At 15:36 -0500 10/2/03, Don Drake wrote:
>Alan,
>
>I've received this and similar requests from my customers as well as users
>of this script.
>
>My opinion on this has always been: If you want to not see the messages
>marked as spam, feel free to modify the return codes.  I won't support it,
>and I won't implement it on my server.  The reason being is that all it
>takes is *one* message that's a false positive to screw everything up.  I
>don't want to be responsible for that mistake.
>
>This script is GPLed, so modify it all you want and assume the
>responsibility of potentially lost emails.

To all,
First off, I must thank Don for an incredibly effective, well running script.

What I've done, which I feel is a decent compromise, is to block 
truly egregious spam (with a score of 10 or higher) by checking for 
the star level in the X-SPAM Level heading, marking this as SPAM TO 
DELETE, and exiting with a reject message exit code. We've found that 
with Spamassassin, you really have to work at it to get a score of 10 
(even if the message is explicitly only SPAM related). Non-spam 
content at this score is essentially unheard of. All other spam 
simply blocks the remaining filters, but is passed on to the user. We 
then educate our clients about setting up appropriate filters in 
their email client to direct the SPAM messages to a special folder, 
and review it periodically.

This kind of fix is certainly something that a user would want to be 
very sure of before implementing, so it may not want to end up in 
your script. I imagine there should be any easy enough way of setting 
this up as a configurable option, if enough people are clamoring for 
it.

I've also modified the command line params in the spamc line in the 
Perl file so I can have separate user configuration files for each 
user (allowing for custom set spam scoring, custom auto-whitelisting 
and blacklisting, etc on a user-by-user basis).  Unfortunately, it 
doesn't seem possible to have user configured rules (but we hardly 
ever get requests from individuals to block individual words anyway).

One thing you may consider for a future upgrade is a configuration 
file, like in Peter Lindeman's AntiVirus filter. That would allow you 
to offer different configuration options (i.e. for more spam hating 
mailing hosts).

Thanks again for a great script,
Toby Reiter


-- 
Toby Reiter                          mailto:[EMAIL PROTECTED]
Breezing Internet Communications     http://www.breezing.com
1106 West Main St                    phone:434.295.2050
Charlottesville, VA 22903            fax:603.843.6931
-
To unsubscribe from this list: send the line "unsubscribe xmail" in
the body of a message to [EMAIL PROTECTED]
For general help: send the line "help" in the body of a message to
[EMAIL PROTECTED]

Reply via email to