To this end, the authors carried out two observational studies based on the deployment of a public display application in two party settings: a book launch and a welcome party for students.

The application developed was the Opinionizer-a public display that allowed people to type in opinions and comments about a theme relevant to the gathering (the theme was previously selected by the authors). The theme was represented by images and a "provocative phrase or question". The idea was that people could type their comments via a laptop computer near the projection. Comments could be associated with backgrounds (e.g., 'techie', 'softie', 'designer', 'student') and could also be associated with an avatar (the comment was displayed as speech bubble coming from the avatar). The display was designed this way "to add 'colour' and 'personality' to their opinions [...]"

The Opinionizer was deployed in a book launch party and in a welcome party for new students of a university. The display was setup in a prominent place so that it could easily be noticed and a helper stood near the laptop computer to explain people what it was for. Two observers watched how the crowd moved and interacted with the system. In the welcome party two cameras were also used to record people's movement.

The authors observed that initially people would not interact with the system and were reluctant even when invited to do so. Adding some opinions by themselves seemed to create a momentum effect.

The authors also noticed what they called the "honey-pot effect": an increasing number of people near the Opinionizer, where people were typing their comments. It seemed that people were curious about what was going on near the Opinionizer and would approach it to see.

From the observations of the movements of people at the two parties, the authors identified three 'activity spaces':

• Peripheral awareness activities: activities like eating, drinking, talking, etc where people were only peripherally aware of the display but not paying attention to it.
• Focal awareness activities: activities triggered by the display itself like talking about it, watching it, gesturing to it, etc.
• Direct interaction activities: typing a personal or group opinion to the display.

According to the proposed framework, the transition between these activity spaces is what causes resistance to the interaction with the display. People need to be stimulated in order to transition and they need to be provided with information about what they will get in return. Specifically, the

"key information that needs to be made readily available to the person to decide whether to cross the threshold is:

• How long an interaction takes
• What they will get out of it
• What steps are involved
• If it will be a comfortable experience
• If there is a quick let out, where they can walk away gracefully, without it disturbing the ongoing public activity"

People need to be encouraged to cross the thresholds to focal awareness and to participation. To encourage people to cross the threshold to focal awareness, systems should be designed to show the display and the activities around it. It is necessary to get the first people to interact with it. To encourage people to pass the threshold to participation it is necessary to use a lightweight interaction mechanism so that users don't need to waste time learning how to use it and it should make it easy to recover from mistakes so that users don't feel embarrassed.

 [Brignull and Rogers, 2003]