Scrivener 3.1.1.0
Currently, when you change a custom metadata from a list-type to a text-type, all the existing data gets replaced with numbers (corresponding to the items in the list) rather than whatever the list items were. This does not seem intended.
Well, they have two very distinct functions… and I doubt they were ever intended to be switched from a format to another one.
Yeah, I don’t think there would be a clean way of doing what you tried to do. Basically each binder item refers to the list item you assigned to it with the ID number for that list item—the numbers you see now. If you then change that field type to something that doesn’t use ID numbers and just uses text, there is no rational way to resolve the request. When you click “OK” on the Project Settings panel, the list labels are gone because the new field type doesn’t support them, they are deleted. From what then would all of the binder items extract the text of the label that now doesn’t exist?