Hi,
On Fri, 06 Aug 2004 11:21:10 -0400 Matt Kettler <[EMAIL PROTECTED]> wrote:
> Something's fishy here.. those numbers don't add up to 5.5.
>
> They actualy add to 3.5.
>
> Sure there's some loss of precision due to truncation, but even adding
> +0.0999 for all 12 rules that hit you only get 4.6988.
>
> Clearly something's not right here. Perhaps the true adjustment applied by
> the AWL isn't being reported correctly due to some bug and was actually
> +1.9 instead of -0.1?
No, the numbers add up; it's a line-wrapping problem.
Let me reformat this to make it clearer (truncating descriptions so it
doesn't happen again...):
> At 01:07 AM 8/6/2004, lists wrote:
> >Content analysis details: (5.5 points, 5.0 required)
> >
> > pts rule name description
> >---- ----------------------
> >--------------------------------------------------
> > 0.2 NO_REAL_NAME From: does not include a real
> > 0.7 RCVD_BY_IP Received by mail server with n
> > 3.4 RCVD_NUMERIC_HELO Received: contains an IP addre
> > 0.0 HTML_80_90 BODY: Message is 80% to 90% HT
> > 0.0 HTML_BADTAG_00_10 BODY: HTML message is 0% to 10
> > -2.6 BAYES_00 BODY: Bayesian spam probabilit
> > [score: 0.0000]
> > 1.2 MIME_HTML_MOSTLY BODY: Multipart message mostly
> > 0.0 HTML_MESSAGE BODY: HTML included in message
> > 0.0 HTML_NONELEMENT_00_10 BODY: 0% to 10% of HTML elemen
> > 2.0 MIME_MISSING_BOUNDARY RAW: MIME section missing boun
> > 0.7 MIME_BOUND_NEXTPART Spam tool pattern in MIME boun
> > 0.0 RCVD_DOUBLE_IP_LOOSE Received: by and from look lik
> > -0.1 AWL AWL: From: address is in the a
( 0.2 + 0.7 + 3.4 + 0.0 + 0.0) = 4.3
(-2.6 + 1.2 + 0.0 + 0.0 + 2.0) = 0.6
( 0.7 + 0.0 + -0.1) = 0.6
Total = 5.5
Arithmetic. Woo.
-- Bob
