Systems Analysis and Design (SIG SAND)

Paper Type

Complete

Paper Number

1416

Description

Context: Technical debt (TD) contextualizes the problem of pending development tasks as a type of debt that brings a short-term benefit to the project, but which may have to be paid with interest later. When development teams explicitly admit these issues, TD is named Self-Admitted Technical Debt (SATD). Objective: Evaluate a SATD identification strategy, based on a SATD conceptual model implemented by the eXcomment tool, identifying comment patterns that are good indicators of the presence of SATD and those that point to false positive items. Method: We conducted an empirical study, considering two large and mature open-source software projects. Results: We identified 25 patterns that are more commonly found in code comments that point to SATD items and patterns that point to false-positive comments, indicating that those patterns can be excluded from the SATD conceptual model. Conclusion: We provide new evidence on how software engineers can use code comments to detect SATD items automatically.

Share

COinS
 
Aug 9th, 12:00 AM

On Comment Patterns that are Good Indicators of the Presence of Self-Admitted Technical Debt and those that Lead to False Positive Items

Context: Technical debt (TD) contextualizes the problem of pending development tasks as a type of debt that brings a short-term benefit to the project, but which may have to be paid with interest later. When development teams explicitly admit these issues, TD is named Self-Admitted Technical Debt (SATD). Objective: Evaluate a SATD identification strategy, based on a SATD conceptual model implemented by the eXcomment tool, identifying comment patterns that are good indicators of the presence of SATD and those that point to false positive items. Method: We conducted an empirical study, considering two large and mature open-source software projects. Results: We identified 25 patterns that are more commonly found in code comments that point to SATD items and patterns that point to false-positive comments, indicating that those patterns can be excluded from the SATD conceptual model. Conclusion: We provide new evidence on how software engineers can use code comments to detect SATD items automatically.