Bonnie Nardi

Bonnie Nardi

Having studied local cultures from Samoa to Silicon Valley, Bonnie Nardi is one of the best-known ethnographers working in the corporate world. Have you heard the term "information ecology" quite a bit of late? Bonnie is largely responsible, having co-authored Information Ecologies: Using Technology with Heart (MIT Press, 1999). Having worked at Apple, AT&T, and now Agilent, Bonnie has had ample opportunity to study complex corporate information ecologies. Not surprisingly, those experiences have made her seek "a middle ground between technophilia and technophobia," and have caused her to question the traditional view of user-centered design (UCD).

-- Lou Rosenfeld

Lou: Bonnie, your recent work seems focused on balancing the needs of social groups with those of individuals. And at Agilent, you're trying to figure out how ubiquitous computing devices impact this tension. Please tell us about your work and what you hope to learn.

Bonnie: Yes, a theme in my work is how to balance the increasing pervasiveness of technology with deep needs for privacy and time away from technology that must be protected. It is very easy for us to believe that we are so indispensable that we have to have our cell phones and Blackberries at the beach, but if we do that two things happen. One, we miss what's going on around us--the shift in the light in the sky, the pelicans flying past, the dampness of the sand as the tide comes in. Two, we degrade the experience of others at the beach by bringing our offices with us. Now the beach is a workplace. That's not what it's for!

Lou: It's not surprising then that your book, Information Ecologies (MIT Press, 1999), is subtitled "Using Technology with Heart". In it you and Vicki O'Day describe a robust collection of components that comprise the information ecologies found in most organizations. These include:

  • people
  • practices
  • technologies
  • values
If you look at any basic information architecture "blueprint," it'll become immediately obvious that our field has a difficult enough time visually representing a few related (and quite basic) areas, namely content and tasks. Is there a "visual vocabulary" to model the broader ecologies that you describe?

Bonnie: That's a great question. I don't know of a visual vocabulary but it would be very useful to develop one. Might be a good pilot project for an information science/library science student to tackle.

Lou: Might the field of social network analysis be a good place to look?

Bonnie: Network diagrams can be a start in this area for laying out the players, but I think visualizations of ecologies need to include available technical resources, for which a map might be a good visualization. There's also the question of values, which are more difficult to visualize, though perhaps some hypertext could be useful here.

Lou: Please discuss this concept of values a bit; perhaps you could provide an example?

Bonnie: Most institutions and groups have clearly articulated values; it's part of the group identity and a guideline for action. So libraries, for example, have values of free information access and service to clients. These values should, in my opinion, be a reference point for how to introduce and deploy new technologies. One library I visited at the University of Illinois-Chicago refused to put telephone menus on their phones--they always have a human answer the phone--as they believe that provides the best service to clients. I love that!

Lou: The ethnographic approaches that you and your colleagues promote are hugely important parts of the user-centered design (UCD) toolkit. Yet for some reason ethnography doesn't seem to receive the attention it deserves as part of that toolkit. In fact, UCD is often equated with usability engineering. Why do you think this is?

Bonnie: I like to talk about activity-centered design instead of user-centered design, for exactly the reason you stated: UCD has come to be equated with usability. The essence of activity-centered design is a combination of usefulness and usability. Good applications should have both.

Activity-centered design focuses on the activities for which the technology will be used in an information ecology. Ethnographic methods are effective for finding out what would be useful technology. Ethnographic methods can be combined, in the iterative design cycle, with experimental usability methods, or even heuristic evaluations, once a good design has been produced, to see if it's working well.

You might be interested in my work on ContactMap in this context, which will be published in the December 2001 issue of CACM. You might also be interested in the ethnographic study that informed ContactMap; it's available from First Monday.

Lou: I enjoyed your discussion of ContactMap at the IA 2000 conference, but for readers who couldn't attend, could you describe what your goals are for this software?

Bonnie: ContactMap is a redesign of the computer desktop interface based on people rather than computer-centric objects like folders and hierarchies. It represents the way we really work--with other people. ContactMap integrates communication and information functions in a single UI. It doesn't replace current systems, but is what we hope is a more useful interface to communication and documents than what we currently have.

Lou: Getting back to the respective roles of ethnography and usability: why do you think that for most people UCD equals usability? And are there other fields that should be considered within the context of UCD? Merchandising and market research come to mind.

Bonnie: I think UCD is equated with usability for historical reasons--it was the problem most amenable to the cognitive science methods that were prominent in the early days of human-computer interaction study. Now that we are branching out to ethnographic methods, looking at the world sociologically, we can think more about usefulness. I think market research is a great area for us to explore. In fact, I'm learning to do competitive analysis as part of my job at Agilent. I hope to pick areas of study more strategically by doing so.

Lou: Let's conclude with some practical advice for information architects: are there two or three ethnographic techniques that we can start using right away?

Bonnie: Ethnography is a simple set of techniques (interviews, observations, participant-observation) but a mind set that takes a while to develop. The reason one uses ethnographic methods is to understand, from the user's point of view, their activities, in their social context. So the techniques are easy, but the analysis takes thought and reflection.

The best way to learn ethnographic methods is to apprentice with an ethnographer. This is quite doable in any university setting where you can take an anthropology course or do an independent study with an ethnographer (usually found in anthropology or sociology departments, but also scattered around elsewhere).

Doing the interviews and observations will be easy; distilling meaning will be harder and takes some time and practice. It's great to do this kind of thing in a group where you can discuss your findings (with privacy safeguards in place of course) and your field experiences as you are having them. I'm going to teach a course at Stanford in the spring, hopefully, in the computer science department, to give design students a chance to learn ethnographic methods while they are still students.

Lou: Yet another reason to be back in grad school... Thanks Bonnie!

Subscribe to our bi-weekly newsletter for notification of new ACIA articles and interviews.