the day, but have low occupation at night. For example, the Financial District is a vibrant place in the daylight (work) hours but distinctly quieter and dark at night. This kind of information is useful for potential daters venturing to new parts of town for a night out with someone they don’t really know.

Perhaps there is a whole space of possibility in the creation of “anxiety maps.” I know of more than one person who avoids entire routes on the basis of bad memories or the possibility that they may encounter a known, but no longer favored, person. More whimsically, given my love of high heels, I also designed a map and route/navigation tool for perambulating San Francisco, irrespective of footwear practicality. The map charts routes based on the height and style of your shoes with rules like: Five-inch platform boots should not be worn on steep slopes, and stylish stilettos are a no-no on potholed, grated Mission Street. I call such maps “ socio-topographical.” Even further up the whimsical chain, I’m thinking of designing the Prescient Plug-in for a navigation device. It will tell you when you are on a fool’s errand and send you elsewhere—somewhere that may be more enjoyable or rewarding.

No, I am not entirely serious, but I am trying to illustrate that maps may be adapted to a range of terrain experience, proclivity, and accessibility beyond simply car, bike, and hike routes. Maps can address a sense of a place not rooted in objective fact or mathematical measurement. Some maps should be more reflective of experience rather than “flat” fact; some maps

should be warm and “living” not cold. Online maps with stories, comments, and conversation attached to representations of location—Flickr maps, Twitter vision, Community Walk, Platial, and my group’s MapChat prototype that lets people insert chat bubbles onto locations in Yahoo! maps—are examples of living, warm maps.

Aside from the representation issues that I have been elaborating, there are some interactional issues that my story of visual nothingness and vocal inappropriateness brings up. Why don’t technologies know when they don’t know, or know that they are perhaps confused? What is so hard about building in some reflective capability that lets an interactional device know when it’s the victim of a service interruption, that lets it then handle that knowledge in a socially appropriate way? Like, “Whoops, I was offline for a while and I’m now discombobulated. What was it you wanted to do?” Or, perhaps there is a different, socially appropriate action to take when clueless—shut up and don’t spout out random suggestions. For Jane, who was clearly confused after having been offline that would have been a better course of action than telling me to turn around. Just maybe, at times like this one, I know more than she does.

Yes, yes, I am anthropomorphizing Jane beyond her capabilities. I attribute this requirement of mine that she actually be a conversational navigational partner to her “sounding” so human—stilted, but human. As if she really were a person. So I do get lulled into a false sense of security about her knowledge

and about her understanding of what I know. But this is not just a personal pathology. In looking at the ways in which we interact with humanlike conversational agents, Stanford researchers Byron Reeves and Cliff Nass a decade ago pointed out in their book The Media Equation that we get very engaged with these agents, as if they were human. And we expect them to be as good as a human in communicating with us. Further research suggests that when these conversational agents slip up, it can actually cause us to be quite annoyed with them.

Indeed.

Could Jane and the map she was reading from not have collectively told me that they just didn’t know what was happening or where the heck they were? That they had a dearth of information?

Apparently not. The illusion of a navigator-friend embodied in a small box occupying the passenger seat crumbled.

Jane and I came to an agreement. I would find my own way out of the valley, in my own time. And she would shut up. I turned the Tom Tom off and navigated myself to the beach.

uRLs of interest http://www.flickr.com/map http://twittervision.com http://www.communitywalk.com http://platial.com

ABOUT ThE AUThOr Dr. Elizabeth Churchill is a principal research scientist at Yahoo! Research leading research in social media. Originally a psychologist by training, for the past 15 years she has studied and designed technologies for effective social connection. At Yahoo, her work focuses on how Internet applications and services are woven into everyday lives. Obsessed with memory and sentiment, in her spare time Elizabeth researches how people manage their digital and physical archives. Elizabeth rates herself a packrat, her greatest joy is an attic stuffed with memorabilia.

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without the fee, provided that copies are not made or distributed for profit or commercial advantage, and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on services or to redistribute to lists, requires prior specific permission and/or a fee. © ACM 1072-5220/08/0700 $5.00

July + August 2008