[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Bug#226593: ITP: prime -- a Japanese PRedictive Input Method Editor

On Sat, 2004-01-17 at 15:22, Hidetaka Iwai wrote:

[ Adding debian-i18n-english to the CC list for obvious reasons. ]

> Upstream author suggests the following description.
>  PRIME is a Japanese PRedictive Input Method Editor.

This looks like a good short description by dropping "PRIME is a".

>  PRIME predicts user's input words using the knowledge of natural

PRIME predicts the user's input using knowledge of natural

>  language and the history of user's operations, and reduces the cost of

languages and the user's previous input to reduce the difficulty of

(Rationale: Languages, because multiple languages can be supported.)

>  typing by the user.  For example, if a user wants to input

typing. For example, when the user

>  "application" and types "ap" as the beginning characters of the word,

types "ap", intending "application",

>  PRIME might predict some candidate words like "apple", "application",

PRIME might suggest "apple", "application",

>  "appointment", etc...  And then the user can input "application"

and "appointment." The user then

>  easily by selecting the word from the candidate words by PRIME.

easily selects "application."

>  PRIME works for Japanese at this stage, although it essentially can
>  deal with other languages.

Currently, PRIME only supports Japanese.

BTW: I am only minimally familiar with Japanese, but does PRIME use
Romanji, or something else? If that isn't a question whose answer is "X,
DUH!" to someone who knows Japanese, you might want to include it here.

Also, if the answer is "yes", how hard would it be to substitute in a
European-language dictionary?

Attachment: signature.asc
Description: This is a digitally signed message part

Reply to: