I like that the internet archive exists, but I don't like how they implement their opt-out policy. Sites can opt-out at any time, which I suppose is the correct thing to allow, but once they opt out then everything the Archive has gathered up to that point also goes away, which seems wrong.
It doesn’t go away, it’s just not public. This is to comply with copyright law. It’s still safely stored on disk.
Actually as I've noted before other archive initiatives often ignore robots.txt, for example the Danish National Archive ignores it and is legally allowed to for gathering all relevant Danish material.