User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1.6) Gecko/20070725 Firefox/2.0.0.6 Build Identifier: 2.0.0.6 (20070728) Email with a club news send from cn_news@clubnautique.net is over and over and over again marked as a possible "Scam." I keep unchecking it every time, but Thunderbird learning module somehow does accept it. I wish I would cut and paste the header of this message, as it was possible with Netscape Messenger, but this is not possible with Thunderbird, or at least I have not discovered how this works. It is amazing in how many aspects Thunderbird is *less* than the Messenger, aside of performance issues. I think that this problem was reported by others in many similar variants, this might even be a duplicate of the other bugs. Reproducible: Always Steps to Reproduce: 1. 2. 3.

I am attaching a screen dump of a header of one such message.

Every time I click on "not a scam," for months if not for years already. Over all these updates to Thunderbird this has never been fixed.

"Scam" detection looks for a few specific patterns, it does not "learn" from your marking the way the spam filters do. The scam/not-scam setting should be saved along with any given mail, but won't affect incoming new mail. The most likely cause of this is either a link to a numeric address instead of a host name, or a link that goes to a site other than the one stated in the link text. In a true scam the mail may say "www.citibank.com" but the link really goes to http://ima.phisher.com/. Unfortunately many newsletters send links through click-tracking sites for marketing purposes which then do redirect to the site stated in the text, but Thunderbird has no way to know that. Future versions of Thunderbird may be enhanced with something like Firefox's phishing detection, which will make it easier to tell the difference between actual scams and mere big-brotherish behavior.

I don't see why it's a good policy not to have the scam-marking function learn from your clicking on "Not a Scam"; otherwise, what purpose does it serve? The fact that it doesn't learn makes me ignore it, and that reduces the possible effectiveness of the marking. I'm even getting the marking on email that I create myself; what's that all about? I just assumed that this function is in its early stage of creation, and am hoping that sometime soon these issues will be addressed. As it stands now, the scam-marking function is not much more than a constant source of irritation.

actually 320351 is a better match for what you're saying here

The simple solution to this is to "learn" which recurring senders are NOT scammers and create a list. If the senders email address is not in the learned "not a scam" list, then by all means check and flag it. Also, make the list user editable.