Is `en_US.utf8` canonically wrong?

The output of locale seems to distinguish between upper and lowercase:

% locale -a 
C
en_AU.utf8
en_US.utf8
POSIX

Almost everywhere else in the wild seems to use the uppercase .UTF8, however glibc normalises locales to lowercase, and comments:

There is no standard for the codeset names.

What is the standard that defines the form <language>_<country>.<codeset>? Is it really silent on code page capitalisation?

Go to Source
Author: Tom Hale