From rob at thetolerplace.com  Sun Mar  2 20:01:29 2014
From: rob at thetolerplace.com (Robert Toler)
Date: Sun, 2 Mar 2014 14:01:29 -0500
Subject: [Spambayes] Compatibility issues
Message-ID: <7bca01cf3649$d1f65cd0$75e31670$@thetolerplace.com>

Hello,

 

I am interested in trying Spam Beyes. I notice it doesn't list Outlook 2013
so far and it indicates 64bit support is limited. I have Windows 7 Pro on a
64bit system with Outlook 2013. Are there any updates or suggestions for
this combination?

 

Regards,

Rob Toler

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/spambayes/attachments/20140302/be484d17/attachment.html>

From dennis.mccunney at gmail.com  Sun Mar  2 23:14:49 2014
From: dennis.mccunney at gmail.com (dmccunney)
Date: Sun, 2 Mar 2014 17:14:49 -0500
Subject: [Spambayes] Compatibility issues
In-Reply-To: <7bca01cf3649$d1f65cd0$75e31670$@thetolerplace.com>
References: <7bca01cf3649$d1f65cd0$75e31670$@thetolerplace.com>
Message-ID: <CAF4AJDyMhbBhHNX_PHHN7P4tek_Jz48v69x+YGgZ7xqwmi1uvA@mail.gmail.com>

On Sun, Mar 2, 2014 at 2:01 PM, Robert Toler <rob at thetolerplace.com> wrote:

> I am interested in trying SpamBayes. I notice it doesn't list Outlook 2013
> so far and it indicates 64bit support is limited. I have Windows 7 Pro on a
> 64bit system with Outlook 2013. Are there any updates or suggestions for
> this combination?

It depends upon your Ofiice version.  If you are running a 32 bit
version of Office, SpamBayes ought to work.

It does have a caveat about *where* you install it:  It needs to be
someplace other than Program Files, like C:\SpamBayes.  Win Vista/7/8
have a more stringent permissions model, and if you install SpamBayes
in Program Files, it won't work because it won't have permission to
run.  Installed outside of Program Files it runs as expected.

If you are running 64 bit Office, SpamBayes won't work as an plugin.
SpamBayes is a 32 bit application, and you can't use a 32 bit plugin
in a 64 bit program.

Your option for SpamBayes with 64 bit Office is to install it as a
proxy server between your Internet connection and Outlook.  64 bit
Windows systems will run 32 bit programs, and SpamBayes will run
stand-alone.  It just won't work as a plugin.

(Getting SpamBatyes to work as a plugin in 64 bit Outlook would
require someone to rebuild it as a 64 bit application.  Thus far, no
one who can has volunteered to do it.)

> Rob Toler
______
Dennis
https://plus.google.com/u/0/105128793974319004519

From garey at gruberagency.net  Tue Mar 18 23:28:59 2014
From: garey at gruberagency.net (Garey Gruber)
Date: Tue, 18 Mar 2014 17:28:59 -0500
Subject: [Spambayes] False spam identification
Message-ID: <008801cf42f9$753a2af0$5fae80d0$@net>

My program continues to call certain e-mail messages junk or junk suspects
even after I mark it as trusted address or domains.  How do I fix this?

 

Garey Gruber

Garey Gruber Insurance Agency

 

Phone: (281) 741-2653

FAX:    (281)  741-9373

e-mail  <mailto:garey at gruberagency.net> garey at gruberagency.net

Web Site  <http://www.gruberagency.net> www.gruberagency.net

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/spambayes/attachments/20140318/310d6107/attachment.html>

From fredex at fcshome.stoneham.ma.us  Wed Mar 19 03:12:13 2014
From: fredex at fcshome.stoneham.ma.us (Fred Smith)
Date: Tue, 18 Mar 2014 22:12:13 -0400
Subject: [Spambayes] False spam identification
In-Reply-To: <008801cf42f9$753a2af0$5fae80d0$@net>
References: <008801cf42f9$753a2af0$5fae80d0$@net>
Message-ID: <20140319021213.GA26391@fcshome.stoneham.ma.us>

On Tue, Mar 18, 2014 at 05:28:59PM -0500, Garey Gruber wrote:
>    My program continues to call certain e-mail messages junk or junk
>    suspects even after I mark it as trusted address or domains.  How do I
>    fix this?

every time you find one of these, move it from the SPAM folder into the HAM
folder, or wherever you have configured SpamBayes to find its non-spam,
for training purposes.

Once you train it on enough of them it'll be less likely to think of
them as spam.


-- 
---- Fred Smith -- fredex at fcshome.stoneham.ma.us -----------------------------
                         God made him who had no sin
                      to be sin for us, so that in him
                 we might become the righteousness of God."
--------------------------- Corinthians 5:21 ---------------------------------

From skip at pobox.com  Wed Mar 19 10:25:46 2014
From: skip at pobox.com (Skip Montanaro)
Date: Wed, 19 Mar 2014 04:25:46 -0500
Subject: [Spambayes] False spam identification
In-Reply-To: <20140319021213.GA26391@fcshome.stoneham.ma.us>
References: <008801cf42f9$753a2af0$5fae80d0$@net>
 <20140319021213.GA26391@fcshome.stoneham.ma.us>
Message-ID: <CANc-5UwOf=X3X5Trvfy1WSdbERMKroGBrhJUka-XiLBPJh7qYA@mail.gmail.com>

On Tue, Mar 18, 2014 at 9:12 PM, Fred Smith
<fredex at fcshome.stoneham.ma.us> wrote:
> Once you train it on enough of them ...

I will also point out one potential issue which we used to see BITD.
Don't train SpamBayes on every single message, only on the ones that
aren't classified correctly. You wind up with a much smaller training
database, are less likely to misclassify messages, and will likely see
stronger effects on the classification accuracy in the situations
where you do add a new message.

Skip

From BCash at guesswho.com  Wed Mar 19 18:21:22 2014
From: BCash at guesswho.com (BCash at guesswho.com)
Date: Wed, 19 Mar 2014 13:21:22 -0400
Subject: [Spambayes] spambayes not learning
Message-ID: <2AA600764E54964491083B1E0EC81A30010A8AA3B02E@EXCLUS.nationala-1advertising.com>

Hello,  I am trouble shooting an issue for the CEO of my company.  There is an email domain that keeps getting flagged in the junk suspects folder.  we keep clicking on "not junk" and it will release the email, but everytime they send again they keep going to junk suspects.  I wanted to reset spambays and have it re-learn or uninstall and reinstall but I have been told this is not an option as there is too much for it to re-learn.  This is an XP system sp3 running outlook 2010.  Any help would be greatly appreciated.

Thank You,

Brent


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/spambayes/attachments/20140319/fc201f15/attachment.html>

From skip at pobox.com  Wed Mar 19 19:39:30 2014
From: skip at pobox.com (Skip Montanaro)
Date: Wed, 19 Mar 2014 13:39:30 -0500
Subject: [Spambayes] spambayes not learning
In-Reply-To: <2AA600764E54964491083B1E0EC81A30010A8AA3B02E@EXCLUS.nationala-1advertising.com>
References: <2AA600764E54964491083B1E0EC81A30010A8AA3B02E@EXCLUS.nationala-1advertising.com>
Message-ID: <CANc-5UxAcJgLq=K7u8+NEHeA7tS+q6akREVaGwPmUsKLr=K4Yw@mail.gmail.com>

On Wed, Mar 19, 2014 at 12:21 PM,  <BCash at guesswho.com> wrote:
> Hello,  I am trouble shooting an issue for the CEO of my company.  There is
> an email domain that keeps getting flagged in the junk suspects folder.  we
> keep clicking on ?not junk? and it will release the email, but everytime
> they send again they keep going to junk suspects.  I wanted to reset
> spambays and have it re-learn or uninstall and reinstall but I have been
> told this is not an option as there is too much for it to re-learn.  This is
> an XP system sp3 running outlook 2010.  Any help would be greatly
> appreciated.

Not sure where the "too much for it to re-learn" opinion comes from.
As I indicated in an earlier message today, training on every message
is almost certainly not what you want to do. In this case, it may very
well be that someone somewhere along the way misclassified mail from
this source as spam. If so, you are now paying the price for that
mistake, as it generally takes a fair bit of effort to overcome such
mistakes. When I was actively still working on SpamBayes, discarding
my training database and starting over was something I did with some
regularity (I would guess at least once a month) as the version of
SpamBayes I was using changed.

I was never an Outlook user, so I'm not quite sure of the proper
protocol in that interface, but my advice would be to empty the
training database, then classify the next group of messages (five to
10) that arrive and land in the unsure pile. That should be enough to
get SpamBayes pointed in the right direction. I wouldn't bother
training on any mails other than those which are misclassified or
those that wind up in the unsure pile.

If you really want Outlook to whitelist mail from a specific set of
email addresses or domains, I'm pretty sure it allows you to do that.
In that case, Outlook would in effect be saying that it knows better
than SpamBayes. Which it might well. There might be mail which is too
important to trust to a spam filter. Such as mail from your wife. :-)

Skip

From Davor.Skeledzija at comping.hr  Fri Mar 28 09:34:24 2014
From: Davor.Skeledzija at comping.hr (Davor Skeledzija)
Date: Fri, 28 Mar 2014 09:34:24 +0100
Subject: [Spambayes] Russaian SPAM
Message-ID: <a1c68f0699fc49d0bc249d209a259201@Atlas.comping.hr>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/spambayes/attachments/20140328/2161203c/attachment.html>