jonathon.blake at gmail.com
Sat Jan 15 18:16:17 CST 2005
>standardized on using the word folders while all command line
a) That is a _style guide_ issue, not a translation issue.
b) That relates to creating documentation, not translating a User Interface.
> A bad translation helps.
If the person is only fluent in the target language, maybe. But if
that is the case, they shouldn't be doing any translation. They can
do the other stuff, like write _original_ documentation, or fix the
language specific bugs.
>Often one of the trouble when doing translations is remembering what
the translation of a word is, not how it's used.
This gets into how fluent are the bilingual speakers. If they don't
know the vocabulary, they aren't fluent. Just because somebody claims
to be bilingual, does not mean that they are.
You have to be willing to test, and reject people who volunteer to
work on an L10N Project. [A case can be made for rejecting 80% of the
volunteers for an L10N Project.]
> most first time translators make a lot more mistakes in their translation than ppl who have translated a lot.
True, experience helps.
>That is why I think it would be important that we could mark
translations ... at least three stages,
[Though I don't really agree with MAT part.]
> Wikipedia has long articles, I understand why you think a long
po file might be the same.
The theory behind Wikipedia, is that with time, articles become more
accurate. That has not turned out to be the case. I could have cited
wiktionary, or a number of other, similar projects with equal ease.
Wiktipedia simply has the best documented "lack of quality control
despite multiple eyeballs".
More eyeballs do not mean greater accuracy.
> It would also help if we had comment fields, so that people could explain why they made the change.
That part does make sense.
Monolingualism is a curable disease.
More information about the rosetta-users