Using Location-aware Multimedia Messaging (LMM) systems as a research testbed, this paper presents an analysis of how ‘fun or playfulness’ can be studied and designed for under mobile and ubiquitous environments. These LMM systems allow users to leave geo-tagged multimedia messages behind at any location. Drawing on previous efforts with LMM systems and an envisioned scenario illustrating how LMM can be used, the authors discuss what playful experiences are and three problems that arise in realizing the scenario: how playful experiences can be inferred (the inference problem), how the experience of capture can be motivated and maintained (the experience-capture maintenance problem), and how playful experiences can be measured (the measurement problem). In response to each of the problems, three design considerations are drawn for playful Human-Computer Interaction: 1) experiences can be approached as information-rich representations or as arising from human-system interaction 2) incentive mechanisms can be mediators of fun and engagement, and 3) measuring experiences requires a balance in testing methodology choice.
TopIntroduction
On a sunny afternoon in mid-July, Nicole and Nick are tourists shopping around Nejmeh Square in downtown Beirut, Lebanon. While Nick insists on seeing the cultural offerings of Saifi Village, a village completely rebuilt as a New Urbanist-style neighborhood after its destruction during the civil war, Nicole has a different notion of what is fun and enjoyable. Familiar with her interests in warm, foreign cities, Nicole’s mobile device sets her to experience ‘fun’ places nearby, suggesting several lively cafés along the Corniche, a seaside walkway with a glittering view of the Mediterranean. Skeptical about the suggestion, she makes a predefined gesture instructing her device to show her different multimedia (photos, songs, videos, text) that reflects people’s experiences there. The device presents her with a dizzying nexus of visual and musical perspectives captured by people enjoying themselves, complementing each multimedia message with related past and future events. Leaving Nick, she makes her way toward the Corniche until she reaches a café, where she sits outdoors, happily absorbing the scorching sun rays. Wondering where Nick went, she decides to capture her current experience. She takes a photo of the clear blue sky and sea (Figure 1), which she annotates with the song by The Cure ‘Play for Today’ and writes: “That’s New Urbanist-style culture too!!” While she awaits her hookah and drink, she scans through other people’s experiences at the café she is at, only to realize the place attracts mainly an older crowd, which is no fun at all.
Figure 1. A mockup illustrating the photo Nicole took of the Corniche seaside and the corresponding annotations she added
The preceding scenario illustrates ongoing research efforts within the MOCATOUR (Mobile Cultural Access for Tourists, http://mocatour.wordpress.com) project. The aim of the project is to define computational methods that facilitate tourists with contextualized and media-based access to information while they freely explore a city. The provision of contextualized information anytime, anywhere, to the right persons as they go about their daily lives is part of an emerging paradigm dubbed as ubiquitous computing (Weiser, 1991), context-aware computing (Dey, Abowd, & Salber, 2001), pervasive computing (Ark & Selker, 1999), embodied interaction (Dourish, 2001), or everyware (Greenfield, 2006). Irrespective of the name given, a central tenet of this paradigm is the promise of populating our everyday lives with context-aware services that make interaction with the world easier, more manageable, and more efficient. This endeavor is made possible through embedding (at times personal and imperceptible) low-cost and low-power sensors and devices into our everyday environment.
A major step in this direction has been the widespread adoption of location-aware technologies such as GPS-enabled mobile devices and automotive GPS. Yet with our cities becoming interfaces for computational experimentation that are intermixed with human activities, we need systems that go beyond location-awareness and towards context-awareness (Dey, 2001). In other words, we need to know more about context (Dey, 2001), its inference from human activity, and how that feeds into our everyday experiences. As Bellotti and Edwards (2001) state, inference and adaptation to human intent in context-aware systems is at best an approximation of the real human and social intentions of people. This raises the need to further explore the kinds of services and usability issues brought forth under real-world usage contexts.