Hypothesize a machine that allows you to change people's emotions directly, one that has seen mass adoption. Sure, people opt-in to using it, but once plugged in, it works as advertised: send the right commands to it, and people's emotions shift.

You could do a lot of good with one of those! The ultimate anti-anxiety treatment, the ultimate in letting people become the cooler heads they need to be for good ideas to prevail in tense situations.

You could also hack into it with almost no effort and turn a whole city into a bad knock-off of Resident Evil; hit the neurons that rabies hits, get people tearing and biting at each other.

Should people be allowed to install it?

I think it's possible that books were a great idea and a machine like the one I described (without proper constraints to put control in the hands of the owner) would be a terrible idea. There's probably some technology along the line between these where we say "Whoa, wait. What this could achieve is grand, but the way it gets there is, uh-oh."

The more powerful a technology is, the more obligation it implies upon the people holding control over its implementation and execution to consider moral and philosophical concerns of its use.