1536-1268/11/$31.00 © 2011 IEEE
Published by the IEEE Computer Society
What Will You Sacrifice?
The pervasive computing field has been acutely aware of privacy challenges since the beginning. Early press stories about technologies such as the Active Badge were often presented in the context of discussions about an Orwellian future in which we were observed all of the time. Of course, the reporters didn't accurately predict the rapid, widespread adoption of smartphones, which have extended our ability to monitor users far beyond the confines of the office. Concerns about monitoring people within the context of their workplaces seem the least of our worries in 2011.
I'm writing this EIC message outside after the last session of the 12th HotMobile workshop in Phoenix, Arizona. The "Conferences" department of a future issue of IEEE Pervasive Computing will feature a full report on the workshop, but I wanted to take this opportunity to reflect on a recurring theme of the workshop—that is, how much privacy will users sacrifice for application functionality? Many discussions at the workshop focused on privacy and how users might or might not be prepared to trade privacy for either increased functionality or reduced cost. Of course privacy concerns extend beyond the field of mobile and pervasive computing, but the combination of ubiquitous connectivity and rich sensing present particular challenges.
As a field, we've been acutely aware of privacy challenges since the beginning. Early press stories about technologies such as the Active Badge were often presented in the context of discussions about an Orwellian future in which we were observed all of the time. Of course, the reporters didn't accurately predict the rapid, widespread adoption of smartphones, which have extended our ability to monitor users far beyond the confines of the office. Concerns about monitoring people within the context of their workplaces seem the least of our worries in 2011.
Perhaps more tellingly, neither the reporters nor the technologists predicted the emergence of technologies such as Facebook and users' seemingly insatiable appetite for disclosing personal information online. This appetite for disclosure is, of course, at odds with many people's stated views on privacy, especially as it relates to images. Beyond the social networking world we are all, of course, being subject to increasing levels of observation. Those of us who live in the UK are particularly conscious of this with the proliferation of cameras in public spaces that record much of our everyday lives. Many people raise concerns about the privacy implications of these cameras. In practice, most of these public surveillance devices are in the hands of responsible authorities and their use is subject to appropriate regulation. In contrast, the cell phones we all carry can capture the same images, which many users then share without regulation or control—and these images are often helpfully tagged with the subjects' identities. It's in areas such as this that there is extensive public and subject specialist debate about how to manage privacy. Most people agree that the current approaches aren't working, but few can agree on an appropriate way forward.
Pervasive computing presents even richer challenges to our privacy. Sensing that extends beyond simple records of location to smart public and private spaces containing tens or hundreds of sensors offers the possibility of radically new applications in areas as diverse as intelligent diagnostics and security. However, it also raises the possibility of radically increased levels of privacy intrusion. Because it arose with the emergence of social networking, many areas of computer science see privacy as a "new" problem. Our field, however, has long recognized both the threat and the possibility of interesting research in this area. Indeed, the many tutorials offered in this area at conferences (for example, at the Pervasive 2007 Conference by Langheinrich) helped educate an entire generation of researchers on the issues. As a result, the pervasive computing community is well placed to suggest solutions to the "new" set of problems in this space, and we have the opportunity to demonstrate real impact from our work.
For this issue, guest editors Chandra Narayanaswami, Antonio Krüger, and Natalia Marmasse have assembled a collection of articles on the topic of pervasive retail. Advertising and retail have traditionally been early adopters of new technologies, and this will likely continue to be the case as pervasive computing technologies become more widespread.
In addition to the three theme articles, we have three nontheme articles. In "What's in the Eyes for Context-Awareness?" Andreas Bulling, Daniel Roggen, and Gerhard Tröster explore how eye movements can serve as a source of context information in pervasive computing. They hypothesize that the dynamics of eye movements as users engage in different activities reveals information about the activities themselves. If this line of research proves fruitful, the possibilities for capturing new levels of context information are really exciting.
Our second nontheme article, "LifeMap: A Smartphone-Based Context Provider for Location-Based Services," by Yohan Chon and Hojung Cha, describes a system that fuses accelerometer, digital compass, Wi-Fi, and GPS to track and automatically identify points of interest with room-level accuracy.
Finally, in "The Escort System: A Safety Monitor for People Living with Alzheimer's Disease," Daniel Taub and his colleagues describe a system for monitoring patients with Alzheimer's who are prone to wandering.
All of the reader surveys we conduct, both official and unofficial, suggest that IEEE Pervasive Computing's departments are a major contributor to the magazine's success. As part of a major refresh of this part of the magazine, I'm delighted to report that we have three new departments starting this issue.
In "Pervasive Health," Anind Dey (Carnegie Mellon University) and Deborah Estrin (University of California, Los Angeles) will report on new developments in pervasive and mobile computing as they apply to the health domain. Given this sector's importance, having a regular, dedicated department will let the magazine report on key developments in this area in a timely fashion. For the first installment, Dey and Estrin have solicited contributions from researchers in the field to help define the research area's scope and to provide a snapshot of the state of the art.
In the second of our new departments, "Experimental Methodology," Paul Lukowicz (University of Passau) and Stephen Intille (Massachusetts Institute of Technology) consider the role of experimental methods in pervasive computing. The overall aim of this department is to provide a forum for exploring best practices in experimental pervasive computing research. We hope that the department will be of value to all researchers in the field but especially to those early-career researchers who are looking for guidance on how to use experimental methods in the context of pervasive computing.
Albrecht Schmidt (University of Stuttgart) edits our last new department, "Innovations in Ubicomp Products." It replaces the "New Products" department, most recently edited by Maria Ebling and Ramón Cáceres. I would like to thank them for all of their work. Rather than report on new products (there are many online sources of information for readers interested in product reviews), the new department will focus on the underlying trends and innovations that are driving pervasive computing from the laboratory to the marketplace.
I'm absolutely delighted that we have been able to put together such a strong set of new departments. To make space, we're suspending the "Standards" department that Sumi Helal has edited since the magazine's inception. This department has proved immensely popular over the years and I would like to thank Sumi for all of his work.
After tremendous service to IEEE Pervasive Computing, including acting as a guest editor for this issue, Chandra Narayanaswami is stepping down. IEEE Pervasive Computing AEICs carry a significant workload as they are responsible for finding reviewers for all of the regular submissions as well as serving as guest editor on several special issues. On behalf of the IEEE staff and editorial board members, I would like to thank Chandra for his service to the community.