TY - GEN
T1 - On automatically collectable metrics for software maintainability evaluation
AU - Ostberg, Jan Peter
AU - Wagner, Stefan
N1 - Publisher Copyright:
© 2014 IEEE.
PY - 2014/12/30
Y1 - 2014/12/30
N2 - In our work with industry partners as well as with students in seminars we noticed that many people often stick to 'aged' metrics when they want to evaluate the maintainability of software. They consider these metrics for this purpose without second thoughts, because the metrics are so present that almost every developer has at least heard the name and has some kind of knowledge about it. This smattering leads to an unreflected usage of theses metrics today without fully understanding them. To find suitable metrics for maintainability evaluation, we did a preliminary study, which indicated that still McCabe, Halstead and some kind of lines of code measurement dominate maintainability evaluation. We will discuss examples that question the ability of those metrics to reliably and automatically evaluate maintainability of software. Instead we will present metrics we consider more suitable to make solid statements about software maintainability. We conclude that our proposed metrics are better evaluators for maintainability because their impact on maintenance is clearer to interpret. Future work on metrics should carefully consider the use of 'aged' metrics.
AB - In our work with industry partners as well as with students in seminars we noticed that many people often stick to 'aged' metrics when they want to evaluate the maintainability of software. They consider these metrics for this purpose without second thoughts, because the metrics are so present that almost every developer has at least heard the name and has some kind of knowledge about it. This smattering leads to an unreflected usage of theses metrics today without fully understanding them. To find suitable metrics for maintainability evaluation, we did a preliminary study, which indicated that still McCabe, Halstead and some kind of lines of code measurement dominate maintainability evaluation. We will discuss examples that question the ability of those metrics to reliably and automatically evaluate maintainability of software. Instead we will present metrics we consider more suitable to make solid statements about software maintainability. We conclude that our proposed metrics are better evaluators for maintainability because their impact on maintenance is clearer to interpret. Future work on metrics should carefully consider the use of 'aged' metrics.
KW - Metrics
KW - Software Maintenance
KW - Software Quality Measurement
UR - http://www.scopus.com/inward/record.url?scp=84929623294&partnerID=8YFLogxK
U2 - 10.1109/IWSM.Mensura.2014.19
DO - 10.1109/IWSM.Mensura.2014.19
M3 - Conference contribution
AN - SCOPUS:84929623294
T3 - Proceedings - 2014 Joint Conference of the International Workshop on Software Measurement, IWSM 2014 and the International Conference on Software Process and Product Measurement, Mensura 2014
SP - 32
EP - 37
BT - Proceedings - 2014 Joint Conference of the International Workshop on Software Measurement, IWSM 2014 and the International Conference on Software Process and Product Measurement, Mensura 2014
A2 - Vogelezang, Frank
A2 - Daneva, Maya
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - Joint Conference of the 24th International Workshop on Software Measurement, IWSM 2014 and the 9th International Conference on Software Process and Product Measurement, Mensura 2014
Y2 - 6 October 2014 through 8 October 2014
ER -