[Spambayes] Spammers try to get around BAYES spam filters - withexample

[Tony Meyer]

> i noticed that there is increasingly SPAM with a small
> amount of SPAM text and a large amount of unsuspicious
> text (in HTML formatted mail in extremely small font or
> in a font color same as background).
>
> The result is a very low SPAM rating and a high likelyhood 
> for the message to get through.
> Unfortunately, I don't really have an idea how to address
> this.  I attach one example.

I see the occasional one of these (getting through) - enough that I have
vague plans to look into this at some point (maybe trying to adapt the
Bayesian Dobly ideas), but not many.

The only effective way to give us an example, though, is to include the
clues for the message.  If you're using the Outlook plug-in, this means
selecting the message (before training on it), and choosing "Show Spam Clues
for current message" from the SpamBayes menu.  This creates a new message
that you can send to the list.

With the clues, we can discuss *why* it's scoring what it is.  Without the
clues, it's rather meaningless to do this.  For example, your message (with
all the "this is from the spambayes list" clues and your text) scored 8% for
me, when messages from this list tend to score 0%.  So it looks fairly
spammy to me.  The clues would tell us why it doesn't to you.

=Tony Meyer

---
Please always include the list (spambayes at python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes. This
way, you get everyone's help, and avoid a lack of replies when I'm busy.