Design for Social Encounters

Abstract

Almost every mundane activity in which we engage during the day (e.g., walking, running, meeting other people etc.) leaves a trace in the form of data that can be collected, measured and shared as input to shape individual and social interactions. This data can be referred to as ‘activity traces’ (Dong 2014). A nascent application domain for these traces is ‘Social sensing’—that is, the ability to sense social structures in a social context. This thesis addressed the main research question: “How can connected object be used to support meaningful experiences of social sensing?” This project focussed on social sensing in situations of co-presence, and specifically on mingling events. A ‘mingling event’ is defined here as a large gathering of people in a relatively confined space. Pop Glass is an example of how activity traces can be used to evoke a meaningful experience of social sensing in a mingling event. When two users of Pop Glass share a toast their glasses will get the same colour. This creates a dynamic network of colours, which you are able to see and change. This concept was used in a study to investigate notions of “meaningfulness”. In giving meaning there is formal meaning (how we make experiences intelligible) and affective meaning (Our emotional responses to experiences). On a level of formal meaning sensing social structure through activity traces can be seen as a way of ordering information that connected objects carry out. An affective meaningful experience in a mingling event will mostly depend on the relevance of the conversation. However, the ability to sense social structure in a mingling event might cue you towards relevant people. The user study with Pop Glass aims on revealing its effects on the social experience of the crowd and how sense making played a role in it. 29 students participated in a staged network event with Pop Glass. They were measured by neck sensors and recorded by cameras and microphones. The lights of the glasses turned on half way through the event, so the experience with or without the lights could be tested. Participants were also asked to fill in a pre- and post-questionnaire. Afterwards 10 of them were interviewed about their experiences. Due to time constraints the Pop Glasses were programmed individually. This made them have lights that followed a random colour sequence with a random time interval. This did not harm the intentions of the test, because it was never tested if people saw a structure in the lights in the initial concept of Pop Glass. Two external research parties contributed in preparing the test and in analysing data. Researchers from the Faculty of Electrical Engineering, Mathematics and Computer Science (EEMCS), from Delft University of Technology led by Dr. Hung analysed proximity data and annotations of group size over time quantitatively. Dr. Day, ethnomethodologist and expert in interaction analysis from the department of Language and Communication at the University of Southern Denmark, conducted a conversation analysis based on scenes of the video and audio recordings of the event. A cluster method generated outcomes from the post-interviews. These served as the starting point for the results in this thesis. In the end Pop Glass seemed to have impacted the user’s social experience. People switched groups more often after the lights went on and the groups of participants became bigger for a short period. Both effects stagnated after some time. According to Day’s analysis the extraordinary behaviour of the glass seems to affect the way people use the glass and also how they?mingle. The glasses proved to be a topic of talk, a ‘toasting device’ and a boundary object even without the interactive lights inside the glass being on. When the glass turned on, new interactions seemed to occur. When finding meaning behind the ever-changing lights, Pop Glass was a perfect ‘talkable’ for participants. Besides, it did evoke people to tap their glasses with others, which seemed the key to make quick dis- or re-engagements. Insights from this user study resulted in a shift in perspective. Mingling events were not instigated by people’s formal understanding of the system, but more about their lack of. Their search for meaning proved to be a perfect networking tool. However the question remained how to keep a user interested in a product that it will never fully be understood. This revelation resulted in a set of paradoxical design guidelines that were aimed to support designers to design for meaningful experiences of social sensing in mingling events. In the first place the system should enable equal involvement from all people in the interaction space. Next to that there are three paradoxical design guidelines. The guideline ‘shaping interest’ aims on facilitating a sense making loop throughout their experience with the system. A product that compromises between a product that is transparent and ambiguous can do this. Second the system should ‘provide control’. Therefore it should ask for attention, but also be invisible. Lastly a product should ‘regulate belonging’ by making people feel one in a group, but also let them be the outlier. These guidelines were used in an iterative design process to generate a concept that is meant to inspire designers and researchers. This concept shared the same goal as Pop Glass. This resulted in Anthers. Anthers is a connected pin that will be worn by an attendee at a mingling event. Every attendee has a set of light particles in one colour; these will be exchanged with the people they talk to. In this way people will pollinate their conversation partners and therefore reveal their social traces. A video and mock-up of the concept were created and used in a review with 8 potential users. This resulted in mixed reactions. Most likely the way the concept was presented influenced people’s perspective. After all the video clarified all the functionalities of the concept. That is where the real challenge for ambiguous connected objects in mingling events lies: How to ‘sell’ a concept without revealing its core functionality, but only its potential result?