Yeah, I saw that article. The guy was a scapegoat, pure and simple. The fact he used the freaking manual to do his set up in any rational company would mean that:
1) He'd be in a safe playpen, able to screw around to his heart's content and never even see production data, much less have full access to it (especially unawares!).
2) Following the manual's instructions should have protected him from repercussions, even from such a monumental disaster. The firing was somebody high up covering their behind. Not cool.
Clearly the manual he was given was intended for production set up, not development, and that was the true trigger of the disaster. Along with a chain of other stupid decisions on the IT architect's part.
For instance our backup system has evolved over time into a differential DR system with two geographically isolated redundant (and encrypted) backups for the onsite backup system. All three hold *months* worth of our data, we can restore any file as it existed on any date in the past. Including SQL Server databases.
That gives me the warm and fuzzies. Triple safety nets are cool. 😎