The paper discusses modelling temporal variation of data in the context of the relational data model. This is done by employing time-stamps, which leads to two possible approaches." adding time-stamps to tuples and using first normal-form (1NF) relations, and adding time-stamps to attributes and using non-first normal form (N1NF) relations. Modelling temporal data according to each approach with their respective data-manipulation operations are covered. This is followed by a discussion that includes the advantages and expressive power of each approach, the "classification of temporal databases, and the handling of retroactive and postactive changes.
databases, temporal data, modelling, relational data model
A database contains data pertaining to an organization and its activities. It forms a data repository from which information is extracted for various purposes. Databases in general carry the most recent data. As new data values become available through updates, existing data values are discarded from the database. Such databases are called snapshot databases as they only contain current data, which are a snapshot of the current reality. Many applications, however, need both current and past data, and possibly future as well. In the broadest sense, a database that maintains past, present, and future data is called a temporal database.
The activities of any organizations are an ongoing process, and its information needs and processing capabilities should be considered in a time perspective. That is, to support managerial information needs, as well as others, the database should possess a temporal dimension to store and manipulate time-varying data. Most data models, however, do not address issues of maintenance and processing of temporal data. In these models, it is possible to carry the time reference of an attribute as another special attribute. This approach is an ad hoc and limited solution. It either creates undue data redundancy and/or provides limited time-processing capacity. There are two possible directions that can be followed for handling temporal data. One is to develop a new model to support time dimension and the other to augment existing data models to support time dimension in a coherent way. In this article the second approach is followed and