DeeDeeG
commented
6 years ago That sounds like a step up compared my idea: I was thinking we could do a best guess with Google Translate, then find a reviewer to see if this is correct. (The only advantage there is if we can't find a reviewer, we still have something that might be close.) But assuming we do have a reviewer/translator, I think your suggestion is better (contacting the reviewer/translator as the first step, and see if they can do a fluent translation.)
Maybe (as another suggestion:) we could just put out a general call for translators on social media when we have new text, but I think having a list of translators is probably the most reasonable approach.
P.S. If we ever need to ship commits with untranslated English text, for whatever reason, it would be a good move to paste the untranslated English text into the non-English translation files, which would show the original English text rather than the translation key names (e.g. "resources-dropdown-label" -> "Resources").
Edit: Actually, I think all strings fall-back to the default language (in our case English) in production. (Strings don't auto-fallback during development, so as to make the lack of translation more obvious.) So the suggestion in the crossed-out text above is not needed.
Intimaria
Every time we write new text, we're going to need to add translated versions of it. I'm curious how we want to handle this. My first thought it to have a list of translators and ping them every once in a while to do a bulk update of new text.