A study of the structural composition of PL/I programs

Authors:
James L. Elshoff
Affiliations:
General Motors Research Laboratories, Warren, Michigan
Venue:
ACM SIGPLAN Notices
Year:
1978

Citing 3
Cited 4

Algorithm dynamics

ACM '73 Proceedings of the ACM annual conference
An investigation into the effects of the counting method used on software science measurements

ACM SIGPLAN Notices
Measuring commercial PL/I programs using Halstead's criteria

ACM SIGPLAN Notices

Surveyor's Forum: Heads I Win, Tails You Lose

ACM Computing Surveys (CSUR)
Surveyor's Forum: Projecting Problems

ACM Computing Surveys (CSUR)
Integrated program measurement and documentation tools

ICSE '84 Proceedings of the 7th international conference on Software engineering
Software metrics: an introduction and annotated bibliography

ACM SIGSOFT Software Engineering Notes

Quantified Score

Hi-index	0.00

Visualization

Abstract

Some research on a theory of the structural composition of an algorithm has recently been reported. The theory embodies models of operator and operand distributions that have been proposed by Bayer, Zipf, and Zweben. This paper reports a study in which the three models of operator distributions are compared with the measured distributions found in 34 PL/I programs. Two of three models, one by Zipf and one by Zweben, are shown to correlate highly with the measured distributions. Two variations of the Zipf model are then formulated and tested with good results. Also, the relationship between the method of counting operators and the models is investigated.