Recent work by our group at Motorola Labs has focused on applying the principles of ambient interfaces to the domain of mobile communications. Our methods incorporate both formative ethnographic studies and field evaluation of prototypes to investigate how people make use of ambient information in the course of everyday communication. Our goal is to enable applications that provide rich presence information for close friends and family that is appropriate to their particular tasks and social conventions. In this paper, we briefly summarize the results of two prior field studies of ambient awareness on mobile phones: shared motion presence and music listening history. We then discuss our current efforts in context-aware photo sharing on the phone. Together, these studies provide insights about how people understand, use, and contextualize presence information. We also show how this information helps people stay connected and strengthens relationships. We hope to use these insights to establish how contextual information can be gathered, synthesized and displayed to more effectively promote connectedness.
INTRODUCTION
To date, researchers have found some success in the
development of ambient information tools for such
environments as the home and office. These settings are
characterized by situations where there is a high potential
for users to be distracted from their primary activities by
any number of environmental or social sources. Ambient
interfaces allow these sources to communicate information
in a lightweight, non-distracting manner and can be
attended to selectively by users. We see similar benefits
RELATED WORK
Our work has been influenced by several existing systems.
WatchMe [
While our focus is to support existing close relationships
between people, other systems use context such as
proximity to allow users to access information about less
familiar others in their environment. DigiDress [
Participants used the service on their own phones for one week. In addition, they left voicemail diaries documenting their use of the system, and participated in two brief interviews during the study, with a longer final interview at the end of the study.
Motion Presence
The Motion Presence application (Figure 2) detected when
a user was in motion between places (e.g., driving home
from work or walking to the grocery store), and conveyed
this information to close family or friends through an
augmented phone book application. Whenever a user
entered this phonebook, they could see at a glance if their
close family or friends were in one place or moving
between places and the amount of time that they had been
in that state. We used GSM cell ID information to
determine when a participant was in motion, and these
changes matched well with participants’ own concepts of
“moving.”
We deployed this application to a total of 10 users from the
Chicago area, three couples and one group of four friends.
To understand use of the application and explore privacy
concerns raised by sharing motion information, participants
used the application for two weeks, left voice mail diaries,
and participated in two interviews with researchers. The
application and study are further explained in [
Micro-moments
We often found our users checking on the status of others in
small breaks during the day, similar to the breaks that
Anttila and Jung noticed in mobile media usage [
Abigail reported checking her phone during these micromoments at her data entry job where “I stare at a computer all day, so I figure it’s better to stare at my phone. So I’d constantly go to my purse and check [the music that my friends were playing on] my phone.” Likewise, in the Motion Presence study, Participants often needed a distraction from their current task and saw the Motion Presence application as an entertaining source of diversion. Harold reported looking at the application “at work when I don’t feel like doing my actual work.” Other users saw the application as a game and would try to catch each other moving or not moving throughout the day. Chris: “I looked at it mainly out of curiosity, mostly it was a game for me this afternoon to see if I could find a time when I could see her moving.” George reported checking for his “own amusement.” These micro-moments provided an escape from current tasks, even if just for a second, and helped users get through tasks that challenged their concentration.
Boredom While checking presence information during micromoments can be seen as offering a quick distraction from a current task, our participants also checked for longer periods of time when they considered themselves to be bored. These interactions typically involved a prolonged interaction with the application, looking into past history to try to learn more about the other participants.
In the music study, Bianca reported that “it was only when I was bored that I like went out of my way to see what they
were listening to or like cared a lot” and “If I was like…doing anything more interesting then checking it, like eating or at work or watching a movie on TV. But if I wasn’t doing anything that exciting like on the computer checking my email, I would always check it.” We also noticed our Motion Presence participants checking the application when they were bored. Several of our participants lived away from the city center and had to take busses or trains to get home. These participants frequently reported checking the application on public transportation as a way to pass the time. Individuals within our group of four friends often checked for when other members of the group were leaving work or going out to lunch. These interactions helped them to learn more about each other and in some cases led to topics for later communication. Purposeful As we had hoped, many interactions with these systems served a greater purpose than curiosity. Often users had specific tasks that they wanted to accomplish, such as determining if a friend was at home or trying to coordinate arriving at a location at the same time. In these cases, participants would check the status of others to determine the best course of action.
Music context was commonly used as a proxy for location
context. Since only participants’ home computers could
broadcast music context information, seeing that a friend
was playing music meant that they were at home (or had left
their audio player running). Early in the study, Bianca said
“[I’m] bored because nobody could go out and do things
this weekend, so there’s nothing for me to do now. But
maybe if someone is listening to music, I’ll know they’re
home…I was thinking if maybe they played music, I could
call them because I know they’d be home.” Caroline
reported looking at the application for this same purpose: “I
did pay attention to [timing] because I did wonder ‘oh, are
they home right now?’”
Abigail knew that her friend Bianca was out for the night
and “wanted to see what time Bianca got home, so I was
looking at her messages then to see when she’d gotten back
from going out.”
Checking to see if a friend was home could be seen as a
simple example of what Ling and Yttri call
microcoordination [
We saw our participants use the application to give themselves more time to spend in their current location. Ebony would check the status of her partner towards the end of the day to see if she had left work yet. “If she didn’t leave yet, that means I can go do whatever I’m doing, like at work stay later.” James also described using the application in this way: “If you knew someone was going to go pick you up or if someone was going to go someplace and you knew that and you know about what time, you could see if they were actually on their way or if they were running late. … Kind of lets you know when you should be ready for things like that.” Other participants used motion information to try to arrive at the same place at the same time. Harold and Ian were going to meet for lunch. Harold originally was going to call Ian when he was leaving, but then reconsidered: “I’ll call you, or I’ll just see that you’re moving!” Later that week when they were actually getting together, Harold reported using that information: “I could tell when he was leaving work by when he went off of ‘not moving.’ … It was like, ok, I saw that he was already on his way and we’d get there about the same time.” These purposeful uses show that much can be inferred from simple presence information. Oftentimes it is not necessary to share precise data like location that can lead to many privacy concerns. Something as simple as music playing or being stationary could be enough for a close friend of family member to be able to discern one’s location or activity while still allowing people to feel like they have not exposed information that could be misused by casual friends or strangers.
Not for weekends We found our participants mainly used these applications during weekdays when their schedules were busier and were often tying to meet or communicate with friends or family. In fact, one participant in the Music Study said that she sort of forgot about the system over the weekend, but it became important when she “had rehearsals [Monday and Tuesday] and I was really bored so I took out my phone like every two minutes looking at it.” In the Motion Presence study our participants also used the application less frequently on weekends. For participants that lived with their significant other, they often spent most of the weekend with them and therefore didn’t have a need to use the application. For friends, weekend time was either spent with family or scheduled in advance such that participants knew when their friends would be available and when they were all meeting to go out. Ian said that he “knew where my friends were going to be all day. So I had no reason to know where they were or what they were doing or to contact them.” This data shows that mobile presence is most useful when a person’s time is scheduled, but prone to variations (e.g., the exact time they leave work). On weekends, a participant would either be unavailable or have their plans made in advance. In either case, the application was not needed. When participants could not use the information for purposeful interaction, were not bored, or did not desire a distraction, they simply did not view the information presented on their phones.
CURRENT WORK
Currently, we are investigating the use of contextually
relevant photographs displayed on the idle screen of the
phone. Using the ZoneTag [
This indirectness of data used in our studies allows for a
larger variety of information to be sent with fewer privacy
concerns since there is still a large amount of plausible
deniability. By contrast, most home presence applications
have provided direct information such as physical presence
in a room [
Finally, mobile phones are unique in being able to provide a large amount of rich context that can be shared with others.
Many phones today are able to determine their location through GSM cell ID or GPS, capture rich media such as photos and videos, and are aware of the media that a user is playing. This rich context provides many opportunities to share this information and allow for a greater sense of awareness as well as increased availability management. Having this information is increasingly important as people lead more hectic lives while maintaining a strong desire to feel connected to others.
QUESTION For us, a question we continually face is “How ambient is ambient in a mobile setting?” Is information that appears in the course of a normal phone interaction (e.g. an augmented phone book) enough to be considered ambient? What about information that is always visible on the idle screen or the external display? This leads us to a larger question of “What types of ambient affordances should a phone provide?” We hope that through this workshop we can explore some of these questions with others as well as explore the differences between mobile and home-based ambient applications.