data classification scheme

Suggest a scheme for classifying data. What are each of the categories and what types of information would go into each category? What is the actual risk that may be associated with the loss of information in this category? (i.e., not hyperbole or FUD-based risk)

What process did you use to come up with this? Did you think about risk first or examples first? Once you determined the actual risk, did you reclassify any examples?