APL Extended Compared With Other Languages According to Halstead's Theory

Authors:
F. H. D. van Batenburg;Joseph de Kerf
Affiliations:
Institute of Theoretical Biologie, Kaiserstraat 63, 2311 GF Leiden, The Netherlands;Belgian APL-CAN Users Society, Rooienberg 72, B-2570 Duffel, Belgium
Venue:
ACM SIGPLAN Notices
Year:
1998

Citing 5
Cited 0

Probability, statistics, and queueing theory with computer science applications

Probability, statistics, and queueing theory with computer science applications
APL compared with other languages according to Halstead's theory

ACM SIGPLAN Notices
Elements of Software Science (Operating and programming systems series)

Elements of Software Science (Operating and programming systems series)
Extending Halstead's software science for a more precise measure of APL

APL '82 Proceedings of the international conference on APL
APL and Halstead's theory of software metrics

APL '81 Proceedings of the international conference on APL

Quantified Score

Hi-index	0.00

Visualization

Abstract

In a previous paper, Halstead's theory was applied to first generation APL. Those implementations only support simple arrays (APL/SA). In this paper we extend the application of Halstead's theory to second generation APL; the main difference being that those implementations support arrays of arrays or nested arrays (APL/NA). In both investigations, Halstead's theory had to be adjusted in order to account for the fact that in APL operations may be functions or operators. It is shown that the language level increases from 3.07 for APL/SA to 3.62 for APL/NA This means that language efficiency of APL versus assembler raises from about 12 for APL/SA to nearly 17 for APL/NA.