Rejected posting to MAILSCANNER@JISCMAIL.AC.UK (fwd)

Thu Dec 18 21:36:29 GMT 2003

Could you switch me from ballard at cae.wisc.edu (or something similar to that) 
to ballard at engr.wisc.edu so I can post?

Thanks,

-Jeff

------- Forwarded Message

Received: from cae.wisc.edu [144.92.240.11]
	by localhost with IMAP (fetchmail-5.9.11)
	for ballard at localhost (single-drop); Thu, 18 Dec 2003 14:43:41 -0600 (CST)
Received: from engr.wisc.edu (snapple.engr.wisc.edu [144.92.12.24])
	by cae.wisc.edu (8.12.9/8.12.9) with ESMTP id hBIKhKom020216
	for <ballard at mail.cae.wisc.edu>; Thu, 18 Dec 2003 14:43:20 -0600 (CST)
Received: from smtp.jiscmail.ac.uk (smtp.jiscmail.ac.uk [130.246.192.48])
	by engr.wisc.edu (8.11.6+Sun/8.11.6) with ESMTP id hBIKhDG26332
	for <ballard at ENGR.WISC.EDU>; Thu, 18 Dec 2003 14:43:13 -0600 (CST)
Received: from LISTSERV.JISCMAIL.AC.UK (jiscmail.ac.uk) by smtp.jiscmail.ac.uk 
(LSMTP for Windows NT v1.1b) with SMTP id <0.003CEB13 at smtp.jiscmail.ac.uk>; 
Thu, 18 Dec 2003 20:43:13 +0000
Date: Thu, 18 Dec 2003 20:43:12 +0000
From: "L-Soft list server at JISCMAIL (1.8e)" <LISTSERV at JISCMAIL.AC.UK>
Subject: Rejected posting to MAILSCANNER at JISCMAIL.AC.UK
To: ballard at ENGR.WISC.EDU
Message-ID: <LISTSERV%2003121820431283 at JISCMAIL.AC.UK>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="GGXXCRNBEARXMZXEbCFGOUQLFROSeI"
X-CAE-MailScanner: Found to be clean, Found to be clean
X-CAE-MailScanner-Information: Please contact helpdesk at cae.wisc.edu for more 
information or to report problems.
X-Spam-Checker-Version: SpamAssassin 2.61 (1.212.2.1-2003-12-09-exp) on 
	starburst.cae.wisc.edu
X-Spam-Status: No, hits=0.0 required=5.0 tests=none autolearn=no version=2.61
X-Spam-Level: 

- --GGXXCRNBEARXMZXEbCFGOUQLFROSeI

You  are  not  authorized to  send  mail  to  the  MAILSCANNER list  from  your
ballard at ENGR.WISC.EDU account. You might be authorized to send to the list from
another of  your accounts,  or perhaps  when using  another mail  program which
generates slightly  different addresses, but  LISTSERV has no way  to associate
this other account or address with yours. If you need assistance or if you have
any question regarding  the policy of the MAILSCANNER list,  please contact the
list owners: MAILSCANNER-request at JISCMAIL.AC.UK.

- --GGXXCRNBEARXMZXEbCFGOUQLFROSeI
Content-Type: message/rfc822

Return-Path: <ballard at engr.wisc.edu>
Received: from 130.246.192.53 by JISCMAIL.AC.UK (SMTPL release 1.0i) with TCP; 
Thu, 18 Dec 2003 20:43:12 GMT
X-RAL-MFrom: <ballard at engr.wisc.edu>
X-RAL-Connect: <starburst.cae.wisc.edu [144.92.240.2]>
Received: from cae.wisc.edu (starburst.cae.wisc.edu [144.92.240.2])
	by fili.jiscmail.ac.uk (8.12.8/8.12.8) with ESMTP id hBIKhAYY027114
	for <MAILSCANNER at JISCMAIL.AC.UK>; Thu, 18 Dec 2003 20:43:10 GMT
Received: from jalopy.cae.wisc.edu (root at jalopy.cae.wisc.edu [144.92.12.93])
	by cae.wisc.edu (8.12.9/8.12.9) with ESMTP id hBIKh2om020116
	for <MAILSCANNER at JISCMAIL.AC.UK>; Thu, 18 Dec 2003 14:43:02 -0600 (CST)
Received: from guinness.cae.wisc.edu (guinness.cae.wisc.edu [144.92.240.245])
	(authenticated bits=0)
	by jalopy.cae.wisc.edu (8.12.3/8.12.3/Debian-6.6) with ESMTP id hBIKh11I005588
	(version=TLSv1/SSLv3 cipher=EDH-RSA-DES-CBC3-SHA bits=168 verify=NOT)
	for <MAILSCANNER at JISCMAIL.AC.UK>; Thu, 18 Dec 2003 14:43:02 -0600
Received: from guinness.cae.wisc.edu (guinness [127.0.0.1])
	by guinness.cae.wisc.edu (8.12.9/8.12.9/Debian-5) with ESMTP id hBIKh1j0031225
	for <MAILSCANNER at JISCMAIL.AC.UK>; Thu, 18 Dec 2003 14:43:01 -0600
Received: from guinness (ballard at localhost)
	by guinness.cae.wisc.edu (8.12.9/8.12.9/Debian-5) with ESMTP id hBIKh1Wt031220
	for <MAILSCANNER at JISCMAIL.AC.UK>; Thu, 18 Dec 2003 14:43:01 -0600
X-Mailer: exmh version 2.5 07/13/2001 (debian 2.5-1) with nmh-1.1-RC1
To: MailScanner mailing list <MAILSCANNER at JISCMAIL.AC.UK>
Subject: Re: Bayes Poisoning? Spam with negative BAYES Scores 
In-reply-to: Your message of "Thu, 18 Dec 2003 12:10:46 
 PST."<CC6008BDBF458A4DB8456380C1BAB3800BA404 at server1.in.tcpnetworks.com>
X-url: http://www.cae.wisc.edu/~ballard
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Thu, 18 Dec 2003 14:43:01 -0600
Message-ID: <31219.1071780181 at guinness>
From: Jeff Ballard <ballard at engr.wisc.edu>
X-CAE-MailScanner-Information: Please contact helpdesk at cae.wisc.edu for more 
information or to report problems.
X-CAE-MailScanner: Found to be clean
X-Scanned-By: MIMEDefang 2.38

On Thu, 18 Dec 2003 12:10:46 -0800  Nathan Johanson wrote:

> Hello,
> 
> A recent thread made brief reference to "bayes poisoning". We're
[snip]
> Is anyone else seeing this sort of thing? Any recommendations for
> combating it? I noted the links to BigEvil custom rules (posted
> earlier), but I'm starting to wonder if Bayes usefulness is starting to
> dwindle. Is there some way to prevent this poisoning?

Bayes is a good tool but you have to watch it.  If SPAM is getting 
auto-learned as HAM, then the Bayes is acting in the wrong direction.  This is 
exactly why I won't use Bayes on a system-wide level until I can find a clear 
way of allowing my users to retrain mismarked messages.

I've watched this happen on one of my high-spam mailboxes and only after a bit 
of retraining Bayes was behaving as expected again.

Bayes is not a aim once and fire forever type of weapon.

- -Jeff

- -- 
- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
=-
Jeff Ballard <ballard at engr.wisc.edu>   608-265-5090
Unix Systems Manager, Computer-Aided Engineering Center

- --GGXXCRNBEARXMZXEbCFGOUQLFROSeI--

------- End of Forwarded Message

-- 
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Jeff Ballard <ballard at engr.wisc.edu>   608-265-5090
Unix Systems Manager, Computer-Aided Engineering Center