Recently, crowd computing has been receiving interest from researchers, scientists, and practitioners as well (Barkuus and Jørgensen, 2008 and Brignull and Rogers, 2003). Crowd computing is a field in which interaction takes place between many people that are physically collocated, sharing a large display, and possessing collaborative or distributed control in public settings (Kaviani et al., 2009, Schieck and O’Neill, 2009 and Sieber et al., 2009).
Crowd computing systems can be seen as a type of locative media. Locative media is digital media applied to real places that mediate interaction in real society (Willis et al., 2009). As a special type of locative media, crowd computing considers the users of the system as a crowd. This means that many, unspecified users are gathered in the same location, possibly with the same purpose. This gathering of a large number of users in the same place and time is what differentiates crowd computing from other locative systems. For example, an arrival information system installed at a city bus stop is an example of locative media but not of crowd computing. Most locative systems are based on the same location, but users do not always have to interact in large numbers. Crowd computing is also different from traditional CSCW systems, where many users collaborate together, but they do not necessarily have to be co-located or collaborate at the same time.
Existing studies on crowd computing and locative media have constructed a well-founded basis for the field by aiming for three common themes: location, community, and context (Willis et al., 2009). Location is about “where the system is” or “what happens at which the system is installed”. Community is about “who uses the system there”. The two themes form together an important superordinate concept known as “context”. There are several studies that propose a contextualized crowd computing system in a certain location which focuses on the community of users. For example, Vajk et al. (2008) developed a multiplayer game application as a form of media which utilized a public display and mobile phones at the venue of a forum on mobile phones. José et al. (2008) developed a system which supported implicit interaction among people, enhancing awareness of each other through automatically detected individual Bluetooth devices in a campus bar. McDonald et al. (2008) suggested three applications that support one-to-one, one-to-many, and many-to-many interactions among attendees of academic conferences. Attendees were able to interact by accessing information about one another based on their location, thereby promoting serendipitous interactions that may not have been able to take place without the use of the applications. Schieck and O’Neill (2009)’s system visualized pedestrians stepping on and looking at the system embedded in a walkway so that users can play on it for fun. These studies reveal the three recurring themes of crowd computing: utilizing locations for the user community in specific contexts.
Based on this foundation of the field, further studies in this area should be strengthened in three ways to expand upon existing contributions, which also relate to the motivations of this study. First, studies need to examine the basic characteristics of their potential users, or the crowd. This allows the system to be designed appropriately according to crowd behavior and can offer deep understanding of a crowd’s motivation so that their behavior can be explained. This study, therefore, starts by examining existing studies on the motivation of crowds. Specifically, this study views crowds from the perspective of optimal distinctiveness theory, which presents a theoretical basis on understanding a crowd’s innate motivation and desire (Brewer, 2003 and Hornsey and Hogg, 1999).
Second, either individual users or small groups of users can be viewed as interaction agents in crowd settings. Numerous prior studies in swam behavior (e.g., Bonabeau et al., 1999) and complexity theory (e.g., Helbing, 2008) investigate the self-organizing and indirect-coordination functions of individuals in a crowd setting. However, relatively little research has been carried out on group interaction, especially in the crowd context. However, the group is where the action takes place in a crowd (Goffman, 1967) and to ignore the group level is to ignore the level of analysis in which meaning and behavior of a crowd are established and solidified (Harrington and Fine, 2000). Therefore, a group may be treated as a single entity with its own characteristics, not just as a simple sum of individuals (Bisker and Casalegno, 2009 and Stott and Reicher, 1998).
Group culture has been pointed out to be an important contextual factor in prior studies on communication (Gudykunst et al., 1988 and Leonard et al., 2009) and is defined as what the members have, the things they do, and what they think in common (Fine, 1979 and Herskovits, 1948). Communication processes, agents, and even contents depend on the cultural context of the group (Leonard et al., 2009). This study focuses on a specific type of group culture known as idioculture as a means for exploring interaction of small groups within a large crowd. This is because idioculture is a unique cultural product of small groups which provides distinct group characteristics (Fine, 1979). Because crowd computing systems involve a large number of people in the same place at the same time, the crowd has an inclusive effect on the small groups that gather together. Therefore, such small groups form a natural desire to distinguish themselves from other groups. Idioculture is utilized as a tool by small groups to obtain certain amounts of distinctiveness within a highly inclusive crowd context. Therefore, based on the influence of idioculture on group communication and distinctiveness, it should be closely examined in crowd computing research.
Third, the common goal of systems from the HCI perspective is to enhance the quality of user experience. User experience is generally defined as “all aspects of how a user uses, understands, and perceives when he or she uses an interactive system” (Law et al., 2008). Depending on the specific characteristics of systems and contexts, diverse factors of user experiences have been investigated (Hassenzahl et al., 2010, Hassenzahl and Tractinsky, 2006, Law and Schaik, 2010 and O’Brien, 2010). However, it is not yet known which aspects of user experience is important in crowd computing systems. Therefore, this study seeks to identify user experience elements specific to a crowd computing system in order to investigate the impact of idioculture on user experience.