Talk:Interface guidelines
I want to contribute for the "Language Strings" title.
The prepositions in English usually does not correspond to meaningfull words in other languages. Say "3 of 10 is being displayed" string should be coded in the language file as is.
Wrong Approach:
Phrase 1: of
Phrase 2: is being displayed
Instead, the correct approach is:
Phrase: %s of %s is being displayed
Because the preposition may or may not exist in the translation as a word, therefore the translated string may also be shuffled, i.e. may not be starting with "3" for the specific example above.
I think, this problem can be avoided by setting a guideline for it like:
"Use whole sentences or sub-sentences as language strings, do not divide your sentence into words of prepositions to avoid dublication in translations. Because not all the languages have the same grammer and linguistics that English built on."
This way, the the translated UI would be much more understandable.
- Actually, this is already part of the coding guidelines - at least as I understand them - I have to say that Coding is not very clear on this point. Anyway, I think that Internationalisation should be a separate section of the coding guidelines, and we should keep Interface guidelines more focussed towards usability. I think I may take this discussion to the forums.--Tim Hunt 03:34, 5 April 2009 (UTC)
Language strings?
While I appreciate the modularity of keeping the language strings separate, I am not convinced that should be a hard and fast rule. If a valid language string exists in the core code then I see no reason to duplicate it. I get the idea, I'm just find myself not convinced. Peace - Anthony
