RBL checks and Spamcop

[Paul Welsh]

----- Original Message -----
From: "Matt" <hciss at HCIWS.COM>
To: <MAILSCANNER at JISCMAIL.AC.UK>
Sent: 01 December 2002 20:55
Subject: Re: RBL checks and Spamcop


> I think it is much better to bounce then just quarantine the messages.
If
> it does get a false positive at least the sender will know that
something
> went wrong and the message failed to go through.  Tagging is a good
option
> for power users but your typical user just gets annoyed and never
figures it
> out. We block in sendmail.cf with Spamcop and Ordb with decent
results.
>

Glad someone has something positive to say about SpamCop!  Using ordb
and spamcop is also what I intend doing.  I can quarantine (and delete
later) with ease because my MailScanner box relays incoming messages to
an NT box running MailSweeper which does the virus scanning.  This gives
me an easy graphical interface to quarantine and then delete or release
messages marked as spam.  Note that we had already invested in
MailSweeper before discovering MailScanner, so it seemed churlish not to
make use of it.

I agree completely about typical users never figuring out how to setup
rules to delete spam tagged messages, hence the idea of trying to find a
service such as SpamCop that can stop most spam from ever reaching them.

The problem with SpamAssassin is that it does tend to mark mailings from
reputable companies that they wish to receive as spam.  This kind of
mailing clearly isn't spam since the user has signed up for it.
Presumably, such mailings won't find their way into SpamCop's list.