Closed msesboue closed 2 years ago
It is really "relationships between data and definitions of data. This one is simple, "definitions" here really means "classes", so identifying a node as an instance of a class. "Relationships within data" can be anything where one item of data being linked to (related to) another gives it some meaning (e.g., "part of" relationship may influence how you interpret a particular node/object). "Hard-wired in software" is what we have done traditionally: you have data, you write some code to manipulate that data (i.e., "interpret" that data).
Chap 1, page 14, section 1.3.2