On Issue crawler, key informants and black boxes

One of the arguments that regularly crops up in my work is to call for the integration of various ‘traces’ in ethnographic research. This seemingly simple point actually poses a number of quite intricate challenges to the ways in which ethnography is traditionally practiced. One of the strands of this exploration of the potential of traces, the consideration of hyperlinks, leads to the wild and wonderful jungle of maps and mapping tool.

Among the many fruitful discussions begun during our visit to Barcelona, one had to do with the use of various mapping tools in our analysis of networking practices in women’s studies, as addressed in our lecture ‘Sisterhood, EU Funding and Networked Research: a virtual ethnography of the making of the Athena Thematic Network in Women’s Studies’. There are several aspects to this exchange, such as the point raised by Adolfo Estallela in a working session, about the use by his informants of certain types of maps. What does it mean when analysts and informants use similar tools? This question is especially interesting if they are not using them in the same way! What is then the ‘privileged’ position of the analyst? Is the analyst supposed to deconstruct (or debunk?!) the transparency of the tools and their representations?

Another related topic is the extent to which ethnographers, as analysts, can use these maps in constituting their objects. Elisenda Ardevol raised this issue in relation to a presentation about issue crawler, at the recent EUROQUAL meeting on digital qualitative methods in Cardiff. Elisenda describes, if I follow correctly, her relation to this mapping tool as ultimately disappointing for her purposes. The specifics of that disappointment are what I find so fascinating and useful in reflecting on virtual ethnography.

In our work on ATHENA (together with my colleague Johannes von Engelhardt) , we went several rounds with various tools. The joint effort had the beneficial effects of not only of pooling our complimentary expertise (Johannes delights in the exquisite statistical modulations of formal analysis), but also the added effect that since every decision needed to be communicated and coordinated between us, we hashed things out rather thoroughly.

Where disillusion threatened, we sometimes (though not always) came to a shared appreciation of the possibilities offered by various tools. One of the fruitful motifs that was repeated across these exchanges had to do with the notion of the ‘black box’. Can one hold against a mapping tool that it is a black box? It would seem disingenuous to do so… In doing ethnographic work, one works with black boxes all the time! Furthermore, I’m one of the black boxes too! The point is that at certain points, we all readily accept black boxes. No tool is entirely transparent. The question is not ‘is this a black box or not’, but rather, what do we find acceptable to find black boxed or not.

Different tools allow different angles of scrutiny and have different kinds of transparency, and as researchers, we have different demands for the accountability of tools. Whether we require published algorithms, easy to use/non command-line interfaces, stable databases, evocative forms of output, or tolerance for messiness, all these criteria have to do with the particular purposes we want these maps to fulfill, and the relative (privileged) place we want to grant them in constituting our object. Mapping tools as key informants really…


2 responses to “On Issue crawler, key informants and black boxes

  1. Yeah! the question about maps as tools for representing “the field” are really intriguing and fascinating. The problems we were discussing were of different textures. As analysts, if maps represent relationships among “things” in terms of space and distances between “locations”: first, we have to know and understand the parameters by which software like Issue Crawler or Touch Graph identify and relate different “sites”, and how these relations are graphically represented. Second, we have to learn how we must interpret these graphics. And, third, we have to give meaning to the results. As ethnographers, we can also study how people are using these kind of maps, for example, to legitimate some claims about collective identity, influence, power or social networking. In this sense, we cannot interpret a network of hiperlinks as a representation of a social network, but may be some people we study with, do. For me, another interesting example realted with Adolfo’s work is to see how the “blogsphere” can be “materialized” by this exercise of mapping, and how maps can be used to pretend that they describe its shape.

    By the way, I am just getting familiar with Issue Crawler and I cannot evaluate it, yet. I only have been playing with Touch Graph, and until now, the results show no clear evidences that it will be useful for our online ethnographies, more than Google itself can be. But…

  2. Hi there,
    Just a quick reaction:
    As far as I know Issue Crawler is a great tool for tracing Web debates, but what I missed in this blog is a link to the actual tool, developed by Richard Rogers et al. Please, see for more info at: http://www.govcom.org/publications.html


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s