Data are essential products of scientific work that move among and through research infrastructures over time. Data constantly changes due to evolving practices and knowledge, requiring improvisational work by scientists to determine the effects on analyses. Today for end users of datasets much of the information about changes, and the processes leading to them, is invisible—embedded elsewhere in the work of a collaboration. Simultaneously scientists use increasing quantities of data, making ad hoc approaches to identifying change difficult to scale effectively. Our research investigates data change by examining how scientists make sense of change in datasets being created and sustained by the collaborative infrastructures they engage with. We examine two forms of change, before examining how trust and project rhythms influence a scientist’s notion that the newest available data are the best. We explore the opportunity to design tools and practices to support user examinations of data change and surface key provenance information embedded in research infrastructures.

Scientific collaboration is a long-standing subject of CSCW scholarship that typically focuses on the development and use of computing systems to facilitate research. The research presented in this article investigates the sociality of science by identifying and describing particular, common forms of organizing that researchers in four different scientific realms employ to conduct work in both local contexts and as part of distributed, global projects. This paper introduces five prototypical forms of organizing we categorize as coordinative entities: the Principal Group, Intermittent Exchange, Sustained Aggregation, Federation, and Facility Organization. Coordinative entities as a categorization help specify, articulate, compare, and trace overlapping and evolving arrangements scientists use to facilitate data intensive research. We use this typology to unpack complexities of data intensive scientific collaboration in four cases, showing how scientists invoke different coordinative entities across three types of research activities: data collection, processing, and analysis. Our contribution scrutinizes the sociality of scientific work to illustrate how these actors engage in relational work within and among diverse, dispersed forms of organizing across project, funding, and disciplinary boundaries.

Software is an integral element of the work of science yet it is not commonly an object of inquiry in studies of scientific infrastructures. This paper presents findings from an ethnographic study of a cosmology group's collaborative scientific software production. We demonstrate how these cosmologists use plots to simultaneously test their software and analyze data while interrogating multiple layers of infrastructural components. We broaden perspectives on scientific software development using a sociotechnical, software studies lens to examine this work of scientific discovery as a creative and embodied, yet exacting and methodical, activity that requires a 'human in the loop'. We offer a new reading of scientific software practices to convey how creating scientific software is often really the act of doing science itself--an intervention we believe is necessary to more successfully support scientific software sharing and infrastructure production.

This paper considers sensemaking as it relates to everyday software engineering (SE) work practices and draws on a multi-year ethnographic study of SE projects at a large, global technology company building digital services infused with artificial intelligence (AI) and machine learning (ML) capabilities. Our findings highlight the breadth of sensemaking practices in AI/ML projects, noting developers' efforts to make sense of AI/ML environments (e.g., algorithms/methods and libraries), of AI/ML model ecosystems (e.g., pre-trained models and "upstream"models), and of business-AI relations (e.g., how the AI/ML service relates to the domain context and business problem at hand). This paper builds on recent scholarship drawing attention to the integral role of sensemaking in everyday SE practices by empirically investigating how and in what ways AI/ML projects present software teams with emergent sensemaking requirements and opportunities.

Scientific exploration generates expanding volumes of data that commonly require High Performance Computing (HPC) systems to facilitate research. HPC systems are complex ecosystems of hardware and software that frequently are not user friendly. The Usable Data Abstractions (UDA) project set out to build usable software for scientific workflows in HPC environments by undertaking multiple rounds of qualitative user research. Qualitative research investigates how individuals accomplish their work and our interview-based study surfaced a variety of insights about the experiences of working in and with HPC ecosystems. This report examines multiple facets to the experiences of scientists and developers using and supporting HPC systems. We discuss how stakeholders grasp the design and configuration of these systems, the impacts of abstraction layers on their ability to successfully do work, and the varied perceptions of time that shape this work. Examining the adoption of the Cori HPC at NERSC we explore the anticipations and lived experiences of users interacting with this system’s novel storage feature, the Burst Buffer. We present lessons learned from across these insights to illustrate just some of the challenges HPC facilities and their stakeholders need to account for when procuring and supporting these essential scientific resources to ensure their usability and utility to a variety of scientific practices.

Scientific datasets are continually expanding and changing due to fluctuations with instruments, quality assessment and quality control processes, and modifications to software pipelines. Datasets include minimal information about these changes or their effects requiring scientists manually assess modifications through a number of labor intensive and ad-hoc steps. The Deduce project is investigating data change to develop metrics, methods, and tools that will help scientists systematically identify and make decisions around data changes. Currently, there is a lack of understanding, and common practices, for identifying and evaluating changes in datasets since systematically measuring and managing data change is under explored in scientific work. We are conducting user research to address this need by exploring scientist's conceptualizations, behaviors, needs, and motivations when dealing with changing datasets. Our user research utilizes multiple methods to produce foundational, generative insights and evaluate research products produced by our team. In this paper, we detail our user research process and outline our findings about data change that emerge from our studies. Our work illustrates how scientific software teams can push beyond just usability testing user interfaces or tools to better probe the underlying ideas they are developing solutions to address.

Scientifc software development processes are understood to be distinct from commercial software development practices due to uncertain and evolving states of scientifc knowledge. Sustaining these software products is a recognized challenge, but under-examined is the usability and usefulness of such tools to their scientifc end users. User research is a well-established set of techniques (e.g., interviews, mockups, usability tests) applied in commercial software projects to develop foundational, generative, and evaluative insights about products and the people who use them. Currently these approaches are not commonly applied and discussed in scientifc software development work. The use of user research techniques in scientifc environments can be challenging due to the nascent, fuid problem spaces of scientifc work, varying scope of projects and their user communities, and funding/economic constraints on projects. In this paper, we refect on our experiences undertaking a multi-method user research process in the Deduce project. The Deduce project is investigating data change to develop metrics, methods, and tools that will help scientists make decisions around data change. There is a lack of common terminology since the concept of systematically measuring and managing data change is under explored in scientifc environments. To bridge this gap we conducted user research that focuses on user practices, needs, and motivations to help us design and develop metrics and tools for data change. This paper contributes refections and the lessons we have learned from our experiences. We ofer key takeaways for scientifc software project teams to efectively and fexibly incorporate similar processes into their projects.

Along with a number of other computing technologies, cloud computing services are increasingly being promoted as a way of enabling openness, reproducibility, and the acceleration of scientific work. While there have been a variety of studies of the cloud in terms of computing performance, there has been little empirical attention to the changes going on around cloud computing at the level of work and practice. Through a qualitative, ethnographic study, we follow a cosmology research group’s transition from a shared high performance computing cluster to a cloud computing service, and examine the cloud service as a coordinative artifact being integrated into a larger ecology of existing practices and artifacts. We find that the transition involves both change and continuity in the group’s coordinative work and maintenance work, and point out some of the effects this adoption has on the group’s larger set of practices. Finally, we discuss practical implications this has for the broader adoption of cloud computing in university-based scientific work.