News for hackers. Stuff that matters.
Re:Japanese Explained to Programmers by visarga on Tuesday January 31, 2023 @09:00AM (#34595052)
> copious use of lookup tables or prefix trees.

or language models

Re:Japanese Explained to Programmers by ickoonite on Tuesday January 31, 2023 @09:05AM (#34595116)
Sure, but it’s a messy model that tells you that 生 is pronounced nama or sometimes ki, 生きる ikiru, 生じる shōjiru, 学生 gakusei, 生む umu…
Re:Japanese Explained to Programmers by wst_ on Tuesday January 31, 2023 @09:21AM (#34595295)
All the examples you gave are pretty obvious to read because of suffixes, called okurigana, or the other kanji next to it. Problem arises only when the kanji stands alone and that doesn't happen so often as people think. Even when it does, most often than not, you can infer the reading from the context.
Re:Japanese Explained to Programmers by gray_-_wolf on Tuesday January 31, 2023 @09:47AM (#34595624)
わたし (私) and わたくし (私) is probably my favorite example. Good bless furigana. きょう (今日) and こんにち (今日) is sometimes fun as well.
Re:Japanese Explained to Programmers by glandium on Tuesday January 31, 2023 @12:53PM (#34598748)
だいにんき(大人気) and おとなげ(大人気) can be fun too. The latter usually is followed by ない, which wouldn't usually happen with the former, but the usual way I'm hit with these is in sentences where the word is at the end of the line (or better, when the line ends at 大人).
Re:Japanese Explained to Programmers by visarga on Tuesday January 31, 2023 @12:03PM (#34597814)
In some cases you need to read the whole phrase to know what a word meant and how to read it.