Some people have been preaching #SafetyByDesign principles. I would like to point out a huge problem with one of them.
Namely, the principle that profiles should be hidden by default.
This feature exists on Reddit right now. You know who uses it most? Dishonest people, who use the feature to conceal their history of dishonesty.
This allows them to do the same thing as the snake oil salesmen of the Wild West, going from town to town and running the same dishonest scheme over and over again.
Another “safety by design” principle is that platforms are to somehow automatically block all harmful content before it is posted.
Billions of dollars and decades of work have already been done on that, and it all failed miserably. Policy-obeying content gets randomly censored and its innocent authors banned. Bad actors speak in paper-thin euphemisms and get away scot-free.
It's impossible. Completely, absolutely impossible.
I will give these people some credit for going after the recommendation algorithms that most major platforms have.
Harmful content existing is a fact of life. There's no way to permanently get rid of all of it without shutting down all methods of communication entirely.
But recommendation algorithms trying to hold our attention by analyzing and manipulating our emotions? That's not inevitable. That's not a fact of life. We had perfectly good chat long before those algorithms were invented.
ARGVMI~1.PIF