Revisiting Technical Debt Types and Indicators for Software Systems


Caglayan D., Özcan Top Ö.

39th Annual ACM Symposium on Applied Computing (SAC), Avila, İspanya, 8 - 12 Nisan 2024, ss.834-841 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1145/3605098.3636043
  • Basıldığı Şehir: Avila
  • Basıldığı Ülke: İspanya
  • Sayfa Sayıları: ss.834-841
  • Anahtar Kelimeler: TD categorization, TD identification, technical debt, technical debt indicators, technical debt types
  • Orta Doğu Teknik Üniversitesi Adresli: Evet

Özet

Technical Debt (TD) 1 term in software systems was introduced over two decades ago and remains a critical concern in software development. It has the potential to evolve into a liability that necessitates refactoring or rewriting code over time. Regardless of its significance, there exists a notable gap in literature concerning a comprehensive list of technical debt indicators. The purpose of this study is to re-evaluate existing TD categorization and extend TD indicators and offer a complete and validated TD Type and TD Indicator list. In this study, we adopted a qualitative research approach and used mapping and expert opinion techniques as the research approach. The number of TD indicators extracted from existing formal literature was 60 which was extended to 92 by reviewing gray literature. This list was then subjected to the expert review, and with their feedback, grew by an additional 21%. Consequently, we present 10 distinct TD types, accompanied by 120 TD indicators that would aid in TD identification, resolution and minimizing the risks and costs associated with technical debt in software development.