On the diffuseness of technical debt items and accuracy of remediation time when using SonarQube

Tutkimustuotos › › vertaisarvioitu

Standard

On the diffuseness of technical debt items and accuracy of remediation time when using SonarQube. / Baldassarre, Maria Teresa; Lenarduzzi, Valentina; Romano, Simone; Saarimäki, Nyyti.

julkaisussa: Information and Software Technology, Vuosikerta 128, 106377, 2020.

Tutkimustuotos › › vertaisarvioitu

Author

Baldassarre, Maria Teresa ; Lenarduzzi, Valentina ; Romano, Simone ; Saarimäki, Nyyti. / On the diffuseness of technical debt items and accuracy of remediation time when using SonarQube. Julkaisussa: Information and Software Technology. 2020 ; Vuosikerta 128.

Bibtex - Lataa

@article{d535e67f3460460d896a33fe32b9b583,

title = "On the diffuseness of technical debt items and accuracy of remediation time when using SonarQube",

abstract = "Context. Among the static analysis tools available, SonarQube is one of the most used. SonarQube detects Technical Debt (TD) items—i.e., violations of coding rules—and then estimates TD as the time needed to remedy TD items. However, practitioners are still skeptical about the accuracy of remediation time estimated by the tool. Objective. In this paper, we analyze both diffuseness of TD items and accuracy of remediation time, estimated by SonarQube, to fix TD items on a set of 21 open-source Java projects. Method. We designed and conducted a case study where we asked 81 junior developers to fix TD items and reduce the TD of 21 projects. Results. We observed that TD items are diffused in the analyzed projects and most items are code smells. Moreover, the results point out that the remediation time estimated by SonarQube is inaccurate and, as compared to the actual time spent to fix TD items, is in most cases overestimated. Conclusions. The results of our study are promising for practitioners and researchers. The former can make more aware decisions during project execution and resource management, the latter can use this study as a starting point for improving TD estimation models.",

keywords = "Case study, Effort estimation, Remediation time, Sonarqube, Technical debt",

author = "Baldassarre, {Maria Teresa} and Valentina Lenarduzzi and Simone Romano and Nyyti Saarim{\"a}ki",

note = "EXT={"}Lenarduzzi, Valentina{"}",

year = "2020",

doi = "10.1016/j.infsof.2020.106377",

language = "English",

volume = "128",

journal = "Information and Software Technology",

issn = "0950-5849",

publisher = "Elsevier",

}

RIS (suitable for import to EndNote) - Lataa

TY - JOUR

T1 - On the diffuseness of technical debt items and accuracy of remediation time when using SonarQube

AU - Baldassarre, Maria Teresa

AU - Lenarduzzi, Valentina

AU - Romano, Simone

AU - Saarimäki, Nyyti

N1 - EXT="Lenarduzzi, Valentina"

PY - 2020

Y1 - 2020

N2 - Context. Among the static analysis tools available, SonarQube is one of the most used. SonarQube detects Technical Debt (TD) items—i.e., violations of coding rules—and then estimates TD as the time needed to remedy TD items. However, practitioners are still skeptical about the accuracy of remediation time estimated by the tool. Objective. In this paper, we analyze both diffuseness of TD items and accuracy of remediation time, estimated by SonarQube, to fix TD items on a set of 21 open-source Java projects. Method. We designed and conducted a case study where we asked 81 junior developers to fix TD items and reduce the TD of 21 projects. Results. We observed that TD items are diffused in the analyzed projects and most items are code smells. Moreover, the results point out that the remediation time estimated by SonarQube is inaccurate and, as compared to the actual time spent to fix TD items, is in most cases overestimated. Conclusions. The results of our study are promising for practitioners and researchers. The former can make more aware decisions during project execution and resource management, the latter can use this study as a starting point for improving TD estimation models.

AB - Context. Among the static analysis tools available, SonarQube is one of the most used. SonarQube detects Technical Debt (TD) items—i.e., violations of coding rules—and then estimates TD as the time needed to remedy TD items. However, practitioners are still skeptical about the accuracy of remediation time estimated by the tool. Objective. In this paper, we analyze both diffuseness of TD items and accuracy of remediation time, estimated by SonarQube, to fix TD items on a set of 21 open-source Java projects. Method. We designed and conducted a case study where we asked 81 junior developers to fix TD items and reduce the TD of 21 projects. Results. We observed that TD items are diffused in the analyzed projects and most items are code smells. Moreover, the results point out that the remediation time estimated by SonarQube is inaccurate and, as compared to the actual time spent to fix TD items, is in most cases overestimated. Conclusions. The results of our study are promising for practitioners and researchers. The former can make more aware decisions during project execution and resource management, the latter can use this study as a starting point for improving TD estimation models.

KW - Case study

KW - Effort estimation

KW - Remediation time

KW - Sonarqube

KW - Technical debt

U2 - 10.1016/j.infsof.2020.106377

DO - 10.1016/j.infsof.2020.106377

M3 - Article

VL - 128

JO - Information and Software Technology

JF - Information and Software Technology

SN - 0950-5849

M1 - 106377

ER -

TUTCRIS