Abstract
An empirical study of 447 operational commercial and clerical Cobol programs in one Australian organization and two U.S. organizations was carried out to determine whether program complexity, programming style, programmer quality, and the number of times a program was released affected program repair maintenance. In the Australian organization only program complexity and programming style were statistically significant. In the two U.S. organizations only the number of times a program was released was statistically significant. For all organizations repair maintenance constituted a minor problem: over 90 percent of the programs studied had undergone less than three repair maintenance activities during their lifetime.
| Original language | English |
|---|---|
| Pages (from-to) | 128-134 |
| Number of pages | 7 |
| Journal | Communications of the ACM |
| Volume | 26 |
| Issue number | 2 |
| DOIs | |
| Publication status | Published - 1 Feb 1983 |
Keywords
- experimentation
- modular programming
- program complexity
- program maintenance
- programmer quality
- programming management
- repair maintenance
- structured programming