I just saw a message that was forwarded to the ipv6@xxxxxxxx list by
spamassassin.
Apparently spamassassin decided it was spam but forwarded to the list
anyway with explanation of why it thought it was spam. Not only was
the message legitimate (it was an I-D announcement) but it reveals just
how bogus the criteria used by spamassassin were:
- it says the From field does not have a real name. Guess what, that's
perfectly legitimate.
- it incorrectly says that the To field has a malformed address because
it uses group syntax, even though this is a perfectly legitimate use of
group syntax.
- it trusts blacklists, even though these have been known to be
unreliable, to use dubious criteria, or even to misrepresent the facts
- it claims that the MIME boundary is indicative of spam, even though
it's also perfectly legitimate.
In other words, there's nothing wrong with the message, and it's not
spam by any stretch of the imagination, and yet it's being labelled as
"probably spam".
The last thing we need is to have such poorly chosen criteria used to
filter IETF list traffic.
Please consider this a formal objection to this practice. Even if
spamassassin is being configured to forward the messages, it's doing so
in such a way as to make the actual message very difficult to read.