Вячеславовна"(=Russian), В’ячеславівна (=Ukrainian) and Вячеславівна (was Russian, but now Ukrainian).
Is it ok?
 

[DalekiyObriy]

Але є одна примітка. Краще за замовчуванням не вантажити level3(ralt_switch), це не дає змоги встановити перемикач мов на ralt - або порадьте як це зробити (ralt - us/ua, rctrl - 3lv). Коли я закоментував оте include - все завелося як треба (через grp:toggle,lv3:switch).

насправді, це була вимога координатора xkeyboard-config, хоч це і неправильне, але у всіх інших розкладках завжди було типове lv3:ralt_switch тому прийшлося в ua(unicode) теж поставити, але в останніх X.org є прапорець, щоб цю опцію скасувати: lv3:ralt_alt (Right Alt never chooses 3rd level). Тоді йому можна призначити іншу функцію.