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

Re: Temporarily mapping non-ascii->ascii characters in gettext

On 09/08/2007, at 3:58 PM, Christian Perrier wrote:

Do you think that finding such a "mapping" for non-latin languages it
also possible? What about all Cyrillic, Chinese, Japanese, etc. - based

Even for Latin languages.... French does not have such mapping system
for its accented characters. And I would bet that several other Latin
languages don't either. Indeed, among the Latin-based languages I know
about and where non ASCII characters are meaningful (and not only
there to show the tonal accent),

Um, Christian, the tonal accents are meaningful, at least in Vietnamese. They actually distinguish meaning. Words with the same letters but different tonal accents have different meanings.

only German uses such as complete
mapping system for its umlauts and "ss".

I actually feel somewhat the same than Damyan. Working the issue
around is probably the best way to...hide it under the rug

Solving the issue is far preferable. Inaccurate accents in my language would make it unreadable.

It's actually a challenge for students, to give them a piece of text in Vietnamese which has no accents, and ask them to predict from possible context which accents should be used. It makes reading a puzzle, rather than a straight-forward process.

from Clytie (vi-VN, Vietnamese free-software translation team / nhóm Việt hóa phần mềm tự do)

Attachment: PGP.sig
Description: This is a digitally signed message part

Reply to: