If a service doesn't provide an easy and obvious way to unsubscribe from unwanted emails—by my definition, not theirs—I personally have no qualms about marking

If a service doesn't provide an easy and obvious way to unsubscribe from unwanted emails—by my definition, not theirs—I personally have no qualms about marking them as spam and letting them take the reputation hit. The same applies if I never subscribed in the first place; I'm not going to click an "unsubscribe" link (which could well be malicious, and at the very least leaks the fact that I received the message) for a service I didn't sign up for in the first place and have no other connection with. If not subscribing didn't work then there is no reason to think that merely unsubscribing will prove any more efficacious. It gets more complicated when the unwanted spam is from a service which I do want to receive notifications from (just not the junk mail). Most email clients offer rule-based classification which can help, but that is obviously a bit more work than just marking messages as spam.