Our current support for the various two-byte Unicode encodings is deficient in
the following ways:

  (1) We can read unmarked two-byte Unicode streams but we cannot write them.
      That is, we have CharToByte converters for "UnicodeLittleUnmarked" and
      "UnicodeBigUnmarked", but the corresponding ByteToChar converters are
      missing.

  (2) We should add the following aliases in order to better support MIME
      types and the drag-and-drop subsystem:

          unicode-1-1-utf-8 == utf-8
unicode-1-1 == UnicodeBigUnmarked
iso-10646-ucs-2 == UnicodeBigUnmarked
x-utf-16be == UnicodeBigUnmarked
x-utf-16le == UnicodeLittleUnmarked

      These names come from the IANA charset registry with the exception of
      "unicode-1-1-utf-8", which is recommended in RFC 1641.

-- mr@eng 1999/5/26