Semantics of Time-Varying Attributes and their Use for Temporal Database Design

Based on a systematic study of the semantics of temporal attributes of entities, this paper provides new guidelines for the design of temporal relational databases. The notions of observation and update patterns of an attribute capture when the attribute changes value and when the changes are recorded in the database. A lifespan describes when an attribute has a value. And derivation functions describe how the values of an attribute for all times within its lifespan are computed from stored values. The implications for temporal database design of the semantics that may be captured using these concepts are formulated as schema decomposition rules.