I mean, you get toxic behavior even on something like Nextdoor, where you pretty much know it's the neighbors across the street. Technology has just made things more convenient -- social media removes any curation, and technology also has made some means of harassment much easier to execute.

Myself, personally, I avoid social media that encourages toxic behavior (which usually means, smaller, special interest type sites; social circles that you know; etc.). This involves some degree of moderation or self-selection.

I don't see a good way around limited moderation for Reddit either, which is unfortunate in that it is hard to moderate something that size well (it's usually inconsistent and often arbitrary-ish).

That is not at all clear, especially as a percentage of comments. There are plenty of sociopaths who are perfectly willing to troll under their real name. Meanwhile, more reasonable people may quite rationally be worried that expressing any opinion on a controversial issue will lead to online mobs trying to get them fired from their jobs, kicked out of school, or otherwise ostracized. Not to mention scenarios like being a gay teenager in a very socially conservative environment.

I don't think it would. Lots of people post horribly objectionable material under their own names on a regular basis, depending on their level of financial security, peer group, and social milieu.

What makes a "normal" person take an alter-ego/view/counter view, solely based on lack of face to face interaction...

Some people are horrible, and are just as unpleasant in real life as they are online.

Is this true? I've seen various studies saying that the opposite is actually true, but I can't currently find any of those studies. Does anyone have any sources?

EDIT: some sources, though I don't know the strength of their validity:

  - https://techcrunch.com/2012/07/29/surprisingly-good-evidence-that-real-name-policies-fail-to-improve-comments/

  - http://www.slate.com/blogs/future_tense/2014/07/17/google_plus_finally_ditches_its_ineffective_dangerous_real_name_policy.html

The cost is that it would prevent people from anonymously reporting abuses, which means that fear of retaliation will have a chilling effect. We've already seen this where people get death threats, houses burned down, etc, when they do things like report sexual assault.

I can see at the cost of making stolen identities worth even more.

As it is websites completely suck at keeping our 'anonymous' identity secure. Our email and passwords are hacked so commonly there are websites dedicated to tracking it. Now you're just adding 'real identities' to the brokered data. Any real trolls will be able to use this data from the dark net pretending to be you. Even worse, since real names are required, any employers will look for you on the internet, they will see your "I'm an anti-gay right wing pro-russian" profiles online and say "you are not a cultural fit for our company". You will have to take the time and effort to clean up what is said about you. Since it's a real identity, it's not going to change, and they already have all the information they need on you.

Good luck in that terrible future.

Those who post offensive content and do care, find it completely trivial to get a fake name. FB’s enforcement of their policy is basically nonexistent. This is the second-biggest group of users.

I honestly don't think this will ever be the case. Because there is a lot of profit from an unmoderated comment store. Also because it'd be monumentally hard to actually make all sites compliant.