This paper borrows ideas from social science to inform the design of novel "sensing" user-interfaces for computing technology. Specifically, we present five design challenges inspired by analysis of human-human communication that are mundanely addressed by traditional graphical user interface designs (GUIs). Although classic GUI conventions allow us to finesse these questions, recent research into innovative interaction techniques such as `Ubiquitous Computing' and `Tangible Interfaces' has begun to expose the interaction challenges and problems they pose. By making them explicit we open a discourse on how an approach similar to that used by social scientists in studying human-human interaction might inform the design of novel interaction mechanisms that can be used to handle human-computer communication accomplishments.

Abstract: Large displays are increasingly being placed in public places to support community and social activities. However, a major problem that has been observed with this new form of public interaction is the resistance by the public to participate. A main reason is due to the prominence of the affective aspect of the user experience. In particular, feelings of social embarrassment often act as a barrier. Our paper is concerned with understanding why this is the case and considering how we can attempt to overcome these aspects through improving the design of public interaction. Our focus is on how groups of people socialize around large public displays, the way they move towards them, congregate around them and change from being onlookers to participants and back again. We describe a system – the Opinionizer – which we designed and placed in two authentic social gatherings, intended to encourage socializing and interaction. We present our findings in terms of the patterns of physical and social engagement that take place around it. We then present a model of public interaction flow, which we use as the basis from which to provide design recommendations for encouraging public participation.

Middleware is software designed to support the development or operation of other software. A particular middleware system may afford certain styles of applications atop it, and may even determine the features of applications built using it. This poses a challenge: although we have good techniques for designing and evaluating interactive applications, our techniques for designing and evaluating middleware systems intended to support these applications are much less well formed. In this paper, we reflect on case studies of three middleware systems for interactive applications. We look at how traditional user-centered techniques, while appropriate for application design and evaluation, fail to properly support middleware design and evaluation. We present a set of lessons from our experience, as well as opportunities for further work.