This paper reports experimental measurements of productivity and quality in pair programming. The work complements Laurie Williams' work on collaborative programming, in which Pair Programming and Solo Programming student groups wrote the same programs and then their activities were measured to investigate productivity, quality, etc. In this paper, Pair and Solo industrial programmer groups are requested to complete algorithm-style aptitude tests so as to observe the capability of solving algorithms in singles and in pairs. So doing is independent of the familiarity of a programming language. Besides, we also take another approach to examining pair programming. A single group of industrial programmers carries alternately out Pair Programming and Solo Programming. All these demonstrate that productivity in pair programming hinges upon algorithm design at all levels from understanding problems and implementing solutions. In addition, we reach similar conclusions to Williams. Our findings indicate that simple design, refactoring, and rapid feedback provide an excellent continuous-design environment for higher productivity in pair programming.
|Number of pages||9|
|Journal||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Publication status||Published - 1 Dec 2003|
ASJC Scopus subject areas
- Computer Science(all)
- Biochemistry, Genetics and Molecular Biology(all)
- Theoretical Computer Science