I don't think much of these experiments. They are fun, of course. But they seem irrelevant to me as the emails to be categorized are essentially not what I receive. How do I know how a real PayPal message looks like if I'm not a PayPal customer. How should I say whether a newsletter is ham or spam if I never registered for it?
Such experiments are completely useless unless they are personalized to my Inbox. They should ask me about some faked ham mails derived from my very own emails. Then, I could probably wonder whether something looks legitimate or not. But categorizing my neighbors Inbox is (if at all) just fun, nothing more.
Slashdot Mirror
I don't think much of these experiments. They are fun, of course. But they seem irrelevant to me as the emails to be categorized are essentially not what I receive. How do I know how a real PayPal message looks like if I'm not a PayPal customer. How should I say whether a newsletter is ham or spam if I never registered for it? Such experiments are completely useless unless they are personalized to my Inbox. They should ask me about some faked ham mails derived from my very own emails. Then, I could probably wonder whether something looks legitimate or not. But categorizing my neighbors Inbox is (if at all) just fun, nothing more.