About

Melanie Feinberg is a professor at the School of Information and Library Science (SILS) at the University of North Carolina at Chapel Hill. Her scholarship examines the human character of data, synthesizing approaches from information science, design, and the humanities. Her research focuses on data modeling, information organization, critical data studies, and the history and foundations of information science. Melanie has a background in humanities with a BA from Stanford University, a MIMS from the University of California at Berkeley, and a PhD in Information Science from the University of Washington. She has worked in technical communication and content strategy at companies including Apple Computer, PeopleSoft, and Scient. Her academic contributions include teaching courses such as Foundations of Information Science, Organizing Information, and Metadata Architectures. She has received awards such as the Marie Skłodowska-Curie Individual Fellowship and was a Faculty Fellow at the Institute for Arts and Humanities.

Research topics

• Computer Science
• Data science
• Sociology
• Artificial Intelligence
• Epistemology
• Art
• Management science
• Engineering
• Aesthetics
• Knowledge management
• Mathematics

Selected publications

• Everyday Adventures with Unruly Data

  The MIT Press eBooks · 2022 · 14 citations

  1st authorCorresponding
  • Computer Science
  • Sociology
  • Epistemology

  Paired informal and scholarly essays show how everyday events reveal fundamental concepts of data, including its creation, aggregation, management, and use. Whether questioning numbers on a scale, laughing at a misspelling of one's name, or finding ourselves confused in a foreign supermarket, we are engaging with data. The only way to handle data responsibly, says Melanie Feinberg in this incisive work, is to take into account its human character. Though the data she discusses may seem familiar, close scrutiny shows it to be ambiguous, complicated, and uncertain: unruly. Drawing on the tools of information science, she uses everyday events such as deciding between Blender A and Blender B on Amazon to demonstrate a practical, critical, and generative mode of thinking about data: its creation, management, aggregation, and use. Each chapter pairs a self-contained main essay (an adventure) with a scholarly companion essay (the reflection). The adventure begins with an anecdote—visiting the library, running out of butter, cooking rice on a different stove. Feinberg argues that to understand the power and pitfalls of data science, we must attend to the data itself, not merely the algorithms that manipulate it. As she reflects on the implications of commonplace events, Feinberg explicates fundamental concepts of data that reveal the many tiny design decisions—which may not even seem like design at all—that shape how data comes to be. Through the themes of serendipity, objectivity, equivalence, interoperability, taxonomy, labels, and locality, she illuminates the surprisingly pervasive role of data in our daily thoughts and lives.

  PublisherDOI

• Reading databases: slow information interactions beyond the retrieval paradigm

  UNC Libraries · 2020-04-18

  articleOpen access1st authorCorresponding
  PublisherDOI

• The New Reality of Reproducibility: The Role of Data Work in Scientific Research

  Proceedings of the ACM on Human-Computer Interaction · 2020 · 19 citations

  1st authorCorresponding
  • Computer Science
  • Computer Science
  • Data science

  Although reproducibility--the idea that a valid scientific experiment can be repeated with similar results--is integral to our understanding of good scientific practice, it has remained a difficult concept to define precisely. Across scientific disciplines, the increasing prevalence of large datasets, and the computational techniques necessary to manage and analyze those datasets, has prompted new ways of thinking about reproducibility. We present findings from a qualitative study of a NSF--funded two-week workshop developed to introduce an interdisciplinary group of domain scientists to data-management techniques for data-intensive computing, with a focus on reproducible science. Our findings suggest that the introduction of data-related activities promotes a new understanding of reproducibility as a mechanism for local knowledge transfer and collaboration, particularly as regards efficient software reuse.

  PublisherDOI

• Human-Centered Study of Data Science Work Practices

  2019-04-30 · 50 citations

  article

  With the rise of big data, there has been an increasing need to understand who is working in data science and how they are doing their work. HCI and CSCW researchers have begun to examine these questions. In this workshop, we invite researchers to share their observations, experiences, hypotheses, and insights, in the hopes of developing a taxonomy of work practices and open issues in the behavioral and social study of data science and data science workers.

  PublisherDOI

• At the Intersection of Culture and Method

  2019-06-18 · 4 citations

  article1st authorCorresponding

  Boundaries and borders continue to draw political attention and create conflict on a global scale. They interrupt or facilitate people and money, and influence how designers and politicians alike work in opposition to oppression. This one-day workshop will explore the use of design research and computing practice in resisting and reifying inequalities. Recognizing that technology production is only a partial response to caring for ourselves and our environment, participants of this workshop will consider the design commitments, pedagogies, and labor that produce new strategies for more equitable futures. We will collectively ask and grapple with the following questions: 1) what are the current concerns and pressures of our community in producing feminist means and ends, 2) how do we work as academics, activists, allies and advocates through our design activities; and 3) whose opportunities, hopes, fears, innovations, and futures are we building?

  PublisherDOI

• Understanding Change in a Dynamic Complex Digital Object: Reading Categories of Change Out of Patch Notes Documents

  Lecture notes in computer science · 2019-01-01 · 1 citations

  book-chapterSenior author
  PublisherDOI

• Factotem: What is Information Access for?

  Cataloging & Classification Quarterly · 2018-11-09 · 1 citations

  articleOpen access1st authorCorresponding

  Inspired by Allyson Carlyle’s advice to explain fundamental concepts clearly and simply, this article seeks to explain the notion of information access, and what it means to provide information access in a responsible way. Specifically, this essay looks at the idea of facts. How should providers of information deal with facts? To examine this question, the essay considers the 2017 protest slogan “Librarians for Facts.” What does this slogan really mean? Ultimately, the essay contends that information providers need to determine what they are for, and orient information access mechanisms toward that goal.

  PublisherOA PDFDOI

• The Politics of classification

  Ergon Verlag eBooks · 2018-01-01

  book-chapter

  Classification schemes are powerful constructions in our information environments. They represent the intentional and unintended biases of those who construct them and are structurally bound by the technical systems in which they are built. As the study of what concepts exist in the world and how expressions of these concepts interconnect , classifications are fundamental to how we come to imagine and subsequently position ourselves embedded within a system of entities. On a more practical level, classification systems become the means by which we materially structure information within knowledge delivery systems, especially in the digital realm. Given the expressive qualities of classifications, they are also unavoidably political entities: they include and exclude, express and repress, facilitate and restrict. Actively identifying and addressing the political aspects of KO systems is a necessary activity within the KO community. This workshop will articulate possible courses of action to address the political aspects of classification in all subdomains of KO, inclusive of both research, pedagogy, and future disciplinary emphases. This workshop will address the political aspects of classifications through the following short paper presentations, which will help guide structured discussions. Abstracts of these presentations are available and, if this proposal is accepted, they will be included in the final workshop description: • Meaning through many hands: the neglected craft of classification implementation • name / change: Towards a politics of classification • Mechanisms of Cultural Bias in Classificatory Activity • The politics of biodiversity classification: Reconciliation and Interoperability • Expressing and obscuring throughout: Politics, values, and semantics derived from a conceptual analysis of classification terms

  PublisherDOI

• Understanding tag functions in a moderated, user-generated metadata ecosystem

  Journal of Documentation · 2018-02-08 · 13 citations

  articleSenior author

  Purpose The purpose of this paper is to investigate tag use in a metadata ecosystem that supports a fan work repository to identify functions of tags and explore the system as a co-constructed communicative context. Design/methodology/approach Using modified techniques from grounded theory (Charmaz, 2007), this paper integrates humanistic and social science methods to identify kinds of tag use in a rich setting. Findings Three primary roles of tags emerge out of detailed study of the metadata ecosystem: tags can identify elements in the fan work, tags can reflect on how those elements are used or adapted in the fan work, and finally, tags can express the fan author’s sense of her role in the discursive context of the fan work repository. Attending to each of the tag roles shifts focus away from just what tags say to include how they say it. Practical implications Instead of building metadata systems designed solely for retrieval or description, this research suggests that it may be fruitful to build systems that recognize various metadata functions and allow for expressivity. This research also suggests that attending to metadata previously considered unusable in systems may reflect the participants’ sense of the system and their role within it. Originality/value In addition to accommodating a wider range of tag functions, this research implies consideration of metadata ecosystems, where different kinds of tags do different things and work together to create a multifaceted artifact.

  PublisherDOI

• What does it mean to adopt a metadata standard? A case study of Omeka and the Dublin Core

  Journal of Documentation · 2018-04-25 · 28 citations

  articleSenior author

  Purpose The purpose of this paper is to employ a case study of the Omeka content management system to demonstrate how the adoption and implementation of a metadata standard (in this case, Dublin Core) can result in contrasting rhetorical arguments regarding metadata utility, quality, and reliability. In the Omeka example, the author illustrate a conceptual disconnect in how two metadata stakeholders – standards creators and standards users – operationalize metadata quality. For standards creators such as the Dublin Core community, metadata quality involves implementing a standard properly, according to established usage principles; in contrast, for standards users like Omeka, metadata quality involves mere adoption of the standard, with little consideration of proper usage and accompanying principles. Design/methodology/approach The paper uses an approach based on rhetorical criticism. The paper aims to establish whether Omeka’s given ends (the position that Omeka claims to take regarding Dublin Core) align with Omeka’s guiding ends (Omeka’s actual argument regarding Dublin Core). To make this assessment, the paper examines both textual evidence (what Omeka says) and material-discursive evidence (what Omeka does). Findings The evidence shows that, while Omeka appears to argue that adopting the Dublin Core is an integral part of Omeka’s mission, the platform’s lack of support for Dublin Core implementation makes an opposing argument. Ultimately, Omeka argues that the appearance of adopting a standard is more important than its careful implementation. Originality/value This study contributes to our understanding of how metadata standards are understood and used in practice. The misalignment between Omeka’s position and the goals of the Dublin Core community suggests that Omeka, and some portion of its users, do not value metadata interoperability and aggregation in the same way that the Dublin Core community does. This indicates that, although certain values regarding standards adoption may be pervasive in the metadata community, these values are not equally shared amongst all stakeholders in a digital library ecosystem. The way that standards creators (Dublin Core) understand what it means to “adopt a standard” is different from the way that standards users (Omeka) understand what it means to “adopt a standard.”

  PublisherDOI

Frequent coauthors

• Daniel Carter

  Wichita State University

  4 shared

• Julia Bullard

  4 shared

• Ayse Gursoy

  The University of Texas at Austin

  3 shared

• Karen M. Wickett

  University of Illinois Urbana-Champaign

  2 shared

• Katherine Schomer

  University of Washington

  2 shared

• Eryn Whitworth

  META Health

  2 shared

• Kurt Johnson

  University of Washington

  2 shared

• Pat Brown

  2 shared

Awards & honors

• Marie Skłowdoska-Curie Individual Fellowship (2019-2021)
• Institute for Arts and Humanities (IAH) Faculty Fellow (2018…