Could anybody please recommend me a kana touch typing tutorial software? Either for windows or linux, that doesn't matter. I don't need to learn how to write like a pro, just enough not to spend 30s looking for every character.
Huh, this was harder than it looked. I thought "what a retard, justfuckinggoogleit" and started plugging in all sorts of things into Google, but came up with approximately nothing. Nothing that I understood, anyway.
Maybe you should ask in /nihongo/ instead. (But of course, they'll probably only say "duh, nobody uses JIS layout, use romaji".)
but wouldn't direct kana input be twice as fast?
Yeah, but you have four rows of keys to deal with instead of three.
You better have big hands and a good memory.
The argument I've heard against kana input is that you will need multiple keystrokes anyway (for がざだ... instead of かさた...; to get small ゃゅょ; and for any combinations involving small っ), so it isn't really faster. I'm not sure I believe that line of reasoning myself, but who knows.
However, if you want to be macho and really impress the girls, you might look into learning T-Code. Two keystrokes to type any character, including kanji. That ought to speed things up quite a bit, I think. I believe you can use it in Microsofts IME, and it's obviously there in the various open-source Unix implementations too.
This site mentions it, and the reasons why kana-kanji comversion is bad: http://www.m17n.org/ntakahas/npx/aggressive/index.en.html
All that said, I'd really like to know how direct kana input compares too. :-)
Oh jesus. Even if you could enter any kanji with two keystrokes, remembering which two keystrokes corresponds to which kanji would be enough to keep you busy for a year.
Incidentally there is a reason why kana-kanji conversion is good. For one, it means that software could remember what reading was input when the kanji was created. This means that if the reading were remembered by the computer (which in most word processors it sadly isn't) then at a later point in time a computer could read it without having to resort to advanced AI.
>>7
True. (OTOH, I suppose kanji-kana conversion could in theory accomplish the same thing, if you wanted it to...)
Is there anything today that bothers with storing both the reading and the glyph? It would be very cool, but I don't see any reason why anyone would have bothered.