http://www.iana.org/assignments/character-sets
and RFC 1428 have an encoding (charset) "unknown-8bit".
There is also the widely recognized "x-user-defined",
which means the same thing, afaik.
Both designate an 8-bit encoding where *nothing* is known
about its characters. Especially, unknown-8bit is *not*
necessarily a superset of US-ASCII.
Wouldn't it be useful (e.g. for certain HTML documents)
to have a charset parameter that is "half unknown"?
Such a charset parameter should denote an 8-bit encoding
that is a superset of US-ASCII and only the remaining half
(128 characters) is unknown.
Or am I mistaken and is this actually meant by
"x-user-defined"?