Consider the accident.
The common way to think about it is to consider it something that happens despite all precautions of safety. You try to do everything right, but sometimes, out of statistical necessity, things just go wrong. Horribly, horribly wrong.
This is a rather backward way of going about it.
Paul Virilio proposed we see accidents happening because, not despite. Train accidents don't happen because we failed to do those things that prevent train accidents; they happen because we're running a train system.
The accidents are built in to the system.
And so it goes for all other things. We have automobile accidents because we have automobiles. We have nuclear power plant accidents because we have nuclear power plants.
We have computer accidents because we have computers.
The power of this way of looking at things lies in its shifting of focus. It's very easy to fall into the thinking that things only work in the way they're supposed to work, and everything else is the exceptions that proves the rule. A limited number of uses are identified as legitimate, and the rest are illegitimate - accidents.
Needless to say, this limits one's analysis of things. For instance, file sharing becomes piracy. Those who designed the internet didn't envision the file sharing behaviors we see today, but as accidents would have it, it happens. And it won't go away, no matter how hard one tries, as it is inherent in the very nature of things.
Another computer accident: people gathering together spontaneously become a terrorist threat. As they suddenly move too fast to keep track of.
It has been said that the best ways to prepare for an accident is to do things that are good even if it doesn't happen.
There might be something to it.