First of all, the hardest part is defining spam, in that way it's eerily like "obscenity". Of course the "real" name for spam is UCE (unsolicited commercial email) but only true geeks really use that definition and we're all such pedants that a discussion based on it that way tends to get really boring. In the real world (populated mostly by non-geeks) spam is anything that comes into an inbox that you don't want  (and as anyone who's worked in an admin capacity in an ISP-like setting will tell you, at least half of the things people report as spam are a result of them asking for it in the first place (left their email address with a company they ordered something from and didn't read carefully and uncheck the relevant checkboxes), and most of those spams actually have valid "unsubscribe" links in them.   But of course we try (above all else) to teach people not to click on links in emails from unknown sources (thx so much for that unnecessary contribution to the SOP, microsoft), so it's hard to blame folks for forwarding on to postmaster@ or abuse@ instead of just unsubscribing themselves.

But putting aside the definition itself (and agreeing that we all have a pretty good understanding of what makes something "spam-like"), the reason I'm writing here now is because of the thought I just had.   It's certainly easy to define good spam vs. bad spam  (in fact incredibly trivial).  Good spam has a working URL in it and bad spam has a broken URL :-]