< In section 3, the sentence "However, conformant software is encouraged to
< check the version number and at least warn the user if an unrecognized
< MIME-version is encountered." is poor implementation advice and should be
< removed. The value of MIME-Version: is never going to change and readers
< should not waste code checking for its existence, much less verifying its
< value.
I disagree very strongly with this statement. There are other mail systems
out there that predate MIME AND use the Content-Type header in incompatible
ways. These systems are NOT going away in the near future.
If you receive a mail message that contains a Content-Type: header, and you
want to treat it as a MIME message, the ONLY way to start is to look for the
Mime-Version: header. If it is not present, then ALL BETS ARE OFF as to the
format of the Content-Type: header, and the format of the message. Such a
message may follow the MIME spec, and it may not. More than likely, it
won't. (After all, it didn't follow it enough to put out the Mime-Version:
header, how can you expect it to follow it enough to do any of the other
things that Mime specfies.)
If your mail reader doesn't check for at least the presence of the
Mime-Version: header, then you can expect it to break once in a while when
one of these messags crosses your system.
I won't comment on the usefulness of checking >>the value<< of the
Mime-Version header.
Tony Hansen
hansen(_at_)pegasus(_dot_)att(_dot_)com,
tony(_at_)attmail(_dot_)com
att!pegasus!hansen, attmail!tony