You click a number in a scanned contract and nothing happens. No cursor, no selection. The page looks like a document and behaves like a photograph — because that is what it is: one image, and every word on screen is part of it. There is nothing to edit because there is no text, only pixels arranged to look like some. Two things still work there: whatever you add sits on top as usual — a signature, a note, a mark, a box that covers something — and some scans already carry a hidden recognised layer behind the image, which is why they can be searched. Making more of a scanned page editable is open work here, and hard; I would rather say that than give you a date. The full explanation, including how to tell the three kinds of page apart, is in Guides. 🔎