Representing measured properties in digital systems: going beyond the SI

This paper argues that digital support for various forms of expression of scientific data should extend beyond the traditional framework of formal unit systems. Ambiguities arising when only units are recorded can be addressed by a framework that identifies the nature of the measured property – its aspect – and the scale on which it is expressed. Within this framework, transformations to legitimate alternative forms of expression can be recorded and indexed by aspect–scale identifier pairs. A brief review of formal unit systems highlights inherent limitations of unit notation as a basis for representing data in digital systems. Two applications of particular relevance to chemistry are considered: the expression of elementary entities associated with measures of amount of substance, and expressions for logarithmic ratio quantities . The paper demonstrates that this approach offers significant benefits in data representation.