Ian Smith

Social Mobile Computing

Biography
Ian Smith is a senior researcher at the Intel Research Seattle lab in Seattle, Washington. His work focuses on having a big bowl of ubicomp technology, social science, and some software engineering. Stir vigorously and don’t forget to drizzle on some privacy. He previously stirred the pot at the Palo Alto Research Center in Palo Alto, California. He was granted a Ph.D. and a chef’s hat from the Georgia Institute of Technology in 1998.

Position Paper
It may come as a shock to some that not all of the people one overhears on mobile phones are talking about their next sales presentation. Similarly, those millions of people sending billions of text messages all over the world, they are not all discussing the document that is due to the boss tomorrow. It has even come to our attention that there are people using mobile devices like BlackBerrys and Hiptops with mobile electronic mail and instant messaging to (gasp!) socialize or plan their next rendezvous with friends!

While the last paragraph may seem facetious, we trying to drive home the point that mobile communications technology has many uses beyond the obvious “work” and “efficiency” purposes, although these purposes are often the reason that mobile devices are ostensibly sold. One could even argue that for many people—particularly younger users—mobile devices have only a social purpose. At Intel Research Seattle, we are studying a new class of programs, social-mobile applications. Social-mobile applications are applications that support users in their social interactions with other people. Although there are social interactions at both home and work, the setting of social interactions are commonly “out in the world” at places like restaurants, cafes, concert halls, bars, and sporting events. Because of the varied settings of these social interactions, we believe that mobility is a key component of supporting users in their social milieu.

As we have been exploring the area of social mobile computing, we have found several research challenges that appear to key issues in allowing social mobile applications to reach mass adoption.

• Privacy. Applications that do not give users sufficient control of their personal data will be ones that at best attract significant negative publicity and at worst are rejected by users as agents of big brother.
   
• Limited Input And Output: Mobile devices, notably phones, are devices with extremely limited capabilities for input and output, are often used in unusual circumstances and almost always with only one hand. Applications that support social interactions must perform their function easily even under these constraints—or users will simply fall back to more “general purpose” applications such as telephony and SMS.
   
• Utility per dollar networking. SMS bandwidth, measured in dollars per bit, is some of the most expensive in the world; in general there is a substantial cost to bandwidth on mobile devices. Since social applications by definition will need to share data between people, these applications must be carefully crafted to share information in ways that maximizes utility to the end user for the cost paid.

While these challenges may seem daunting, the Social Mobile Computing group at Intel Research Seattle is pressing ahead with applications to test solutions to these and other problems. We are building a variety of applications in the areas of supporting social coordination and “meetup,” social networks for health and fitness, and social uses of collaborative filtering.

Back to Social Computing Symposium 2005