increasing the comfort level for commenting on other group member’s behavior. In fact, lack of participation was often called out to the whole group by other group members: “[User name] why aren’t you slamming?” To compensate, it is critical that these systems support naturally occurring social group boundaries and their always evolving dynamic nature. New groups need to be easy to make, and systems might leverage naturally occurring physical world social interactions, such as social events, to suggest subgroups to the user. Flexible memberships for ongoing membership modification can also help.
The somewhat controlled nature of the study provided the opportunity to attain critical mass among small, existing social groups along with other benefits to the research process, such as being able to meet face to face with each study group. The downside to this methodology was that the system is somewhat removed from the broader context of users’ full social networks. In fact, this was noted by several participants, such as one person who commented that he would need a much larger portion of his social network in the system, “so [he] could seriously separate the large group into smaller social circles,” in order for it to be really useful. Regarding group memberships and group formation, how might Slam or something similar be used in a less controlled setting with a wider net cast across user’s social networks? Given that even in this limited trial, the naturally occurring subgroups arose quite quickly, such as the group of three friends from high school or the woman somewhat isolated in a group of men who could “easily see a girls group”, we would expect to see many more groups and subgroups in a more comprehensive social context. The ability to make these groups easily should smooth out some of the sensitive social issues surrounding group membership, and issues of exclusion due to the closed nature of the system, such as participants failing to text message with their other friends, presumably would subside in the case of an ubiquitous group-based messaging system. However, digital communication groups are not likely to be tremendously different from their physical counterparts, complete with their share of membership and status issues. Fleshing out the degree to which these two are similar regarding social status presents a fascinating research opportunity. Although this may be done with web-based groups, like Yahoo! Groups, mobile messaging groups arguably are better digital analogues of the physical group dynamics of interest given their close tie to the physical groups themselves and their focus on communication and experience sharing, which influence group cohesion and other indicators of the health of a social group.
In summary, this paper presented an experimental system for group-wide text and photo messaging. In a user study, system usage revealed that the presence of the group significantly increased the number of messages sent by group members. Participants used group-based messaging for roughly the same purposes as one-to-one messaging, but the increase in message volume generated improvements in social relationship metrics. Given that the message types all reflect some aspect of the social side of leisure, the group context augmented both participant’s social relationships and their interpersonal