%0 Journal Article
%T An Empirical Study on the Impact of Duplicate Code
%A Keisuke Hotta
%A Yui Sasaki
%A Yukiko Sano
%A Yoshiki Higo
%A Shinji Kusumoto
%J Advances in Software Engineering
%D 2012
%I Hindawi Publishing Corporation
%R 10.1155/2012/938296
%X It is said that the presence of duplicate code is one of the factors that make software maintenance more difficult. Many research efforts have been performed on detecting, removing, or managing duplicate code on this basis. However, some researchers doubt this basis in recent years and have conducted empirical studies to investigate the influence of the presence of duplicate code. In this study, we conduct an empirical study to investigate this matter from a different standpoint from previous studies. In this study, we define a new indicator ¡°modification frequency¡± to measure the impact of duplicate code and compare the values between duplicate code and nonduplicate code. The features of this study are as follows the indicator used in this study is based on modification places instead of the ratio of modified lines; we use multiple duplicate code detection tools to reduce biases of detection tools; and we compare the result of the proposed method with other two investigation methods. The result shows that duplicate code tends to be less frequently modified than nonduplicate code, and we found some instances that the proposed method can evaluate the influence of duplicate code more accurately than the existing investigation methods. 1. Introduction Recently, duplicate code has received much attention. Duplicate code is also called as ¡°code clone.¡± Duplicate code is defined as identical or similar code fragments to each other in the source code, and they are generated by various reasons such as copy-and-paste programming. It is said that the presence of duplicate code has negative impacts on software development and maintenance. For example, they increase bug occurrences: if an instance of duplicate code is changed for fixing bugs or adding new features, its correspondents have to be changed simultaneously; if the correspondents are not changed inadvertently, bugs are newly introduced to them. Various kinds of research efforts have been performed for resolving or improving the problems caused by the presence of duplicate code. For example, there are currently a variety of techniques available to detect duplicate code [1]. In addition, there are many research efforts for merging duplicate code as a single module like function or method, or for preventing duplications from being overlooked in modification [2, 3]. However, there are precisely the opposite opinions that code cloning is a good choice for design of the source code [4]. In order to answer the question whether duplicate code is harmful or not, several efforts have proposed comparison methods
%U http://www.hindawi.com/journals/ase/2012/938296/