I've rarely felt more urbane than the last time I arrived in Tokyo. I had the foresight to bring my SUICA (Super Urban Intelligent Card), an RFID-based smartcard already loaded with yen left over from my last trip. Card in hand I swept through the gates of the JR transit line at Shinjuku station to the nearest automatic noodle restaurant, where I selected my meal from pictures on a vending machine positioned at the entrance. I touched my card to the machine and received a ticket that I handed to the staff at the desk, where the hot noodles were already waiting for me. Meal eaten, I proceeded to check out which bottled water was No. 1 at ranKing ranQueen (a local chain of stores, where items are ranked), and again paid using a simple gesture with my card. The only thing that could have made the experience more complete was if the RFID chip had been integrated in my keitai (mobile phone). Although the SUICA is now at least seven years old, and many similar systems are in use in Hong Kong, Seoul, London, and elsewhere, it still is a remarkable example of how ubiquitous technologies such as sensing, networking, and data access can be used in everyday applications.
It is interesting to note how most of this technology is currently used to increase convenienceand reduce human contact. SUICA eliminates the fumbling with coins at shops and subway gates, which could lead to embarrassing lines. But it also cuts down on interaction with other people, especially when combined with various vending-machine services. In ubiquitous computing research, on the other hand, we have seen RFID tags and other near-field communication technologies such as Bluetooth often being used with a view to increase human contact. Projects such as Meme Tags, an interactive nametag that communicates the wearer's interests to act as an icebreaker , or Ticket2Talk, which presents a person's interests on a public display to initiate conversation , or tunA, a system for listening to other people's music , all promote a vision that technology can and should encourage communication, especially among strangers.
In fact, it seems researchers more often than not want to imagine a utopia, where everybody loves everybody and wants to hang out with everybody else. "Wouldn't it be great," we say, "if a device on my body could broadcast my interests to everyone around, so that I could strike up exciting conversations during my commute!" Never mind that the person whose taste in music most matches my own might not be the cute girl sitting in the opposite aisle, but the slightly scary middle-aged divorced guy who can't keep upright because he already had a few before breakfast. Fortunately, no mobile social networking app has reached enough users to let us find out... yet.
Even though the study participants were very curious about who they were sharing music with, they never crossed over from the technologically mediated communication into actual conversation. This showed that the barriers between electronic and real connections are harder to cross than many seem to think.
This is not for lack of trying. There have already been several attempts at social networking software for standard mobile phones, such as Germany's akaaki (www.aka-aki.com) and U.S.-based Loopt (www.loopt.com). But they all struggle with the same two basic problems: battery life and critical mass.
The first may seem trivial, and more than one startup seems to simply shake it offisn't everything in electronics getting better all the time anyway, according to Moore's law?
No, this is actually a real killer. A device that pings its surroundings wirelessly with regular intervals, using Bluetooth or Wi-Fi, will drain any sensibly sized battery in a few hours at most. Continuously pulling up and reporting a GPS location can be even worse. It is highly unlikely that users will stand for carrying half a dozen replacement batteries, and barring an unprecedented breakthrough in battery technology, the only fix on the horizon is some kind of push solution based on network cell location. Unfortunately, to be useful this requires a degree of cooperation between network service providers that is still a long way off.
The second issue stems less from a lack of user interest and has more to do with the extremely fragmented mobile device market. Whereas signing up for a Facebook account can be done in a matter of minutes, downloading and installing a mobile application has been lots of hard work and beyond the reach of most normal users. This might change very fast, however. With Apple's iPhone 3G and the accompanying Appstore, there is now for the first time an attractive platform and sales channel for mobile software. In response the rest of the market is likely to finally consolidate around a small number of standard operating systems (including Android, Symbian, and Windows Mobile). This means that quite soon, we will see people downloading and using social software on their phonesand those that hook into existing networks will have a head start. Already, iPhone versions of AIM and Facebook are among the Appstore's top downloads, with others such as Twitteriffic and MySpace also gaining headway.
This is an exciting time: Suddenly, all those research prototypes we have dreamed up over the past decade will have a chance to become real products. All previous deployments of mobile social applications have been limited by access to hardwareresearch labs like mine typically have only a few dozen terminals at most to give out, and this makes gaining critical mass impossible. Soon it will be possible to distribute mobile software freely just like for desktop computers. Then we will finally see if people really want to strike up conversations about music on the subway, or perhaps even start swapping household tasks with each other . So what can researchers add to the big corporations and startups in the imminent mobile social software feeding frenzy? A sense of perspective, perhaps.
In my own projects, I have tried to approach the problem in an open-ended and explorative, rather than dogmatic, way. By building working prototypes and putting them to real-world use, it is possible to go beyond the idealized cases and get some real knowledge on mobile social software, even with a fairly small set of users. With the Hummingbird, we built the first device for mobile awareness that required no infrastructure, in a time when mobile phones were still not widely used . To get around the critical mass issue, we used it to enhance the interactions of a group of friends, rather than act as an icebreaker for strangers. It turned out to be particularly interesting to use the Hummingbird in situations that spanned work and social life, pointing toward current use of mobile phones. However, we also found that battery life (as always!) and the burden of carrying an extra device prevented people from taking it up for extended use. A startup company that was formed to commercialize the concept crashed and burned, mainly because of the costs of developing dedicated hardware.
When handheld computers with Wi-Fi became available, my group constructed a number of games in collaboration with students to explore new forms of wireless play . All the games were constructed so that to win, players had to communicate with one another directly, rather than through the technology. For instance, in a racing game, each player had a unique view on their device. One player was the driver who controlled the car, and the other was the map reader who continuously fed information on upcoming turns. To succeed, they had to share information through speech rather than the device interface. Here we learned that the physical context is a powerful resource for mobile applications, one that is still not used in any products I know of, even though technically, Nintendo DS and other mobile gaming platforms could easily handle it.
Media sharing is another interesting domain for mobile software. With Push!Music, we tried to create an application in which music would spread among users almost virally, both through automatic and manual recommendations . Similar commercial products, e.g., Microsoft's Zune, paint a picture of how music sharing would open up new avenues for social contact. However, we found that even through an extended period of use, during which hundreds of songs were shared, not a single person initiated real-world contact with someone they didn't already know! It turned out that even though the study participants were very curious about who they were sharing music with, they never crossed over from the technologically mediated communication into actual conversation. This showed that the barriers between electronic and real connections are harder to cross than many seem to think, which could spell trouble for the more naïve social applications that will flood Apple's Appstore in the coming months.
What this teaches us is that while the mechanics of our journeys may be automated almost completely, the connections along the way will be a much tougher nut to crack. Although brushing by someone in virtual spaces like MySpace and Facebook does not carry a lot of weight, most of us are a lot more particular about whom we socialize with on the subway. Even when it comes to people we know, we might not always be in the mood to see them. A new system under development in my group, meant to be used for keeping track of the position of friends on public transport, turned out to be just as useful for avoiding people as for meeting them!
But perhaps a new generation, reared on FriendFeed and Twitter, will move more fluidly between modalities. It is possible that soon we will not even notice the transitions from digital media to real-world conversations and back again. And maybe that guy on the subway will turn out to be not so scary after all, but your new best friend. Until then, I am happy with a system like the SUICA, which simply makes my travels and noodle purchases a little simplerno human contact required.
For more thoughts on Automated Journeys, see the workshop of the same name at UbiComp 2008: http://www.inbetweeness.org/automated-journeys.
1. Borovoy, R., F. Martin, S. Vemuri, M. Resnick, B. Silverman, and C. Hancock. "Meme Tags and Community Mirrors: Moving from Conferences to Collaboration." In the Proceedings of CSCW 1998. Seattle, Washington, 1998.
3. Bassoli, A., J. Moore, and S. Agamanolis. "tunA: Socialising Music Sharing on the Move." In Consuming Music Together: Social and Collaborative Aspects of Music Consumption Technologies, edited by Kenton O'Hara and Barry Brown. New York: Springer, 2006.
7. Håkansson, M., M. Rost, and L.E. Holmquist. "Gifts from friends and strangers: A study of mobile music sharing." In the Proceedings of ECSCW 2007, edited by L.J. Bannon, I. Wagner, C. Gutwin, R.H.R. Harper, and K. Schmidt. 311330, New York: Springer, 2007.
Lars Erik Holmquist is a research leader at the Mobile Life Centre at Stockholm University and the Swedish Institute of Computer Science. The Centre collaborates with the telecom industry to develop and study future mobile services that span all areas of life, from entertainment and socialization to work and society. His group, the Future Applications Lab, has developed and published work on many novel interactive technologies, including mobile media, visualization, and robotics.
Figure. Images of the SUICA penguin, well-known mascot for Japan's Super Urban Intelligent Card, can be found throughout Japan where a quick, contactless swipe allows cardholders to pay for mass transit tickets and many other purchases.
©2009 ACM 1072-5220/09/0100 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without 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 servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2009 ACM, Inc.