Document Encodings can be placed on a continuum depending on the easy of machine extraction of semantic information:
Digitized collections are level 1, most online research literature is level 2 or 3.