Laura Wynholds

Data, data use, and scientific inquiry: two case studies of data practices

Data are proliferating far faster than they can be captured, managed, or stored. What types of data are most likely to be used and reused, by whom, and for what purposes? Answers to these questions will inform information policy and the design of digital libraries. We report findings from semi-structured interviews and field observations to investigate characteristics of data use and reuse and how those characteristics vary within and between scientific communities. The two communities studied are researchers at the Center for Embedded Network Sensing (CENS) and users of the Sloan Digital Sky Survey (SDSS) data. The data practices of CENS and SDSS researchers have implications for data curation, system evaluation, and policy. Some data that are important to the conduct of research are not viewed as sufficiently valuable to keep. Other data of great value may not be mentioned or cited, because those data serve only as background to a given investigation. Metrics to assess the value of documents do not map well to data.

Follow the data: How astronomers use and reuse data

Author(s): Borgman, Christine L.; Sands, Ashley; Wynholds, Laura; Traweek, Sharon | Abstract: We analyze the people and infrastructure involved in the building, sustaining, and curation of large astronomy sky surveys. Our research assesses what new infrastructures, divisions of labor, knowledge, and expertise are necessary for the proper care of data. Between May 2011- February 2012, we conducted fourteen interviews employing Sloan Digital Sky Survey (SDSS) data use as the focus. SDSS is a multi-faceted, multi-phased data-driven telescope project with hundreds of collaborators and thousands of users of the open data. The Follow the Data interview protocol identifies a single publication authored by each interviewee and uses it as a lens looking backward and forward to identify data uses leading into and out of the publication. The interviews revealed the ways these astronomers discover, locate, retrieve, and store external data for their research. Any given astronomy research project may employ multiple methods to discover, locate, retrieve, and store multiple datasets. Our research finds that informal and formal methods are used to discover and locate data, including person-to-person contact. Data retrieval and storage methods are often determined by the size of the dataset and the amount of infrastructure available to the researcher. Astronomy research practices are evolving rapidly with access to more data and better tools. The poster presentation will report further on how those data are used and reused in astronomy. Sands, A., Borgman, C. L., Wynholds, L., a Traweek, S. (2012, October 29). Follow the Data: How astronomers use and reuse data. Poster presented at the ASISaT; 75th Annual Meeting, Baltimore, MD. Retrieved from http://www.asis.org/asist2012/abstracts/341.html

When use cases are not useful: data practices, astronomy, and digital libraries

As science becomes more dependent upon digital data, the need for data curation and for data digital libraries becomes more urgent. Questions remain about what researchers consider to be their data, their criteria for selecting and trusting data, and their orientation to data challenges. This paper reports findings from the first 18 months of research on astronomy data practices from the Data Conservancy. Initial findings suggest that issues for data production, use, preservation, and sharing revolve around factors that rarely are accommodated in use cases for digital library system design including trust in data, funding structures, communication channels, and perceptions of scientific value.

Awash in stardust: data practices in astronomy

One of several major research initiatives into the grand challenge of data curation, the Data Conservancy (DC), funded by the National Science Foundations DataNet Initiative, is investigating data use, sharing, and preservation in multiple fields of science. Our group at the University of California, Los Angeles is conducting a deep case study of astronomy and astrophysics. DC partners at Cornell, Illinois, the National Center for Atmospheric Research, and the National Snow and Ice Data Center are studying data practices in several other science domains. The DC is a collaborative multi-sited research project that will offer new insights into data practices in an array of physical and life sciences. The mandate of the project is to research, design, implement, deploy and sustain data curation infrastructure for cross-disciplinary discovery with an emphasis on observational data. [4].n This poster will summarize findings from the first year of UCLAs research on astronomers and astronomy data. Our approach to studying data practices is complementary to that of our DC project partners, most of whom are surveying a broader set of fields less deeply. The UCLA team is part of Data Conservancy information science and computer science (IS/CS) team, which will share methods and findings. Our overall goal is to compare comparative data practices and data curation requirements across a range of physical and life science fields.n Astronomy is considered to be at the forefront of data-driven science. Hanisch and Quinn, in explaining the development of the Virtual Observatory, wrote, Astronomy faces a data avalanche. Breakthroughs in telescope, detector, and computer technology allow astronomical instruments to produce terabytes of images and catalogs...These technological developments will fundamentally change the way astronomy is done. These changes will have dramatic effects on the sociology of astronomy itself.[7].n Over the course of the last ten years, astronomy data projects have grown from terabyte scales to petabyte scales, and the data deluge has affected many more sciences, large and small. Long predicted by the science community [8], not only has Nature, a premier science journal, published feature sections on big data [2] so have Wired Magazine [1], and the Economist [5].n However, significant tensions surrounding big data projects are present in the field, as expressed by two Nature editors: Astronomy is in an era of unprecedented change...more and more astronomy papers are showing evidence that familiarity with the essential dirtiness of data and models is being lost. ...Worries that the centuries-old culture of astronomy is being eroded have been voiced in the community for several years, especially in cosmology where the big-science approach now dominates. [12]n Data curation of these complex digital objects presents a significant challenge facing both scientific research and scholarly record keeping institutions. Bowker and Star [14] argued that of the problems of aggregating data within an information system are reflective of the sociotechnical systems that yielded the data. Following that argument, the quest to build repositories for data becomes largely a quest to fold the practices of an established community into evolving technological solutions. Thus it is essential to study the data practices of communities whose data is to be curated. Astronomy is a rich domain in which to study data practices, and the Data Conservancy offers a diverse environment in which to compare data curation challenges across the sciences.n We approach astronomy data practices with three questions:n 1. What are the data management, curation, and sharing practices of astronomers and astronomy data centers, and how have they developed?n 2. Who uses what data when, with whom, and why?n 3. What data are most important to curate, how, for whom, and for what purposes?n The first question focuses on what people do, how they manage data, and what counts as relevant research data to generate, use, keep, and discard. The second question addresses the social contexts, networks, and communities within which these practices occur. The third question focuses on specific aspects of data curation, such as deciding what data will be of future use to others, assigning responsibilities for organizing and describing datasets for use, identifying incentives and disincentives for individuals or groups to curate their data, and developing tools and services necessary to exploit those data.n At the core of our astronomy case study is an analysis of the large sky surveys, as these generate massive amounts of data that fuel both inquiry and the tensions outlined above. The first year of the project has been concerned with capturing a broad perspective of the empirical and theoretical research that can be accomplished with astronomical observations, comparing data activities associated with sky surveys to other types of inquiry.n Our starting point has been the Sloan Digital Sky Survey (SDSS) [13], which began data capture in 2000 and recently completed its final data release of the SDSS-II project. This groundbreaking optical survey telescope and accompanying digital dataset provides distributed access to data for one quarter of the sky. We are studying the development, practices and challenges of data management and curation in the SDSS, as well as the projects impact on astronomy. Our study of subsequent sky survey projects, such as PanSTARRS [11] and LSST [10], will offer insights to the role and value of synoptic surveys in physical science research.n Our methods follow from our three research questions about data practices, social contexts, and curation requirements in these astronomy settings:n 1. Examining data practices through qualitative ethnography, including in-depth interviews and site observations; andn 2. Mapping the social context of projects by analyzing documents about projects and their history, and peoples networks of professional affiliations and research activities.n Within the context of qualitative ethnographies, we are interviewing people who have worked in multiple roles in sky surveys and who use sky survey data in their own research. These interviewees include software developers, university faculty, postdocs, and other researchers using data from networked astrophysical instruments. We are comparing the range of curation requirements for managing large-scale archives and smaller collections of research data.n We are examining the extensive documentation of the SDSS project, including an archived listserv discussion group of its builders and users.n Our initial fieldwork on astronomy sites has found broad differences in curation practices and requirements between projects, data centers, academic collaborations, and domains of research. Identifying generalizable comparisons is a core challenge. We see historical and cultural changes at large and small levels, including the professionalization of data management and the role of informatics in astronomy. Adoption of computational approaches to knowledge discovery appears uneven across the astronomy community. Science-driven research has exhibited tensions with computer engineering approaches to data archives, according to some of our respondents.n We are seeing considerable variation in the use of sky surveys, from scientific inquiry to calibration of other instruments. In conjunction with a considerable variation in use, we see significant diversity in what counts as data among those studying each wavelength, and between observational and theoretical approaches. Among the interviewed theoretical astrophysicists who rely on computational modeling, some archive the results of simulations, while others retain the algorithms but discard the data generated by simulations. Data archiving practices for sky surveys appear to vary widely by wavelength, partially due to differences in data volume, format and complexity. Similarly, astronomy data use may be further divided by practices of ground-based versus space-based instruments. Data practices and data curation requirements within astronomy are far less homogeneous than they may appear from the outside. Similarly, the computation- and data-intensive methods that characterize modern astronomical research are not embraced universally.n Our poster will compare our initial results to those of our Data Conservancy partners analyses of data practices in other science domains. We may see similar practices of data management and preservation practices among fields; however, early reports by DC partners at Illinois show no field-wide norms for sharing data among the researchers they interviewed, and diverse use of data repositories even within a research field. [3] Data practices appear to vary widely within disciplines in the physical and life sciences, and even more so between them.

Curators to the stars

We are introducing to the ASIS&T community what will be, to date, the most extensive study of data practices for astronomy and astrophysics from the Information Science field. We approach astronomy data curation with three questions: 1) What are the data management, curation, and sharing practices of astronomers and astronomy data centers, and how have they developed? 2) Who uses what data when, with whom, and why? 3) What data are most important to curate, how, for whom, and for what purposes? The first question is about what people do, how they manage data, and what counts as relevant research data to generate, use, keep, and discard. The second question addresses the social contexts, networks, and communities within which these practices occur. The third question focuses on tasks of data curation, such as deciding what data will be of future use to others, assigning responsibilities for organizing and describing datasets for use, identifying incentives and disincentives for individuals or groups to curate their data, and developing tools and services necessary to exploit those data. The poster will summarize findings from our first year of research. n nOur research team, based at the University of California Los Angeles Center for Embedded Networked Sensing is part of a five-year project, the Data Conservancy (DC), funded by the National Science Foundations DataNet Initiative (Data Conservancy, 2010). DCs partner institutions are investigating data use, sharing, and preservation in multiple fields of science. UCLA is conducting a deep case study of astronomy and astrophysics. DC partners at Cornell, Illinois, the National Center for Atmospheric Research, and the National Snow and Ice Data Center are studying data practices in several other science domains. The DC is a research project that will offer new insights into data practices in an array of physical and life sciences. Research will be translated into practice via the design of a data repository. The poster will summarize findings from the first year of research on astronomers and astronomy data. n nWhat can the information sciences learn from astronomy about data curation? Astronomy is a pioneer in big science projects with large-scale digital datasets. Telescopes on the ground and in orbit can stream data instantly and internationally. Large instruments can stream data directly into institutional data centers; those data may be available immediately or in periodic data releases some months later. Collaborative use of instruments has fostered the ongoing development of standards for data formats and interoperability. New technologies brought changes to the profession and to research practices accompanying data production, analysis, sharing, and preservation. The investments in shared research instruments and information technologies that characterize big science also support smaller-scale projects by astronomers using virtual observatories from their offices, making data management a more personal responsibility. Astronomy offers a rich and complex setting to study data curation practices and to identify challenges applicable to many fields of science. n nAt the core of our astronomy case study is an analysis of the large sky surveys, as these generate massive amounts of data that serve multiple scientific purposes. We are comparing data activities associated with sky surveys to those of point-based observations and to other types of astronomical inquiry. The first year of the project is concerned with capturing a broad view of empirical and theoretical research that can be accomplished with astronomical observations. Our starting point is the Sloan Digital Sky Survey (SDSS) (Sloan Digital Sky Survey, 2010), which recently completed its final data release of the SDSS-II project. This pioneering optical telescope and accompanying digital dataset, online since 2000, provides distributed access to data for one quarter of the sky, surveyed across eight years. The Johns Hopkins University libraries will curate these data as part of the Data Conservancy project. We are studying the development of the SDSS, its practices of data management and curation, hurdles overcome and remaining, and its impact on astronomy. Our study of subsequent sky survey projects (PAN-STARRS, 2009; Large Synoptic Sky Telescope, 2010) will offer insights to the role and value of synoptic surveys in physical science research. n nTo better understand the social contexts of these projects, we are interviewing people who have worked in multiple roles in sky surveys and who use sky survey data in their own research. These people include software developers, university faculty, postdocs, and other researchers using data from networked astrophysical instruments. We are examining practices and curation issues of data centers that support virtual astronomy projects, and are investigating the work of the International Virtual Observatory Alliance (IVOA) (Hanisch & Quinn, 2002) to build standards and tools for interoperable data archives and instruments. We are comparing the range of curation requirements for managing large-scale archives and smaller collections of research data. n nOur methods follow from our three research questions about data practices, social contexts, and curation requirements in these astronomy settings: 1) We examine data practices through qualitative ethnography, including in-depth interviews and site observations; and 2) we map the social context of projects by analyzing documents about projects and their history, and peoples networks of professional affiliations and research activities. n nTo support this multi-modal research, we are examining the extensive documentation of the SDSS program, including an archived listserv discussion group of its builders and users. We are also building a research database to help integrate our qualitative analysis of interviews and project documentation (Wynholds, Borgman, Traweek, Fearon Jr & Fidler, 2010). Our approach to studying data practices is complementary to that of our DC project partners, most of whom are surveying a broader set of fields in less detail. The UCLA DC subgroup participates within the DC information science and computer science team, collaboratively sharing methods, findings, and enabling a comparative analysis of practices across a range of physical and life science fields. n nOur initial fieldwork at astronomy sites has found broad differences in curation practices and requirements between data centers and smaller university faculty groups, and also significant diversity among data centers and among areas of astronomy research. Identifying generalizable comparisons is an ongoing challenge. We see historical and cultural changes at large and small levels, including the professionalization of data management and informatics roles in astronomy. n nWe see significant diversity in what counts as data among those studying each wavelength, and between observational and theoretical approaches. For example, among the astrophysicists we have interviewed who primarily use computational modeling, some must archive the results of supercomputer runs, while others retain algorithms but discard data generated by simulations. Data archiving practices for sky surveys appear to vary widely by wavelength, due to differences in data volume, format and complexity. Similarly, astronomy data use may be subdivided by practices of ground-based and space-based instruments. n nAdoption of computational approaches to knowledge discovery is uneven across the astronomy community. Requirements for rich science-driven investigation may conflict with engineering approaches to data archives, according to some of our respondents. We also are seeing considerable variation in the use of sky surveys, from scientific inquiry to calibration of other instruments. n nOur poster will compare our initial results to those of our Data Conservancy partners analyses of data practices in other science domains. We may see similar practices of data management and preservation practices among fields; however, early reports by DC partners at Illinois (Cragin, Palmer, Carlson & Witt, 2010) show no field-wide norms for sharing data among the researchers they interviewed, and diverse use of data repositories even within a research field. Data practices appear to vary widely among disciplines within the physical and life sciences, and even more so between them. Data integration in astronomy, for example, relies upon established constants and physical laws to calibrate instruments and to determine measurement standards. Such standards are rare for fields with diverse phenomena like biology. Also of interest to ASIS&T will be our initial efforts to translate our research findings into design features for a data repository.