Let's learn about input methods

1 réponse [Dernière contribution]
Connochaetes

I am a member!

Hors ligne
A rejoint: 12/13/2017

> I wonder how about other language setting. Is that as much as difficult/troublesome to type Cyrillic or Hangle?

The great thing is that you can usually choose:

For Russian, for example, GNU offers input methods that approximate an actual Russian standard keyboard layout. These have Cyrillic letters arranged randomly (much like the QWERTY layout has Latin letters in random order), completely unrelated to regular Latin layouts and hence difficult to memorize. But you don't have to use this. Instead, you can choose an input method that works "phonetically" from the Latin keyboard, which means, for example, pressing the key labelled "M⃣" will input the Cyrillic character м which in Russian sounds like an English M; pressing "F⃣" will input the letter ф that sounds like an English F, and so on. I think this "phonetic" method based on Latin letters is what most people in Japan prefer to type Japanese, no?

You have the same choice for Hangul:
You can choose an input method working "phonetically" from Latin letters (apt install ibus-hangul → log out → log in → run ibus-setup → "input method" tab → Add → Korean → Hangul → Preferences → dropdown list → "Romaja"). Or you can choose one resembling the standard layout used in Korea:

The most popular input method in Korea doesn't use Latin letter sounds, so it's not of the "Latin-phonetic" kind. It's called dubeolsik (2벌式) and it's the default for ibus-hangul. If I may simplify things a bit, this is how dubeolsik works: You start by pressing a key on the left half of the keyboard (where the consonants are, in random order), and next you press a key on the right (where the vowels are, in random order). By alternately hitting the left and right halves of your keyboard, you form syllable blocks. Let's say you want to write 니다 = [nida], consisting of ㄴ = [n], ㅣ = [i], ㄷ = [d], ㅏ = [a]. The first letter, ㄴ, is a consonant, so its key must be somewhere on the keyboard's left half; it's the S⃣ key, completely unrelated to the sound [n]. The second letter, ㅣ, is a vowel, so its key must be on the keyboard's right half; it's the L⃣ key, completely unrelated to the sound [i]. The sounds [n] plus [i] form one syllable [ni]; because each syllable gets its own imaginary rectangle in normal writing, people squeeze all letters deemed to belong to this syllable (in our case: the letters ㄴ plus ㅣ) into a single imaginary rectangle. On modern computer systems, the software will choose the corresponding precomposed syllable block (in our case: 니). Instead of using precomposed syllable blocks, you could also let the software do all the work of composing: With a suitable font installed (Trisquel 9 comes with Noto CJK ≈ Source Han), non-precomposed 니 should look more or less like precomposed 니. (If you're reading this from a browser, it probably won't show in your choice of font for Korean because the Trisquel forum discards <span lang="ko">…</span> markup. If they do look alike, try copying them and backspacing each, and you'll notice a difference in behaviour.) Some input method editors let you fine-tune how a string of key presses gets divided into syllable blocks, but whatever your settings, division is 100 % predictable, so there is no need for any internal candidate word lists unless you want Chinese characters.

To get 니다, you hit the following key sequence: slek On a QWERTY keyboard, the keys S⃣ and E⃣ are on the left half (to enter the consonants ㄴ [n] and ㄷ [d]), L⃣ and K⃣ on the right half (to enter the vowels ㅣ [i] and ㅏ [a]).

Syllables starting with a vowel sound are written beginning with a silent dummy "consonant" letter: sk dk ↳ 나 [na] 아 [a].

"Consonant" letters can also appear at the bottom of a syllable block, meaning this syllable ends in that consonant (or the next one starts with it): ek eks dk dks ↳ 다 단 아 안.

As we have seen with the silent "consonant", the terms "consonant" and "vowel" in the context of Hangul don't mean exactly the same as in the context of spoken Korean phonology, so a more precise terminology for the parts that make up a written Hangul syllable block is "lead" (初聲) or "L" for initial consonant, and "trail" (終聲) or "T" for bottom consonant. All written Hangul "syllable" blocks have the structure LV(T), where each of L, V and T may be either a simple letter (such as ㄷ or ㅏ) or a complex one (such as ㄸ = ㄷ + ㄷ or ㅐ = ㅏ + ㅣ).
Some complex letters are usually entered using the shift key: ek Ek ↳ 다 따.
Other complex letters are usually entered just by the sequence of their constituents, without the shift key: qnpfr ↳ 뷁.

To enter a very special character such as Chinese characters, you use a conversion key (ibus-hangul uses F9 by default; Korean keyboards have a key just for this, in the lower right corner of the keyboard, where many non-Korean keyboards have a "Right Ctrl" key instead). Example: To write 山 [san], you type its pronunciation 산 (T⃣K⃣S⃣), hit the conversion key (F9) and choose from the list of candidates (navigate using the arrow keys, press Enter or number key to confirm). With ibus-hangul, you can simply convert whole words, whereas on some systems you may have to convert each syllable block individually. If you like, have a look at the hanja dictionary used by ibus-hangul in /usr/share/libhangul/hanja
It contains very few mistakes, as far as I can tell. The other file in that directory is for non-Chinese special characters. For example: A⃣, F9, 6⃣ → ※

Another, less popular input method is called sebeolsik (3벌式) which comes in different versions.

When dealing with Hangul as used to write the Korean language, it helps to keep in mind that letters and sounds rarely have a 1:1 correspondence.

Post-1933 standard orthographies for Korean tend to spell morphemes consistently with little regard to how their phonetic realization is affected by preceding or following sounds, whereas pre-20th century writing was more of the "written as pronounced" kind.

A third input method for Korean is called cheonjiin (天地人 천지인), seen in very old Samsung mobile phones. By chance, all Korean "vowels" of the post-1933 orthographies can be unambiguously represented by the sequence of short strokes, horizontal strokes and vertical strokes used to write them. For example, ㅏ is the only "vowel" to be written using the sequence "vertical, short". Thus, three keys (one for each type of stroke), pressed in sequence, are enough to unambiguously enter all "vowels". Pre-1933 Hangul had ᅟᆔ or ㆌ (now obsolete) whose stroke sequence "horizontal, short, short, vertical" is identical to that of ᅟᅯ or ㅝ, so any mobile phones before 1933 would have needed a solution to this problem.

Connochaetes

I am a member!

Hors ligne
A rejoint: 12/13/2017

Oh, wait, you meant just the initial setup, not its use, right? Ha, sorry.