[Spambayes] [spambayes-dev] address book

[Jesse Pelton]

This subject was discussed quite a bit in April.  If you're interested,
look for messages with "Feature Request" in the subject in the archives
(http://mail.python.org/pipermail/spambayes/2005-April/thread.html#16790
).

Assuming you're running Outlook (and most other modern mail clients),
you can set up rules that have the effect of whitelisting people in your
address book.  If you configure background filtering on the SpamBayes
Manager Advanced tab, SpamBayes strives to let Outlook do any rule
processing on messages before classifying them.  You could set up a rule
that moves whitelist messages to a folder called "Whitelist", for
instance.  As long as you don't tell SpamBayes to process messages in
that folder, it'll leave them alone.  This is a simple if inelegant
approach.  More elegant approaches are possible, at the cost of
complexity.  A few possibilities are mentioned in the archive messages
referred to above.

> -----Original Message-----
> From: spambayes-bounces at python.org 
> [mailto:spambayes-bounces at python.org] On Behalf Of Kenny Pitt
> Sent: Tuesday, May 17, 2005 2:30 PM
> To: 'Big Max'
> Cc: spambayes at python.org
> Subject: Re: [Spambayes] [spambayes-dev] address book
> 
> Questions about the use of SpamBayes are better asked on the
> "spambayes at python.org" mailing list.  The 
> "spambayes-dev at python.org" mailing
> list is primarily for issues regarding the development of SpamBayes.
> 
> Big Max wrote:
> > Spambayes does not pay attention to my address book email 
> addresses and
> > mark as spam mail that should  not as they come from friends.
> 
> Correct.  SpamBayes does not implement any form of 
> whitelisting, either from
> your address book or from a custom list.  Please see FAQ 6.6:
> http://spambayes.sourceforge.net/faq.html#why-don-t-you-add-wh
> itelisting-bla
> cklisting-to-spambayes
> 
> > How can I train Spambayes to accept as Ham all emails 
> coming from anybody
> > in my address book ?
> 
> You cannot tell SpamBayes in one shot that any mail from 
> anyone in your
> address book is ham.  However, if you continue to train 
> individual messages
> that SpamBayes doesn't get right, then it will quickly learn 
> to recognize
> all of the good mail.  In addition, it will learn to 
> recognize spam messages
> even if they claim to come from one of the people in your 
> address book.
> 
> -- 
> 
> Kenny Pitt