the Transparent Crowd

Ed, Annie, and Brooke have each mentioned our discussion of whether to allow users to mark parts of a text or picture in more than one color. In the grad lounge on Tuesday morning, Ed and I went back and forth about the implications for interpretation and for possible visualizations. He was concerned about preserving and representing the particular instances when a user marks something as more than one color while I suggested that this would not matter nearly as much as the aggregate markings in each color. He drew on the dry erase board. I raised questions about his ideas and then questioned my own questions and probably contradicted myself. We were each of course projecting assumptions about what kinds of texts we will have in Prism, what we might ask people to highlight, and how the resulting interpretation might be used. I don’t think we realized all that, though; in retrospect, I can see how conversations like this one put pressure on the ideas and images of Prism that I’ve been forming since August. The same goes for the transparency exercise (quite a metaphor!) we did in our full group meeting Tuesday afternoon. I hadn’t thought much about interpreting images (Ed’s example) or about highlighter categories created to emphasize overlapping interpretations (Alex’s example). And for the first time, I recognized that Prism could be used to interpret the highlighter categories as much as or even more than the text (“modernism” in Sarah’s example). This exercise made me feel a little foolish for having spent so much time on the issue of overlap without a fuller sense of what kinds of categories and texts might be possible, but it also contributed to my realization that exercises like this are what we need to remind us that we still don’t know what exactly Prism is, and maybe we never will. Fittingly, I think the Fight Club reference finally emerged last week (first rule of Prism…)

Talking even hypothetically about multiple markings taught me something about crowdsourcing as well. Preserving the individual instances of multiple highlights does indeed seem like it could serve many intellectual interests, but it does not seem like “crowdsourced interpretation” as I understand it. If we privilege crowdsourcing as the method that Prism makes possible (or at least much, much easier), should we limit the questions researchers can ask Prism to those that a large number of non-individuated responses can answer? When Bethany piled all our transparencies on top of one another, no one’s individual transparency was easy to see. One person’s double-markings were indistinguishable from the markings of one transparency beneath another. Will Prism recreate this exercise on a larger scale, or will it allow for a kind of crowdsourcing that preserves individual interpretation somehow? In my mind, this question parallels the ideas I was struggling with recently about equality and collaboration versus the solitary, masterful scholar. I think I’m coming down on the side of the wisdom of the crowd instead of the nuance of the individual, so maybe we can call that progress.