Universal Temporal Data Languages

Temporal reasoning and temporal query languages present diicult research challenges , which are slowly yielding to the combined attack of many investigations motivated by the theoretical interest and practical import of the problem. In this paper, we subscribe to TSQL2 insofar as practical requirements for a query language are concerned, but we propose a solution that overcomes its shortcomings, particularly the lack of universality whereby TSQL2 temporal extensions can not be easily applied to other query languages, such as QBE and Datalog. In this paper, we use Datalog as a framework to develop a new language | Temporal Data Language (TDL). To support our claim of universality, we argue that TDL constructs and semantics can be directly applied to derive temporal extensions of languages, such as QBE and SQL. Finally, we evaluate alternative approaches to the implementation of TDL, using as the basis for implementation the LDL++ system with extended aggregates developed at UCLA.