A Hypothesis
I wonder if our seemingly decreasing tolerance for real life inappropriateness is at all related to our increasing exposure to such inappropriateness online, via comments, IM, etc?
Though we can block someone online or even get them banned from websites, in the back of our minds, we know that trolls will never really go away, that they'll just make a new username or use a different machine and be back at it within a week. To our POV, there are no real consequences for the trolls, asshats and hateful people for posting what they do online except they get shut down from that one site for a while, and that's even if they care.
They basically get a slap on the wrist, if that, and sometimes they get a bunch of supporters crawling out of the woodwork to make an unassailable wall of ass that modders/admins can do nothing about short of shutting the site down and killing their own web traffic, which would probably be financial suicide for them and could result in the website going down permanently.
I've experienced that aggravation before, seeing a full page of nothing but racial slurs, trolling and hate speech, and hit the "report user" buttons, only to see them still commenting and posting similar crap months later. It makes me cry out in despair sometimes. So from my own experiences, the frustration can be real, and it carried with me through at least part of my day, so it affected me outside of just the website.
So when that sort of thing happens around us in real life - where consequences are much more serious and long lasting, and would therefor be far more satisfying to witness - would it not become harder to keep from lashing out and overreacting as part of a desire to try and make up for the numerous injustices we constantly witness online? Is it possible we go overboard because we know that although we can't stick it to all the trolls online, we can at least stick it to this one sitting right next to us?
...I don't know, I'm asking because it just occurred to me to ask. I know Correlation does not equal Causation, but if studies have been done before that could show this kind of link - using samples large enough to matter (i.e. over 1K-10K) and peer-reviewed - it'd be interesting to know.