As a computer engineering student, I'd have to say this is a truism. Anyone who has been through an electronics lab has probably had an experience with what I learned in my first lab class to call "Releasing the Magic Smoke". It's the magic smoke which causes the ICs and transistors to function. Indeed, you first see smoke, then it stops functioning. (Usually. Sometimes the smoke is delayed.)

Now I have to figure out how exploding transistors and sweating transistors fit into the picture.

(for the solution to this "dilemma", read correlation does not prove causation).