I saw a commit (#5422) that contained a lot of spelling changes, such as "Unpublished" to "Un-published". In many cases, the addition of the hyphen is within a gettext() function. However, I didn't see the locale files get updated. Would they not require updating to add those hyphens also, since the original text key will no longer match?
A translator's work is never done! I was just thinking that small changes like the hyphenation would break the translated words, when the translation itself is often "close enough", or may not need need the hyphen in the target language anyway.
Are there enough active translators to make those changes to the officially-included languages before the 1.3.0 release? If not, maybe changing the hyphenation of the original strings in the translation file so the translations still match up would be helpful, to preserve as many of the translated messages as possible? Someone using a language other than English will have a lot of English strings showing up otherwise.
Then again, I suppose it's true that the translations should be "right", not just "close enough", so it's still a necessary step.
Anyhow, that's my thoughts on the matter - you're the boss! :-)
I know I am maintaining the German translation myself for two years...;-) (And I really would not want to start a new one yet with around 3000 strings!)
The gettext translation program Poedit I use and we recommend is actually clever enough to notice most strings with minor changes and marks them as "fuzzy" instead of "untranslated". It keeps a kind of internal catalog, although not always 100% accurate.
We do not have a template translation file, it is recommend that every one updates the strings via parsing themselves. That is quite easier than updating those files manually. If you are interessted you can read on the translation tutorial.
The official included languages actually vary, depending on if they are ready. But we have a great group of some real frequent translators.
I ran POEdit and it gave me a list of strings it was going to delete as obsolete - I feared this would delete strings that were close but not exact, rather than letting me match 'em up again. I suppose the "fuzzy" strings would not get deleted? I'll play with this more to learn it.
The only language I speak fairly well is English, so I'm afraid I won't be of much help with the others!
Sure, since Zenphoto speaks US English mainly you could do a UK English variant..;-) (my English is probably an "awful" mix of both..;-))
Yes. Poedit removes obsolete strings but it most always readds the changed ones. And if it has some similar stuff in the internal catalog it marks them as fuzzy, which means he suggests a translation. Sometimes good, sometimes really wrong..;-)
A great tool and being on Mac the only one anyway.