For things like centralNotices and fundraising banners, the software doesn't support
automatic fallback or variant conversion (yet). So we usually translate a simplified
version and a traditional version, then copy&paste then into all the variants.
Otherwise users with their interface set to things like zh-cn or zh-tw will get their
messages in English.
IIRC Liangent was looking into variant fallback (or conversion, can't remember) for
CentralNotice, but said to me once that the relevant code was deep inside and there was no
easy way to implement them.
Benjamin Chen / [[User:Bencmq]]
On 17 Nov, 2012, at 7:17 PM, Quark Zhang <wulaquark(a)gmail.com> wrote:
I can not quite understand why should Chinese divided into 9 parts in translation. they
are zh, zh-cn, zh-hans, zh-hant, zh-hk, zh-mo, zh-my, zh-sg, zh-tw. when I translate an
information to Chinses, I seem to must edit all this 9 versions. I want to say that is the
waste energy of translator. Because this 9 versions are consistent in essence. The
differences between them just are individual words. It just like metro and subway in
english. Editing 9 times for one information not only reduce efficiency in 9 times, but
also increase the chance of errors in 9 tims. I may not know some simplified operation
behind this, but I think the problem may indeed exist. It is just word conversion between
simple and tradition Chinese in the Chinese Wikipedia, instead of editing many versions of
My suggestion is translating zh-hans, zh-hant at frist. And finished or published,
default zh-hans=zh-cn, zh-my, zh-sg; zh-hant=zh-hk, zh-mo, zh-tw. At the same time, zh-cn,
zh-my, zh-sg, zh-hk, zh-mo, zh-tw's status is need to proofreading, in order to amend
the wording of them. At last, I really do not know where should be zh version used.
Translators-l mailing list