> What about
> \xHHHH or for double byte encodings
> \uHHHH or \UHHHHHHHH for unicode (or whatever java and c99 syntax is).

There are three problems with this.

1. Currently unvis(3) can always reproduce original string regardless
  of current locale setting. But above proposal make it impossible.
2. If \xHHHH means a wchar_t value, then it is not portable at all
  (for example, there are at least two major ways to represent
   multibyte EUC character as wchar_t).
  If \xHHHH just means multibyte sequence, it's better to just use
  current format of vis(3), IMHO.
3. How does unvis(3) handle \uHHHH in non-unicode locale which doesn't
  have any corresonding character with the \uHHHH?

My current opinion is it's better to just leave vis(3) and unvis(3)
as is. :-)
--
soda