Towards a model for optimizing technical debt in software products

Abstract

There is a growing interest in applying the technical debt metaphor to investigate issues related to the tradeoff of the likely long-term costs associated with software design shortcuts for expected short-term business benefits in terms of increased earlier functionality. We propose an optimization model that contrasts the patterns of technical debt accumulation in a software product with the patterns of consumer adoption of the product throughout its evolution. This facilitates a rigorous and balanced analysis of the pros and cons of accumulating technical debt at various lifecycle stages of a software product. We discuss the use of the optimization model to derive policies for managing technical debt and the potential for empirical tests of the model and other future interdisciplinary research.

Identification and management of technical debt: A systematic mapping study

“…Benchmarking-Based Model [Mayr et al, 2014] 1 Model for optimizing technical debt [Ramasubbu and Kemerer, 2013] 1 Finance and accounting practices [Conroy, 2012] 1…”

RT 128: New Project Incubator

“…Over the past three decades, about a hundred of studies have been published, with their topics ranging from TD conceptual analysis (e.g., (Siebra et al, 2012;Schmid, 2013), detection (e.g., (Marinescu, 2004;Wong et al, 2011;Marinescu, 2012;Zazworka et al, 2014), to evaluation (e.g., (Izurieta et al, 2013;Ktata and Lévesque, 2010;Nugroho et al, 2011). Although the increasing number of studies produced significant benefits in defining and assessing TD as well as improving software quality (Sharma, 2012;Ramasubbu and Kemerer, 2013;Griffith et al, 2014), there lacks of a consistent and consolidated view of the definitions and determinant factors of TD, which may result in confusion on its detection and evaluation in both academic research and industrial practice. For instance, in the state-of-the-art TD research, code smell is often considered as a cause of TD (Fontana et al, 2012;Guo et al, 2010;Ligu et al, 2013) that could be naturally classified as the code debt.…”

A simulation study of practical methods for technical debt management in agile software development

“…Currently, several basic methods for managing technical debt in practice have been proposed, yet there is little empirical work supporting these claims (Ramasubbu and Kemerer 2013), due to the nature of the problem making empirical studies prohibitive. Thus, simulation provides an excellent alternative to evaluate proposed technical debt management methods, within the context of agile development processes, in a cost and time sensitive way.…”