While technical debt grows in absolute numbers as software systems evolve over time, the density of technical debt (technical debt divided by lines of code) is reduced in some cases. This can be explained by either the application of refactorings or the development of new artifacts with limited Technical Debt. In this paper we explore the second explanation, by investigating the relation between the amount of Technical Debt in new code and the evolution of Technical Debt in the system. To this end, we compare the Technical Debt Density of new code with existing code, and we investigate which of the three major types of code changes (additions, deletions and modifications) is primarily responsible for changes in the evolution of Technical Debt density. Furthermore, we study whether there is a relation between code quality practices and the ‘cleanness’ of new code. To obtain the required data, we have performed a large-scale case study on twenty-seven open-source software projects by the Apache Software Foundation, analyzing 66,661 classes and 56,890 commits. The results suggest that writing “clean” (or at least “cleaner”) new code can be an efficient strategy for reducing Technical Debt Density, and thus preventing software decay over time. The findings also suggest that projects adopting an explicit policy for quality improvement, e.g., through discussions on code quality in board meetings, are associated with a higher frequency of cleaner new code commits. Therefore, we champion the establishment of processes that monitor the density of Technical Debt of new code to control the accumulation of Technical Debt in a software system.

During the software development life cycle, assumptions are an important type of software development knowledge that can be extracted from textual artifacts. Analyzing assumptions can help to, for example, comprehend software design and further facilitate software maintenance. Manual identification of assumptions by stakeholders is rather time-consuming, especially when analyzing a large dataset of textual artifacts. To address this problem, one promising way is to use automatic techniques for assumption identification. In this study, we conducted an experiment to evaluate the performance of existing machine learning classification algorithms for automatic assumption identification, through a dataset extracted from the Hibernate developer mailing list. The dataset is composed of 400 "Assumption" sentences and 400 "Non-Assumption" sentences. Seven classifiers using different machine learning algorithms were selected and evaluated. The experiment results show that the SVM algorithm achieved the best performance (with a precision of 0.829, a recall of 0.812, and an F1-score of 0.819). Additionally, according to the ROC curves and related AUC values, the SVM-based classifier comparatively performed better than other classifiers for the binary classification of assumptions.

Software reuse is a well-established software engineering process that aims at improving development productivity. Although reuse can be performed in a systematic way (e.g., through product lines), in practice, reuse is performed in many cases opportunistically, i.e., copying small code chunks either from the web or in-house developed projects. Knowledge sharing communities and especially StackOverflow constitute the primary source of code-related information for amateur and professional software developers. Despite the obvious benefit of increased productivity, reuse can have a mixed effect on the quality of the resulting code depending on the properties of the reused solutions. An efficient concept for capturing a wide-range of internal software qualities is the metaphor of Technical Debt which expresses the impact of shortcuts in software development on its maintenance costs. In this paper, we present the results of an empirical study on the relation between the existence of reusing code retrieved from StackOverflow on the technical debt of the target system. In particular, we study several open-source projects and identify non-trivial pieces of code that exhibit a perfect or near-perfect match with code provided in the context of answers in StackOverflow. Then, we compare the technical debt density of the reused fragments, obtained as the ratio of inefficiencies identified by SonarQube over the lines of reused code, to the technical debt density of the target codebase. The results provide insights to the potential impact of small-scale code reuse on technical debt and highlight the benefits of assessing code quality before committing changes to a repository.

Source code refactorings and transformations are extensively used by embedded system developers to improve the quality of applications, often supported by various open source and proprietary tools. They either aim at improving the design time quality such as the maintainability and reusability of software artifacts, or the runtime quality such as performance and energy efficiency. However, an inherent trade-off between design- and run-time qualities is often present posing challenges to embedded software development. This work is a first step towards the investigation of the impact of transformations for improving the performance and the energy efficiency on software quality metrics and the impact of refactorings for increasing the design time quality on the execution time, the memory and the energy consumption. Based on a set of embedded applications from widely used benchmark suites and typical transformations and refactorings, we identify interrelations and trade-offs between the aforementioned metrics.

During software evolution technical debt (TD) follows a constant ebb and flow, being incurred and paid back, sometimes in the same day and sometimes ten years later. There have been several studies in the literature investigating how technical debt in source code accumulates during time and the consequences of this accumulation for software maintenance. However, to the best of our knowledge there are no large scale studies that focus on the types of issues that are fixed and the amount of TD that is paid back during software evolution. In this paper we present the results of a case study, in which we analyzed the evolution of fifty-seven Java open-source software projects by the Apache Software Foundation at the temporal granularity level of weekly snapshots. In particular, we focus on the amount of technical debt that is paid back and the types of issues that are fixed. The findings reveal that a small subset of all issue types is responsible for the largest percentage of TD repayment and thus, targeting particular violations the development team can achieve higher benefits.