Why do I want ambient intelligence?

“Pushing the Envelope” from ACM interactions, March/April 2005

By Fred Sampson
wfreds@acm.org

I usually think of ambience in terms of the delightful surroundings at my favorite restaurant: It’s one of the reasons I feel comfortable there. But restaurant ambience and ambient intelligence are two very different things (aren’t they?).

I may choose to dine at a favorite restaurant because of its fine food and equally attractive environment. Can I count on intelligence being an element of that ambience in the future? Will the intelligent system know what I want to order as soon as I arrive? Do I get to change my mind?

Philips Electronics defines ambient intelligence as “The presence of a digital environment that is sensitive, adaptive, and responsive to the presence of people.” Isn’t it refreshing to see people (not “users” or “consumers”) as the focus?

Scenarios promoting ambient intelligence remind me at times of scenes from the film version of Fahrenheit 451. In one scene, Linda Montag (one of Julie Christie’s characters) has been invited to appear in an interactive play, giving her the opportunity to be part of the action. But, embarrassingly, she really hasn’t a clue about what to do.

Will I have the same experience with ambient intelligence?

At the 1964 New York World’s Fair, there were various exhibits and demonstrations purporting to show what the future might hold: astounding new means of transportation, communication, and interaction. I recall clearly thinking, “OK, fine, this all seems wonderful and even desirable, but who’s going to pay for it? This stuff looks expensive!” (During my research for this article I noted with some interest that Disneyland’s House of the Future, unveiled in 1958, highlighted not electronics, but plastics.)

Forty years later I’m still not riding a monorail to work, but I do have all sorts of personal technology devices to help fill my life with information and misinformation. We have been promised, ever since the industrial revolution, that new technology will give us more leisure time, but it hasn’t worked out that way, now has it? We’re chained to our personal electronics instead of being liberated. We’re inundated by information, plagued by what one writer terms “media obesity.”

I already have too many electronic devices—and television, print, and Web advertising encourage me to acquire even more. What are the chances that any of the devices I might acquire will successfully interact? Can they do so securely, or must I fear for my data at every turn? How is ambient intelligence going to help me manage the glut of unstructured information?

One of the more fascinating recent technological developments has been smart dust: tiny microelectronic devices that can communicate in an ad-hoc network, transmitting data from battlefields or vineyards. Where do smart dust motes fit into ambient intelligence? And what about radio-frequency ID (RFID) tags? Retailers and the military want them for inventory control, but they can also persist as identifiers of things we wear or carry with us. Does RFID have a place in ambient intelligence? What if I want to be anonymous in a smart environment? We have ongoing issues with establishing identify on the Web; how will identity be established and confirmed in a smart environment? How do we establish trust?

Now here’s a scary thought: If we expect ambient intelligence environments to respond appropriately to our emotions, including knowing when to keep quiet, then they should have better social skills than many geeks, including me. If they don’t, I’ll be looking for the off switch. How do ambient-intelligence proponents plan to respond to that challenge? Here’s another one: While children are well-known to readily deduce the workings of interfaces that challenge adults, they are also well-known for their experimentally destructive tendencies. My children know how to operate a VCR, but they also know how to stick a peanut-butter sandwich into it on a whim. What happens when children—or adventurous adults—try to “game” the system, fooling it into inappropriate responses? Is it possible to hack an ambient intelligence into misidentifying us for nefarious purposes?

The January 2005 issue of Wired includes photos and descriptions of some very wired (and wireless) homes in the United States using the latest in high-budget technology. But nowhere to be seen are the words “ambient intelligence,” or even “ubiquitous computing.” And just cleverly hiding the controls for conventional electronics behind attractive doors and hiding displays in the ceiling doesn’t meet the ambient intelligence vision of being inconspicuous as well as pervasive. What are we missing? Is the United States falling behind here as elsewhere?

The July/August issue of <interactions> will focus on ambient intelligence. Here’s your chance to reveal to the SIGCHI community your vision of the ambient future, and specifics on how we’ll get there. But more importantly (to me), here’s your chance to convince me that ambient intelligence is feasible and desirable. Why do I want to be surrounded by this ubiquitous, responsive, interactive, intelligent network of devices? What’s in it for me besides too much information?

About the Author

Fred Sampson is a co-chair of BayDUX, www.baydux.org, a member of SIGCHI, and a senior member of STC. In his spare time, Fred works as an information developer at at IBM’s Silicon Valley Lab in San Jose, California. Contact him at wfreds@acm.org.

URLs

The following URLs were referenced in this article:

Philips on Ambient Intelligence

http://www.research.philips.com/technologies/syst_softw/ami/background.html

A 21st Century Affliction: Media Obesity

http://www.peterme.com/archives/000428.html

Smart Dust

http://robotics.eecs.berkeley.edu/~pister/SmartDust/

Wired’s “Own Your Own Remote-Control Castle Today!”

http://www.wired.com/wired/archive/13.01/remote.html?pg=1

Copyright Notice

© ACM 2005. This is the author’s version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in ACM <interactions>, Volume XII.2, ISSN 1072-5520 (March/April 2005) http://doi.acm.org/10.1145/1052438.1052448.