DocumentCode :
3277996
Title :
Design Patterns in Software Maintenance: An Experiment Replication at University of Alabama
Author :
Nanthaamornphong, Aziz ; Carver, Jeffrey C.
Author_Institution :
Dept. of Comput. Sci., Univ. of Alabama, Tuscaloosa, AL, USA
fYear :
2011
fDate :
21-21 Sept. 2011
Firstpage :
15
Lastpage :
24
Abstract :
Design patterns are widely used within the software engineer community. Researchers claim that design patterns improve software quality. In this paper, we describe two experiments, using graduate student participants, to study whether design patterns improve the software quality, specifically maintainability and understandability. We replicated a controlled experiment to compare the maintainability of two implementations of an application, one using a design pattern and the other using a simpler alternative. The maintenance tasks in this replication experiment required the participants to answer questions about a Java program and then modify that program. Prior to the replication, we performed a preliminary exercise to investigate whether design patterns improve the understandability of software designs. We gave the participants the graphical design of the systems that would be used in the replication study. The participant received either the version of the design containing the design pattern or the version containing the simpler alternative. We asked the participants a series of questions to see how well they understood the given design. The results of two experiments revealed that the design patterns did not improve either the maintainability or the understandability of the software. We found that there was no significant correlation between the maintainability and the understandability of the software even though the participants had received the design of the systems before they performed the maintenance tasks.
Keywords :
Java; computer graphics; computer science education; continuing education; software maintenance; software quality; student experiments; Java program; University of Alabama; controlled experiment; design patterns; experiment replication; graduate student participants; graphical design; maintenance tasks; replication experiment; software designs; software engineer community; software maintainability; software maintenance; software quality; software understandability; Java; Libraries; Maintenance engineering; Software design; Software maintenance; Software quality; Controlled experiment; design pattern; maintenance; software designs;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Replication in Empirical Software Engineering Research (RESER), 2011 Second International Workshop on
Conference_Location :
Banff, AB
Print_ISBN :
978-1-4673-0972-1
Type :
conf
DOI :
10.1109/RESER.2011.11
Filename :
6148333
Link To Document :
بازگشت