I don't think I understand the question. Training SpamBayes incrementally is the preferred approach and designed to be easy. You train on anything that is not classified correctly, either by clicking the Spam or Not Spam buttons as appropriate or by dragging an incorrectly classified message to the folder it belongs in. You take no training action on anything that's correctly classified. So, let's say a message shows up in your inbox that should have been classified as spam. You can either select it and click the Spam button or drag it to your spam folder. If you find ham in your spam or unsure folder, click Not Spam or drag it to your inbox. Ham in your inbox you read; spam in your spam folder you gleefully delete. (That's what I do, anyway.) If that doesn't help, maybe someone else can take a crack at answering. ________________________________ From: spambayes-bounces@python.org [mailto:spambayes-bounces@python.org] On Behalf Of Gil Hurlbut Sent: Monday, April 24, 2006 3:44 PM To: spambayes@python.org Subject: Re: [Spambayes] Incremental Training for ham in Outlook Plugin? My concern is the ongoing training. Rather than restarting the configuration I want to use new ham to keep up with the new spam I have identified. I'm looking for confirmation that the approach to take to do incremental training is to move filtered ham out of my Inbox and then select "Recover from Spam".