Large Open Source Software Research Articles

We’re in This Together: Visible Social Support Actions in Virtual Teams Using Enterprise Social Media

Enterprise social networks (ESNs) are a communication standard within virtual teams. Among other affordances, ESNs enable colleagues to provide each other with social support. In this paper, we analyzed the message logs of virtual teams in a large open-source software project to determine how virtual teams use ESNs to provide particular forms of social support to each other and, secondly, to determine how the visibility of these interaction patterns influences team functioning. Our findings reveal distinct ESN use patterns in relation to four types of social support, each described using a metaphor: a dynamic notice board for information sharing, a community of practice for teaching and knowledge sharing, a team huddle for emotional support, and a job board for instrumental support. The findings provide structure to the diverse set of social support actions in this context and identify the higher-order functions that the visibility of social support produces.

A Microstructural Approach to Self-Organizing: The Emergence of Attention Networks

A recent line of inquiry investigates new forms of organizing as bundles of novel solutions to universal problems of resource allocation and coordination: how to allocate organizational problems to organizational participants and how to integrate participants’ resulting efforts. We contribute to this line of inquiry by reframing organizational attention as the outcome of a concatenation of self-organizing, microstructural mechanisms linking multiple participants to multiple problems, thus giving rise to an emergent attention network. We argue that, when managerial hierarchies are absent and authority is decentralized, observable acts of attention allocation produce interpretable signals that help participants to direct their attention and share information on how to coordinate and integrate their individual efforts. We theorize that the observed structure of an organizational attention network is generated by the concatenation of four interdependent micromechanisms: focusing, reinforcing, mixing, and clustering. In a statistical analysis of organizational problem solving within a large open-source software project, we find support for our hypotheses about the self-organizing dynamics of the observed attention network connecting organizational problems (software bugs) to organizational participants (volunteer contributors). We discuss the implications of attention networks for theory and practice by emphasizing the self-organizing character of organizational problem solving. We discuss the generalizability of our theory to a wider set of organizations in which participants can freely allocate their attention to problems and the outcomes of their allocation are publicly observable without cost. Funding: Financial support for this work was provided by the Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung [Grant 100018_150126] (“Relational event modes for bipartite networks with application to collaborative problem solving,” P.I. Alessandro Lomi) and by the Deutsche Forschungsgemeinschaft [Grant 321869138] (“Statistical analysis of time-stamped multi-actor events in social networks,” P.I. Jüergen Lerner). Supplemental Material: The supplemental video containing the dynamic visualization of the data is available at https://zenodo.org/record/7564503 and in the e-companion (available at https://doi.org/10.1287/orsc.2023.1674 ).

A taxonomy for mining and classifying privacy requirements in issue reports

Digital and physical footprints are a trail of user activities collected over the use of software applications and systems. As software becomes ubiquitous, protecting user privacy has become challenging. With the increasing of user privacy awareness and advent of privacy regulations and policies, there is an emerging need to implement software systems that enhance the protection of personal data processing. However, existing privacy regulations and policies only provide high-level principles which are difficult for software engineers to design and implement privacy-aware systems. In this paper, we develop a taxonomy that provides a comprehensive set of privacy requirements based on two well-established and widely-adopted privacy regulations and frameworks, the General Data Protection Regulation (GDPR) and the ISO/IEC 29100. These requirements are refined into a level that is implementable and easy to understand by software engineers, thus supporting them to attend to existing regulations and standards. We have also performed a study on how two large open-source software projects (Google Chrome and Moodle) address the privacy requirements in our taxonomy through mining their issue reports. The paper discusses how the collected issues were classified, and presents the findings and insights generated from our study.

A Software Requirements Ecosystem: Linking Forum, Issue Tracker, and FAQs for Requirements Management

User feedback is an important resource in modern software development, often containing requirements that help address user concerns and desires for a software product. The feedback in online channels is a recent focus for software engineering researchers, with multiple studies proposing automatic analysis tools. In this work, we investigate the product forums of two large open source software projects. Through a quantitative analysis, we show that forum feedback is often manually linked to related issue tracker entries and product documentation. By linking feedback to their existing documentation, development teams enhance their understanding of known issues, and direct their users to known solutions. We discuss how the links between forum, issue tracker, and product documentation form a requirements ecosystem that has not been identified in the previous literature. We apply state-of-the-art deep-learning to automatically match forum posts with related issue tracker entries. Our approach identifies requirement matches with a mean average precision of 58.9% and hit ratio of 82.2%. Additionally, we apply deep-learning using an innovative clustering technique, achieving promising performance when matching forum posts to related product documentation. We discuss the possible applications of these automated techniques to support the flow of requirements between forum, issue tracker, and product documentation.

Does working from home work? A natural experiment from lockdowns

Using tracked changes from a large open-source software platform, this paper studies how working from home affected the output of individuals working in tech. The basis of the natural experiment comes from idiosyncratic and state-imposed workplace closures during the COVID-19 pandemic. I find a negative but almost-negligible change in individual-level output of 0.5 percent (standard error of 0.091 percent). Overall, and based on descriptive analyses of the time-stamped data, tracked changes in software development cadences approximate regular work activity and provide a useful avenue for future studies of work.

Turnover of Companies in OpenStack: Prevalence and Rationale

To achieve commercial goals, companies have made substantial contributions to large open-source software (OSS) ecosystems such as OpenStack and have become the main contributors. However, they often withdraw their employees for a variety of reasons, which may affect the sustainability of OSS projects. While the turnover of individual contributors has been extensively investigated, there is a lack of knowledge about the nature of companies’ withdrawal. To this end, we conduct a mixed-methods empirical study on OpenStack to reveal how common company withdrawals were, to what degree withdrawn companies made contributions, and what the rationale behind withdrawals was. By analyzing the commit data of 18 versions of OpenStack, we find that the number of companies that have left is increasing and even surpasses the number of companies that have joined in later versions. Approximately 12% of the companies in each version have exited by the next version. Compared to the sustaining companies that joined in the same version, the withdrawn companies tend to have a weaker contribution intensity but contribute to a similar scope of repositories in OpenStack. Through conducting a developer survey, we find four aspects of reasons for companies’ withdrawal from OpenStack: company, community, developer, and project. The most common reasons lie in the company aspect, i.e., the company either achieved its goals or failed to do so. By fitting the survival analysis model, we find that commercial goals are associated with the probability of the company’s withdrawal, and that a company’s contribution intensity and scale are positively correlated with its retention. Maintaining good retention is important but challenging for OSS ecosystems, and our results may shed light on potential approaches to improve company retention and reduce the negative impact of company withdrawal.

Mining GitHub to Identify Open-Source Software Health in Blockchain Projects

GitHub represents the largest open-source software (OSS) community hosting over 100 million repositories by 56 million developers. Since the rise of Bitcoin [1], blockchain projects have grown dramatically and 84% of the top 200 are hosted on GitHub. Growth inspires many new initiatives and clones, but few projects survive. Identifying healthy projects can assist developers looking to contribute to OSS and researchers seeking innovative teams in addition to reducing fraudulent activity. This presentation is part of a larger ongoing study investigating OSS blockchain performance and health. The hypothesis to be studied is that the health of an OSS project can be determined through publicly available data. This will be investigated by answering the research question: How can factors that influence the health of OSS be identified? The methodology is adapted from a framework for extracting, processing, and analysing OSS [2] that combines eleven variables into a measurement model. Latent variables are introduced to represent the constructs of community engagement, project robustness, and public interest. Statistical analysis methods are employed both for the development and specification of the measurement model and the construction and evaluation of its structural efficacy. Validation is by two mechanisms: confirmatory factor analysis applied to the measurement model and structural equation modelling to estimate the validity of the relationships [3]. In the broader study a tool is to be developed to monitor health and highlight areas for innovation. Preliminary findings indicate that community engagement and project interest are positively correlated to robustness, which, in turn, is expected to be a predictor of software project health. A future hypothesis is that healthy projects can be a leading indicator of innovative technology and practice. If OSS communities and projects can be monitored for health, stake holders can identify areas of strength and weakness, extending to industries beyond blockchain.

Companies’ Participation in OSS Development–An Empirical Study of OpenStack

Commercial participation continues to grow in open source software (OSS) projects and novel arrangements appear to emerge in company-dominated projects and ecosystems. What is the nature of these novel arrangements? Does volunteers’ participation remain critical for these ecosystems? Despite extensive research on commercial participation in OSS, the exact nature and extent of company contributions to OSS development, and the impact of this engagement may have on the volunteer community have not been clarified. To bridge the gap, we perform an exploratory study of OpenStack: a large OSS ecosystem with intense commercial participation. We quantify companies’ contributions via the developers that they provide and the commits made by those developers. We find that companies made far more contributions than volunteers and the distribution of the contributions made by different companies is also highly unbalanced. We observe eight unique contribution models based on companies’ commercial objectives and characterize each model according to three dimensions: contribution intensity, extent, and focus. Companies providing full cloud solutions tend to make both intensive (more than other companies) and extensive (involving a wider variety of projects) contributions. Usage-oriented companies make extensive but less intense contributions. Companies driven by particular business needs focus their contributions on the specific projects addressing these needs. Minor contributors include community players (e.g., the Linux Foundation) and research groups. A model relating the number of volunteers to the diversity of contribution shows a strong positive association between them.

Being a Mentor in open source projects

Mentoring is a well-known way to help newcomers to Open Source Software (OSS) projects overcome initial contribution barriers. Through mentoring, newcomers learn to acquire essential technical, social, and organizational skills. Despite the importance of OSS mentors, they are understudied in the literature. Understanding who OSS project mentors are, the challenges they face, and the strategies they use can help OSS projects better support mentors’ work. In this paper, we employ a two-stage study to comprehensively investigate mentors in OSS. First, we identify the characteristics of mentors in the Apache Software Foundation, a large OSS community, using an online survey. We found that less experienced volunteer contributors are less likely to take on the mentorship role. Second, through interviews with OSS mentors (n=18), we identify the challenges that mentors face and how they mitigate them. In total, we identified 25 general mentorship challenges and 7 sub-categories of challenges regarding task recommendation. We also identified 13 strategies to overcome the challenges related to task recommendation. Our results provide insights for OSS communities, formal mentorship programs, and tool builders who design automated support for task assignment and internship.

Exploring and Expanding GSE Education with Open Source Software Development

Global software engineering (GSE) courses traditionally require cooperation between at least two universities so as to provide a distributed development environment to the students. In this study, we explore an alternative way to organize a global software engineering course where students work on open source software development (OSSD) projects rather than in a multi-university collaboration setting. The results show that the new setup may provide core GSE challenges as well as challenges associated with software development outsourcing and challenges related to working on large open source software. The present article compares the experiences gained from running a combined GSE and OSSD course against the experiences gained from running a traditional GSE course. The two alternatives are compared in terms of students’ learning outcomes and course organization. The authors found that a combined GSE and OSSD course provides learning opportunities that are partly overlapping with, and partly complementary to, a traditional GSE course. The authors also found that the combined OSSD and GSE course was somewhat easier to organize because most of the activities took place in a single university setting. The authors used the extended GSE taxonomy for the comparison and found it to be a useful tool for this, although it had some limitations in expressive power. Therefore, two additional relationship dimensions are proposed that will further enrich the extended taxonomy in classifying GSE (and OSSD) projects.

A Study of Measurement for Development Efficiency in Large Scale Open Source Software Projects

A Study of Measurement for Development Efficiency in Large Scale Open Source Software Projects

Discovering Program Topoi Through Clustering

Understanding source code of large open-source software projects is very challenging when there is only little documentation. New developers face the task of classifying a huge number of files and functions without any help. This paper documents a novel approach to this problem, called FEAT, that automatically extracts topoi from source code by using hierarchical agglomerative clustering. Program topoi summarize the main capabilities of a software system by presenting to developers clustered lists of functions together with an index of their relevant words. The clustering method used in FEAT exploits a new hybrid distance which combines both textual and structural elements automatically extracted from source code and comments. The experimental evaluation of FEAT shows that this approach is suitable to understand open-source software projects of size approaching 2,000 functions and 150 files, which opens the door for its deployment in the open-source community.

A Method for Recommending Bug Fixer Using Community Q&A Information

It is a very time-consuming task to assign a bug report to the most suitable fixer in large open source software projects. Therefore, it is very necessary to propose an effective recommendation method for bug fixer. Most research in this area translate it into a text classification problem and use machine learning or information retrieval methods to recommend the bug fixer. These methods are complex and overdependent on the fixers’ prior bug-fixing activities. In this paper, we propose a more effective bug fixer recommendation method which uses the community Q & A platforms (such as Stack Overflow) to measure the fixers’ expertise and uses the fixed bugs to measure the time-aware of fixers’ fixed work. The experimental results show that the proposed method is more accurate than most of current restoration methods.

Following the Code: Spillovers and Knowledge Transfer

Abstract Knowledge spillovers in Open Source Software (OSS) can occur via two channels: In the first channel, programmers take knowledge and experience gained from one OSS project they work on and employ it in another OSS project they work on. In the second channel, programmers reuse software code by taking code from an OSS project and employing it in another. We develop a methodology to measure software reuse in a large OSS network at the micro level and show that projects that reuse code from other projects have higher success. We also demonstrate knowledge spillovers from projects connected via common programmers.

Collaboration in OSS Communities: Who Solves Whose Problems?

In open collaboration, an economically important form of organizing technology development in online communities, both producers and users, either firm or independent users, invest their resources to develop and share technological problems as well as solutions freely accessible to the public. Existing literature draws ambivalent conclusions about whether and how the different actors engage in joint problem solving. Given potential conflicts of interest and variance in problem types, in terms of who shares the problem and how much generative efforts a problem demands, we know little about the role and nature of joint problem solving in open collaboration. Taking a dynamic view towards knowledge collaboration, this paper uses granular data about problem solving in OpenStack, a large and successful open source software development community, to examine joint problem solving in open collaboration. We show that not all problem types channel participation in the same way. Both producers and users prefer firm user problems. Secondarily, they also focus more on generative problems unless they are independent users. Our findings contribute to the theoretical puzzle about whether and when open collaboration is actually about “collaboration” and informs managers about opportunities for greater joint problem solving.

A Comparison Among ARIMA, BP-NN, and MOGA-NN for Software Clone Evolution Prediction

Software evolution continues throughout the life cycle of the software. During the evolution of software system, it has been observed that the developers have a tendency to copy the modules completely or partially and modify them. This practice gives rise to identical or very similar code fragments called software clones. This paper examines the evolution of clone components by using advanced time series analysis. In the first phase, software clone components are extracted from the source repository of the software application by using the abstract syntax tree approach. Then, the evolution of software clone components is analyzed. In this paper, three models, Autoregressive Integrated Moving Average, back propagation neural network, and multi-objective genetic algorithm-based neural network, have been compared for the prediction of the evolution of software clone components. Evaluation is performed on the large open-source software application, ArgoUML. The ability to predict the clones helps the software developer to reduce the effort during software maintenance activities.

The Effect of 'Following' on Contributions to Open Source Communities

Online communities have been growing rapidly in the last two decades and at the same time becoming more diverse. Millions of internet users actively contribute to various types of online communities ranging from user generated contents and social media to collaborative content communities, on a daily bases. Although numerous studies have examined members’ motivation to contribute to such communities, the positive effect of social factors has not been unanimously confirmed in different settings. In this study, we estimate the effect of “online following,” a basic form of online social interaction, on members’ contributions in open source software (OSS) communities. This estimation is based on analysis of a large-scale dataset of 4 million online members and their interactions over 7 years, collected from the largest OSS community. The results have implications for online community designers and OSS scholars.

A metric for package level coupling

Several software development activities are based on object oriented paradigm. Such software projects comprise of packages that are collection of correlated classes which can be reused. This reusability of a package depends on its cohesion and coupling strength. To determine the coupling and cohesion, there are several established metrics at the class level, but at the package level, such reliable and established metrics rarely exist. To address this important issue, a new metric, termed as package level coupling (PLC) is proposed in this paper. The metric estimates the quality, as well as reusability of the object-oriented software. The connections among these packages and sub-connections among the classes, related to different packages are used to determine the proposed metric, i.e., PLC. To establish the proposed metric, it is successfully experimented over four large java-based open source software comprising of multiple packages. Further, it is statistically correlated with the help of various parameters, i.e., number of sub-connections and number of classes of the package. The experimental results show a very weak correlation between the proposed PLC and the number of classes in the package. Spearman’s Rank correlation coefficient is used to illustrate the correlation between sub-connection and PLC. A smaller value of 0.108 is obtained signifying that there is no strong correlation between these two parameters. Further, a correlation between PLC and number of classes obtained is −0.061, signifying that these two parameters tends towards independency while other existing approaches are highly dependent on number of classes.

SMPLearner: learning to predict software maintainability

Accurate and practical software maintainability prediction enables organizations to effectively manage their maintenance resources and guide maintenance-related decision making. This paper presents SMPLearner, an automated learning-based approach to train maintainability predictors by harvesting the actual average maintenance effort computed from the code change history as well as employing a much richer set of 44 four-level hierarchical code metrics collected by static code analysis tools. We systematically evaluated SMPLearner on 150 observations partitioned from releases of eight large scale open source software systems. Our evaluation showed that SMPLearner not only outperformed the traditional 4-metric MI model but also the recent learning-based maintainability predictors constructed based on single Class-level metrics, demonstrating that single Class-level metrics were not sufficient for maintainability prediction.

Conducting quantitative software engineering studies with Alitheia Core

Quantitative empirical software engineering research benefits mightily from processing large open source software repository data sets. The diversity of repository management tools and the long his...