Showing papers by "James D. Herbsleb published in 2016"

How to break an API: cost negotiation and community values in three software ecosystems

Abstract: Change introduces conflict into software ecosystems: breaking changes may ripple through the ecosystem and trigger rework for users of a package, but often developers can invest additional effort or accept opportunity costs to alleviate or delay downstream costs. We performed a multiple case study of three software ecosystems with different tooling and philosophies toward change, Eclipse, R/CRAN, and Node.js/npm, to understand how developers make decisions about change and change-related costs and what practices, tooling, and policies are used. We found that all three ecosystems differ substantially in their practices and expectations toward change and that those differences can be explained largely by different community values in each ecosystem. Our results illustrate that there is a large design space in how to build an ecosystem, its policies and its supporting infrastructure; and there is value in making community values and accepted tradeoffs explicit and transparent in order to resolve conflicts and negotiate change-related costs.

How to Hackathon: Socio-technical Tradeoffs in Brief, Intensive Collocation

Abstract: Hackathons are events where people who are not normally collocated converge for a few days to write code together. Hackathons, it seems, are everywhere. We know that long- term collocation helps advance technical work and facilitate enduring interpersonal relationships, but can similar benefits come from brief, hackathon-style collocation? How do participants spend their time preparing, working face-to- face, and following through these brief encounters? Do the activities participants select suggest a tradeoff between the social and technical benefits of collocation? We present results from a multiple-case study that suggest the way that hackathon-style collocation advances technical work varies across technical domain, community structure, and expertise of participants. Building social ties, in contrast, seems relatively constant across hackathons. Results from different hackathon team formation strategies suggest a tradeoff between advancing technical work and building social ties. Our findings have implications for technology support that needs to be in place for hackathons and for understanding the role of brief interludes of collocation in loosely-coupled, geographically distributed work.

Measuring Similarity Similarly: LDA and Human Perception

Abstract: Several intelligent technologies designed to improve navigability in and digestibility of text corpora use topic modeling such as the state-of-the-art Latent Dirichlet Allocation (LDA). This model and variants on it provide lower-dimensional document representations used in visualizations and in computing similarity between documents. This article contributes a method for validating such algorithms against human perceptions of similarity, especially applicable to contexts in which the algorithm is intended to support navigability between similar documents via dynamically generated hyperlinks. Such validation enables researchers to ground their methods in context of intended use instead of relying on assumptions of fit. In addition to the methodology, this article presents the results of an evaluation using a corpus of short documents and the LDA algorithm. We also present some analysis of potential causes of differences between cases in which this model matches human perceptions of similarity more or less well.

Building a socio-technical theory of coordination: why and how (outstanding research award)

Abstract: Research aimed at understanding and addressing coordination breakdowns experienced in global software development (GSD) projects at Lucent Technologies took a path from open-ended qualitative exploratory studies to quantitative studies with a tight focus on a key problem – delay – and its causes. Rather than being directly associated with delay, multi-site work items involved more people than comparable same-site work items, and the number of people was a powerful predictor of delay. To counteract this, we developed and deployed tools and practices to support more effective communication and expertise location. After conducting two case studies of open source development, an extreme form of GSD, we realized that many tools and practices could be effective for multi-site work, but none seemed to work under all conditions. To achieve deeper insight, we developed and tested our Socio-Technical Theory of Coordination (STTC) in which the dependencies among engineering decisions are seen as defining a constraint satisfaction problem that the organization can solve in a variety of ways. I conclude by explaining how we applied these ideas to transparent development environments, then sketch important open research questions.

Transactivity as a Predictor of Future Collaborative Knowledge Integration in Team-Based Learning in Online Courses.

Abstract: To create a satisfying social learning experience, an emerging challenge in educational data mining is to automatically assign students into effective learning teams. In this paper, we utilize discourse data mining as the foundation for an online team-formation procedure. The procedure features a deliberation process prior to team assignment, where participants hold discussions both to prepare for the collaboration task and provide indicators that are then used during automated team assignment. We automatically assign teams in a way that maximizes average observed pairwise transactivity exchange within teams, whereas in a control condition, teams are assigned randomly. We validate our team-formation procedure in a crowdsourced online environment that enables effective isolation of variables, namely Amazon’s Mechanical Turk. We compare group knowledge integration outcomes between the two team assignment conditions. Our results demonstrate that transactivity-based team assignment is associated with significantly greater knowledge integration (p < .05, effect size 3 standard deviations).

Intelligently Transparent Software Ecosystems

Abstract: Today's social-coding tools foreshadow a transformation of the software industry, as it relies increasingly on open libraries, frameworks, and code fragments. Our vision calls for new intelligently transparent services that support rapid development of innovative products while helping developers manage risk and issuing them early warnings of looming failures. Intelligent transparency is enabled by an infrastructure that applies analytics to data from all phases of the life cycle of open source projects, from development to deployment. Such an infrastructure brings stakeholders the information they need when they need it.