Quoth Christian Schulte <cs@schulte.it>:
> 
> I did not mean to be disrespectful with this. The same way OpenBSD went
> from 32 bit time_t to 64 bit time_t, OpenBSD would need to go from 8 bit
> char to 32 bit char to get "full" unicode support. That's not that easy.

It's also not a very good solution; keeping the native encoding in utf8 is
more performant, less error prone, and doesn't encourage people to assume
one codepoint is one glyph (it isn't -- you can get arbitrary numbers of
codepoints that need to be treated as a unit by text editing routines).