With apologies to Monty Python…
I’ve had a Yahoo email address for about 10 years, so I get quite a bit of spam. A few weeks ago, I noticed they tweaked their spam filter.
The old filter wasn’t great, but the new one has been a pain in the butt. It routinely puts legitimate emails in the spam folder, even when the sender is in my Yahoo address book.
At the same time, there are certain spam emails I get again and again in my inbox. These messages have the same subject line and same originating address. I flag them as spam each time, but they still end up in my inbox.
Here’s what I don’t understand: Yahoo can store gigs and gigs of email, a bunch of profile data, inbox rules and a lot of other data that is specific to me… and yet they can’t remember any personalized data about what I think is spam? What gives?
I just can’t see how hard it is to have a personal blacklist based on when I press the spam button, and a personal whitelist based on (a) my address book and (b) other emails in my inbox.
As a rough estimate, those two measures would decrease false positives and false negatives by 60-75% for me.
We all lose a lot of time to spam. It seems to me that the next big leap in anti-spam will be driven more by common sense and user behavior than a clever algorithm…