In challenges on this site, particularly code-golf, answers are sometimes given that you can neither read nor test. If the answer is, for example, the shortest, should you still accept it?
As an example, consider code-golf answers written in APL. It is likely that I won't be able to read any of it and there is no free APL compiler/interpreter as far as I know so I can't test it either. Should I just believe that it does whatever the answerer says it does and award the win?
I don't think you can exclude these by default. For example, if an asker only knows C-like syntax languages, they're going to balk on many, many answers here, regardless of whether there's an interpreter "somewhere". Answerers can't read minds (well, most of them), so they won't even know up front which languages are "good". That's simply not fair.
The community in general frowns on language-specific challenges (unless there's a good reason), and I don't see how this is much different. The vast majority of answers are in languages that have an interpreter easily available (glare at Mathematica).
If you need an interpreter/compiler/environment, there's a large list of them here on meta. If there's an answer in a language you don't see on the list (APL?), consider leaving a comment asking them to add it to the list.
{} style functions which are not supported by GNU APL. Use tryapl.org.
\$\endgroup\$