A flexible approach to user-defined symbolic granularities in temporal databases

User-defined granularities, calendars and periodicities are gaining an increasing relevance in the area of temporal databases (TDB). In this paper, we discuss the advantages of adopting for TDBs a modular framework, based on a family of symbolic languages, defined starting from a set of mutually orthogonal properties that characterize periodicity. First, we show that one of the languages covers the notion of granularity (as defined in the TDB glossary), so that it can be used in order to define validity times of facts in a TDB. Second, we discuss the usefulness of using more expressive languages in the family to achieve such a goal. Third, we explore how such languages can be used in order to deal in an intensional way with infinite periodic data in temporal (relational) databases.