Today we have one big language file for any string in Limesurvey.

However, often people need to use an exotic language (for example Inuktitut - https://www.limesurvey.org/forum/can-i-do-this-with-limesurvey/119171-language-not-available).

So someone who needs to use this language might happily translate the respondent facing text strings into the respective language, and maybe even maintain it. But I doubt that someone who will just do a couple of surveys in such an "exotic" language like Inuktitut we surely desist to translate our language file, given that there is no easy way for them to distinguish between strings that are seen by survey respondents (which should be a lot less) from those that are part of the backend.

Even for other languages, it would make it a lot easier to keep Limesurvey ready for use, because translators can focus on the strings that face the respondents, which of course are a lot more important than the ones that face the survey admin.

For example for PT-BR there are missing strings. I won't be able to translate everything quickly, but I am sure that the respondent facing strings could be translated in a couple of days without stress.