More than 10 years ago I have seen a study in Taiwan that shows that when entering the correct pronunciation and tone a hitrate of more than 90% for sentences could be reached. This means the technology is there. Chinese in Taiwan who are trained to use computers tend to use CangJie and therefore do not have the issue. Chinese in Taiwan who are using phonetic input would normally never think that entering something like "ren4" would be interpreted by the computer as Chinese. They would automatically enter ㄖㄣˋ and get a selection list (users of ZhuYin) or ㄖㄣˋㄕˋ and get 認識 (users of Chewing). Microsoft has similar solutions, but I have given up on them years ago and so cannot tell you much about them. Uli On Tuesday, November 15, 2005 10:15, Tobias Burnus wrote:

Hello,

naively I'd expect that the hits using traditional Chinese and using simplified Chinese would be the same - except for the different character. But somehow the simplified version guesses quite well, which character I want whereas the traditional Chinese version does not.

Example: ren4shi (to know): a) Simplified: Typing "rensh" (or "renshi") suggest as first hit: 认识 b) Traditional: Typing "renshi" gives: 人士, 人事, 人世 and 人氏 and then comes the list of all "ren" hits followed by all "shi" hits; that way one can finally get 認識, but it takes much longer.

Can one improve the traditional version? Or do I do something wrong?

Tobias who is a novice in terms of Chinese

-- To unsubscribe, e-mail: m17n-unsubscribe@suse.com For additional commands, e-mail: m17n-help@suse.com