Fragile, flowing data and aspirational labor - Originally Posted February 24, 2016

Our Feb 22nd HSCollab conversation asked "who has the rights?" with respect to data ownership, invisible labor in digital economies, and agency with respect to data production, analysis, etc.

Fragility was a major theme of the discussion. Data is fragile in a number of ways:

  • it is subjective and therefore subject to discrepancies across similar datasets
  • the data creators and curators are often single individual or small groups of people whose work is subject to time, energy, and resource constraints. Very powerful datasets sometime just stop getting collected because someone - a person - just can't anymore.
  • Classification schemes are contingent and therefore are subject to variation over time and across communities.
  • Data flows are subject to gatekeeping practices - making data flow itself fragile
  • Data as information captured/maintained/transmitted through material media are subject to decay and deformation
  • Data can be lost, hidden, and/or destroyed

We also spent a fair bit of time talking about "prosumption" and "prosumers" in the context of invisible and often precarious labor. Examples included young women who are using social media platforms (like pinterest) to promote themselves in beauty industries, often without knowledge of how that work can be appropriated. In this context we talked about the many sides of "aspirational labor" - work that gets your name out there but might also make you vulnerable to threat and or appropriation. This included the largely unstudied psychological effects this an have on aspirants themselves.

A really fascinating discussion on digital lives and selves as a matter of public health, security as a matter of public health followed. This included noting that trans people can be particularly insecure in a data-based, social media culture. We talked about people who don't want to be documented - whether to promote "safety" or to "fit" existing legal frameworks. In many cases this arises from long histories of documentation-as-violence. This led us to want to know more about the general histories of disenfranchised people's relationship to data. We also discussed the ways in which efforts around sports injury testing might be themselves harming and imagined what a more holistic approach to thinking through issues like head injury might look like (hint: it's not all accelerometers).

We had several really productive forays into the idea that data is not an artifact (or not only an artifact) but an episteme or scenario. Every piece of data is an act of theory or ideology. This dovetails nicely with efforts elsewhere to think about intersectional or deeply multivariate data. In this context we noodled around thinking about how data lives in our own lives - often as stories. This got us thinking about the temporality of narrative or story and a comparison to the "allatonceness" of the visual representation of data. We also found ourselves intrigued by the richness of "scenario" given some of our interests in performing data - thinking about scene and scenic repetition. This, of course, has certain resonances with our conversation about data as evidence. In particular, we spent some time thinking through the questions we'd like to ask of any data: produced by whom? for whom? to what end? and in which theater of proof?

We also touched on time as suggested above - balancing an observation that some data uses are a way for us to stop time, to take a slice of life and reflect/subject it to scrutiny, with the idea that humanistic data actually flows diachronically, through time. Perhaps this is one of the many ways in which humanists can intervene in our collective understandings of data -- by bringing the diachronic back to what has largely been an effort to freeze/fix and study.

Our third luncheon will be held on Monday March 21 - topic: Healthy Data: health, data and healthy practices in the age of the quantified-self. To RSVP send an email to