The need for advanced automation and artificial intelligence (AI) in various fields, including text classification, has dramatically increased in the last decade, leaving us critically dependent on their performance and reliability. Yet, as we increasingly rely more on AI applications, their algorithms are becoming more nuanced, more complex, and less understandable precisely at a time we need to understand them better and trust them to perform as expected. Text classification in the medical and cybersecurity domains is a good example of a task where we may wish to keep the human in the loop. Human experts lack the capacity to deal with the high volume and velocity of data that needs to be classified, and ML techniques are often unexplainable and lack the ability to capture the required context needed to make the right decision and take action. We propose a new abstract configuration of Human-Machine Learning (HML) that focuses on reciprocal learning, where the human and the AI are collaborating partners. We employ design-science research (DSR) to learn and design an application of the HML configuration, which incorporates software to support combining human and artificial intelligences. We define the HML configuration by its conceptual components and their function. We then describe the development of a system called Fusion that supports human-machine reciprocal learning. Using two case studies of text classification from the cyber domain, we evaluate Fusion and the proposed HML approach, demonstrating benefits and challenges. Our results show a clear ability of domain experts to improve the ML classification performance over time, while both human and machine, collaboratively, develop their conceptualization, i.e., their knowledge of classification. We generalize our insights from the DSR process as actionable principles for researchers and designers of 'human in the learning loop' systems. We conclude the paper by discussing HML configurations and the challenge of capturing and representing knowledge gained jointly by human and machine, an area we feel has great potential.

Software developers use many different communication tools and channels in their work. The diversity of these tools has dramatically increased over the past decade and developers now have access to a wide range of socially enabled communication channels and social media to support their activities. The availability of such social tools is leading to a participatory culture of software development, where developers want to engage with, learn from, and co-create software with other developers. However, the interplay of these social channels, as well as the opportunities and challenges they may create when used together within this participatory development culture are not yet well understood. In this paper, we report on a large-scale survey conducted with 1,449 GitHub users. We discuss the channels these developers find essential to their work and gain an understanding of the challenges they face using them. Our findings lay the empirical foundation for providing recommendations to developers and tool designers on how to use and improve tools for software developers.

One of the many effects of social media in software development is the flourishing of very large communities of practice where members share a common interest, such as programming languages, frameworks, and tools. These communities of practice use many different communication channels but little is known about how these communities create, share, and curate knowledge using such channels. In this paper, we report a qualitative study of how one community of practice---the R software development community---creates and curates knowledge associated with questions and answers (Q&A) in two of its main communication channels: the R-tag in Stack Overflow and the R-users mailing list. The results reveal that knowledge is created and curated in two main forms: participatory, where multiple members explicitly collaborate to build knowledge, and crowdsourced, where individuals work independently of each other. The contribution of this paper is a characterization of knowledge types that are exchanged by these communities of practice, including a description of the reasons why members choose one channel over the other. Finally, this paper enumerates a set of recommendations to assist practitioners in the use of multiple channels for Q&A.

Bots are used to support different software development activities, from automating repetitive tasks to bridging knowledge and communication gaps in software teams. We anticipate the use of Bots will increase and lead to improvements in software quality and developer and team productivity, but what if the disruptive effect is not what we expect?

The software development community has embraced GitHub as an essential platform for managing their software projects. GitHub has created efficiencies and helped improve the way software professionals work. It not only provides a traceable project repository, but it acts as a social meeting place for interested parties, supporting communities of practice. Recently, educators have seen the potential in GitHub's collaborative features for managing and improving---perhaps even transforming---the learning experience. In this study, we examine how GitHub is emerging as a collaborative platform for education. We aim to understand how environments such as GitHub---environments that provide social and collaborative features in conjunction with distributed version control---may improve (or possibly hinder) the educational experience for students and teachers. We conduct a qualitative study focusing on how GitHub is being used in education, and the motivations, benefits and challenges it brings.