Evaluation of code and data spatial complexity measures

Abstract

The comprehension of a program is largely dependent on the spatial abilities of the programmers. Some spatial complexity measures have been proposed in the literature to estimate the effort required in the process of software comprehension. Two of the widely discussed and important spatial measures for procedure-oriented software are code spatial complexity and data spatial complexity. This paper evaluates these two measures using Weyuker's properties. The paper further validates these two measures on the basis of one of the most acceptable framework proposed by Briand et al. The results of this study show that these two spatial metrics satisfy all properties and parameters required by these two formal evaluation frameworks and thus these spatial measures are practical and useful. © 2009 Springer Berlin Heidelberg.