In "Responsibilities and Implications: Further Thoughts on Ethnography and Design continues to elaborate on the use of ethnography in human-computer interaction and the "implications for design" issues he addressed at CHI2006 (see my notes here). In the CHI paper, he argued how the use of ethnographic investigation in HCI is often partial since it underestimated, misstated, or misconstrued the goals and mechanisms of ethnographic investigation. Which is problematic since researchers aims a deriving "implication for design" from these investigations. The DUX paper continues on that topic to show how ethnography is relevant but not in the bullet-point "short term requirements" way some use to think about. As he says, "the valuable material lies elsewhere" or "beyond the laundry list", which is described through 2 case studies about emotion and mobility.
Then what should be these implications for design (voluntarily skipping the examples, see the paper pls)?
"The implications for design, though, are not of the “requirements capture” variety. They set constraints upon design, certainly, but not in terms of operationalizable parameters or specific design space guidance. What they tend to do, in fact, is open up the design space rather than close it down, talking more to the role of design and of technology than to its shape. (...) A second observation about the implications is that they are derived not from the empirical aspects of ethnographic work but from its analytic aspects. That is, the ethnographic engagement is not one that figures people as potential users of technology, and looks to uncover facts about them that might be useful to technologists (or to marketers). Instead, ethnographic engagements with topics, people, and fieldsites are used to understand phenomena of importance to design, and the implications arise out of the analysis of these materials. (...) the theoretical contributions that the studies provide have a considerably longer shelf life, and a relevance that transcends particular technological moments.
Is it a cop-out to say that what these studies provide is a new framing for the questions rather than a specific set of design guidelines? Hardly.
In addition, his discussion about the responsibilities is also important:
"The engagement between ethnography and design must be just that – an engagement. Ethnography and ethnographic results are part of that engagement. (...) I’d argue that it is no more the ethnographer’s responsibility to speak to design within the context of each specific publication than it is the designer’s responsibility to speak likewise to ethnography. Rather, the responsibility for ethnographically grounded design results is a collective one.
Why do I blog this? This is a topic Paul Dourish will address at LIFT08 in Geneva. Beyond that, this article echoes a lot with both reviews I received from academic papers (criticisms towards implications for design that are too broad and not short term requirements) and what can be observed from designers' practices at the Media and Design Lab I joined 6 months ago.
Closer to my own research, I like the way he frames this notion of implication; and indeed ethnography can bring more than sort term recommendations as it can uncover motivations for action, needs and deeper human rationale. In my research about location-awareness, we explored the differences between self-disclosure of one's location and automatic positioning; in this case, the crux issue was not to oppose the two sort of interfaces but rather, to show how each of them was different and had different implications in terms of human motivations (for example, self-disclosure of one's location is linked to communication intentionality).
Dourish, P. 2007. Responsibilities and Implications: Further Thoughts on Ethnography and Design. Proc. ACM Conf. Designing for the User Experience DUX 2007