> Maybe, maybe not.  Why can't listserv just check to see if all message
> lines are valid listserv commands?  If just one line is found that is
> not a valid listserv command, then the whole thing is mail.  This
> would take care of 99% of the bogus things that are posted to lists by
> ignorant (read, uneducated) users.  (I can hear the objection now...
> "it'll mean having to scan the message body twice."  Only seldom.
 
I don't want my LISTSERV eating any more CPU than it really has to.
 
Since most of the junk is usually one line, can we settle for testing the
number of lines of text (how easy is that?) and if exactly 1, then test to
see if the first word is a valid command name/abbreviation.
 
If any mail then arrives that is junk, then we as a group can teach the
offending user about how to use LISTSERV in a way he will never forget :-)
It *IS* documented.  If the user is told by someone else, then we need to
get THAT person's name.