Its algorithms are as infused with bias as any human decision, but neither it nor its audience thinks of Facebook as a news organization with editorial responsibilities.

What makes a record a "human rights record"? What types of records fall under this umbrella term? How and why might we develop a typology of such records? What is at stake—ethically, theoretically, and practically—in the ways in which and the reasons why we define and classify records as such? This article seeks to answer these questions by delineating a typology of human rights records. First, this article will provide a literature review exploring the history of conceptions of human rights records in archival studies, as well as the ongoing discussion in information studies more broadly about the politics of the organization of information. Next, this paper will outline the chosen methodology of conceptual analysis and describe the ways such methodology will be employed to de/construct the term “human rights record.” This paper will then provide a typology of human rights records, positing that such records can be examined according to five interlocking vectors: who created them, why, and when, where they are currently stewarded, and how they are being put to use. This paper will then analyze two keys examples of human rights records using the proposed typology. Finally, this paper will conclude by examining the ethical, political, and professional implications of the proposed typology and suggest ways in which this rubric can be used in the future.

Post contributed by Matthew Farrell, Digital Records Archivist. At the University Archives, we work hard to dispel the stereotype that we are merely reactive documenters of Duke’s history, that we wait to receive evidence of activity reflected in the records of the offices, organizations, and bodies that donate or transfer materials to us. We pursue student … Continue reading Archiving Social Media about Duke Activism →

This is the text of a presentation I made yesterday at a wonderful Columbia University symposium called Insuetude (still ongoing), which is bringing media archaeologists together with stones-and-bones archaeologists. I started my talk with a bit of film, as a way of time-traveling to the middle...

Coming soon: An edit-a-thon to add articles on the people who are sorely lacking on Wikipedia.

Unless you're a white guy from Brooklyn, the publishing industry has a huge problem when it comes to representation. But no one is sure how to fix it.

From software that records your every keystroke, to GPS tracking, to ignition kill switches—lenders have more power over their customers than ever.

AAAAAAAAAAAAARGH.

Surveillance and public-benefits programs gather large amounts of information on low-income communities, feeding opaque algorithms that can be used to deny them opportunities.

Depending on how algorithms are trained, they could be significantly more accurate when identifying white faces than African American ones.

Fairness in Machine Learning Delip Rao

New research shows that social-media users who said they have “nothing to hide” from the government often avoid posting controversial opinions on Facebook.

M. NourbeSe Philip combs history for the black American experience.