The work was done. Three years later, the proof is not.

“Show me the room-14 prototype, on the day of the test”

Fifteen prototypes, none at the same revision level. Three years later the auditor asks which one passed which test. The answer is in a spreadsheet and in the memory of three people, two of whom have left.

never at the same revision level

« Proving that the room-14 prototype was at that document level on the day of the test is computerised nowhere. »

How do you answer that today?

The question about one line of the standard, still open

Everybody is comfortable with the standard in outline and nobody knows it to the letter. So how one clause should be read gets asked, gets escalated, and three months later has still never been settled — while design carries on around it.

one clause, escalated, never settled

« Nobody knows the standard to the letter. You end up on some tiny detail that was never settled. »

The standard everyone is sure about is the 2015 one

Standards get revised a handful of times in a career, so people stay confident in the version they learned. Then two thousand units have to be built to it, by technicians nobody trained on the current text.

to a standard that ages in people's heads

« He was completely sure of himself — on a version of the standard that was true fifteen years ago. »