Well, true, the MNEMONIC issue is the only one people are arguing about,
but given that metamail just treats "charset" as a parameter to be
passed on to the viewing program, it really makes no difference to
metamail what the final set of "approved" character sets is, so I didn't
worry about that. If the literal string "charset" were changed to
"character-set", that would make the early release of metamail more
problematic. Changes to the set of approved values for the charset
parameter, however, don't affect the metamail code at all. I didn't
mean to belittle this issue, it just doesn't affect the metamail
implementation. -- Nathaniel