Depressing to see that the Hare language specification https://harelang.org/specification/ only includes identifiers made from a-zA-Z0-9, we are in 2022 and UTF identifiers are fine... #harelang
@norbu
Unicode source code has an extreme amount of weird cases to deal with. Right-to-left, variable width, semanticly equivilent sequences, etc. For a serious low-level langauge which has subtle aspirations to become a lingua franca like C, I think they made a sensible choice.
If the Unicode consortium were much more conservative and sensible, it might have been possible.
No emoji variables, sorry Javascripturds.
@norbu
I would entertain some basic extended latin, along with some Greek, Cyrillic, Hiragana, Katagana... but then we'd have to draw the line somewhere, which might rub some peoples the wrong way.
I can see that enforcing ASCII is anglophone-centric. Considering all things, I think assuming English as a meta lingua franca for programming is probably the least problematic way of doing things for everyone. Consider that Hare libraries will be used by others.
@torresjrjr thanks for the comments, but I am not thinking of say accented characters, or CJK, which would be extremely useful and aren't hard to deal with. I don't think it is a good decision, but for sure it is a safe and simple decision.