Issue No. 06 - December (1994 vol. 6)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/69.334884
<p>In commercial DBMSs, it is possible to model only primitive numeric data types and perform arithmetic between them. In practice, however, organizations need to store and manipulate more complex numeric elements that could be interpreted as representing quantities of a non-numeric measurement system like that in which a distance is expressed in feet and inches, or like that in which a weight is expressed in quarters, stones, pounds, and ounces. This implies that users have to choose between two options, either to abandon the nonmetric system and completely adapt their applications to the limited capabilities of the DBMS, or to write their own pieces of code for the management of such more complex numeric data types. The first approach is principally unacceptable, and at the same time, there is a loss in the precision of arithmetic operations, because quantities have to be expressed as real numbers. The second one is tedious, because distinct pieces of code have to be written for handling different nonmetric units. Furthermore, integrity checking for these pieces of data has to be performed by application programs rather than by the DBMS. To overcome these problems, a new generic data type is proposed, the composite number, whose support automatically enables the use of any nonmetric measurement system. Functions and operations are defined for the management of composites. Because, in practice, time is usually expressed in many distinct nonmetric measurement units whose choice depends on the particular application, temporal databases represent one of the many application areas of the proposed formalization.</p>
database management systems; temporal databases; arithmetic; units (measurement); data structures; abstract data types; DBMS; nonmetric measurement systems; numeric data types; nonnumeric measurement system; precision; arithmetic operations; integrity checking; generic data type is; composite number; temporal databases
N. Lorentzos, "DBMS Support for Nonmetric Measurement Systems," in IEEE Transactions on Knowledge & Data Engineering, vol. 6, no. , pp. 945-953, 1994.