Drucker:

I agree strongly with Drucker’s central argument: that data are not simply collected but created and interpreted, and that a critical lens is rarely applied to data visualization–often in the rush to adopt digital platforms and make eye-catching visuals. However, I take issue with many of the alternative visualizations she uses as examples to illustrate that argument. Trying to demonstrate migration, “gender ambiguity,” and cultural constructionism in one chart seems unnecessarily ambitious. And the image itself communicates little without its long accompanying text (15). In contrast, the “hours as a function of time pressures” (19) chart doesn’t suffer these problems and is perhaps the best and most concise illustration of Drucker’s argument. The publishing timeline is also great (24).

Of course, these are not thoroughly researched projects but rather quick sketches to make a point. But I could easily see that being better accomplished by citing others’ successful work in this area, such as Native Land or the Zuni map project. As Drucker’s writing suggests, critical data visualization is not just about messing with the plot and the coordinate system. It is also about making interventions into the way we think about data collection, what rhetorical modes we unconsciously adopt, and how we treat the materials (and people) we work with. And in chapters 3 and 4, Drucker shows us that it is also about deconstructing some fundamental ontologies. That’s much more experimental work that actually benefits from her more unconventional techniques.

One other note: I also think Drucker’s usage of the word “affect”(/affective) is a bit confusing, or maybe it is just too different from how I usually see it used. She writes that it “is used as shorthand for interpretative construction, for the registration of point of view, position, the place from which and agenda according to which parameterization occurs” (25). I prefer the following from Shaka McGlotten paraphrasing Eric Shouse:

affect is related to but distinct from emotions and feelings. As Eric Shouse parses these distinctions, feelings are personal and biographical, emotions are socially performed and circulating forms of feelings, and affects are pre‐subjective or pre‐personal “experience[s] of intensity.”

McGlotten, Shaka. 2013. Virtual Intimacies: Media, Affect, and Queer Sociality. Albany, NY: State University of New York Press, page 64.

boyd and Crawford:

As for the boyd and Crawford piece, it’s interesting to see what’s changed in only the past 8 years. They were ahead of their time in many respects, as the public seems much more wary of Big Data now, especially after high profile breaches, leaks, and scandals. Their point number 1 bothers me a bit because it seems to discourage programmatic approaches to knowledge construction. Following Wolfgang Ernst, I would rather say that researchers should build their own research tools or be highly critical and skeptical of the ones they are relying on. But computational approaches aren’t all of one piece.

Point number 2 seems to be related to the Drucker piece. Point 3 and 4 still offer valuable methodological insights, especially the idea of “small data.” Points 5 and 6 get at issues of data ethics and data justice before those terms were coined, or at least popularized. It seems like this is what everyone is talking about right now, so it was quite prescient in 2011.

A piece I also enjoyed was Andre Brock’s response to this article, which can be read here: Brock, Andre. 2015. “Deeper Data: A Response to boyd and Crawford” Media, Culture & Society 37 (7): 1084–88.

PS:

Note, I definitely wrote way more than what’s necessary. So don’t feel like I’m setting an example you have to follow… everyone has been writing the proper amount so far!