I’m interrupting work to post this because I’m stunned.

I understand that spam detection is an art. When you are in the software profession you can appreciate the complexity of it by just reading the interesting articles on the subject.

In fact, given how complex it is, I’m always gracefully surprised to see how good GMail is at it. Granted, from time to time one spam email leaks in, but considering that it correctly detects hundreds of them per week, it’s definitely more that good.

On the other hand, I still have a hotmail account.

If I can only search and replace one for the other in everyone’s contact list.

And I’m just equally surprised about how bad hotmail is at detecting spam. Roughly, 90% of the spam makes it into the inbox.

But I figured that it was mainly my fault for not taking the time to train hotmail by teaching it what was spam and what wasn’t (though I never did such thing for GMail FYI).

Recently, however, I started to think that hotmail uses AS techniques. Now I’m almost convinced: I just received a message from…hmm, a guy named “VIAGRA”!

OK, ok, I shouldn’t be so harsh on it… after all, how would it know that was spam?