Improving Software Sustainability Through Data-Driven Technical Debt Management

Abstract

We define technical debt as a software design issue that: 1) Exists in an executable system artifact, such as code, build scripts, automated test suites; 2) Is traced to several locations in the system, implying ripple effects of impact of change; 3) Has a quantifiable effect on system attributes of interest to developers, such as increasing number of defects, negative change in maintainability and code quality indicators are symptoms of technical debt and 4) We initially focus on detecting indicators in the form of violating known architectural pattern and maintainability rules to trace such symptoms.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 07, 2015
Accession Number
AD1128247

Entities

People

  • Ipek Ozkaya

Organizations

  • Carnegie Mellon University

Tags

Communities of Interest

  • Autonomy
  • Biomedical

DTIC Thesaurus Topics

  • Acquisition
  • Case Studies
  • Communication Terminals
  • Computer Programs
  • Contractors
  • Contracts
  • Data Sets
  • Databases
  • Department Of Defense
  • Detection
  • Engineering
  • Governments
  • Guarantees
  • Indicators
  • Materials
  • Software Design
  • Software Development
  • Technical Debt
  • Test Beds
  • Universities

Fields of Study

  • Computer science
  • Engineering

Readers

  • Circadian Sleep-Wake Regulation and Chronobiology
  • Computer Science.
  • Systems Analysis and Design