While most of the Unicode specs looked quite great, it seems like no implementation actually implemented the "grapheme" level of it.
Please, please, please language implementers do it.
I know it's hard, it's difficult, it's frustrating, but please stop treating Strings as "a bunch of bytes/chars/ints in an Array".
I would really love to see a language which gets Unicode right. An implementation where getting the length of a String is _always_ correct, not just "we just take the array length" or "use only ASCII then". The same for indexed access, comparisons, etc.
Perl 6 could be the first programming language with Unicode support ever. That would be really great.
The Perl 6 Tablets provide documentation of how Perl 6 works, not a specification for how it should work. What do you think the tablets should contain to help deal with your concerns?