Re: Now: SpamAssassin and Evolution Was: [SLUG] Spam Lotto

From: Paul M Foster (paulf@quillandmouse.com)
Date: Wed Feb 09 2005 - 22:29:13 EST


On Wed, Feb 09, 2005 at 09:31:42PM -0500, Logan Tygart wrote:

> On Wed, 2005-02-09 at 12:34 -0500, Eben King wrote:
> > On Wed, 9 Feb 2005, Richard Smoot wrote:
> >
> > > Did everybody on the list win 1,500,000 Euros?
> <snip>
> > I really need to set up a Bayesian filter... any recommendations?
>
> I have been running spamassassin with Evolution 2.0, for about two
> months now. It works flawlessly and allows you to dump procmail. I use
> fetchmail, to retrieve my mail from a multitude of servers, and allow
> Evolution to parse my mail.
>
> For Debian folks, to engage SpamAssassin with Evolution 2.0, apt-get
> install both (for anything but Debian Woody), then edit your
> spamassassin file. (For .rpm users, install Evolution 2.0 and
> SpamAssassin, then find the files I am talking about, in what ever
> manner you follow the insanity of .rpm distros.) Debian users
> edit /etc/defaults/spamassassin, and ensure these lines are in the file:
>
> ENABLED=1
> OPTIONS="-L --create-prefs --max-children 1 --helper-home-dir"
> PIDFILE="/var/run/spamd.pid"
>
> The -L is for local use and set the max children number to how many
> people actually sit and read mail at the box. The default is 5; if you
> are the only one with an account at the computer, set it to 1.
>
> For Debian folks, restart SpamAssassin (spamd) with:
>
> /etc/init.d/spamassassin restart
>
> In my inbox, I had a few 419 and lottery spam messages. I clicked on
> each of the messages and then click on the junk icon.
>
> It is as simple as that. Evolution moves the messages to the junk
> folder and SpamAssassin learns from the junk mail folder.
>
> Now, when Evolution parses my mail, it runs each message through
> spamassassin. Using the aforementioned method, I have had zero false
> positives.

I use SpamAssassin, but as someone who gets 400+ messages a day, I can
say that I get 10% of that amount in my "maybe" folder (SA isn't sure
about it). I've set SA down to a threshhold of 3, and I don't use its
Bayesian filtering capacity. And I only feed it mail that's definitely
not from lists, etc. Great program; I wouldn't be without it, but it
only catches 90% of my spam.

Hooking it up to Evolution and dropping procmail??!! Argh!! One of my
criteria in selecting an MUA is whether it plays nice with procmail.
Nothing can filter with the complexity that procmail's capable of.

(Of course, I don't use GUI email clients either. ;-)

Paul

-----------------------------------------------------------------------
This list is provided as an unmoderated internet service by Networked
Knowledge Systems (NKS). Views and opinions expressed in messages
posted are those of the author and do not necessarily reflect the
official policy or position of NKS or any of its employees.



This archive was generated by hypermail 2.1.3 : Fri Aug 01 2014 - 18:54:42 EDT