Software Engineering Practices Research Articles

What Could Possibly Go Wrong: Undesirable Patterns in Collective Development

Software development, often perceived as a technical endeavour, is fundamentally a social activity requiring collaboration among team members. Acknowledging this, the software development community has devised strategies to address possible collaboration-related shortcomings. Various studies have attempted to capture the social dynamics within software engineering. These studies developed methods to identify numerous teamwork issues and proposed various approaches to address them. However, there is a need for a comprehensive bottom-up exploration from practitioner’s perceptions to common patterns. This paper introduces the concept of undesirable patterns in collective development, referring to potential teamwork problems that may escalate if unaddressed. Through 38 in-depth exploratory interviews, we identify and classify 42 patterns, revealing their origins and consequences. To the best of our knowledge, some patterns, like Teamwork pipeline bottleneck , were never reported before. Subsequent surveys, 436 and 968 participants each, explore the significance and frequency of the undesirable patterns, and evaluate potential tools and features to manage these patterns. The study contributes a nuanced understanding of undesirable patterns, evaluating their impact and proposing pragmatic tools and features for industrial application. The findings provide a valuable foundation for further in-depth studies and the development of tools to enhance collaborative software engineering practices.

Empowering Diversity by Building Inclusive Software Engineering Projects with Large Language Models

The integration of emerging AI technologies, particularly Large Language Models (LLMs), is fundamentally reshaping the landscape of software engineering. LLMs offer a wide array of capabilities that can enhance various aspects of software development, including design assistance, automated code analysis and synthesis, and testing. Consequently, their integration into software engineering practices holds the potential to greatly improve efficiency and code quality, marking a notable paradigm shift towards the creation of more intelligent and adaptive systems. This allows the definition of an extended perspective when establishing requirements for building software engineering solutions by incorporating evidence from literature resources and practice. By synthesizing requirements and evidence from scholarly and practitioner sources using LLMs, software engineers can ensure that their solutions are not only technically sound but also align with best practices in fostering social norms and values like inclusivity. Adopting such an approach supports the creation of responsible and inclusive educational software that caters to diverse learning needs and promotes equitable access to educational resources. Moreover, by using LLMs to inform decision-making throughout the software development lifecycle, software engineers can iteratively refine and enhance their solutions based on emerging research findings, thereby ensuring continuous improvement in fostering inclusive educational environments. Hence, this research aims to develop a novel evidence-based software engineering method informed by insights from scientific literature. As a use case, we design and implement a dyslexia-oriented educational software application that supports children in learning to read, guided by this new methodological approach.

The well-being of software engineers: a systematic literature review and a theory

For decades, software engineering research and practice has focused primarily on technological and process-related factors. Today, there is a growing interest in organizational, social, and psychological factors, including well-being. Organizational studies show that well-being contributes to work outcomes, including creativity, performance, and productivity. But despite its importance, the predictors and outcomes of software engineers’ well-being as a multidimensional construct to date are under-researched. This paper reports on the multidimensional well-being of professional software engineers and generates insights for the future research in this area. 44 quantitative survey studies published between 2000 and 2023 were selected and synthesized both quantitatively and qualitatively through a systematic literature review. The results of the review were further analyzed to construct a quantitatively-testable theory, detailing the predictors and outcomes of well-being in software engineering organizations. The total number of research participants included in the selected studies is 16,086 software engineering professionals from at least 42 countries. The literature review identified various measures, constructs, and indicators of well-being, as well as its predictors and outcomes. The theory, based on cumulative results of carefully selected quantitative studies, is an attempt to “correct the record” by establishing well-being in software engineering as a meta-construct of hedonic, eudaimonic, and integrated or hedaimonic qualities predicted by different individual, team and organizational factors and impacting the functioning of software engineers and their organizations. The review highlighted the under-researched aspects of well-being in software engineering and confirmed the need for more advanced quantitative studies. We hope that the theory will benefit researchers in conducting future studies and practitioners in developing nuanced and science-based interventions for improving software engineers’ well-being.

Enhancing software engineering practices with generative AI: A framework for automated code synthesis and refactoring

This paper is based on how software development has been revolutionized using AI in automation, mainly dealing with code synthesis and rewrite frameworks. While there is no focused definition for software development with AI technologies, their application in development processes is unlikely to remain marginal as they mature and provide higher productivity, improved code quality, and enhanced ability for automating repetitive tasks in development. Automating coding means that predefined tools can bring code suggestions, show and apply refactoring features, and enforce coding standards so developers can devote their efforts to other aspects of software development. The following paper describes the practical systems that should be adopted before AI can be incorporated into software development. It also covers essential factors such as ethical, security, and quality aspects regarding the proper use of artificial intelligence. Lastly, the paper focuses on the direction and innovations in the future, including domain-specific AI models, better explainability of the AI solutions, and collaborative tools that can work according to improved and modified development practices. The framework is intended to balance machine learning with human experience to help developers utilize the benefits that an AI-Based Software Development Framework can render.

From Code to Care: The Impact of Software Engineering and QA on Healthcare Delivery Systems

The rapid digitalization of healthcare has led to an unprecedented reliance on software systems to improve patient care, streamline operations, and ensure regulatory compliance. This article presents a comprehensive analysis of the critical role that software engineering and quality assurance (QA) play in three key areas of modern healthcare: Electronic Health Records (EHRs), telemedicine platforms, and medical device software. Through an examination of current practices, challenges, and case studies, we demonstrate how robust software engineering methodologies and rigorous QA processes contribute to the development of reliable, secure, and efficient healthcare applications. Our findings highlight the importance of data integrity, interoperability, user experience, and regulatory compliance in healthcare software development. Furthermore, we explore emerging trends such as artificial intelligence, blockchain, and the Internet of Medical Things (IoMT), discussing their implications for future software engineering and QA practices in healthcare. This article provides valuable insights for healthcare organizations, software developers, and policymakers, offering recommendations to enhance the quality and safety of software-driven healthcare solutions.

Using rapid reviews to support software engineering practice: a systematic review and a replication study

Using rapid reviews to support software engineering practice: a systematic review and a replication study

Large-scale agile security practices in software engineering

Purpose Security in large-scale agile is a crucial aspect that should be carefully addressed to ensure the protection of sensitive data, systems and user privacy. This study aims to identify and characterize the security practices that can be applied in managing large-scale agile projects. Design/methodology/approach A qualitative study is carried out through 18 interviews with 6 software development companies based in Portugal. Professionals who play the roles of Product Owner, Scrum Master and Scrum Member were interviewed. A thematic analysis was applied to identify deductive and inductive security practices. Findings The findings identified a total of 15 security practices, of which 8 are deductive themes and 7 are inductive. Most common security practices in large-scale agile include penetration testing, sensitive data management, automated testing, threat modeling and the implementation of a DevSecOps approach. Originality/value The results of this study extend the knowledge about large-scale security practices and offer relevant practical contributions for organizations that are migrating to large-scale agile environments. By incorporating security practices at every stage of the agile development lifecycle and fostering a security-conscious culture, organizations can effectively address security challenges in large-scale agile environments.

VeriFog: A Generic Model-based Approach for Verifying Fog Systems at Design Time and Generating Deployment Configurations

Fog Computing is a paradigm decentralizing the Cloud by geographically distributing computation, storage, network resources and related services. It provides benefits such as reducing the number of bottlenecks, limiting unwanted data movements, etc. However, managing the size, complexity and heterogeneity of the Fog systems to be engineered is challenging and can quickly become costly. According to best practices in software engineering, verification tasks could be performed on a system design prior to its implementation and deployment. We propose a generic model-based approach for verifying Fog systems at design time, also enabling the automatic generation of corresponding deployment configuration files. Named VeriFog, this approach is notably based on a customizable Fog Modeling Language (FML). We experimented in practice by modeling three use cases, from three different application domains, and by considering three main types of non-functional properties to be verified. From this modeling and verification effort, we show that we are able to automatically generate usable deployment configuration files for different deployment tools. In direct collaboration with our industrial partner Smile, the approach and underlying language presented in this paper are necessary steps towards a more global model-based support for the complete life cycle of Fog systems.

Performance Comparison of AI Platforms in Solving Computer Science Problems

In the rapidly evolving landscape of artificial intelligence (AI), its significance and accelerated development are undeniable. AI has emerged as a cornerstone technology with profound implications across various domains, driving innovation and reshaping the way we approach complex problems. Particularly, the utilization of AI in coding tasks has garnered substantial attention, given its potential to streamline development processes and enhance the efficiency of software engineering practices. Against this backdrop, this paper presents a detailed comparative analysis of four different AI platforms, namely ChatGPT, Gemini, Blackbox, and Microsoft Copilot, in addressing key challenges within the realm of computer science, spanning natural language processing, image processing, and cybersecurity. The study focuses on leveraging the C++ programming language to develop solutions for these multifaceted problems across the aforementioned platforms. Each platform's outputs are meticulously evaluated on various parameters including accuracy, execution time, code size, and time complexity to provide a comprehensive understanding of their performance. Furthermore, an iterative optimization methodology is employed, entailing three rounds of refinement for the code produced by each platform, with the resultant outputs subjected to comparative analysis in each iteration. Through this rigorous approach, the paper not only elucidates the efficacy of different AI platforms in addressing diverse computational challenges but also underscores the iterative enhancement process on AI platforms for refining code quality and performance across multiple domains within computer science.

Transformer-based models application for bug detection in source code

This paper explores the use of transformer-based models for bug detection in source code, aiming to better understand the capacity of these models to learn complex patterns and relationships within the code. Traditional static analysis tools are highly limited in their ability to detect semantic errors, resulting in numerous defects passing through to the code execution stage. This research represents a step towards enhancing static code analysis using neural networks. The experiments were designed as binary classification tasks to detect buggy code snippets, each targeting a specific defect type such as NameError, TypeError, IndexError, AttributeError, ValueError, EOFError, SyntaxError, and ModuleNotFoundError. Utilizing the «RunBugRun» dataset, which relies on code execution results, the models – BERT, CodeBERT, GPT-2, and CodeT5 – were fine-tuned and compared under identical conditions and hyperparameters. Performance was evaluated using F1-Score, Precision, and Recall. The results indicated that transformer-based models, especially CodeT5 and CodeBERT, were effective in identifying various defects, demonstrating their ability to learn complex code patterns. However, performance varied by defect type, with some defects like IndexError and TypeError being more challenging to detect. The outcomes underscore the importance of high-quality, diverse training data and highlight the potential of transformer-based models to achieve more accurate early defect detection. Future research should further explore advanced transformer architectures for detecting complicated defects, and investigate the integration of additional contextual information to the detection process. This study highlights the potential of modern machine learning architectures to advance software engineering practices, leading to more efficient and reliable software development.

Sustainable systematic literature reviews

Context:Systematic Literature Reviews (SLR) have been recognized as an important research method for summarizing evidence in Software Engineering (SE). At the same, SLR still presents several problems, such as the high resource consumption (mainly human resources) and lack of effective impact on SE practitioners, although much research has already been done. Objective:The main goal of this paper is to explore the concept of sustainability in the SLR area, intending to contribute to understanding better and solving such problems in an integrated way. More specifically, this paper characterizes what sustainable SLR are, their core characteristics, critical factors (i.e., sensitive points in the SLR process), and guidelines for conducting such SLR. Methods:We performed a meta-ethnographic study to find key concepts of sustainable software systems and transpose them to sustainable SLR. For this, we systematically selected 16 studies about sustainable software systems and 14 distinguished studies about SLR. Following, we extracted the main keywords and metaphors, determined how both areas are correlated, and transposed them to obtain a set of core characteristics of sustainable SLR as well as critical factors and guidelines. Additionally, we validated them with specialists using the Delphi method. Results:We found 15 core characteristics that offer a broad view of sustainable SLR, 15 critical factors in the SLR process that should be carefully addressed when conducting and updating SLR, and also 16 guidelines to manage SLR from the sustainability perspective. Conclusion:The concept of sustainability in SLR can contribute to solving SLR problems in a more integrated way, while this work could change the mindset of the SLR community about the need to conduct sustainable SLR.

How do software practitioners perceive human-centric defects?

Context:Human-centric software design and development prioritises the way users prefer to complete their jobs, rather than expecting users to adapt to the software. Software users can have different genders, ages, cultures, languages, disabilities, socioeconomic statuses, and educational backgrounds, among many other differences. Due to the inherently varied nature of these differences and their impact on software usage, preferences and issues of users can vary, resulting in user-specific defects that we term as ‘human-centric defects’ (HCDs). Objective:This research aims to understand the perception and current management practices of such HCDs by software practitioners, identify key challenges in reporting, understanding and fixing them, and provide recommendations to improve HCDs management in software engineering. Methods:We conducted a survey and interviews with software engineering practitioners to gauge their knowledge and experience on HCDs and the defect tracking process. Results:We analysed fifty (50) survey- and ten (10) interview-responses from SE practitioners and identified that there are multiple gaps in the current management of HCDs in software engineering practice. There is a lack of awareness regarding human-centric aspects, causing them to be lost or under-appreciated during software development. Our results revealed that handling HCDs could be improved by following a better feedback process with end-users, a more descriptive taxonomy, and suitable automation. Conclusion:HCDs, given their diverse end-user base, present a major challenge to software practitioners. In the software engineering domain, research on HCDs has been limited and requires effort from research and practice communities to create awareness and support for human-centric aspects.

Evolving Software Architecture Design in Telemedicine: A PRISMA-based Systematic Review.

This article presents a systematic review of recent advancements in telemedicine architectures for continuous monitoring, providing a comprehensive overview of the evolving software engineering practices underpinning these systems. The review aims to illuminate the critical role of telemedicine in delivering healthcare services, especially during global health crises, and to emphasize the importance of effectiveness, security, interoperability, and scalability in these systems. A systematic review methodology was employed, adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses framework. As the primary research method, the PubMed, IEEE Xplore, and Scopus databases were searched to identify articles relevant to telemedicine architectures for continuous monitoring. Seventeen articles were selected for analysis, and a methodical approach was employed to investigate and synthesize the findings. The review identified a notable trend towards the integration of emerging technologies into telemedicine architectures. Key areas of focus include interoperability, security, and scalability. Innovations such as cognitive radio technology, behavior-based control architectures, Health Level Seven International (HL7) Fast Healthcare Interoperability Resources (FHIR) standards, cloud computing, decentralized systems, and blockchain technology are addressing challenges in remote healthcare delivery and continuous monitoring. This review highlights major advancements in telemedicine architectures, emphasizing the integration of advanced technologies to improve interoperability, security, and scalability. The findings underscore the successful application of cognitive radio technology, behavior-based control, HL7 FHIR standards, cloud computing, decentralized systems, and blockchain in advancing remote healthcare delivery.

SERP4IoT'24 Workshop Report

We face a new software crisis. In 1968, computer scientists learned that developing robust software requires skills, methods, and tools. Today, software and hardware engineers realize that developing a robust Internet of Things (IoT) also pushes the states of their art and practice. Recent news illustrates the many problems IoT faces: from lack of interoperability to broken updates to massive security attacks. In this context, the 6th International Workshop on Software Engineering Research and Practices for the Internet of Things (SERP4IoT) aims to provide a highly interactive forum for researchers and practitioners to address the challenges of, find solutions for, and share experiences with the development, release, and testing of robust software for IoT systems.

A Disruptive Research Playbook for Studying Disruptive Innovations

As researchers today, we are witnessing a fundamental change in our technologically-enabled world due to the advent and diffusion of highly disruptive technologies such as generative Artificial Intelligence (AI), Augmented Reality (AR) and Virtual Reality (VR). In particular, software engineering has been profoundly affected by the transformative power of disruptive innovations for decades, with a significant impact of technical advancements on social dynamics due to its socio-technical nature. In this paper, we reflect on the importance of formulating and addressing research problems in software engineering through a socio-technical lens, thus ensuring a holistic understanding of the complex phenomena in this field. We propose a research playbook with the aim of providing a guide to formulate compelling and socially relevant research questions and to identify the appropriate research strategies for empirical investigations, with an eye on the long-term implications of technologies or their use. We showcase how to apply the research playbook. Firstly, we show how it can be used retrospectively to reflect on a prior disruptive technology, Stack Overflow, and its impact on software development. Secondly, we show how it can be used to question the impact of two current disruptive technologies: AI and AR/VR. Finally, we introduce a specialized GPT model to support the researcher in framing future investigations. We conclude by discussing the broader implications of adopting the playbook for both researchers and practitioners in software engineering and beyond.

Can GPT-4 Replicate Empirical Software Engineering Research?

Empirical software engineering research on production systems has brought forth a better understanding of the software engineering process for practitioners and researchers alike. However, only a small subset of production systems is studied, limiting the impact of this research. While software engineering practitioners could benefit from replicating research on their own data, this poses its own set of challenges, since performing replications requires a deep understanding of research methodologies and subtle nuances in software engineering data. Given that large language models (LLMs), such as GPT-4, show promise in tackling both software engineering- and science-related tasks, these models could help replicate and thus democratize empirical software engineering research. In this paper, we examine GPT-4’s abilities to perform replications of empirical software engineering research on new data. We specifically study their ability to surface assumptions made in empirical software engineering research methodologies, as well as their ability to plan and generate code for analysis pipelines on seven empirical software engineering papers. We perform a user study with 14 participants with software engineering research expertise, who evaluate GPT-4-generated assumptions and analysis plans (i.e., a list of module specifications) from the papers. We find that GPT-4 is able to surface correct assumptions, but struggles to generate ones that apply common knowledge about software engineering data. In a manual analysis of the generated code, we find that the GPT-4-generated code contains correct high-level logic, given a subset of the methodology. However, the code contains many small implementation-level errors, reflecting a lack of software engineering knowledge. Our findings have implications for leveraging LLMs for software engineering research as well as practitioner data scientists in software teams.

Challenges and strategies of software engineering in big data environment

Software engineering in a big data environment involves developing and maintaining large-scale, scalable, data-intensive applications in a rapidly evolving technology domain. As data volumes grow exponentially, so do the requirements for software systems to process and analyze this data. These systems need to be able to efficiently store, retrieve, process and analyze data collected from sources such as social media, IoT devices, enterprises and scientific research. Software engineers must address challenges including data heterogeneity and quality, scalability of data storage and computation, and real-time processing of high-speed data streams. With the evolution of technology, new architecture and database technologies have emerged, such as NoSQL and distributed computing frameworks. These technologies have challenged traditional methods of software engineering practice and also brought new design and implementation strategies.

A comprehensive review of techniques for documenting artificial intelligence

PurposeCompanies are increasingly benefiting from artificial intelligence (AI) applications in various domains, but also facing its negative impacts. The challenge lies in the lack of clear governance mechanisms for AI. While documentation is a key governance tool, standard software engineering practices are inadequate for AI. Practitioners are unsure about how to document AI, raising questions about the effectiveness of current documentation guidelines. This review examines whether AI documentation guidelines meet regulatory and industry needs for AI applications and suggests directions for future research.Design/methodology/approachA structured literature review was conducted. In total, 38 papers from top journals and conferences in the fields of medicine and information systems as well as journals focused on fair, accountable and transparent AI were reviewed.FindingsThis literature review contributes to the literature by investigating the extent to which current documentation guidelines can meet the documentation requirements for AI applications from regulatory bodies and industry practitioners and by presenting avenues for future research. This paper finds contemporary documentation guidelines inadequate in meeting regulators’ and professionals’' expectations. This paper concludes with three recommended avenues for future research.Originality/valueThis paper benefits from the insights from comprehensive and up-to-date sources on the documentation of AI applications.

WASMICO: Micro-containers in microcontrollers with WebAssembly

The Internet-of-Things (IoT) has created a complex environment where hardware and software interact in complex ways. Despite being a prime candidate for applying well-established software engineering practices, IoT has not seen the same level of adoption as other areas, such as cloud development. This discrepancy is even more evident in the case of edge devices, where programming and managing applications can be challenging due to their heterogeneous nature and dependence on specific toolchains and languages. However, the emergence of WebAssembly as a viable solution for running high-level languages on some devices presents an opportunity to streamline development practices, such as DevOps. In this paper, we present WASMICO — a firmware and command-line utility that allows for the execution and management of application lifecycles in microcontrollers. Our solution has been benchmarked against other state-of-the-art tools, demonstrating its feasibility, novel features, and empirical evidence that it outperforms some of the most widely used solutions for running high-level code on these devices. Overall, our work aims to promote the use of well-established software engineering practices in the IoT domain, helping to bridge the gap between cloud and edge development.

Current trends in digital twin development, maintenance, and operation: an interview study

Digital twins (DTs) are often defined as a pairing of a physical entity and a corresponding virtual entity (VE), mimicking certain aspects of the former depending on the use-case. In recent years, this concept has facilitated numerous use-cases ranging from design to validation and predictive maintenance of large and small high-tech systems. Various heterogeneous cross-domain models are essential for such systems, and model-driven engineering plays a pivotal role in the design, development, and maintenance of these models. We believe models and model-driven engineering play a similarly crucial role in the context of a VE of a DT. Due to the rapidly growing popularity of DTs and their use in diverse domains and use-cases, the methodologies, tools, and practices for designing, developing, and maintaining the corresponding VEs differ vastly. To better understand these differences and similarities, we performed a semi-structured interview research with 19 professionals from industry and academia who are closely associated with different lifecycle stages of digital twins. In this paper, we present our analysis and findings from this study, which is based on seven research questions. In general, we identified an overall lack of uniformity in terms of the understanding of digital twins and used tools, techniques, and methodologies for the development and maintenance of the corresponding VEs. Furthermore, considering that digital twins are software intensive systems, we recognize a significant growth potential for adopting more software engineering practices, processes, and expertise in various stages of a digital twin’s lifecycle.