Rhythm Awareness:
Predicting future presence when not currently available

When members of a team are remote from each other, as more and more are, real-time awareness systems, like Awarenex, can help determine if a person you want to reach is available right now. But what if they're away? Will they be back soon? Are they out to lunch? Should you send an email, and if you do, when will they see it? The modern workforce faces these and similar questions in coordinating work and communication among remote team members. The Rhythm Awareness research explores what information can be gleaned from the history of a person's activity and presence information and applications of that information to help distributed team members contact each other.

As a simplified example of the complex rhythm inferencing, suppose the current time is 12:15 and John is currently away. The figure below shows his "presence probability" throughout the day on Mondays in his office. From that, the system predicts that John will likely return at 12:45 and be available until 2:00. He will also probably be available again between 3:15 and 5:15. This is just an example - the prediction is actually a more complex process based on more information described in the papers below.

Rhythm Awareness applications complement real-time awareness systems, like Awarenex, by providing information about coworkers' future availability. The applications are potentially useful for distributed coworkers who do not have a strong sense of each other's rhythms and may also be useful to coworkers who are newly introduced and have not yet had time to form awareness of each other's rhythms. The techniques may also be applied to other computer-mediated communication technologies.

Project Status

Papers

For more details about Rhythm Awareness, refer to the following papers.

Rhythm Modeling, Visualizations and Applications (PDF)
James "Bo" Begole, John C. Tang and Rosco Hill
UIST 2003, to appear.

People use their awareness of others' temporal patterns to plan work activities and communication. This paper presents algorithms for programatically detecting and modeling temporal patterns from a record of online presence data. We describe analytic and end-user visualizations of rhythmic patterns and the tradeoffs between them. We conducted a design study that explored the accuracy of the derived rhythm models compared to user perceptions, user preference among the visualization alternatives, and users' privacy preferences. We also present a prototype application based on the rhythm model that detects when a person is away for an extended period and predicts their return. We discuss the implications of this technology on the design of computer-mediated communication.

When Can I Expect an Email Response? A Study of Rhythms in Email Usage (PDF)
Joshua R. Tyler (HP Labs) and John C. Tang
ECSCW 2003, to appear.

A study of email responsiveness was conducted to understand how the timing of email responses conveys important information. Interviews and observations explored users perceptions of how they responded to email and formed expectations of others responses to them. We identified ways in which users maintain and cultivate a responsiveness image for projecting expectations about their email response. We also discuss other ways people discover contextual cues for responsiveness, which include using tools such as the calendar and phone, accounting for the amount of work time overlap available, and establishing a pacing between email correspondents. These cues help users develop a sense of when to expect a response and when breakdown has occurred, requiring further action.

Activity Rhythm Detection and Modeling (PDF)
Rosco Hill, James "Bo" Begole
CHI 2003, Conference on Human Factors in Computing Systems, short paper, Ft. Lauderdale, FL, USA, April 5-10, 2003.
Presentation slides

We present an algorithm for detecting and modeling rhythmic temporal patterns from the record of an individual's computer activity, or online "presence." The model is both predictive and descriptive of temporal features and is constructed with minimal a priori knowledge.

Work Rhythms: Analyzing Visualizations of Awareness Histories of Distributed Groups (PDF, ACM Digital Library)
James "Bo" Begole, John Tang, Randall Smith and Nicole Yankelovich,
Proceedings of the 2002 ACM conference on Computer-Supported Cooperative Work (CSCW 2002), New Orleans, LA, USA, Nov 16-20, 2002, ACM Press, NY, pp.334-343.
Presentation slides

We examined records of minute-by-minute computer activity coupled with information about the location of the activity, online calendar appointments, and e-mail activity. We present a number of visualizations of the data that exhibit meaningful patterns in users' activities. We demonstrate how the patterns vary between individuals and within individuals according to time of day, location, and day of the week. Some patterns augment the schedule information found in people's online calendars. We discuss applications for group coordination (especially across time zones) plus opportunities for future research. In light of the popularity of presence and awareness services, this work identifies some of the benefits and privacy risks associated with the uses of online awareness information.