Dealing with Granularity of Time in Temporal Databases

A question that always arises when dealing with temporal information is the granularity of the values in the domain type. Many different approaches have been proposed; however, the community has not yet come to a basic agreement. Most published temporal representations simplify the issue which leads to difficulties in practical applications. In this paper, we resolve the issue of temporal representation by requiring two domain types (event times and intervals), formalize useful temporal semantics, and extend the relational operations in such a way that temporal extensions fit into a relational representation. Under these considerations, a database system that deals with temporal data can not only present consistent temporal semantics to users but perform consistent computational sequences on temporal data from diverse sources.